Radc TR 88 110 - 1988

Downloaded from http://www.everyspec.
com
Mir,
RADC-TR-88-1 10
Final Technical Report
May 1988
704
RELIABILITY/MAINTAINABILITY/
TESTABILITY DESIGN FOR DORMANCY
Lockheed Electronics Company, Inc. DT[7

Robert M. Seman, Julius M. Etzl and Arthur W. Purnell LL.
. , -
APPROVED FOR PUBLIC RELEASE; DISTRIBU7ION UNLIMITED.
ROME AIR DEVELOPMENT CENTER

Air Force Systems Command
Griffiss AFB, NY 13441-5700
Downloaded from http://www.everyspec.com
Best
Available
Copy
UNCLAS SIFlIED
SECURITY CLASSIFICAT;ON OF THIS PAGi7
IForm Approved
REPORT DOCUMENTATION PAGE OMB NO. 0704-0188
ia REPORT SECURITY CLASSIFICATION lb RESTR;CTIVE MARKINGS
UNCLASSIFIED N/A
2. S CURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION /AVAILABILITY OF REPORT
0,A Approved for public release; distribution
2b OECLASSIFICATION/'DOWNGRADiNG SCHEDULE unlimited.
N/A__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
4 PERFORMING ORGANIZATION REPORT NUMBER(S) 5 MONITORING ORGANIZATION REPORT NUMBER(S)
N/A RADC-TR-88-1 10
6a NAME OF PERFORMING ORGANIZATION 6b OFF'CE SYMBOL 7a. NAME OF MONITORING ORGANIZATION
applicabit)
(if
Lockheed Electronics Company, Rome Air Development Center (RBE-2)
Tnc_ ____________1___________
6L ADDRESS (City, State, and ZIP Code) 7b ADDRESS (City, State, and ZIP Code)
1501 US Highway #!22 Cifs;V Y14150
Plalnfield NJ 07061Crfi. YN 34-50
8a. NAME OF FUNDING/SPONSORING 8b OFFICE SYMBOL 9 PRCCUREMENT INSTRUMENT IDENTIFICATION NUMBER
Rome Air Development Center RJXE-2 P30602.-85-C-0040

9cADDRESS (City, Stat*, and ZIP Code) 10 SOURCE OF FUNDING NUMBERS
PROGRAM PROJECT TASK WORK UNIT
Criffiss AFB W, 13441-5700 ELEMENT NO NO NO ACSINNO
Security Cawication)620F 1 3802G

I1 TITLE (include
RELIABILITY/MAINTAINABIL ITY/TESTABIL ITY DESIGN FOR DORMANCY
12 PIERSONAL AUTHOR(S)
Robert M. Seman, Julius M. Etzl and Arthur W. Purnell
13a. TYPE OF REPO-RT, 13b TIME COVERED 14~ DATE OF REPORT (Year, Month~,Day) 15 PAGE COUNT
IinI FROM ~a L&L.TOEF .. 8 I Max 1988

16 SUPPLEMENTARY NOTATION
N/A
17 COSATI CODES 18 SUBJECT TERMS (Continue on rovers* if necessary and identify by block number)
FIELD GROUP SUB-GROUP IReliability Dormancy Electronic Equipmer
09 01 ?'intainability
13 I 08
19 ABSTRACT (Continue on reverse if necessary
ITestability
and identify by block numnber)
Non-Operating
Design Guidelines
This document has been prepared as a tool for designers of dorm~ant military equipment and
systems. The purpose of this handbook is to provide design engineers with Reliability/
Maintainability/Testability design guidelines for systems which spend significant portions
of their life cycle in a dormant state. The dormant state is defined as a nonoperating
mode where a system experiences ver-y little or no electrical stress.
The guidelines in this report present design criteria in the following categories: (1) Part
Selection and Control; (2) Derating Practices; (3) Equiprment/System Packaging; (4) Trans-
p-rtation and Handling; (5) Maintainability Design; (6) Testability Design; (7) Evaluation
Methods for In-Plant and F'ield Evaluation; and (8) Product Performance Agreements. Where-
ever applicable, design guidelines for operating systems were included with the dormant
design guidelines. This was done in an effort to produce design guidelines for a more
22a NAME OF RESPONSIBLE INDIVIDUAL 22b TFLEPHONE (Include Area Code) I?2c OFFICE SYMBOL
Anthony N. Ciancto (315) 330-2101 RD RE2
DO Form 1473. JUN 86 Pripvioui editions are obsolete SECURITY CLASSIFICATION OF T-41S PAGE
UNCLAM 1 SIFIED
UNCLASSIFIED
complete life cycle. Although dormant systems spend significant portions of their
life cycle in a nonoperating mode, the designer must design the system for the
complete life cycle, including nonoperating as well as operating modes.
The guidelines aie primarily intended for use in the design of equipment composed
of electronic parts avd components. However, they can also be used for the design
of systems which encompass both electronic and nonelectronic parts, as well as for
the modification of existing systems.
UNCLASSIFIED
TABLE OF CONTENTS
ectionge
i.0 INTRODUCTION 1-1
1.1 The Need for Dormancy Design Guidelines 1-1

1.2 The Guideline Development Program 1-2
2.0 HOW TO USE THESE GUIDELINES 2-1
3.0 DORMANCY ENVIRONMENTS 3-1
3.1 System Life Cycle Profile 3-1

3.2 Environmental Considerations 3-3
3.2.1 Natural Environmental Factors 3-14

3.2.2 Climate 3-14
3.2.3 Natural Environment 3-19
3.2.3.1 Temperature 3-19

3.2.3.2 Humidity 3-24
3.2.3.3 Atmospheric Pressure 3-25
3.2.3.4 Solar Radiation 3-26
3.2.3.5 Rain 3-26
3.2.3.6 Solid Precipitation 3-28
3.2.3.7 Wind 3-30
3.2.3.8 Salt Fog 3-32
3.2.3.9 Sand and Dust 3-32
3.2.3.10 Fungus 3-33
3.2.3.11 Space Environment 3-34
3.2.4 Induced Environmental Factors 3-37

3.2.5 Induced Environment 3-41
3.2.5.1 Mechanical Factors 3-41
3.2.5.1.1 Vibration 3-41

3.2.5.1.2 Shock 3-42
3.2.5.1.3 Acceleration 3-44
TABLE OF CONTENTS (Continued)
Se-tio PageM
3.2.5.2 Nuclear Radiation 3-45

3.2.5.3 Atmospheric Pollution 3-47
3.3 Dormant Storage 3-51
3.3.1 Storage Types, Capacities, and Limitations 3-51

3.3.2 Storage of Explosive Material 3-53
3.4 References 3-54
4.0 DESIGN GUIDELINES 4.1-1
4.1 Part Selection/Control 4.1-1
4.1.1 Resisters 4.1-7
4.1.1.1 General Characteristics of Fixed Resistors 4.1-13
4.1.1.1.1 Fixed Composition Resistors 4.1-13

4.1.1.1.2 Fixed Film Resistors 4.1-20
4.1.1.1.3 Fixed Wire-Wound Resistors 4.1-21
4.1.1.2 Variable Resistors 4.1-22
4.1.1.2.1 Low Precision Variable Composition

Resistors 4.1-23
4.1.1.2.2 Semi-Precision Variable Resistors 4.1-30
4.1.1.2.3 Precision Variable Resistors 4.1-30
4.1.1.2.4 Variable Non-Wire-Wound Trimmer Resistors 4.1-30
4.1.1.2.5 Variable Wire-Wound Trimmer Resistors 4.1-30
4.1.1.3 Special Resistors 4.1-31
4.1.1.3.1 Fixed Film Network Resistors 4.1-31

4.1.1.3.2 Thermistors 4.1-31
4.1.2 Capacitors 4.1-32
4.1.2.1 General Characteristics of Fixed Capacitors 4.1-35
4.1.2.1.1 Fixed Paper and Plastic Film Dielectric

Capacitors 4.1-35
ii
Set ion Ae
4.1.2.1.2 Fixed Glass and Mica Dielectric

Capacitors 4.1-46
4.1.2.1.3 Fixed Ceramic Dielectric Capacitors 4.1-46
4.1,2.1.4 Fixed Electrolytic Dielectric Capacitors 4.1-4"/
4.1.i.2 General Characteristics of Variable

Capacitors 4.1-52
4.1.2.2.1 Variable Ceramic Dielectric Capacitors 4.1-53

4.1.2.2.2 Variable Capacitors Tubular Type 4.1-53
4.1.2.2.3 Ai." Dielectric Capacitors 4.1-53
4.1.3 Microci.cuits 4.1-53
4.1.3.1 Characteristics of Hermetic Microcircuits 4.1-66
4.1.3.1.1 Hermetic Monolithic Microcircuits 4.1-67

4.1.3.1.2 Hermetic Hybrid Microcircuits 4.1-74
4.1.3.2 General Characteristics of Non-Hermetic

Microcircuits 4.1-80
4.1.3.2.1 Plastic Encapsulated Devices 4.1-81

4.1.3.2.2 Polymer-Sealed Devices 4.1-85
4.1.3.3 VLSI/VHSIC Microcircuits 4.1-85

4.1.3.4 Microcircuit Stress Screening 4.1-87
4.1.4 Discrete Semiconductors 4.1-91

4.1.5 Inductive (Magnetic) Devices 4.1-92
4.1.6 Relays 4.1-96
4.1.6.1 Solid-State Relays 4.1-96

4.1.6.2 Electromechanical Relays 4.1-96
4.1.6.2.1 Design Guidelines 4.1-97

4.1.b.2.2 Process Control Guidelines 4.1-98
4.1.7 Switches 4.1-99
4.1.7.1 Guidelines 4.1-99
iii
Sect ion
4.1.7.1.1 Design Guidelines 4.1-100

4.1.7.1.2 Process Control Guidelines 4.1-100
4.1.7.1.3 Test Guidelines 4.1-101
4.1.7.2 Environmental Effects/Guidelines 4.1-102
4.1.8 Connectors 4.1-105
4.1.8.1 Guidelines 4.1-106

4.1.9 Cables/Wires 4.1-110
4.1.10 Batteries 4.1-113
4.1.10.1 Design Guidelines 4.1-117

4.1.10.2 Process Control Guidelines 4.1-117
4.1.10.3 Test Guidelines 4.1-124
4.1.11 Fiber Optic Components 4.1-124

4.1.12 Quartz Crystals 4.1-128
4.1.13 Electromechanical Devices 4.1-130
4.2 Derating 4.2-1

4.3 Maintainability Design 4.3-1
4.3.1 Maintenance Concepts for Dormant Equipment 4.3-1
4.3.1.1 "Wooden Round" Maintenance Concept 4.3-2

4.3.1.2 Periodic Monitoring/Repair Maintenance
Concept 4.3-4
4.3.2 Maintainability Design Considerations for

Dormant Systems 4.3-4
4.3.2.1 "Wooden Round" Design Considerations 4.3-4

4.3.2.2 Periodic Monitoring/Repair Design
Considerations 4.3-7
iv
Section 0u
4.3.2.2.1 Optimum Test Intervals to Minimize

Dormant System Unavailability 4.3-10
4.3.2.2.2 Optimum Test Interval to Achieve Minimum
Dormant Reliability Requirement 4.3-11
4.4 Testability Design 4.4-1
4.4.1 testability Concepts 4.4-1

4.4.2 Testability Design Considerations for
Dormant Systems 4.4-3
4.4.2.1 Maintenance Concept Impact on Testability 4.4-3

4.4.2.2 Test Effectiveness Impact on Dormant
Systems 4.4-4
4.4.2.3 "Go/No-Go" Versus "Parameter" Testing 4.4-6
4.4.2.4 Testability Design Guidelines for Dormant
Systems 4.4-7
4.5 Packaging 4.5-1
4.5.1 Material Deterioration 4.5-3
4.5.1.1 Metal Deterioration 4.5-3

4.5.1.2 Organic Deterioration 4.5-5
4.5.2 Cleaning 4.5-6

4.5.3 Dielectric Embedding 4.5-6
4.5.4 Assembly Packaging 4.5-7
4.5.4.1 Moisture Protection 4.5-7

4.5.4.2 Sand and Dust Protection 4.5-11
4.5.4.3 Vibration and Shoc) Protection 4.5-13
4.5.4.4 Pressure Protectio, 4.5-14
4.5.4.5 Solar Radiation 4.5-15
4.5.4.6 Electromagnetic Radiation 4.5-15
4.5.4.7 Nuclear Radiation 4.5-15
4.5.5 Container Design 4.5-16

4.5.6 In-Storage Cyclic Inspection 4.5-19
4.6 Transportation and Handling 4.6-1
4.6.1 Transportation 4.6-1

SectionLA
4.6.1.1 Transportation Equipment, Capacity, and

Limitations 4.6-1
4.6.1.1.1 Truck Transport 4.6-1

4.6.1.1.2 Rail Transport 4.6-5
4.6.1.1.3 Air Transport 4.6-6
4.6.1.1.4 Sea Transport 4.6-7
4.6.1.1.5 Combination Transport 4.6-7
4.6.1.2 Transportation Environment 4.6-8

4.6.1.3 Protection During Transportation 4.6-11
4.6.2 Handling 4.6-14
4.6.2.1 Handling Equipment, Capacity, and

Limitations 4-6-15
4.6.2.2 Human Factors Considerations 4.6-16
4.7 References 4.7-1
5.0 EVALUATION METHODS 5-1
5.1 Introduction 5-1

5.2 :--Plant Evuation ML-Lho.s 5-1
5.2.1 Worst Case Circuit Tolerance Analysis

(Ref. 1) 5-1
5.2.2 Life Cycle Reliability Assessment 5-4
5.2.3 Accelerated Life Testing 5-7
5.2.4 Environmental Stress Screeuing : I
5.3 Field Evaluation Methods 5-18
5.3.1 Readiness/Maintainability Assessments

Using Field Data 5-18
5.3.2 Periodic Test and Recertification 5-27
5.4 Special Evaluation Methods 5-28
5.4.1 Stockpile Laboratory Testing Programs 5-28

5.4.2 Stockpile Flight Test Programs 5-30
5.4.3 Real-Time Component Aging Programs 5-31
vi
Section Page
5.5 References 5-33
6.0 PRODUCT PERFORMANCE AGREEMENTS 6-1
6.1 Product Performance Agreement Objectives 6-1

6.2 Product Performance Agreement Selection,
Analysis, and Structure 6-3
6.2.1 Establishment of PPA Working Group 6-3

6.2.2 Data Collection 6-5
6.2.3 Alternative PPA Selection 6-8
6.2.4 Analysis of Alternatives 6-17
6.2.4.1 Identification of Relevant Cost Elements 6-20

6.2.4.2 Identification of Analysis Methodology
and Data Requirements 6-20
6.2.4.3 Performing the Required Analyses 6-26
6.2.4.3.1 Life-Cycle Cost Analysis 6-26

6.2.4.3.2 Reliability Analysis 6-29
6.2.4.3.3 Cost/Benefit Analysis 6-31
6.2.4.3.4 Risk Analysis 6-35
6.2.4.3.4.1 Sensitivity Analysis 6-35

6.2.4.3.4.2 Confidence Interval Analysis 6-36
6.2.4.3.4.3 Beta Distribution Analysis 6-37
6.2.4.4 Final PPA Selection 6-38
5.2. PPA SLr~icturing 6-41
6.3 Product Performance Agreement Management 6-42

6.4 References 6-51
Appendix A Reliability Checklist A-I

Appendix B Maintainability Checklist B-I
Appendix C Testability Checklist C-i
Appendix D List of Acronyms D-1
vii
LIST OF ILLUSTRATIONS
F icur-P Page
3.1-1 Typical Dormant System Life Cycle Profile 3-2

3.2.2-1 Geographic Distribution of Climate by AR 70-38
Categories 3-17
4.1-1 Operating vs. Non-operating Equipment

Reliability Bathtub Curves 4.1-6
4.1.3.2.1-1 Effectiveness of Sealant Materials (Ref. 31) 4.1-82
4.1.10-1 Shelf Life Characteristics of Various Battery
Systems (Ref. 76) 4.1-115
4.1.10-2 Shelf Life Characteristics of Primary Battery
Systems (Ref. 76) 4.1-116
4.3.1.2-1 Periodic Monitoring/Repair Maintenance Concept 4.3-5
4.3.2.1-1 "Wooden Round" Reliability Curve 4.3-8
4.3.2.2-1 Periodic Monitoring/Repair Reliability Curve 4.3-9
4.3.2.2.1-1 Parameters and Definitions for Optimal Test
Policies Minimizing Dormant System Unavail-
ability (Ref. 79) 4.3-12
4.3.6-1 Inspection Frequency Matrix (Ref. 1) 4.5-20
6.2.3-1 PPA Selection Matrix 6-11

6.2.3-2 PPA Types Grouped by Major Objective 6-14
6.2.4.1-1 Cost Breakdown Structure for PPA Analysis 6-18
6.2.4.2-1 Availability Guarantee Analysis Procedure and
Data Requirements Summary 6-21
6.2.4.2-2 Captive Carry Guarantee Analysi3 Procedure
and Data Requirements Summary 6-22
6.2.4.2-3 Storage Verification Guarantee Analysis
Procedure and Data Requirements Summary 6-23
6.2.4.2-4 Mission Dependability Guarantee Analysis
Procedure and Data Requirements Summary 6-24
6.2.4.2-5 Incentive Award Analysis Procedure and
Data Requirements 6-25
6.2.4.3.1-1 Cost Accounting Relationships in Life-Cycle
Costing (Reference 5.) 6-28
6.2.4.3.2-1 Example Reliability Curve 6-32
6.2.4.3-3-1 Cost/Benefit Analysis Example (Reference 5) 6-34
6.2.4.3.4.3-1 Example Beta Distribution Analysis 6-39
6.2.5-1 Sections of an Effective PPA 6-43
viii
LIST OF TABLES
Table Page
3.1-1 Association of Factor Importance with

Equipment Life Cycle Stage (Ref. 11) 3-4
3.1-2 Typical Military Equipment (Non-Operating
Modes) 3-5
3.2-1 Environmental Effects on Materials 3-7
3.2.1-1 Beaufort Scale of Wind Force with Velocity
Equivalents 3-15
3.2.2-1 Types and Categories of Climate 3-19
3.2.2-2 Summary of Temperature, Solar Radiation, and
Relative Humidity Diurnal Extremes (Ref. 12) 3-20
3.2.3.1-1 Temperature, Solar Radiation, and Relative
Humidity Diurnal Cycles for Consideration
in the Development of Military Materiel
(Ref. 13) 3-22
3.2.3.1-2 Diurnal Temperature Variation Time Sequence
(Ref. 13) 3-23
3.2.3.5-1 Lffects of Rain (Ref. 13) 3-.27
3.2.3.5-2 Extreme Rain Conditions for General Purpose
Military Equipment Design (Ref. 9) 3-27
3.2.3.6-1 Types of Solid Precipitants and Ice Deposits
(Ref. 13) 3-29
3.2.3.6-2 Dependence of Density on Type of Snow Cover 3-30
3.2.3.7-1 Ninety-Percent Confidence Limits on Extreme
Wind Probability (Ref. 14) 3-31
3.2.3.10-1 Fungus-Inert Materials 3-35
3.2.3.10-2 Commonly Used Fungicide: 3-36
3.2.3.11-1 Decomposition of Polymers in High Vacuum
(Ref. 16) 3-38
3.2.5.1.1-1 Vibration-Induced Damage to Electrical and
Electronic Equipment (Ref. 14) 3-43
3.2.5.1.3-1 Effect of Acceleration on Military Equipment
(Ref. 14) 3-46
3.2.5.2-1 Summary of Nuclear Radiation Environments and
Basic Protection Measures 3-48
3.2.5.3-1 Air Pollution Damage to Various Materials
(Ref. 14) 3-49
3.3.1-1 Typical Environmental Conditions in Storage
Facilities (Ref. 13) 3-52
ix
LIST OF TABLES (Continued)
Table Page
4.1-1 Ground Rules for Part Selection and Control

(Refs. 1, 2) 4.1-2
4.1-2 Typical Latent Manufacturing Defects Causing
Failures During Non-Operating Periods (Ref. 5) 4.1-5
4,1-3 General Part Selection/Control Guidelines for
Dormant Equipment Design 4.1-8
4.1.1.1-1 Environmental Effects,'GuidelIaes for Dormant
Fixed Resistors 4.1-14
4.1.1.2-1 Environmental Effects/Guidelines for Dormant
Variable Resistors 4.1-24
4.1.2.1-1 Environmental Effects/Guidelines for Dormant
Fixed Capacitors 4.1-36
4.1.2.1.4-1 Failure Mechanism Analysis, Solid Tantalum
Capacitors (Ref. 18) 4.1-49
4.1.2.1.4-2 Failure Mechanism Analysis, Non-Solid Tantalum
Capacitors (Ref. 18) 4.1-50
4.1.3-1 Environmental Effects/Guidelines for Dormant
Microcircuits 4.1-55
4.1.3.1.1-1 Microcircuit Defects/Screens (Ref. 1) 4.1-70
4.1.3.1.2-1 Hybrid Thick Film Failure Mecharisms (Ref. 18) 4.1-76
4.1.3.1.2-2 Hybrid Thin Film Failure Mechanisms (Ref. 18) 4.1-78
4.1.3.4-1 Comparison of Screening Methods (Ref. 1) 4.1-38
4.1.5-1 Failure Modes Affected by Various Use and
Storage Conditions 4.1-95
4.1.10-1 Characteristics of Primary Batteries 4.1-118
4.1.10-2 Characteristics of Secondary Batteries 4.1-122
4.3.1-1 Classification of Maintenance Policies and
Goals 4.3-3
4.3.2.2.1-1 Optimal Test Policies for Component in
Dormant (Standby) State (Ref. 79) 4.3-14
4.3.2.2.1-2 Optimal Test Policies for Component in
Dormant (Standby) State (Ref. 79) 4.3-15
4.5.3-1 Advantages and Disadvantages of Embedding
Electronic/Electrical Components (Ref. 52) 4.5-8
4.5.4.1-1 Corrosive Silicone Sealants, Adhesives,
and Coatings (Ref. 69) 4.5-12
4.6.1-1 Relationships Between Transport Modes and
the Environment Experienced by Materiel
(Ref. 64) 4.6-2
5.2.3-1 Activation Energies for Electronic Components

(Ref. 8) 5-9
x
LIST OF TABLES (Continued)
5.2.4-1 Overview of Environmental Stress Screening

Documents for Electronic Assemblies -
Part 1 (Ref. 13) 5-19
5.2.4-2 Overview of Environmental Stress Screening
Documents for Electronic Assemblies -
Part 2 (Ref. 13) 5-21
6.2.2-1 Program Characteristics Data for PPA Selection 6-6

6.2.2-2 Equipment Description and Characteristic
Data for PPA Selection 6-9
6.2.4.3.2-1 Reliability Data Sources (Reference 5) 6-30
6.2.5-1 Typical Performance Measures for Dormant
System PPAs 6-48
xi/fxHi blank)
1.0 INTRODUCTION
This document has been prepared as a tool for designers of dormant

military equipment and systems. The purpose of this handbook is to
provide design engineers with Reliability/Maintainability/Testability
design guidelines for systems which spend zignificant portions of
their life cycle in a dormant state. The dormant state is defined as
a nonoperating mode where a system experiences very little or no
electrical stress.
The guidelines in this report present design criteria in the following

categories:
1. Part Selection and Control

2. Derating Practices
3. Equipment/System Packaging
4. Transportation and Handling
5. Maintainability Design
6. Testability Design
7. Evaluation Methods for In-Plant and Field Evaluation
8. Product Performance Agreements
Wherever applicable, design guidelines for operating systems were

included with the dormant design guidelines. This was done in an
effort to produce design guidelines for a more complete life cycle.
Although dormant systems spend significant portions of their life
cycle in a nonoperating mode, the designer must design the system for
the complete life cycle, including nonoperating as well as operating
modes.
The guidelines are primarily intended for use in the design of

equipment composed of electronic parts and components. However, they
can also be used for the design of systems which encompass both
electronic and nonelectronic parts, as well as for the modification of
existinq systems.
1.1 THE NEED FOR DORMANCY DESIGN GUIDELINES
The nonoperating modes include: dormant storage, captive cerry, ready

alert, or other dormant states. After equipment has been installed in
the field or stored in arsenals and the supporting spares stored in
warehouses, the military has recognized that the dormant mode is a
predominant portion ot the life cycle of military equipment.
1-1
Subsequently, the dormant effects on electronic equipment and

reliability is of significant importance. The dormant mode and
dormancy effects are not always considered during the design phases
because sufficient dormant design information is not readily available
to the design engineer. This document has been prepared to present
design guidelines to mitigate the effects of the dormant environment
on electronic equipment. These design guidelines present design
considerations which will improve the reliability, maintainability,
and testability of dormant systems.
1.2 THE GUIDELINE DEVELOPMENT PROGRAM
The development of design guidelines for dormant systems was divided

into four phases:
Phase I - Data Collection

Phase II - Data Analysis
Phase III - Design Guideline Development
Phase IV - Design Guideline Documentation
In the first phase, information on dormancy and existing dormant

system design guidelines were collected from four primary sources.
The first source was a literature search of documents containing

relevant information. Three hundred and seventeen (317) documents
that contain information relevant to the development of dormant system
design guidelines were collected.
The second source was a survey of Qualified Parts List (QPL) Vendors
to determine information on the effects of various dormant
environments on military components, and also to determine any
guidelines for using these components in dormant systems subjected to
various environments.
The third source was a survey of government agencies and contractors

to identify experience with dormant systems, and to identify any
guidelines that were used in the development of these systems.
The fourth source was an analysis of various dormant military systems

to determine design criteria and dormant experiences of these
systems. Documents containing dormancy information for these systems
and personnel with additional information were identified. Pertinent
system documents were obtained and additional information was obtained
from the identified personnel.
I-9
In addition to these four primary sources of information, additional

personnel were identified and contacted. Meetings were held and
relevant conferences were attended in order to obtain dormancy
information.
From these sources data was collected that contained information on

dormancy, dormancy associated problems, dormant design considerations,
and unique modes associated with dormancy. Data on the environments
experienced by dormant systems and components, the effects of these
environments on dormant systems and components,
Reliability/Maintainability/Testability history and design
considerations, and guarantee/warranty/incentive programs for dormant
systems was collected.
In the second phase the data collected in Phase I was analyzed to

identify infurmation that was pertinent to this program. The analyzed
information from the sources was entered into the databases and/or
documented for subsequent design guideline development. The analyzed
data was classified by environmental factors, maintenance and
testabilty concepts, and design guideline categories.
In the third phase Reliability/Maintainab'.lity/Testability design

guidelines were developed for designing systems which spend
significant portions of their life cycle in a dormant state. The
guidelines present design criteria in the following categories:
1. Part Selection and Control

2. Derating Practices
3. Equipment/System Packaging
4. Transportation and Handling
5. Maintainability Design
6. Testability Design
7. Evaluation Methods for In-Plant and Field Evaluation
In the fourth phase, the design guidelines developed in the third

phase were documented in the format presented in this volume of the
report. In volume 2 of this report, the four phases of this project
are discussed in greater detail.
1-3/(]-4 })lank)
2.0 HOW TO USE THESE GUIDELINES
To assist the user in locating information, this report has been

divided into the following major sections:
3. Dormancy Environments
4. Design Guidelines
4.1 Part Selection/Coitrol

4.2 Derating
4.3 Maintainability Design
4.4 Testability Design
4.5 Packaging
4.6 Transportation and Handling
5. Evaluation Methods

Appendices
Reliability Checklist
Testability Checklist
Maintainability Checklist
Each section is formatted to include a summary of problems encountered

in the dormant mode, some background information, and the guidelines
for dormant systems. The guidelines recommend design criteria,
tradeoff methodologies, and special design tasks for dormant systems.
Figures and tables in each section summarize and supplement the
guidelines.
The checklists provided in the appendices present to the designer a

quick method for checking the reliability/maintainability/testability
requirements of the dormant system.
Wherever applicable, significant design guidelines for operating

systems were included with the dormant design guidelines.
Section 3, Dormancy Environments presents information on the

predominant environments encountered by dormant systems. This section
also contains basic guidelines for protecting equipment from these
environments.
2-1
Section 4.1, Part Selection/Control presents design guidelines for

selecting and designing with the following component types when used
in a dormant environment:
4.1.1 Resistors
4.1.2 Capacitors
4.1.3 Microcircuits
4.1.4 Semiconductors
4.1.5 Inductive Devices
4.1.6 Relays
4.1.7 Switches
4.1.8 Connectors
4.1.9 Cables
4.1.10 Batteries
4.1.11 Fiber Optics
4.1.12 Quartz Crystals
4.1.13 Electromechanical Devices
4.1.14 Packaging Electronic Components for Long Term Storage
Section 4.2, Derating, presents derating criteria for components in a

dormant mode.
Sections 4.3 and 4.4, Maintainability and Testability Design, present

design guidelines for maintainability and testability of dormant
systems.
Section 4.5, Packaging, presents guidelines for protecting dormant

equipment from various environmental factors.
Section 4.6, Transportation and Handling, presents design guidelines

for these phases of the life cycle.
Section 5, Evaluation Methods, presents design guidelines for in-plant

and field evaluation methods for dormant systems.
Section 6, Product Performance Agreements, provides design guidelines

for the selection and application of product performance agreements
for dormant systems.
3.0 DORMANCY ENVIRONMENTS
3.1 SYSTEM LIFE CYCLE PROFILE
The Life cycle of military equipment can be divided into 8 phases:
1. design and development
2. production
3. acceptance and evaluation testing
4. packaging
5. transportation
6. dormant nonoperational modes (storage, ready alert, etc.)
7. pre-operation testing
8. operation/use
These phases are illustrated in figure 3.1-1.
In designing and developing military systems it is vitally important

that the system life cycle be thoroughly understood. While the life
cycle presented in figure 3.1-1 encompasses every major step in the
development and use of a typical military system from concept to final
operation, the actual phases and sequence for a specific system will
depend upon its design characteristics and intended use. For example,
reusable equipment will generally experience the latter phases of the
life cycle repetitively, whereas, expendable (e.g., one-shot) equip-
ment will generally experience a less repetitive life cycle, spend
more time in a non-operational phase, and have its life cycle
terminated by its operation/use.
In addition to understanding the life cycle of a system, the designer

must also understand and consider the effects of the various environ-
mental factors that the system will be subjected to throughout its
life cycle. Equipment in the procurement and production stages, the
first four stages of the life cycle, generally does not experience de-
grading environments for any significant periods of time. After
issue, however, the final stages of the life cycle are experienced and
3-1
I I
r*
- - - - -
"C -C
L -
E
cn
- c I
ro
L)
0.
z~
C->
3-2
the equipment is exposed to a variety of potentially degrading envi-

ronmental factors for extended periods of time. The importance of
these environmental factors during pertinent life cycle stages is
illustrated in Table 3.1-1.
Most military equipment is not used immediately after issue, but

rather must await the proper circumstances before its use. Generally,
it is placed in a non-operational (dormant) mode before and between
use. These non-operational modes may be storage or ready alert con-
figurations, or non-operating periods during use such as unpowered
space flight or standby redundant equipment. Table 3.1-2 identifies
typical non-operating modes for military equipment.
Durmancy and its effect on system reliability should be a primary con-

cern during the design and development of military systems. Systems
put into use after extended periods of dormancy must perform without
mission degrading malfunctions. Dormancy in many systems, such as
missiles and munitions, is a particular concern because these systems
spend the majority of their life in a non-operating (e.g., storage,
ready alert, captive carry, etc.) mode. For example, in a typical
missile system the non-operating time could be as much as two million
times longer than operating time. Although the operating failure rate
may be substantially greater than the non-operating failure rate, the
significant difference in time between these two states makes dormancy
a major factor to consider. (Ref. 17)
The following sections describe the primary degrading environments

that may be encountered by systems during dormant periods. Also pre-
sented are basic design guidelines that will assist the design engi-
neer in designing adequate protection to mitigate the effects of these
environments. Subsequent sections of this report provide more de-
tailed guidelines for designing military systems that will spend
significant portions of their life in dormant modes.
3.2 ENVIRONMENTAL CONSIDERATIONS
This section describes the environmental conditions that a dormant

system may be subjected to throughout its life cycle. Essentially,
the environmental factors can be classified into two groupings: the
natural environment and the induced (man-made) envirorments. The
natural and induced environments are important because of their
effects on equipment. Principal effects and types of failures induced
by various environmental factors are listed in Table 3.2-1. (Refs. 4,
5, 6.)
3-3
0 0000 uu u
UOT eSfl ~ 0 0000 ca CJc 0ac
4)IJOeD 0u u ug 00000 0
'-4
0)
Isnp Du pueg u u 0 U u 0 - 4 6- 0 U
S 0
0 u. su.ntdu 0000 LI00UU U
0.0 to~iedsourjy
cysnbung 0 0000 0 40 ccu

U 9
mU C PU Tt 0 uU
u u ammmm0 u
0
r= 80, 0 Inu I c 0 0 m0 0
0 uoTlvljdTDaId0 Q04 0 0
'a
-4 4O1TV VO 0 g440to0
m go 4
V w)
0
-E-r
0U0 0w
~
.4 -Cm ~ ~
o<
0)C
4) 0 0
0) .- d, V) 4.
AU m 4):
(DEn4 0 )t -4 3:a 0 l
z114 U2
$ . 4 (n 4)->.E a) U 2:M H
w~ >4 000la la '
4)t/~ J 0 ~ 043-4
>
0 a
.0- 0.-
1Id > >

> >
a0 r-044
aV
4) 1a 1~-4hw 4
0 4.a 4) h> 0 41 0 41
$4 u N V >~w 410 .L0
4) 4 a cv- 92
a uu
hi ~G 4~.~ -40aaaaaa J) -4 -4a

41.a 41J~ w4h Aihi w14A. hhJJi. ~
044
o>h
d)
S4
@4aQ
$ ui0C
0(d-04a- a.1~. ~
1i a
0
0JJ.aJ~~~.
0 d) u ~ .) 0a.404.a4.-
uu> aah a vWQ W a
)( )d aW d - 4W (0E-
0~~~ aouo~ oa
u..a~ W W 0 o
0 0
41 v 4.
-)000440 amm-
4. j0 . -40 ) , ,
00 W W.94m
W n 4).(D-4-40 .W 0 )4)d
ONt iL 4 0$ 0. W.~$0.. W
Ad -4 w*7.- n 0 r- w C - - 4 14 r-41-
*.
U) 440 G
00 ~'03-5
*0
.1.
0M
*0 a ) -
1 .4P toA 1 t
'm uto 0 A.)

U -1 (a
0
0
J
0
.0
E
04
(D I )4 0)
U4 4d4 6
.0 J 4 ~~ A0J .
* 0 930.0 W 93 0
I 0 -W 41 A 1 1A 0j4
u -44-4 4b-4
41)~ -4 -4 44
e~) 0 u 0 U
to v u4 vj 0jjj~ 0 13 0
-4 w w wUUUUUUUU
(L il w
a. -4 .40000 0 L t 0
0 - 0% EU 0 L EUE
3-6
FA1 0
413 1.4 C)
0) 1-4 93J>A.
0 o..i wi 0 -4 41
hihi w1 to1 '.45 1 I
4. to CD'414 41 to
EU014 0 4) r1 W.-) 4 C).
-4 -4' > 0 93 A 1
L4M "'-4 0 ~'4.4 V In41
01 u5 04 to 0
00 1 410 0. 4-M 041
41h1h '- 0i4 14 041z
to 41 0 0~i1 v)
E '-4 EU.4
4 S4 dV1.' wi4 1
-4 44 00 hi 4.d10
to M*- 0 41 4)
0. .0%6d4 ) 4 1 41c
EU 4 '.4 39 a- 41
u441
Ca
to 0 4U tULV 4.4 u~'- 4)
In~f '.4 .h i110 44 0.
4141 4-) LA 4U) C: A0 0 *41
U. 4.) 0 .. ,4EIn'h (a 0 'lT.4
to4 'ea *( 0414
M to *E
$ N4%
hi
0to
*-. ~ 0 'o vhh 0 M. 47% 44
U1 I -4 0 01 E44- d
" 4j w4 vV 0 01 w
&20 it 0)( 41
to 4 hO a
0) E
41~~% 0.V1
41 ~0 '4 0 h41 m1
0W12 0 0
I~~~~. . V4~ 0~
0 toV 0
'-4 I41 C U '.4 41 u42 m

>' 00 )'. V
41 .) 04 U
>4 4-4i 14 5
44 3 414 4 ~ U4 4i)V
U)0
u 1 .,4i V 1
414.45 to0
'-4Cl4 ' E hi EU Eh 'o
hi.0 hiUw
0041 -4
011 410.4 OA
*-4 mU 4U'4 1004
- ~(aI~ . '4 U10 Oif0 w
.0 0a E " 1iI " u 04
04.U>10
m I '0 0j U)
44 4)
'.-441
0
0-
.
0
V
hiU)
r-4 -40U410V
0 41. .
A.hi) hi
A1 4-44%W-4
41 41 44
10 0441
0
L*0 v
3-7
@20 Li
go t ,0@ V 0.
si f-4 0 -4 0IE >q(d z
0 43 0 u4 2E to
.,11. 0 0a 4)-44)>U V w
V3 li c0 *.4 *.4.-.1 v2 KU -M 0
S 413 W29ol to 1 03 4 0 934) 410 ad 4
0 q a4 0' 044 0 govg UO
4) 4- 0 4) 0
IV j 0-4 W -. 4 uO
0
w U 04 0 0130 0Li-4 92
ad U
rt j 10 E > Li U04. 921Jtor-la @
W - 43 -M. u3 @EL go
it4
a V
Li 0 0>4 @ U0-4
.- 1. UM:N
0 04 u0 PAi 4) 0%1 )V4 r4 . 0 0 i
.,1 0 U ~4 0 U 0 we-1*4 4OEa OEi.
43
Ai -I u0 -4 w A 0 0 wiO
0 4300 U0 t% MLe.44 Li
0) to
E *434
4 Ad A V)OZ0-4 4 ,4 -40Z w 00 0
0 U 0
ui 0 .,1 $Li 43Mid s D39$V 0" %W .,q
E to UV
to* 4i EU EU 04 to 43 A .3 0 to.00 S3
03 - 0 .L
Ai ii0 0 U0
u 1&.4 to 0 .0
*-4 .0 U~E 16 fO0 i V ) O 0000E V -14
EU
it z 0 0 A 93 ty, .0P4 @24 0-4 3 2
Ca
U4 -4 0 E EU
fa MO4 0V
4) 4 ;3 -Mm
0 m- -4 w'
L 0 0%,4 L 0) 43 Uit 0
1"- 04 A.) w 3 yl0 1 :3 @2
c 0 44 04461 0 oEUC0w0 0
Eo 0 EU 0 - 0 0 4 EU) .- 00 V0%
q.CE )-4 4.4
im -4
1-4i0'J Li 0 .M O-4L W 4.4
4A 4-4 -4 93N. 4) IA $4@ to2LwUi O -'4
,
V tooi4 1 -U34 &i u 0 >4@12* 439) 0% w u4 43
@ i 0 -4 0LiUiE4 41 0 31 0 $3000 L0.0 A@2 4
1-4 0- 4 6A~Uu 4U
cu Lii 1- J20a
W .
0 .0
.4m
441
0 1
4)
44. 0 --
43 93 A
.,1 0
Li 0 0 EU .. *4 @20
0 in01 Vi 43 *.
> 4~ 4 0 1 0 v2E @~ 24
ma 0 0c
a. C3 N~0
3
> -M U EU 4) 0 -4 4
*~A 1- 0 ' 4 0.
mU0 E
104 L i EU i 41E
Ua 0 .q 3 * V4. (d
0mi Li @24)a $W ~ 0
u. U) 04 -,4 - 1,0 -44
Li U 0 0
.0 .0E
a0 u
EUE
-4 $
V 0
>0
o 004
0 4 -4 V
00
w) 0
0 0 >. .-
0) 0 V0 U1 M
'3 0..
Of 04 V) -4 '0
o I0 0 U 0 P-
'34
'V a A4 A.24w - t 4
4310 0M '3 0 Ai to 4 0
U! L (D. V) eu u~~ w. iv go u 'to >
AI) -14 I W4 . U )-4 V Li@
-4 r-3 O4i 93. A
fa u w)' 4' I00~~'~ 4) 4' '3
W vi W 0 A00-
P-4 @0v )to " 40 4 4 -4 I v~ r- 0 44' '30u .0 ' --4
@21 -4
In
IV0.
04
IQ -4
'1" 0
CIV4)At
to
'4
2'3"
."4
3U
0 $414)
1
v@ 0 1.
4 -4
400 o
P-4U
0 .
.o
A
to4
Li -4
4)
Id4
00
0M
0a
4'3 '3j4.
t'3 -W A3
A$
S3 -4
1'a r-4 4. 0 toV)U) w a 4)0 I. 0 4) -.4 0.j
0-00
0 44 0 U0, it>4 4-0 'i4dO LitM Li d.4 a ..4 wi
u o 0 O.04
w4. O0 0. 4 A0 0 0 0'J 43 -4-%W
U0 4) 0 . 0 04 02 '3 E ( 0
0 0 '0-
00 4) 012 w' C
12 i 00 0500W-4
(D44 02 E a 'V w(n 7.
0 Li3 Ad.. wi > -4 -4' 4. -4 .40 to -4 W 0 *'M 0l
(~ 44 0 '3 0 :1 0 4) 'M4) '3 0 '4. %&A 3 0 A 0 4. 4. *-4
fa 0 A a -( A.0 4 u 0@2040Li w 00 0 ~ tO id V( -4 0)0 r.
-4 ta4) V) Z 41 U 1-4At0 4ta
0 )V :10 44 01)
u S U 4.)
)
V- w A) 0 L r A mU4 m0o .0. (a A0 U(A0 $ . 0aw
)0 (au '34.
0)
0 '3.)14 'V
r '30
4JI Li4) CLl> 140 0 0 Z -4 .14N C00 0
-4 E- a.3W~~
U44 ~..U CIOW 4 .3 "~ ci~4
CA AC2 in0 L C
1.4
I$- )
I m) U)
00 I
I- 1.
AJ 0V )
".4 0 U)
03I' 0 4)
U 0 U0
Ad >4 0
> 0' .0 '4 E 4 "
'4.'4 '3 '
0 0 4
.. 4 .4 . u0
0j -.4 010Ai 0 4
'3 to '3
v0 .0 1 1 00 .
to ZI 2 . U 0 IV
4) as i Umu0
U U -
U)" I. 0
oLi '3
a L
3 1>4
U)A'
0 '4...I0
0'34
AJ " O
0 3.-
"4 ..a,.
4a ' 0 ) '
- 0. 0 L0 U) "4 U 'V i .~0_L
Ij - 0j0 0 i
0 0o 0 I'
0 .
0. Q0a (a 4) 1
0 4 -4 C 0 41 00 'a 1.
0* 04-v 'a4 4 3t 0 4j a s
0 0aJ E 0 0t
:00 93 m 414 - 4 1 4$34 0 0Ld

4
&.)J.J 1-1~ 0 1., .4U I~ P- )U& Lu 0
q.l ..0' 0Q A j00 C C d 40 O d
44 -4 4) 0? .4 04 d :s 0 4 -4 - 4) (4)
w ~~ ~ ~ u ~ :1. 1 ~
09 ~. . >
J:3 0 o - a44 M ='a 0
1-4~~~~~- 94) 00 0X004 )A- 0
toA a0 -4 0
4
01 CA 0 3 0 E o Q 'a 1 4
53 (D -.I l *.4 W W 4 -4 0 t *. 5).I
0 )0-4Aj0 - V) 41 4) '4 s 6 M 4 1 .:) 44 -- w 00
tv 0 4 'al 'a u 41 mU 4 0 r -4 4
Car 0 0) 0 0-
u 0a Aj0 1 0 -
w -4 0 41 0i '-4 LWL)U
-4 0 30 4 'a m 'a U wi V'a 0 'a
*-4 U -4 Li '4 ". .4

N~ ~~
. ~~~
"4 0 I -00 0
'- 1a 0 0 Ai 'a
(1) '
41 L 4 ~ . 0 0 0 la0
> i U u .,o
to0 F-
0- 1-
F-4 044
u 4
u A -
10 >q0
f~a I.; V
4ad 0 1 >O4
0 I ". M 001-
0 0 ~ e 0t
ad~4 VI0 .fl E ElM 0
oi ". 4). w 0*L

.,40 4 . 41
.i6i 41 41.0 uI si 0V to 0
v~ In fa a 041 0l 0 0 41 EU) mE
V 93 000% 0 to~x :5 1% 4 Q 4
0 ) 0 p ..4 ua- -4 4)*o 0 ) 93004
04 QI %3 r.' QM *-0
cl (a V% .0I m 0 ~ 4J
a -L0 0 0 o 0E
4 .0 ad EUI 0 Cw 4
q >i -4 x E w
i 'MW0 ie F- -4 .- 4 tr -4E L0adON
VI - 0*.14i to
(L).- 1
a
4j io -4
E w
vIE to
0
~wIV 3 -4
4JVL
tr 0
0' 44 1.
0
w.
02 xi U .40 0 tr
v4. (4) O 0 O
)1LI O3 00
)4 0 .4 S3 0of
0 .i .0 01. a0 41 41 4)V t 4LI W Wi. -4
to .-. 4 go. o 4 4
0.E .0 4 9 0 %?In 4 0 0% ~ 4
0 -).4 0 :3 a a 0 00
3 ta 4) oU w. a3 -M. V . 4j -4
C.) E 41 toUA w 41 -E- 4) 4) U 041 41V0.4
VUE .- 1 toUA1 (a LILv
0 0 0
VI ) 4 1 ai w EU 0 ciAi wmm 4 ) -MEE Ai .0 0 w
I j
6-
aUV P-L
t
40
m
441:5UUEE$4-
E4i
Imv(A ca ca ad
a
w0 u HIIC2.C.4-4J"JA00wAw
Ir4
0at d
41
44 0
VI 0
0 40
0 U
u 9) 0 0 0
oJ U.. a4I 0q
..
EU
L LI V) 0 0
r-4s *.3 ".4 i
0 'a.
v L0 ad Ld
I
04 0 .4 1
a)
02 4
...U)*- 0
'i-4 a-4~ w 0
(-4t 4) 4). $4 to"I ' U*4 V
0 >q VI 0 ul $i 0U .04u
Li EU . U Li0 1Q-4 0 L
4 -4
00
LI4
LI*.E
0 4"
0 0 0 0
00
V) 0% 0 53
V I 04 -. Q ~ 0 V
*) >q E 9004) A 1 04
U) 4) 04 U~ u .
V > U, aW 00
S3 to .0 %4 9' 04 4*
6- La 0%0 - 0 .. 4WU3 4)
-. 0 O.4.
to 0 At04 >~414
01
La 0 0- 0A
x. go 0 0 I)4
0 0 *-I 44 00 4 La 0
0 * 0 0 0 000t
"4 0-4 0V o 0IV .- 044 a
4) b. .4.Wcm 040E0 0 IVO 0a 0 0
ad S3o A' ~..4 U, .. 4 $d -. 4 p-j L
- z 0 iv U 0 41 tLa0 4)A 0 ~ 4d M 4) :
>00 41. V a La ."44 L4 m- .0 0
*-4 LaOV0 "4 4)O 4~ 00) :
4) %"4. A.) 0 Ad ) 04Lad .) 0 La
wI
13 zi w'a. ' 00 0
9 4)j .- -4 0z 4 4)
- ~~.4 0-444 40 b)4 bi U 084 U
.4.'
00
0jw
03 Ai 3 0 0 0A 0
w u I >040
'4' I w a
4 1 .
>-441 0 c 0 4) A
4 0 t .4 .-' o to WI
0) %2 4) 1 0 La
4
wa al .4.IV'
to 0 4 0S 4)
a 1
-1 : *-4
-. 4 U 14 $M 4 0 0M .. I
0 4 W 0 0 0 #a 0 0 03 0
a)
Lac 4 00 0 0w - 0 4 to
4) 0 0U 0 0 0 0 a.' 43 .4
0 4 W U 0 0-4 "1 0 0 0 W
>4 4a4 0 0
-4 -4 0 4 93 i' ' .. 4
I~~~4 .0 ) 0 to go
0 00.4 Lga U 0a E -
* j . 0 1 .J41 0 (A 4.)~.
-4 0 W 0 V u
La 0) 0 q 0
3-1
adC
0 0. 0
a% 00 0 a.,.
IV 4. 0 AJJ
1.4 .a 009)
0 024)2 0AJ 0 -- 4
:1 4J t 5 -. 4 0 0
to m.00 @0 .0 1
U) 43 o 0. 4 101OD W 0 4J 0
0) 0 0% -4 -4~ 0) % 0 w 0 0
-mE4 -1 0 Q) 00 v
04A.) -4 0 %W-M 4-4 . 0 :
S34 Wo 0. 160934) 44.4 0 -0 09

0j re
*4 0 09 0 C 000d9 M.4
03 4 0 44 -4 004 ' 6d.4 -00
0 1.4 44 4 )0) I 1.. -L @ 4 $4.- ON
4... 0 0
iv 04 %63U4to (a.J 0J)1.4 . 00400ama
.4-
C.) 04.OV W 43 4.0@ 1.s0 E*.* ) i.
0 )
4JdU 0A
4
400
-4 0
1
o 4 42
0.1. 0 4J 4J W
0 0 0
U30W
0'
W0004 (A3 4 04m. 0000u
1.4
10
93
0
94
a-' 0
6d 424 u
o 43 ON14@
-4 t
Q, .. 4 0go 4
*-4 03 0 A
-,4 Ad .0
-4 9
C:)
0*
Aj
to)
.0 0
.. 44 u
>4
. 41
'.4 a-I4334)13
3.2.1 Natural Environmental Factors
The natural environment consists of the following general elements:
1. Climate - temperature, humidity, pressure and precipitation
2. Solar radiation
3. Sand and dust
4. Wind loading (Table 3.2.1-1)
5. Fungus
6. Ozone
7. Salt Atmosphere
b. Space
3.2.2 Climate
MIL-STD-210, Reference 8, indicates the probable oxtreme climatic con-

ditions of the natural environment to which military equipment may be
exposed, and provides tabular presentation of world-wide climate.
MIL-STD-210 classifies the land and surface areas of the world into
four regional land type environments and one sea surface and coastal
area. AR 70-38, Reference 9, provides a classification systcm for
climate. The system establishes four broad types of climate which are
subdivided into a total of eight climatic categories.
These types and categories of climate are identified in Table 3.2.2-1.

Figure 3.2.2-1 shows the geographic distribution of the eight climatic
categories of climatic design types on a map of the world. For each
category, the range of temperature and humidity is indicated. Table
3.2.2-2 summarizes the temperature, solar radiation, and relative
humidity, diurnal extremes for the AR 70-38 categories.
AR 70-38 provides detailed coverage of the climates experienced around

the world. The climates in these references are listed in the
categories identified in Tablo 3.2.2-1.
3-14
Table 3.2.1-1. Beaufort Scale of Wind Force with

Velocity Equivalents.
Beaufort
Beaufort Descriptive Velocity,
Number Term my
0 Calm Less than 1
1 Light Air 1 to 3
2 Light Breeze 4 to 7
3 Gentle Breeze 8 to 12
4 Moderate Breeze 13 to 18
5 Fresh Breeze 19 to 24
6 Strong Breeze 25 to 31
7 Moderate Gale 32 to 38
8 Fresh Gale 39 to 46
9 Strong Gale 47 to 54
10 Whole Gale 55 to 63
11 Storm 64 to 72
12 Hurricane 73 to 82
17 Purricane 126 to 136
3-15/(3-16 blank)
60CLIUAIC CAI(GWTI so -
10
CLIMATICCATI.GOST51
I 0108 CIL.I
.i I . 0A
,I
- CLIMATIC
CAITORY 4
-lk "V
al
330 140 1)0 IZ0 110 ..0 s o__ 0 -.. 0.4
I CLIVATCCATECGOOYI
CLIMATIC CATEGORIES
I WET WARM, 10
-- 2 WET HOT
S1 WET WARM 2__.

2 WET NOT SEIOHAt All~l JUKATIC
CATECORY
2.
'~% 3 HUMID NOT COASTAL DESERT -71mlcc
S INTERMEDIATE NOT OR)

11716 INTERMEDIATE COLD 40-. I-l
I EXTREME COLD
.- H isMA SHOUJLD NOT BE USED WITHO%
60 60
/ N
"K~--~CLIMATIC CATUICa
ill ~r~ I i iI~h.I~i~i I CLIMAIC

CVtLGDa
V.. .0 4SIIQU 4
0Z '5 '&
/IA- A'---- - --
SDWTOTREFERENCE TO AR 70.-28. ~-I .
Figure 3.2.2-1. Geoi;

Distribution of Clim4
AR 70-38 Categorii
IFCLFT CAIwMON
a
50
k- CimAlic CAITCMYSI'l
1
- .P - T 0
4w,
-4I0
Figure 3.2.2-1. Geographic

Distribution of Climate by
AR 70-38 Categories.
3-17/03-18 blank) ' .~

Table 3.2.2-1. Types and Categories of Climate
Climatic Type Climatic Category
A. Hot - Wet 1. Wet - Warm

2. Wet - Hot
3. Humid - Hot Coastal Desert
B. Hot - Dry 4. Hot - Dry
C. Intermediate 5. Intermediate Hot - Dry
6. Intermediate Cold
D. Cold 7. Cold
8. Extreme Cold
3.2.3 Natural Environment
This section presents some general design guidelines for the protec-
tion of dormant systems from natural environmental factors. Subse-
quent sections of this report present more detailed design guidelines
for the protection of dormant systems.
3.2.3.1 Temperature
The design of dormant systems to survive the stresses of natural and

induced temperature environments is an important element of design
engineering. Each item of material must be considered separately with
respect to its ability to endure or to provide a useful function
throughout the temperature range.
Generally, dormant systems do not experience the high temperatures

experienced by operating systems. The heat dissipated by eneraized
components in combination with the natural surrounding temperature
results in higher temperatures for operating components in most cases.
This lack of induced heat must be considered for dormant systems.
Although high temperature requirements of dormant systems are less
than those for operating systems, the same operating system heat pro-
tection devices should still be applied to dormant systems to protect
the system when operation is required.
The lack of heat generation in dormant systems presents another dilem-

ma. The generation of heat in operating systems reduces the amount of
moisture present in the system. The lack of heat generation in dor-
mant systems requires components and materials used in these systems
to be adequately protected from moisture. Electronic components used
in dormant systems should be hermetically sealed and corrodable
materials should be coated and preserved (see section 3.2.3.2).
3-19
Table 3.2.2-2. Summary of Temperature, Solar Radiation, and

Relative Humidity Diurnal Extremes. (Ref. 12.)
Ambient Induced
Ambient Air Solar Relative Induced Air Relative
Climatic Temperature Radiation Humidity Temperature Humidity
Category OF BTU/ft 2 /hr % OF %
1. Wet-Walm Nearly Negligible 95 to 100 Nearly 95 to 100

constant 75 constant 80
2. Wet--Hot 78 to 95 0 to 360 74 to 100 90 to 160 10 to 85
3. Humid-Hot 85 to 100 0 to 360 63 to 90 90 to 160 10 to 85

Coastal
Desert
4. Hot-Dry 90 to 125 0 to 360 5 to 20 90 to 160 2 to 50
5. Inter- 70 to 110 0 to 360 20 to 85 70 to 145 5 to 50

medi ate
Hot-Dry
6. Inter- -5 to -25 Negligible Tending -10 to -30 Tending

medi ate toward toward
Cold saturation saturation
7. Cold -35 to -50 Negligible Tending -35 to -50 Tending

toward toward
saturation saturation
8. Extreme -60 to -70 Negligible Tending -60 to -70 Tending

Cold toward toward
saturation s.turation
3-20
Table 3.2.2-1 shows the climatic categories that have been selected to
provide environmental guidance for the preparation of documents co-
vering the design, development, testing and procurement of material.
The conditions that establish each of the eight climatic categories
are summarized in Table 3.2.3.1-1 where the temperature, solar radia-
tion and relative humidity ranges quoted represent diurnal variations
as specified in AR 70-38. Table 3.2.3.1-2 shows the diurnal tempera-
ture variation time sequence suggested for use in designing testing
cycles for operational and storage conditions. (Ref. 13.)
The neutralization of temperature stresses on material is a basic de-

sign consideration. Since combination of temperature and humidity is
a major source of material degradation, protective measures should be
emphasized in the design of equipment and packaging for systems likely
to be exposed to their joint occurrence at high levels for any sus-
tained periods of time. (Ref. 13.) Packaging guidelines to help pro-
tect materials from moisture and temperature are presented in secticn
4.5 of this report.
The prevention and control of temperature stresses can be accomplished

either (1) by adaptation to the thermal environment through the judi-
cious selection of materials and choice of design configuration, or
(2) by control of the immediate environment by modification of heat or
cold levels through the use of temperature regulating devices. The
first method is passive, and the second active. In practice the two
approaches are frequently combined. (Ref. 13.)
In addition to the proper selection of materials and components, pas-

sive prevention and control methods include the following: (Ref. 13.)
1. use insulation materials and techniaues
2. apply heat transfer principles (heat sinks and conduction paths)
Active methods for controlling heat or cold levels consist of devices

that change the temperature levels. This is done either by the forced
dissipation of excess heat or by warming, and is accomplished by the
use of heat ptmps, refrigeration units, fans, blowers, and heaters.
Through various qualification programs, a large variety of parts are

available that are qualified for various temperature stiess levels.
Section 4.1 of this report and applicable component Military specifi-
cations will assist the design engineer in selecting the components
for the required stress levels.
3-21
Table 3.2.3.1-1. Temperature, Solar Radiation, and

Relative Humidity Diurnal Cycles for Consideration
in the Development of Military Materiel. (Ref. 13.)
Climatic Conditions
Storage and Transit
Relative
Climatic Climatic Temperature Humidity,
Type Category OF %
A. Hot-Dry 1. Hot-Dry 90 to 160 2 to 50
B. Hot-Wet 2. Wet-Warm Nearly 95 to 100

Constant 80
3. Wet-Hot 90 to 160 10 to 85
4. Humid-Hot 90 to 160 10 to 85
Coastal
Desert
C. Intermediate 5. Intermediate 70 to 145 5 to 50

Hot-Dry
6. Intermediate -30 to -10 High

Cold
D. Cold 7. Cold -50 to -35 High
8. Extreme Cold -70 to -60 High
3-22
Q 0C 0 CD 0 0 0 0 0
-L.4 0'% t- ( Li! r0 0' 0
QXm 0C LA 0 LA! 0 0
I-iU 0 C4 %0 0
%A '0 '
E 0
-4 A
0 CD0 n(
0n LA 0 'Ln 0
%A $A L
0m
0
0 EQ LA 0 0' F-4 P!ENw
00
0
44
LA4 U 4 C! L! 0 '00 0
o~~ 4tI CD - at
'44
CD 0 0
En ON C'! A
%0 '0 E A LON
-4~~., II 44J4
0 0 0 0
0 - ~ ~~~~
E 0' - 00- 1A
oq -
0 ) $ 0 (aI
0 0U 0 4)** 0 4 -1
0>4 u I) 10 z V
-4 W q A A 0 4d
0d w . ) u i 0 L
-4t 0 cle-0 0 0Vz0wQ 0 oz 4 0U

C'. u' m' u'A '03
W! M- :r
EU -'
I 0 I I I -3-23
3.2.3.2 Humidity
Harmful humidity effects are basically caused by unwanted moisture

penetration and accumulation. Penetration by moisture can be either
in the form of water vapor (and subsequent condensation), or both.
Prevention and control of humidity-induce damage rests essentially, on
methods designed to prevent the intrusion and/or accumulation of mois-
ture, and those involving the selection of materials that are imper-
vious to, or minimize, moisture-induce degradation. Specific applica-
tions are constrained by the type of item and materials used, the na-
ture of the effects guarded against (i.e. electrical, mechanical, or
chemical), and the level of component reliability required.
Prevention and control measures applicable to various classes of

material are: (Ref. 13.)
1. Mechanical items. Use of proper finishes for material, use of

nonabsorbent materials for gasketing, sealing of lubricated
surfaces and assemblies, and use of drain holes for water
run-off.
2. Electronic and electrical items. Use of nonporous insulating

materials, impregnation of cut edges on plastic with moisture-
resistant varnish or resin, sealing of components with moving
parts, perforation of sleeving over cabled wire to avoid the
accumulation of condensed water, encapsulation or sealing, and
use of only pure resin as a flux.
3. Electromagnetic items. Impregnation of windings with moisture

proof varnish, encapsulation, or hermetic sealing, avoidance of
the use of commutators, provision of long creepage distances,
and use of alumina insulators.
4. Thermally active items. Use of nonhygroscopic materials, her-

vietic sealing where possible.
5. Finishes. Avoidance of hygroscopic or porous materials; im-

pregnation with wax, varnish or resin on all capillary edges.
True hermetic sealing will eliminate failures due to humidity, salt

spray, fungus, and rain. True hermetic sealing is a process that in-
cludes vacuum drying by evacuation to the micron range, outgassing,
quantitative leak testing, flushing with dry nitrogen, and backfilling
with an inert gas. (Ref. 13.) More comprehensive packaging guidelines
to reduce moisture related failures are presented in section 4.5 of
this report.
3-24
The proper selection of materials and components is among the most

important means for minimizing humidity-induced deterioration and
failure. The part selection guidelines presented in section 4.1 of
this report and the applicable Military specifications will assist the
designer in selectiug components for humid environments.
3.2.3.3 Aopheric Pressure
Maximum and minimum pressure criteria that military equipment should

be designed to meet at sea level and at other altitudes are given in
MIL-STD-210 (Ref. 8.) and AR 70-38 (Ref. 9.). AMCP-706-116 presents
extreme values of high and low atmospheric pressure as a function of
altitude, latitude and probability of being exceeded 10 percent, 5
percent, and 1 percent of the time, respectively. (Ref. 13.)
Various methods are used to prevent or minimize the adverse effects of

pressure. These practices include the selection of materials, proper
design and utilization of sealing and pressurization techniques. The
following briefly describes these methods:
1. The proper selection of materials within cost and design limi-

tations is an important means of preventing undesirable pres-
sure effects in dormant systems. Materials differ widely in
their ability to withstand pressure-induced stresses of the
type applied to containers and seals by pressure differentials
resulting from rapid changes in pressure. When seeping gases
or liquids are a potential problem, the permeability of
materials in contact with them should be carefully evaluated.
Lubricants and fuels differ in their susceptibility to evapora-
tion and some fluids are less subject to diffusion and seepage
than others. Materials for use as brushes in motors and gene-
rators vary in their resistance to failure under low-pressure
conditions. The destructive effects of corona and arcing can
be minimized by the proper choice of insulating materials.
2. Many adverse pressure effects can be minimized or prevented by

modification of design configuration to compensate for poten-
tial problems. This is particularly true of electrical and
electronic equipment where voltage breakdown at low pressures
is a constant potential hazard. The shape and spacing of con-
tacts on such equipment can affect breakdown resistance at dif-
fering voltages and altitudes.
3. Pressure levels can be maintained by sealing enclosures or con-

tainers at sea level or by bloving or pumping air into them to
obtain the desired pressure. Hermetic sealing of items that
have moving electrical contacts is a means of preventing explo-
sions touched of by sparks from corona or arc-over.
1-2c;
Pressurization of dormant equipment can cause as well as solve pro-

blems. Equipment that is pressurized tends to be larger, heavier, and
more difficult to maintain, transport and handle. Explosive decom-
pression, sudden loss of pressure can be a major hazard. It can re-
sult from mechanical failure of the pressurizing apparatus and or
seals, or from combat damage. Safeguards include the piovision of a
secondary pressurization system and use of pressure walls.
3.2.3.4 Solar Radiation
Ultraviolet radiation is the most destructive solar radiation. Vis-

ible radiation can also be destructive, but to a lesser extent. In-
frared radiation is normally less harmful.
The ultraviolet content of sunshine causes photochentical degradation

of most organic materials. Because of the deleterious effect on the
relatively weak bonds of their molecular structure, the elasticity and
plasticity of certain rubber compounds and plastic materials are
likely to be adversely affected. The degradation is also manifested
by the fading of fabric colors, and checking of paints, natural rub-
ber, and plastics. These minute checkered patterns are exaggerated by
the presence of ozone.
Optical glass exposed to ultraviolet radiation becomes increasingly

opaque with radiant intensity and time. By contrast, ultraviolet
radiation has little effect on the stronger chemical bonds of metals,
ceramics, and inorganic compounds. The effect of ultraviolet radia-
tion on structural metal is negligible. (Ref. 13.)
For dormant systems that are exposed to solar radiation the designer
should select materials that are more resistant to degradation, and
provisions should be made for replacements where degradable materials
are required.
3.2.3.5 Rain
Rain - when falling, upon impact, and as deposited water - has a

variety of effects on material, most of which are well known because
of the pervasiveness of rain in the environment. In Table 3.2.3.5-1
some of these effects are categorized with respect to the life cycle
of rain. Thus, many potential effects are not experienced except when
the protective measures are compromised by error or operation expedi-
ency. (Ref. 13.)
Both AR 70-38 and MIL-STD-210 provide extreme values that general pur-
pose equipment can be expected to encounter, Table 3.2.3.5-2 shows
the rain extremes provided in AR 70-38 as a basis for design
requirements.
3-26
Table 3.2.3.5-1. Effects of Rain. (Ref. 13.)
On Impact:
1. Erodes Surfaces
2. Imposes Large Forces on Structures
After Deposition:
1. Degrades Strength of Some Materials
2. Promotes Corrosion of Metals
3. Deteriorates Surface Coatings
4. Destroys or Deteriorates Many Exposed Items of Materiel --

Rations, Clothing, Repair Parts, etc.
5. Can Render Electrical or Electronic Apparatus Inoperative or

Dangerous if Rain Penetrates
Table 3.2.3.5-2. Extreme Rain Conditions for General Purpose

Military Equipment Design. (Ref. 9.)
Applicable to wet-warm, wet-hot, humid-hot, hot-dry,

intermediate hot-dry.
Drop Size throughout: 0.6-4.0 mm, median of 2.5 mm

Windspeed throughout: 40 mph (35 kt)
Duration, Amount,
hr:min in.
0:01 0.45
0:05 1.00
0:10 1.50
1:00 4-5.5**
12:00* 9.50
* 12 hr. for wet-warm, wet-hot, intermediate hot-dry;

4 hr. for humid-hot, hot-dry.
** 4 in./hr. for humid-hot, hot-dry; 5.5 in./hr. for wet-

warm, wet-hot, intermediate hot-dry.
3-27
Measures for the protection of dormant systems against the effects of

moisture are discussed in section 3.2.3.2 of this text, and packaging
guidelines to mitigate the effects of moisture are discussed in sec-
tion 4.5. These measures are equally applicable to rain. The princi-
pal prevention and control measures discussed in these sections are as
follows:
1. select materials to minimize galvanic corrosion.
2. use protective finishes (platings and/or coatings).
3. use preservatives, water repellants, and impregnants.
4. use insulation, hermetic sealing, and dehumidification.
During the design of dormant systems particular attention should be

paid to the e imination of water traps, to the provision of easy ac-
cess to areas of potential trouble due to moisture, and to the protec-
tion against wind-blown rain. (Ref. 13.)
3.2.3.6 Solid Precipitation
This section is concerned with those forms of ice that are derived
from atmospheric moisture. Included in tl -s section are snow, ice
pellets, freezing rain, and hail. Various forms of solid precipita-
tion are more clearly defined in Table 3.2.3.6-1. (Ref. 13.)
While the effects of the several solid precipitants can be different,

there are common features based on the fact that they are all atmos-
pherically derived frozen water. Furthermore, each of the solid
precipitants can occur in combination with other environmental factors
such as wind, rain, temperature, solar radiation, or humidity, in such
a way as to enhance detrimental effects. (Ref. 13.)
The primary degrading factor of snow or frozen rain is the moisture

that is created when either of these solid precipitants melt. Mois-
ture protection guidelines are presented in section 3.2.3.2. Packag-
ing guidelines to minimize the degradation induced by moisture are
presented in section 4.5 of this report.
Another detrimental effect of solid precipitation is the stress load-

ing on structures. Structures must be able to withstand the antici-
pated loading effect of the solid precipitation for the area. Table
3.2.3.6-2 summarizes the density of snow dependent on the type of snow
cover.
3-28
Table 3.2.3.6-1. Types of Solid Precipitants and Ice Deposits. (Ref. 13.)
Te ri Definition
Snow Precipitation composed of white or translucent ice crystals chiefly in

complex denditic, tabular, stellar, or columnar forms with hexagonal
symmetry. Snow is formed directly from atmospheric water vapor.
Snow Precipitation consisting of white, opaque ice particles, usually round

pellet or but sometimes with traces of crystallinity, ranging from 2 to 5 mm in
graupel diameter, and of a brittle texture. Ice pellets are easily crushed but
can rebound when falling on hard surfaces. They are usually formed by
riming of snow crystals.
Ice Precipitation consisting of transparent or translucent pellets of ice

pellet less than 5 mm in diameter. They may be formed by freezing of raindrops
or consist of snow pellets encased in a thin layer of ice.
Slee , A term formerly applied to ice pellets and now used colloquially and in
British terminology to descripe precipitation that is a mixture of snow
and rain.
Hail Precipitation in the form of rounded or irregular lumps of ice, called

hailstones, which range in size from ') to 50 mm or more in diameter and
which are formed from frozen raindro, s. Hailstones have a concentric
layered structure usually alternatiro between clear and opaque ice.
'Ce(dianond dust, ice needles, frost snow, snow mist)
crystal Precipitation in the form or slowly falling, very small (-30 um)
unbranched ice crystals in the form of needles, columns, or plates. Ice
crystals do not reduce visibility, can be seen only in sunlight or an
artificial light beam, and may precipitate from clouds or clear air.
Glaze A dense, hard, transparent coating of ice formed on exposed objects by

the freezing of supercooled rain or drizzle.
Rime A white opaque granular deposit of ice formed by the rapid freezing of
supercooled water drops as they impinge on an object. Rime is
considerably less dense than glaze.
Hoarfrost A light feathery deposit of ice consisting of interlocking ice crystals

directly sublimed from water vapor onto a cold surface.
3-29
Table 3.2.3.6-2. Dependence of Density on Type of Snow Cover
-3
Density, g cm
Type of Snow Cover
Min Max Usual range
Dry New Snow 0.05 0.11 0.07 - 0.10
Dry Settling Snow 0.09 0.22 0.10 - 0.20
Wet Settling Snow 0.10 0.24 0.15 - 0.20
Dry Settled Snow 0.20 0.43 0.25 - 0.35
Loose, Granulat Snow 0.11 0.30 0.18 -0.28
Dry Old Snow 0.20 0.53 0.25 - 0.45
Wet Old Snow 0.28 0.52 0.35 - 0.45
3.2.3.7 Wind
Atmospheric winds transport vast quantities of heat and moisture,

which ultimately affect the weather conditions distant from the air
mass source regions. Strong winds accor.2anying hurricanes, severe
thunderstorms, and other intense weather systems often damage struc-
tures, and carry materials and objects over long distances. Sand and
dust transported by the wind penetrate into and damage equipment.
Evaporation rates normally increase as the ventilation (wind speed)
increases, resulting in loss of exposed liquids and iTncreased wind
chill. (Ref. 13.)
In designing dormant equipment, the design engineer should use herme-

tically sealed components and conformally coated printed circuit
boards to protect material from moisture, sand and dust, and other
contaminants. These design considerations will adequately protect
dormant equipment from the adverse effects of contaminants transported
by mild winds. The design engineer must adequately protect dormant
equipment and structures from extreme winds. The design engineer
should recognize that the need for protection of dormant equipment
depends on the force of the extreme winds and the probability of ex-
treme winds occurring.
3-30
When designing to withstand wind damage for a certain period of time,

it is necessary to use a relation such as: (Ref. 13.)
Pn = 1 -(F(u)]n
where
n = time interval, years
u = extreme windspeed, miles per hour
F(u) = probability that windspeed u will not be exceeded in 1

year, dimensionless
(F(u)]n = probability that windspeed u will not be exceeded in n

years, dimensionless
Pn = probability of at least one wind occurring with a speed

u in n years, dimensionless.
Since Pn and n are chosen by the designer, it is preferable to solve

for F(u), obtaining:
F(u) = (1 - Pn)l/n
For long recurrence intervals, the estimates of extreme wind probabil-

ity are statistically rather than observationally derived. Hence,
confidence intervals to reflect the variability of observation from
estimates should be placed upon the probability estimates. Table
3.2.3.7-1 gives the extreme wind probabilities.
Table 3.2.3.7-1. Ninety-Percent Confidence Limits on Extreme

Wind Probability. (Ref. 14.)
PROBABILITY LOWER LIMIT UPPER LIMIT
0.50 0.407 0.593
0.90 0.827 0.945
0.96 0.827 0.982
0.98 0.908 0.992
0.99 0.955 0.995
3-31
1
3.2.3.8 Salt Fog
The principal effect of salt on materials is the acceleration of

metallic corrosion. Consequently, corrosion prevention must be a pri-
mary concern of the designer. The protection of metal from corrosion
is discusbed in deLdil in hection 4.5.
3.2.3.9 Sand and Dust
Sand and dust primarily degrade equipment by:
(1) Abrasion leading to increased wear,
(2) Friction causing both increased wcar and heat, and
(3) Clogging of filters, small apertures, and delicate equipment.
Although, friction and abrasion are not concerns for dormant systems
while the systems are in a nonoperational state, sand and dust buildup
on potential moving components must be considered. The buildup of
sand and dust on contact surfaces or moving parts will degrade the
operational read ress of dormant systems. Hermetic sealing of elec-
trical components, conforma coatings on electrical boards and proper
system packaging will eliminate sand and dust buildup in dormant sys-
tems. Equipment having moving parts require special care when design-
ing for sand and dust protection. Sand and .ust will abrade optical
surfaces, either by impact when carried by air, or by physical abra-
sion when the surfaces are improperly wiped during cleaning.
Dust accumulations have an affinity for moisture and, when combined,

may lead to corrosion or the growth of fungus.
In the relatively dry regions, such as deserts, fine particles of dust

and sand are readily agitated into suspension in the air where they
may persist for many hours, sometimes reaching heights of several
thousand feet. Thus, even though there is virtually no wind present,
the speeds of vehicles or vehicle-transported equipment through these
dust clouds can cause surface abrasion by impact, in addition to the
other adverse effects of the sand or dust.
Although dust commonly is considered to be fine, dry particles of

earth, it also may include minute particles of metals, combustion pro-
ducts, solid chemical contaminants, etc. These other forms may pro-
vide direct corrosion or fungicidal effects on equipment, since this
dust may be alkaline, acidic, or microbiological.
3-32
Since dust is present in almost any environment, the question is not

whether to allow dust to enter, but, rather, how much or what size
dust could be tolerated. The problem becomes one of filtering any
circulating air to remove dust particles above a nominal size. The
nature of filters, however, is such that for a given working filter
area, as the ability of the filter to stop increasingly smaller dust
particles is increased, the flow of air or other fluid through the
filter is decreased. Therefore, the filter surface area either must
be increased, the flow of fluid through the filter decreased, or the
allowable particle size increased, i.e., invariably there must be a
compromise. Sand and dust protection, therefore, must be planned in
conjunction with protective measures against other environmental fac-
tors. It is not practical, for example, to specify a protective coat-
ing against moisture if sand and dust will be present, unless the
coating is carefully chosen to resist abrasion and erosion, or is
self-healing.
3.2.3 .10 Fungus
Fungi adversely affect items of equipment utilizing cotton, linen,

natural rubber, hair, cork, leather, paper, felt, wood, cement, glue,
bonding materials and other organic materials th~c are nutrients for
fungi, unless these materials are treated to resist them. Fungi will
grow on all cellulosic material; animal and vegetable oils; waxes;
plastics with linen, cotton, wood flour, or other organic filler; and
thermoplastics which arp highly plasticized, where the plasticizing
liquids frequently are susceptible components. If nutrient is pre-
sent, fungi will grow upon any metal, plastic, ceramic or asphalt, and
may damage them to some extent. Fungi can cause complete degradation
of most cellulosic materials. Chemically altered cottons, such as
acetylated and cyanoethylated, show a high degree of resistance Lo
deterioration caused by fungi.
Fungi adversely affect the functioning of electronic and electrical

equipment because of inherent fungal properties and by aggravating the
effects of moisture. Certain types of insulation and plasticizers,
and fillers of molded and laminated plastics support fungal growth.
The mycelium is electrically conductive, and fungal masses encourage
moisture condensation and retard drying by interfering with air diffu-
sion. Hyphal strands of surface fungi can introduce leakage paths
which unbalance electronic circuits and establish couples which pro-
mote electrolytic corrosion. Absorption of moisture lowers the elec-
trical resistance of dielectrics, films of water can cause surface
leakage, and absorption within a capacitor or coil can bring about
alteration in electrical constants of a circuit.
3-33
Corrosion of metal parts is encouraged through moisture retention by

surface-growing fungi and by organic acids produced in their metabolic
activity.
WherE .oideration must be given to fungi as an environmental factor,

selection of nonfungus supporting materials will suffice in most
cases. Table 3.2.3.10-1 lists materials considered fungus-inert.
Refer also to MIL-STD-454 for more fungus-inert materials.
If critical circuits requiring high fungal resistance are involved it

may be necessary to fungus-proof with appropriate compounds. Table
3.2.3.10-2 lists commonly used fungicides. The following is a list of
design guidelines to protect dormant systems against fungus:
1. Use fungus-resistant materials that also meet strength, weight,

environmental, and mechanical requirements.
2. Apply fungicidal treatments or coat the material with a seal

that is impervious to moisturE, but fungicidal treatments are
only effective for limited periods and must be reapplied.
3. Use hermetic sealing, wherever possible. If hermetic sealing

is nnt possible, use gaskets and other sealing devices to keep
out moisture. Make sure the sealing devices do not contribute
to fungal activity. Detect and eliminate any moisture entry
points.
4. Use materials with low moisture absorption qualities.
5. When treated materials are used, make sure they do not contri-
bute to corrosion or alter electrical or physical properties.
3.2.3.11 Space Environment
Polymers are the materials most likely to be affected by exposure to

the environment of outer space. Most of the organic materials used in
spacecraft are long-chain polymeric compounds which degrade in a
vacuum not by evaporation or sublimation but by breakdown of the com-
pounds into smaller, more volatile fragments. This decomposition
takes place not at the surface but throughout the volume of the
piece. Polymers are commonly more stable at elevated temperatures in
the absence of oxygen. (Ref. 16.)
Coatings may be put over organic materials to reduce their net rate of
breakdown in vacuum. The coatings prevent degradation products from
leaving the underlying material and permits them to recombine to some
3-34
Table 3.2.3.10-1. Fungus-Inert Materials
Acrylics Polycarbonate
Acrylonitrilestyrene Polyester-glass fiber laminates
Acrylonitrile-vinyl Polyethylene, high density

chloride copolymer (above 0.940)
Asbestos Polyethylene terephthalate
Ceramics Polyimide
Chlorinated polyether Polymonochlorotrifluorethylene
Fluorinated ethylenepropylene Polypropylene
Glass Polystyrene
Metals Polysulfone
Mica Polytetrafluoroethylene (TFE)
Plastic laminates Polyvinylidene chloride
Silicone glass fiber Silicone resin
Phenolic nylon fiber Siloxane-polyolefin polymer
Diallyl phthalete Siloxane-polystyrene
Polyacrylonitrile
Polyamide
3-35
Table 3.2.3.10-2. Commionly Used Fungicides.
Gove rnnent-App roved Fungicides Other Fungicides
Textiles, hemp, and jute Textiles
Copper B-qui nol inol ate Copper naphthenate, 2, 21

methylerebis (4-chiorophenol)
Plastics and electronic equipment Pp
Paraphenyl phenol -formial dehyde Chl orinated pheol s

with salicylanilide (pentachlorophenol)
Leather poducts Rubber
Parani trophenol Nitrophenol, zinc salicylate

[l%0, zinc bezonate (1%)]
Textiles
Paints, varnishes, enamels, and
Dihydroxyl dichlorodiphenylvnletthane 1 1aur
Clorinated, phenols, salicylanilide
Coal-tar creosote compounds, carbolineums,

wood-tar creosotes, pentachlorophenol , copper
napthenates, chromated zinc cjloride, chemonite
zinc naphthenates, "Wolman" salt tanol ith, zinc
chloride, zinc metacresenite, and copper chromated
zinc chloride.
3-36
extent. If organic materials are used in partial enclosures contain-

ing temperature gradients and exposed to vacuum, there is a possibil-
ity that they will decompose to volatile products in the warmer areas,
and that the volatiles will redeposit on the cooler surfaces. If the
cooler surfaces must conduct electricity or heat, this may lead to
malfunction. In particular, plasticizers can redeposit on exposed
relay contacts and prevent them from closing properly. (Ref. 16.)
Most polymers will be stable in the vacuum of space at temnpratures as

high as they can withstand in air. Important exceptions are some ry-
lons, polysulfides, cellulosics, acrylics, polyesters, epoxies, and
urethanes. Vacuum stability is sensitiv3. to formulation and curing
procedure; addition agents such as plasticizers, mold lubricants, and
polymerization catalysts are generally detrimental. Exposure to
vacuum will not cause loss of engineering properties unless appreci-
able loss in weight occurs. Table 3.2.3.11-1 gives an estimated order
of merit for the behavior of polymers in vacuum, together with temper-
atures where the estimated weight loss is 10% per year. (Ref. 16.)
The particles of the Van Allen radiation belts will damage engineering
properties of all exposed polymer surfaces; solar flare emissions will
probably affect u.posed surfaces of the materials more sensitive to
radiation damage. Radiation protection guidelines are presented in
section 3.2.5.2. (Ref. 16.)
3.2.4 Induced Environmental Factors
The induced environmental factors, consist of the following elements:
1. Mechanical factors - shock, vibration, and acceleration
2. Radiation - nuclear, thermal, and electromagnetic
3. Atmospheric pollutants
The characteristics describing mechanical factors may be presented in

any of several different ways. Basically, vibration is defined in
terms of repeated displacement versus time; shock is defined in the
same units but is not repetitive. Acceleration is usually presented in
gravitational units, and acoustical vibration is described in decibels
and frequency.
Shock connotes impact, collision, or blow, usually caused by physical

contact. It denotes a rapid change of load, or a rapid change of ac-
celeration with a resultant change of load. Shock occurs when a
structure is subjected to a suddenly applied force, resulting in tran-
sient vibration of the structure at its natural frequencies. The
3-37
Table 3.2.3.11-1. Decomposition of Polymers in

High Vacuum. (Ref. 16)
Temperature for 10% weight

loss per year in vacuum
PoIymer 0C °F
Nylon 30 -- 210 80 - 410

Sulfide 40 100
Cellulose nitrate 40 100
Cellulose, oxidized 40 100
Methyl acrylate 40 - 150 100 - 300
Ester 40 - 240 100 - 460
Epoxy 40 - 240 100 - 460
Urethane 70 - 150 150 - 300
Vinyl butyral 80 180
Vinyl chloride 90 190
Linseed oil 90 200
Neoprene (chloroprene) 90 200
Alkyd 90 - 150 200 - 300
Methyl methacrylate 100 - 200 220 - 390
Acrylonitrile 120 240
Isobutylene-Isoprene (butyl rubber) 120 250
Styrene-buladiene 130 270
Styrene 130 - 220 270 - 420
Phenlic 130 - 270 270 - 510
Butadiene-acrylonitrile (nitrite rubber) 150 - 230 300 - 450
Vinyl alcohol 150 310
Vinyl acetate 160 320
Cellulose acetate butyrate 170 340
Cellulose 180 350
Carbonate 180 350
Methyl styreno 180 - 220 350 - 420
Cellulose acetate 190 370
Propylene 190 - 240 370 - 470
Rubber, natural 190 380
Melamine 190 380
Silicone elastomer 200 400
Ethylene tereplithalaie (nylon, dacron) 200 400
Isobutylene 200 400
Vinyl toluene 200 400
Styrene, cross-linked 230 - 250 440 - 490
Butadiene-styrene (GR-2 = SBR) 240 460
Vinyl fluoride 240 460
Etylene, low density 240 - 280 460 - 540
3-38
Table 3.2.3.11-1. Decomposition of Polymers in High Vacuum.

(Ref. 16) (Continued)
Temperature for 10% weight

loss per year in vacuum
0
Polymer C *F
Butadiene 250 490

Vinylidene flouride-hexaflouropropene 250 490
Chlorotrif luoroethylene 250 490
Chlorotrifluoroethylene-vinylidene flouride 260 500
Vinylidene flouride 270 510
Benzyl 280 540
Xylylene 280 540
Ethylene, high density 290 560
Trivinyl benzene 290 560
Tetraf luoroethylene 380 710
Methyl phenyl silicone resin 380 710
3-39
magnitude of the vibration may become great enough to cause fracturing

of brittle material or yielding of ductile material.
Vibration is an oscillation in which the quantity is a parameter that

defines the motion of a mechanical system. Vibration has also been
described as the variation, usually with time, or the magnitude of a
quantity to a specified reference, when the magnitude is alternately
greater and smaller than the reference. Vibration may be periodic,
consisting of motions at one or more frequencies with the motion at
each frequency being harmonic, or it may be random, in which case thc
amplitudes and various frequencies vary randomly with respect to time.
An additional type of vibration, termed white-noise vibration, has no
defined frequencies of motion. The excitation forces that cause vib-
ration may be mechanical in nature, such as caused by a :eciprocating
motion or they may be acoustic in nature, such as caused by rocket
engine noise.
Acceleration is the change of velocity, or the rate of change of

either speed or direction, or both. Whether displacement, velocity,
or acceleration is used in defir 4 ng shock, thc implication of a rela-
tively sudden change is always present. Acceleration by itself does
not constitute shock. For example, a structure subjected to steady-
state acceleration is not considered to be undergoing shock.
Excessive noise produces adverse effects on equipment and on struc-

tures. It also induces and augments vibration and the adverse effects
of vibration. Malfunction and fatigue failure of electronic equipment
are possible at noise levels above 130 decibels (dB).
Nuclear explosion factors are the phenomenon associated with the ex-
plosion or detonation of nuclear devices. The nuclear explosion fac-
tors of interest in this section are: radiation effects and electro-
magnetic pulse effects. Radiation effects are either pure eneigy such
as X or gamma rays, or particulate energy such as alpha or beta par-
ticles, or neutrons. Nuclear radiation intensity is expressed as unit
energy per unit time, such as erg/sec or ev/sec. Radiation received
is usually measured in rads where a rad is a unit of absorbed dose
equal to 100 ergs/g of absorbing material. Identification of the type
of radiation, intensity, duration (half-life) and dose rate are all of
primary importance.
Nuclear explosions produce electromagnetic pulse (EMP) either through

assymetry in the electric charge distcibution in the region of the
detonation or through the expansion of plasma in the earth's magnetic
field. Voltages generated by this phenomenon are coupled into the
susceptible receiver.
3-40
Atmospheric pollutants are considered to be principally any airborne

man-made solid, gas, or liquid contaminants although naturally occur-
ring contaminants such as both solid and liquid volcanic materials are
encountered. Protection from atmospheric pollutants is achieved in
the same manner as sand and dust protection described in section
3.2.3.9.
3.2.5 .Induced Environmen
This section presents general design guidelines for the protection of

dormant systems from induced environmental factors. Subsequent sec-
tions of this report present more detailed design guidelines for the
protection of dormant systems.
3.2.5.1 Mechanical Factors
Mechanical factors that may be encountered by dormant systems include:

shock, vibration, and acceleration. The following subsections present
basic design guidelines for the protection of dormant systems from
echauical shock environments. Subsequent sections of this report
present more detailed guidelines for the protection of dormant systems
from mechanical environmental factors encountered during transporta-
tion and handling and storage.
3.2.5.1.1 Vibaioin
The most severe vibration encountered by most dormant systems is the

vibration encountered during transportation and handling. The trans-
port vehicle as well as the handling and service systems is often
characterized by severe vibratory stresses. Thus, data on vibrations
encountered in the transportation of the dormant system is essential.
The following is a general sunary of the most common forcing frequen-
cies that will be encountered by different transportation carriers:
(a) Railroad : 2 to 7 Hz
(b) Truck: 5 to 500 Hz
(c) Aircraft: 20 to 60 Hz
(d) Ships: 11 to 100 Hz
The mode of transportation and the applicable frequencies tabulated

provide the required data relating to imposed vibrations. The de-
signer must assure that the dormant system is capable of withstanding
the vibrations imposed by the above transportation vehicles. Packag-
ing, or cushioning guidelines to protect the system from vibration
3-41
encountered during transportation can be found in sections 4.5 and 4.6

of this report.
Vibration may affect material in many ways. Generally, these may be

classified into one or more of the following categories: (Ref. 14.)
1. Malfunction of sensitive electrical, electronic, and mechanical

devices
2. Mechanical and/or structural damage to structures both sta-

tionary and mobile
3. Frothing or sloshing of fluids in containers
Table 3.2.5.1.1-1 summarizes possible vibration induced damages to

electrical and electronic components.
The process of reducing the effects of the vibration environment on

material is known as vibration control. Essentially vibration control
consists of varying the structural properties of systems, such as
inertia, stiffness, and damping properties, in order to attenuate the
amount of vibration transmitted to material or to reduce the effects
of the vibraticn transmitted to the material.
3.2.5.1.2 Shock
Dormant systems must be capable of surviving, without damage, the

shock environment induced by transportation, handling, storage, and
maintenance systems as well as that experienced in the operational
environment of the equipment. Shock is considered a special case of
vibration wherein the excitation is a relatively short-term distur-
bance that has not reached or has ceased to be steady state. The ex-
citation is nonperiodic, usually in the form of a pulse, step, or
transient vibration. Shock excitation is generally described by a
time history of the rapid variation in the force applied to the
system, or by displacement, velocity, or acceleration imposed upon the
system. Acceleration is the most frequently measured shock para-
meter. (Ref. 14.)
One of the major problems in transporting material between any two

locations is the shock environment to which it is exposed. The fol-
lowing is a summary of the most severe shocks encountered in the four
basic transportation modes: (Ref. 14.)
1. aircraft - greater than 12 G with durations less than 0.1 sec

during landing and takeoff
3-42
Table 3.2.5.1.1-1. Vibration-Induced Damage to Electrical and

Electronic Equipment. (Ref. 14.)
Component
Category Damage Observations
Cabinet and Among some 200 equipment cabinet and frame structures subjected to
frame shock and vibration, damage included 30 permanent deformations, 17
Structures fractures in areas of stress concentration, two fractures at no
apparent stress concpntrations, 23 fractures in or near welds, and
26 miscellaneous undefined failures.
Chassis Nearly 300 chassis subjected to shock and vibration experienced 18

permanent deformations, eight fractures in or near welds, nine
fractures at no apparent stress concentrations, 46 fractures at
points of stress concentration, and 12 miscellaneous failures.
Cathode-Ray Cathode-ray (CR) tubes are susceptible to vibration damage if they

Tubes are improperly mounted and supported. CR tubes with screens larger
than 5 in. are particularly susceptible. Of 31 cathode-ray tubes
subjected to shock and vibration, the deflection plates of one tube
became deformed, another had a filament failure, five suffered
envelope fractures, and one had a glass-socket seal break.
Meters and Although the moving coil type of meter comprises the majority of
indicators units in this category, other indicators include Bourdon tubes and
drive-type synchros. Of the latter group, most of the failures were
either erratic performance or zero shift difficulties. Nearly 200
units subjected to shock and vibration. Two suffered permanent
deformation of the case, one had elements loosened, 12 gave erratic
readings, one had the glass face fractured, two developed internal
open circuits, two had loose or damaged pivots, three had deformed
pointers, and 10 others failed from miscellaneous causes.
Wiring Wiring failure from shock and vibration is a serious problem. A

defect not only results in malfunctioning of the equipment but
presents a difficult troubleshooting job in locating the wire break.
Transformers In electronic equipments transformers are probably the heaviest and

densest components found on an electronic chassis. Because of the
weight and size of transformers, shock and vibration are more likely
to produce mechanical rather than electrical failures. While not
all mechanical failures immediately prevent the transformer from
functioning properly, they eventually result in destruction of the
transformer and damage to surrounding components.
3-43
2. rail - 30 to 50 G during humping operation
3. sea - less than 2 G
4. highway - maximum of 10 G, but usually in the range of 1 to 2 G
Handling shocks are usually generated by flat drops on a hard surface,

such as dropping a package on a concrete surface from the tailgate of
a truck during unloading operations. Handling shocks are usually des-
cribed by the impact velocity or the drop height and angular orienta-
tion at impact. The shocks are generally less severe than a 36 inch
drop on concrete, which corresponds to a velocity change of approxi-
mately 165 inches per second. (Ref. 14.)
Storage shock environment includes those shocks experienced by

material while in warehouses or storage areas for any extended period.
Shocks typically occur from being moved by forklift and from being
struck by other packages during stacking operations. The acceleration
values usually don't exceed 4 G. (Ref. 14.)
The resistance of an item to shock constitutes its ability to with-

stand impact without damage. The fragility rating of an item is a
quantitative description of this ability. Fragility rating is ex-
pressed as an acceleration in G units. Specifically it is a measure
of the maximum shock levels that an item can withstand yet still func-
tion properly. Therefore, if the estimated shipping and handling
shock environment is greate- than the fragility rating of the item
some form of shock mitigation bystem must be employed. Fundamentals
of package design, barrier, cushioning, and container design are dis-
cussed in section 4.5 of this report.
3.2.5.1.3 Acceleration
Dormant systems that are subjected to relatively long periods of ac-

celeration will suffer the same types of damage which result from ex-
posure to shock and vibration environments but with much less fre-
quency and severity. Acceleration causes a mechanical force to act on
an item so that its ability to survive acceleration is a measure of
its structural integrity. Equipment designed to operate in shock and
vibration environments will in mort cases survive the operational ac-
celeration environment with no ill effects. However, certain devices
such as panel meters, electromechanical time-delay relays and acceler-
ometers may be affected more seriously by acceleration than by shock
or vibration. (Ref. 14.)
-- 44
When a transport vehicle (rail or highway) accelerates from 0 to 60

miles per hour (mph) in 30 seconds, the acceleration is approximately
2.9 ft/s 2 , to do so in 10 seconds the acceleration is 8.8 ft/sec 2 .
In terms of G, these accelerations are 0.09 and 0.27 respectively.
Transport aircraft do not exceed these G levels in normal operation.
To reach a takeoff speed of 120 miles per hour in 30 seconds, the ac-
celeration is approximately 0.18 G; in normal flight, the maximum ac-
celeration is from 0.04 to 0.06 G; and in maneuvers such as a turn
with a 30 degree bank angle, the maximum acceleration is 0.25 G.
Thus, normal accelerations on land vehicles and aircraft will be below
0.3 G. (Ref. 14.)
Table 3.2.5.1.3-1 summarizes some of the effects of acceleration.
The primary means of protecting dorma"L systemis from the acceleration

environment is through proper packaging. In order to effectively
package an item for shipment or storage, the fragility rating of the
item must be known as well as the environmental levels that are likely
to be encountered. Packaging guidelines are presented in section 4.5
of this report.
3.2.5.2 Nuclear Radiation
Although a natural background level of nuclear radiation exists, the

only terrestrial nuclear radiation that is of interest is that asso-
ciated with man made sources such as reactors, isotope power sources,
and nuclear weapons. The most important of these sources is nuclear
weapons, the effects of which ca, produce both transient and permanent
damaging effects in a variety of materials. In space various levels
of radiation also exist.
X rays, gamma rays, and neutrons are the types of nuclear radiation of
most concern. As opposed to charged nuclear particles, which also
emanate from nuclear reactions, those forms of radiation listed have
long ranges in the atmosphere; thus, they can irradiate and damage a
variety of materials.
Among the nuclear effects that have been of most concern are those
called "Transient Radiation Effects on Electronics". These are due to
the transient radiation pulse from a tactical nuclear burst and are
often referred to as TREE. These transient effects are due primarily
to the ionization and total dose effects of gamma rays and X rays, as
well as induced permanent lattice defects due to neutron fluence. The
separation of upset effects aud permanent effects is made on the basis
of the processing of erroneous data or damage to the device structure.
For example, a large current pulse may be produced by ionizing radia-
tion, and this current pulse may result in permanent damage to a de-
vice by overstress. The permanent damage results from overstress due
3-45
Table 3.2.5.1.3-1. Effect of Acceleration on Military

Equipment. (Ref. 14.)
Item Effect
Mechanical: moving parts, Pins may bend or shear; pins and

structures, fasteners reeds deflect; shock mounts may
break away from mounting base;
mating surfaces and finishes may be
scoured.
Electronic and electrical Filament windings may break; items

may break away if mounted only by
their leads; normally closed
pressure contacts may open; normally
open pressure contacts may close;
closely spaced parts may short.
Electromagnetic Rotating or sliding devices may be
displaced; hinged part may
temporarily engage or disengage;
windings and cores may be displaced.
Thermally active Heater wires may break; bimetallic

strips can bend; calibration may
change.
Finishes Cracks and blisters may occur.
Materials Under load, materials may bend,

shear, or splinter; glue lines can
separate; welds can break.
3-46
to excess current rather than to direct radiation-induced material

property change. Generally, total dose effects from X rays and gamma
rays have no effect on dormant electronic equipment. The designer
must still consider these environments when designing dormant systems
in order to insure operational readiness in total dose environments.
Table 3.2.5.2-1 presents a basic summary of nuclear radiation environ-

ments and basic protective measures.
It is impossible to completely protect a system from nuclear radia-

tion as can be accomplished for some other environmental factors. The
variety of effects produed by nuclear radiation for different
materials and components makes protective design difficult. The most
common procedure employed is to define a radiation hardness level in a
given material item and to design and test the item to that level.
3.2.5.3 Atmospheric Pollution
Air pollution may be defined as the presence of foreign matter sus-

pended in the atmosphere in the form of solid particles, liquid drop-
lets, gases, or in various combinations of these forms in sufficient
quantities to produce undesirable changes in physical, chemical, or
biological characteristics of the air. Table 3.2.5.3-1 summarizes
basic material categories, how they are affected by atmospheric pollu-
tants, and the specific pollutants involved. (Ref. 14.)
The severity with which atmospheric pollutants attack materials will

in general vary with the locale and the season. The rate of attack
is influenced by the relative humidity, the extent of the pollution,
the quantity and frequency of rainfall, air movement characteristics,
the frequency of occurrence and duration of fog, the proximity to the
sea, the amount of solar radiation, and temperature ranges.
There would be very little, if any, atmospheric corrosion, even in the

most polluted atmospheres, without moisture. Most metals have a cri-
tical relative humidity threshold below which the rate of corrosion is
very slow, but when exceeded produces a rapid rise in the rate of cor-
rosion. Rain can increase the rate of some metals while it decreases
that of others. By washing the surface of a specimen, rain can remove
corrosive pollutants that have collected on the surface, thus reducing
the corrosion rate. In other cases, it can remove soluble corrosion
products that had been protecting the material. Fogs and dews are, in
general, damaging because they have no washing effect on the Lirface
but rather form surface films of moisture which absorb pollutants from
the atmosphere. (Ref. 14.)
3-47
Table 3.2.5.2-1. Summary of Nuclear Radiation Environments and

Basic Protection Measures.
II
[Radiation [ Radiation
Source Environment Dormancy Storage Guidelines
Tactical Radioactive debris, Seal circuits to avoid contamination

X-rays, Charged radioactive debris.
Particles, Neutrons
Shield Ground Base Systems
Use devices with narrow junction

size or minimum base width.
Fast switching speeds when possible
Design for reduced gain reduction in

efficiency arid an increase in: leakage
current, rev. breakdown voltage, and
forward voltage.
Space Electrons, Protons, Shielding may be used against light,

X-rays, Gamma rays, charged particles; however; surface
Heavy Ions charging and discharging to internal
circuitry is a serious problem and must
be considered during design.
Gamma rays and X-rays cause total dose

damage in passive and optical materials
(i.e., teflon). Use existing data to
determine severity for specific
application.
Heavy ions affect the atomic structure

of the device. Similar to neutrons and
are usually charged causing electrical
disturbancies in charge sensitive de-
vices. However, this ratz zf incidence
is low and i-ay have no effect.
-48
Table 3.2.5.3-1. Air Pollution Damage to Various Materials. (Ref. 14.)
Typical Principal Other envir-r-

Materials Manifestation Measurement Pollutants mental Factors
Metals Spoilage of Weight gain of corrosion products, S02 , acid gases Moisture,
surface, loss weight loss after removal of corrosion temperature
of metal, products, reduced physical strength,
tarnishing changed reflectivity or conductivity
Building Discoloration, Not usually measured quantitatively S02 , acid gases, Moisture,
materials leaching sticky particulates freezing
Paint Discoloration, Not usually measured SO2 . H 2 S, sticky Moisture,

softened finish particulates fungus
Leather Powdered Observation, loss of tensile strength SO2 , acid gases Physical wear
surface,
weakening
Paper Embrittlement Decreased folding resistance S02 , acid tases Sunlight
Textiles Reduced tensile Reduced tensile strength, altered S02 , acid gases Moisture,
strength, fluidity sunlight,
spotting fungus
Dyes Fading Fading by reflectance measurements NO2 , oxidants, Sunlight,

SO2 moisture
Rubber Cracking, Loss in elasticity, increase in depth Oxidants, 03 Sunlight

weakening of cracks when under tension
Ceramics Changed surface Changed reflectance measurements Acid gases Moisture

appearance
3-44
Temperature influences the rate of those chemical reactions that cause

material deterioration. The temperature also determines the rate of
drying or the length of time that a surface remains wet. During an
atmospheric temperature inversion, exposed objects, especially metals,
lose heat rapidly and cool to temperatures below that of the ambient
air. If their surface temperature falls below the dewpoint, the sur-
fa i becomes moist and, in the presence of corrosive pollutants whose
concentrations are increasing due to the temperature inversion, create
a situation conducive to material damage. (Ref. 14.)
Sunlight energy is an important element in the nitrogen dioxide photo-

lytic cycle in which damaging agents such as ozone and hydrocarbon
free radicals are formed in a series of complex photochemical reac-
tions in the atmosphere. (Ref. 14.)
Air movement tends to disperse pollutants throughout the atmosphere.

Specific design steps for protection aguinst the effects of pollutants
arp not comnmon, since protection is normally obtained by methods em-
ployed for protection against other environmental factors. (Ref. 14.)
Metals exhibit varying degrees of susceptibility to atmospheric pollu-

tants. Therefore, the proper choice of metal or alloy is important
for dormant equipment components that will be exposed to polluted at-
mospheres. Aluminum, stainless steels, and copper generally suffer no
adverse effects from atmospheric pollutioa. Ordinary carbon ste,
require some form of protection when exposed to atmospheric cot.-
tions. Preservatives dnd coatings are discussed in section 4.5 -f
this report. (Ref. 14.)
Complex prer ,'on electronic systems, such as computers and switching

networks, ho - been affected by high concentrations of particulate
matter and sulfur dioxide. Dormant systems that are exposed to air
pollutants should be installed in an air-tight container.
Semiconductor devices, integrated circuits, relays, switches, etc. are

susceptible to air pollutants. If used in dormant systems, these de-
vices should be hermetically sealed to preclude any future damage from
air pollutants. (Ref. 14.)
Electr cal contacts, such as those on relays and switches, are suscep-
tible to particulate matter, hydrogen sulfide, and sulfur dioxide.
These components should be hermetically sealed when possible to pre-
vent damage from air pollutants. (Ref. 14.)
3-50
3.3 DORMANT STORAG
This section describes the dormant storage environment and its effects
on systems. Section 3.2 presented the primary degrading environments
that may be encountered by dormant systems. Section 3.2 also pre-
sented some basic design guidelines that will assist the design en-
gineer in designing adequate protection for dormant equipment. Sub-
sequent sections of this report present more detailed design guide-
lines for protecting systems during dormant storage.
3.3.1 atoraQe Types. Capacities. and Limitations
Dormant storage types can be classified by the type of shelter

facility: (Refs. 2, 3.)
Type 1 - outside storage
Type 2 - sheltered storage
Type 3 - dehumidified structural storage
Type 4 - dehumidified nonstructural storage
Type 1 outside storage consists of a storage area which is exposed to

all extremes of loLal, natural environments. Outside storage can be
an improved storage space or an unimproved storage space. The open
improved storage space is an area that has been graded and hard sur-
faced, or prepared with a topping (such as steel mats) to permit ef-
fective material handling. Open unimproved storage space is an open
area that has not been surfaced for storage purposes. Outside storage
is suitable only for materials that are not readily damaged by weather
conditions. The packaged item must also be protected from fungi,
pests, dust, and pilferage. In many cases, tarpaulins must be pro-
vided to cover stored items. The open unimproved storage space has an
additional limitation in its susceptibility to wet ground conditions
which can limit the satisfactory operation of materials handling
equipment. Material which is expected to be stored in outside storage
should be ickaged, and should be constantly surveyed for evidence of
deterioration. Packaging guidelines to protect dormant systems are
presented in section 4.5 of this report.
Type 2 sheltered storage consists of ventilated or unventilated,

heated or unheated buildings, shelters or enclosures of structural
characteristics that are designed to afford protection from the ele-
ments. The stored material is not protected from dtmospheric changes
of temperature and humidity. Periodic surveillance and maintenance
are required to keep deterioration to a minimum. These types of
3-51
structures include every type of covered storage space including

general purpose warehouses, flammable storage warehouses, transit
sheds, dry tanks, and typical storage sheds.
Type 3 dehumidified structure storage consists of a structure in which

the atmosphere is maintained at a relative humidity of 40 percent or
less. Controlled humidity storage in structures provides the highest
degree of protection and is the most economical method of storage for
items of a critical nature since little surveillance and maintenance
is required. When properly sealed and conditioned, almost any type of
warehouse may be dehumidified. In actual practice, however, the
general warehouse is the structure most often converted for dehumidi-
fied storage. Dehumidification has proved to be the most economical
and efficient method of preservation for many classes of items and is
being used increasingly. Whenever possible, Type 3 storage should be
used for dormant systems because it offers the best protectio)n against
natural environmental factors.
Type 4 dehumidified nonstructural storage consists of complete or par-

tial sealing of the packaged item, with mechanical or static dehumudi-
fication controlling the relative humidity inside the package to a
maximum of 40 percent. A typical dehumidifier which automatically
reactivates the desiccant is specified in MIL-D-16886. This dehumidi-
fier is classified by the type of heat used for dessicant reactiva-
tion. Controls are electrically operated. (Ref. 1.) Other dessi-
cants and dehumidifiers are specified in MIL-D-3263A, MIL.-D-43266,
MIL-D-3716, AND MIL-D-3464. Type 4 storage .s recommended for dormant
system protection whenever Type 3 storage is not possible.
Table 3.3.1-1 presents some of the environmental conditions in storage

facilities.
Table 3.3.1-1. Typical Environmental Conditions in Storage

Facilities. (Ref. 13.)
Relative Water Vapor

Temperature Humidity Partial Pressure,
Environment OF % in. Hg
Open Slab & Shed 61.1 68.9 0.378
Standard Warehouse 66.0 55.0 0.363
50% RH Warehouse 66.5 49.0 0.327
40% RH Warehouse 67.3 39.7 0.27c
3-52
3.3.2 Storage of Explosive Material
The storage of explosive material is determined by the potential haz-

ards of the material. Materials of an explosive nature must be stored
in standard amuunition magazines designed for that purpose, in areas
designated specifically for the storage of explosives, ammunition, or
similar components. These areas are usually not wired for electricity
and generally are not heated. Therefore, the packaged item must be
adequately preserved to protect it from deterioration. The package
should be designed to facilitate inspection required by periodic
monitoring and surveillance without removal of the covering.
Earth-covered magazines offer the greatest protection to explosives

and afford the greatest degree of protection from the results of an
explosion. Such magazines are preferred for the storage of all explo-
sives. Other types of magazines built in accordance with approved
drawings may be used as conditions dictate. Refer to DOD 4145.27M,
DOD Ammunitions and Explosives Safety Standards, for magazine siting
requirements. Outdoor storage is not considered desirable. Priority
for indoor storage should be given to items requiring the most protec-
tion from the weather. High explosives, solid propellants, pyrotech-
nics, and critical items should not be stored outside. The basic
types of magazines are the igloo, and above ground magazines. (Refs.
2, 3.)
The igloo is used for the storage of ammunition and high explosives.
An igloo is a type of magazine and is generally constructed of rein-
force concrete or corrugated steel with an arch type roof covered with
earth. Igloos may or may not be ventilated, and although unheated,,
the temperature in unventilated igloos usually ranges from 40 to 45
degrees F in winter, and 60 to 70 degrees F in summer. The arched
roof of the typical igloo limits the storage height of the explosives.
The above ground magazine is designed and used for the storage of am-
munition and explosives and is constructed with roof, side walls, and
end walls. Because of the nature of the items stored in them, above
ground magazines should be built of fire-resistant materials, and
should be ventillated to prevent temperature build-up. They should be
widely separated to minimize the destructiveness of an explosion,
should one occur.
The propulsive characteristics of explosive items such as rockets,

rocket motors, assist take-off units, and missiles must be taken into
consideration during all phases in order to obtain as much safety as
possible under the circumstances. Rockets and like items should be
stored in cool, dry magazines and never in direct rays of the sun. An
environment as near "optimum" as possible for the particular item will
3-53
afford the safest storage condition, and consideration should be given

to this aspect. However, the design upper and lower temperature
limits specified for the item should not be exceeded. Prolonged expo-
sure of the propellant of these items to either high or low tempera-
tures may increase the normal rate of dcteiioration or makc them morz
susceptible to initiation or damage during handling. Such damage
could result in uneven burning and dangerously high internal pressure
during use.
3.4 REFERENCES
1. Not Used
2. AMCP 706-121 Packaging and Pack Engineering, US ARMY Material

Command, October 1964.
3. TM 743-200/NAVSANDA PUB 284/AFM 67-3/NAVMC 1101/DSAM 4145.1,

Storage and Materials Handling, Departments of the Army, Navy,
the Air Force, the U.S. Marine Corps, and the Defense Supply
Agency, January 1967.
4. AFSC Design Handbook DH 1-5, Environmental Engineering,

Department of the Air Force, 5 October 1983.
5. AFSC Design Handbook DH 1-2, General Design Factors, Department

of the Air Force, 5 August 1986.
6. AMCP 706-115, Engineering Design Handbook, Environmental Series

Part One, Basic Environmental Concepts, U.S. Army Material
7. DARCOM-P 706-298, Engineering Design Handbook, Rocket and

Missile Cont3iner Engineering Guide, January 1982.
8. MIL-STD-210C, Climatic Information to Determine Design and Test

Requirements for Military Systems and Equipment, 9 January 1987.
9. AR 70-38, Research, Development, Test, and Evaluation of

Material for Extreme Climatic Conditions, 1 August 1979.
10. MIL-STD-810D "Environmental Test Methods and Engineering

Guidelines", 19 July 1983.
11. MIL-HDBK-338, "Electronic Reliability Design Handbook",

15 October 1984.
12. MIL-HDBK-157, Transportability Criteria, 6 March 1980.
3-54

Part Two, Natural Environmental Factors, U.S. Army Material
Command, April 1975.

Part Three, Induced Environmental Factors, U.S. Army Material
Command, 20 January 1976.

Part Pour, Life Cycle Environments, U.S. Army Material Command,
31 March 1975.
16. "Effects of Space Environment Upon Plastics and Elastomers",

Jaffe, L.D.; Chemical Engineering Progress Symposium Series,
Vol. 59.
17. "An Approach for Assessing Missile System Dormant Reliability",

R.D. Trapp, et. al., BDM Corporation, DTIC Report No. ADA
107519, 30 January 1981.
3-550' (3-56 blank)

4.0 DESIGN GUIDELINES
4.1 PART SELECTION/CONTROL
Designing reliability into a system requires the careful selection of

the discrete parts that comprise the system. The objective of any
part selection and control program is to obtain parts that will per-
form adequately and reliably during the designed service life of the
equipment in which they will be used and that will be available at
minimum cost whenever needed throughout the equipment's service life.
This basic objective applies to all systems regardless of the amount
of time the system will spend in a non-operating state.
The task of selecting, specifying, applying, and controlling the parts

used in complex electronic systems is a major engineering undertak-
ing. It requires a multidisciplinary approach involving the partici-
pation of component engineers, failure analysts, reliability engi-
neers, and design engineers. Numerous controls, guidelines, and re-
quirements must be developed, evaluated, and implemented, including
the development of meaningful procurement specifications which reflect
a balance among design requirements, quality assurance, and reliabil-
ity needs. A list of general ground rules for a successful parts se-
lection and control program is presented in Table 4.1-1.
The essentials of a successful part selection and control program are

detailed in references 1 and 2 and will not be repeated here. They
are applicable to all systbns and should be followed to provide the
framework for an effective i'rogram. The information presented in this
section provides details concerning the selection and control of parts
for systems that will spend a significant part of their service life
in a non-operating state. Combining this information with the guide-
lines presented in references 1 and 2 will allow the tailoring of a
part selection and control program meeting specific program needs.
Central to any successful part selection and control program is an

understanding of the characteristics of the alternative parts for a
given application and their respective failure modes and mechanisms in
the anticipated operating and non-operating environments. Only by
thoroughly understanding their strengths and weaknesses can a designer
properly weigh the alternatives for a given application and arrive at
effective tradeoff decisions that maximize performance and reliability
at the least cost.
4.1-1
Table 4.1-1. Ground Rules for Part Selection and Control

(Refs. 1,2)
A) Determine part type needed to perform the required function and the
environment (operating and non-operating)in which it is expected to
be used.
B) Determine part criticality.
* Does the part perform critical functions, i.e., safety or

mission-critical?
* Does the part have limited life?
* Does the part have long procurement lead time?
* Is the part reliability sensitive?
* Is the part a high-cost item or does it require formal
qualification testing?
C) Determine part availability.
" Is the part on a preferred list?

* Is the part a Standard MIL item available from a qualified vendor?
" What is the part's normal delivery cycle?
" Will the part continue to be available throughout the life of the
equipment?
* Is there an acceptable in-house procurement document on the part?
* Are there multiple sources available?
D) Determine reliability level required for the part in its

application, both operating and non-operating.
E) Select the appropriate burn-in or other screening methods for

improving the part's failure rate (as required).
F) Prepare an accurate and explicit part procurement specification,

where necessary. Specifications should include specific screening
provisions, as needed, to assure adequate reliability.
G) Determine actual stress level of the part in its intended

application. Include failure rate calculations.
H) Employ appropriate derating factors consistent with reliability

prediction studies.
I) Determine need for non-standard part and prepare a request for

approval as outlined in MIL-STD-965.
4.1-2
In a theoretical study of the possible failure mechanisms and material

degradation processes that will cause electronic components to fail
during long-term non-operating periods (Refs. 3,4), it was concluded
that the most important environmental stresses during non-operating
periods are mechanical, chemical, and low thermal. Since a non-oper-
ating environment is normally free from electrical stresses and the
higher temperatures associated with operating devices, the chemical
and physical stresses associated with electronic devices during non-
operating periods are determined by the ambient conditions and mate-
rials within a package or assembly. The synergism of the three pri-
mary non-operating stresses is critical. While any one of the three
acting alone may not be particularly damaging, the combined effect of
two or three acting together increases the likelihood of device
failure.
The two primary sources of mechanical stress during non-operating

periods are inertial forces and thermal-mechanical interactions.
Inertial forces occur from cyclical accelerations associated with
vibrations and from transient accelerations related to shock from
transportation and handling. Except for large components such as
relays, transformers, connectors, and large hybrids, most electronic
components have very small masses and are not particularly susceptible
to damage from the inertial forces expected during non-operating
periods.
Thermal-mechanical interactions introduce differential expansion be-

tween materials within a device and between subassemblies and inter-
connections, due to temperature changes. These stresses occur in
electronic components and assemblies due to two closely related phe-
nomena. First, the materials in an electronic component or assembly
have different coefficients of linear expansion so that large stresses
can result even from slow, uniform temperature variations. Second,
temperature gradients within the same material can lead to differen-
tial displacements due to separate regions of the material being at
different temperatures. Other effects associated with thermal-
mechanical phenomena inckude residual mechanical stresses which
usually exist in electronic devica structures because oxides,
nitrides, and metal films are grown and bonds are made at elevated
temperatures and then cooled down to ambient conditions. (Refs. 3, 4)
Thermal shock is most critical for materials having high thermal im-
pedance such as ceramics. While electronic equipment will be therm-
ally cycled at various times during non-operating periods, the thermal
mass of the equipment will prevent individual devices from experienc-
ing rapid temperature changes. Previous studies (Refs. 3,4) have
shown that the maximum temperature span for non-operating equipment in
4.1-3
open field storage during a given day due to diurnal temperature cycl-
ing will be less than 70 degrees Celsius so that the rate of change
will be less than 6 degrees Celsius per hour.
Chemical stresses develop in electronic devices from a large number of

potential chemical interactions including solid-solid metallurgical
processes as well as chemical reactions with external contaminants.
The rate of degredation depends upon temperature and a nuber of
metallurgical factors such as specific materials, microcracks, compo-
sitional variations, grain size, dislocation density, and impurities.
Some of the more important sources of chemical stresses in electronic
devices and assemblies during non-operating periods are as follows
(Refs. 3, 4):
" Concentration gradients within device structures.

* Process chemicals remaining from fabrication steps.
" Gases evolved from device materials.
" Contaminants including moisture.
" Environmental gases introduced through deficiencies in hermetic
seals.
* Stress accelerated chemical reactions.
* Galvanic cells.
The overwhelming area of concern for long-term non-operating reliabil-

ity is the influence of chemical contaminants. A large number of con-
taminants have been found to cause failures in electronic devices,
including halogen ions, alkali metal ions, residual process chemicals,
hydrogen, oxygen, atmospheric pollutants and water. The thresholds
for contaminant induced chemical reactions are generally unknown.
Moisture may be the single most important factor in long-term non-

operating reliability and all possible steps should be taken to elimi-
nate it within electronic devices and assemblies. The amount of water
required to degrade the materials within a device package is known to
be very small, and may be as little as one molecular layer. (Refs. 3, 4)
Most failures that occur during non-operating periods are of the same
basic kind as tiose found in the operating mode, though precipitated
at a slower rate. Furthermore, experience has also shown that most
failures that occur during non-operating periods are the result of
latent manufacturing defects rather than specific aging mechanisms.
These two facts have significant implications for the selection and
control of parts for ecquipment that will spend a significant portion
of its service life in a non-operating state. Since most failure
modes for non-operating periods are the same as for operating periods,
the same part selection and control measures used to achieve designs
4.1-4
that exhibit high operating reliability will also achieve designs that
exhibit high non-operating reliability. For example, the use of Es-
tablished Reliability components instead of standard components or the
use of Class S integrated circuits instead of Class B integrated cir-
cuits will result in both higher operating and non-operating reliabil-
ity. In Reference 6, S.M. Cherkasky estimates that the number of
failures during non-operating periods can be reduced by two-thirds to
three-fourths by going from MIL-STD components to Established Relia-
bility (ER) components. In addition, since most non-operating fail-
ures are the result of latent manufacturing defects such as identified
in Table 4.1-2, screens that are effective in weeding out such defects
will be effective in eliminating most non-operating failures.
Table 4.1-2. Typical Latent Manufacturing Defects Causing

Failures During Non-Operating Periods (Ref. 5)
e Contamination of integrated circuits.
* Lifted bonds on transistors and integrated circuits.
o Electrolysis of nichrome metal film resistors due to entrapped

moisture.
* Dielectric breakdown in ceramic capacitors.
* Electrolyte leakage in wet tantalum capacitors.
a Sealed in moisture that creates voids in nichrome film.
* Inadequate cleaning of ceramic loose cores causing flaking of

nichrome elements.
* Loose conductive particles causing shorts in IC and hybrid

packages.
This second point is not surprising since the generally lower thermal
and electrical stress levels which occur during non-operating periods
as compared to operating periods will tend to prolong the infant mor-
tality period of the classical reliability bathtub curve. As shown in
figure 4.1-1, this period, which is characterized by a rapidly de-
czeasing failare rate as latent workmanship defects are worked out of
a system, will continue for years for non-operating equipment as op-
posed to only a few weeks or months for operating equipment.
4.1-5
OPERATING EQUIPMENT
A C
OPERATING
FAILURE
RATE
0 WEEKS" DECADES
MONTHS TIME --
NON-OPERATING EQUIPMENT
A jB C
NON-OPERATINGI
FAILURE
RATEI
j~C/
YEARS DECADES
TI M E
ik = INFANT MORTALITY PERIOD

B = CONSTANT FAILURE RATE PERIOD
C = WEAROUT PERIOD
Figure 4.1-1. Operating vs. Non-O)peratinq Equipment Reliaoility

Bathtub Curves
The term "screen" as used here refers to the application of a stress

test, or tests, to a device with the purpose of revealing inherent
weaknesses (and thus incipient failures) of the device without de-
stroying the integrity of the device. The rationale for such action
is that the inferior devices will fail and the superior devices will
pass, provided the tests and stress levels are properly selected. Of
all of the factors that will reduce failure rates during non-operating
periods, part screening is among those that accomplish it most effec-
tively. (Ref. 6)
In summary, the failure mechanisms of greatest importance during non-

operating periods are those related to latent manufacturing defects,
corrosion processes, and mechanical fracture. Because if the absence
of electrical stress, electrical or !tut2:Itial current induced degrada-
tion processes are not important during non-operating periods. Mois-
ture within a device or assembly package is the most important factor
for both corrosion and mechanically induced failures. Chemicals, in-
cluding moisture trapped within a package due to improper cleaning or
because of evolution from materials such as polymers, are also of
great concern for long-term non-operating reliability. Thermal-
mechanical stresses aided by chemical agents will cause cracks to
propagate in seals, passivation layers, bonds, metallization layers,
silicon chips, and other device structures.
Summarized in table 4.1-3 are general guidelines for selecting and

controlling parts for use in equipment that will spend a significant
part of its service life in a non-operating state. In addition to
these general guidelines, the following sections provide specific part
selection and control guidance.
4.1.1 Resistors
Resistors are functionally classified as fixed or variable. Resistor

classifications are further subdivided into three basic groups defined
by construction as described below: (Ref. 13)
" Composition resistors are made from a mixture of resistive material

and a binder and are molded into a specific shape and resistive value.
* Film resistors consist of a resistive film deposited inside or

outside an insulating cylinder.
" Wire-wound resistors are composed of a resistive wire wound on an

insulating body
Special resistors, such as thermistors and network film resistors are

constructed differently. A detailed description of these resistors is
presented in section 4.1.1.3.
4.1-7
Table 4.1-3. General Part Selection/Control Guidelines

for Dormant Equipment Design
General
" The most important environmental stresses during non-operating

periods are mechanical, chemical, and low thermal. The
synergism of these three stresses is critical.
* The failure mechanisms of greatest importance during

non-operating periods are those related to latent manufacturing
defects, corrosion processes, and mechanical fracture, with
most failures being the result of latent manufacturing defects
rather than specific ag3ng mechanisms.
" Most failures that occur during non-operating periods are of

the same basic kind as those found in the operating mode,
though precipitated at a slower rate.
" Moisture is the single most important factor to long-term

non-operating reliability. All possible steps should be taken
to eliminate it within electronic devices.
* The same part selection and control measures used to achieve

designs that exhibit high operating reliability will also
achieve designs of high non-operating reliability.
Material Selection
* Materials for use within hermetically sealed enclosures should

be selected for minimum outgassing of products detrimental to
equipment and parts. Outgassing products can include moisture
and potentially corrosive chemicals. Pefer to reference 11 fur
a compilation of material outgassing data.
* Hygroscopic materials should be avoided or protected against

accumulation of excess moisture. It has been found on tests
that the moisture to be removed 'rom the materials in
dehumidification is from 5 to 200 tiries the net amount that
must be taken form the air. (Ref. 12)
£ Materials with non-porous surfaces are preferred for their

resistance to entrapment of contaminants. Castings should be
iipregnated to prevent absorption of harmful materials.
" The settlement characteristics of hydraulic and other flui6s

over the non-operating period should be considered.
4.1-8

for Dormant Equipment Design (Con't)
Materi~al Selection Con't
" Avoid materials sensitive to cold flow and creep.
* Avoid metalized and non-metal finishes which have flaking

characteristics.
" Consider age hardeni , and strain relief of materials and

joints.
" In general, amorphous thermoplasLics (particularly if

plasticized) and polymers containing labile bonds (such as
Thiokol polysulfide rubbers) exhibit very high stress
relaxation (or creep) ratios and should be avoided.
* Avoid the use of lubricants. They migrate into undesirable

locations and may outgas harmful products. If required, dry
lubricants, such as graphite, are preferred.
" Do not use teflon gaskets. Teflon will flow under pressure
resulting in seal loss after extended periods of time.
" Use silicone based rubber gaskots in lieu of conventional

rubber gaskets. Silicone gaskuts have superior flexibility
retention over time and will not dry out, become brittle, or
crack with aging as conventional rubber gaskets will.
e Do not use corrosive solder fluxes unless detailed cleaning

procedures are specified along with appropriate verification
met- s to ensure complete removal of residual contaminants.
9 Do not use dissimilar metals, as defined by MIL-STD-889, in

intimate contact with each other unless suitably protected
against electrolytic corro:,ion an described in MIL-HDBK-729 a.d
MIL-STD-889.
4.1-9
Table 4.1-3. General Part Selection/Control Guidel.-jes

Part Selection
" Use parts that have demonstrated successful age histories and
that have inherently good storage characteristics. Avoid parts
which exhibit time-dependent deterioration.
" Do not use non-hermetically sealed semiconductors and

microcircuits, such as plastic (organic or polymeric)
encapsulated or sealed devices. (See sections 4.1.3 and 4.1.4.)
* Do not use semiconductors and microcircuits that contain

nichrome-deposited resistors. They are suceptable to
electrolysis with trace amounts of moisture and they are
subject to deterioration and failure at con'-act interfaces with
aluminum conductors due to interdiffusion.
" Do not use semiconductors and microcircuits with gold-alninum

systems that have not been determined to be free of "purple
plague" problems.
" Select parts that use mono-metallization to avoid galvanic

corrosion and other corrosion related problems.
" Do not seal chlorine or other halogen-containing materials

within any circuitry components or assemblies because chlorine
and other halogen ions can rapidly degrade electronic devices
when moisture is prevent. Polymers used should be simple
hydrocarbons or compounds of carbon, hydrogen, and oxygen.
Nitrogen containing polymers should be considered with
skepticism. Ammonia and amines from curing epoxy resins can
react with ions to form con.plexes and with acids or water to
form cations. Sulfur-cured rubbers and plasticized plastics
can generate volatile sulfur and plasticizer compounds that.
attack not only plastics and rubbers but also c. tain metallic
components (e.g. silver and copper contacts.).
Do not use electronic par.s. *,porients employing polymers tor

package seal) Polymers will transmit moistu,' and other gases.
* Do not use wet electrolytic capacitors. (See section

4.1.2.1.4.)
.lI

Part Selection Con't
" Do not use variable resistors, potentiometers, variable

capacitors, or variable inductors. (See sections 4.1.1.2,
4.1.2.2, and 4.1.5.)
" Avoid the use of electromechanical relays. (See section

4.1.6.2.)
* Where possible, avoid attachments and connections that depend

on spring action for effectiveness, unless the spring is
normally in its relaxed position.
" Use molded cables in preference to bundled wire runs to relieve

stress in individual wires and to reduce the area available for
moisture and other contaminant absorption.
* Do not usc solid tantalum capacitors in circuit applications

without at least 3 ohms/volt series resistance to limit current
irto the capacitor.
" Do not use non-hermetically sealed Lilm capacitors. (See

section 4.1.2.)
* Do not use prohibited parts identified iiA MIL-STD-454.
Evaluation/Control
* Part screening is among the most effective ways of reducing

failures during non-operating periods.
" Particulate matter is one of the dominant concerns as a

non-operating failure mechanism.
Therm-l shock should never be used as a screen test stress for

hermetically sea..ed devices that will be in a non-operating
state for long periods of time. The stresses induced by
thermal shock can result in package seal failure over time.
State-of-the-art technologies must be carefully evaluated for

potential non-operating failure modes and mechanisms prior to
use. Where sufficient field experience data is not available,
accelerated aging tests and material compatability studies
should be performed.
4I-1 i
Dormancy is not a degradation factor for resistors, however, dormant

resistors, especially dormant variable resistors may degrade due to
environmental factors experienced during the dormant period. Moisture
is the most significant degradation factor for non-operating resis-
tors. Most resistors are encapsulated in a molded plastic case or are
conformally coated to provide moisture protection, but no plastic is
equivalent to hermetic sealing.
Shock and vibration are significant degrading environmental factors on

variable resistors. Significant changes in resistance value may be
induced by shock and vibration. Section 4.1.1.2 presents a more de-
tailed examination of dormant variable resistors.
Listed below are several general resistor selection/control guidelines

that are applicable to both operational and dormant resistors (Refs.
1, 13, 20, 21). Subsequent subsections present more detailed part
selection/control guidelines for the various resistor categorizations.
Tables 4.1.1.1-1 and 4.1.1.2-1 present a more detailed examination and
corresponding guidelines for environmental effects on dormant
resistors.
" Select a resistor for each application from the list of standard
types and values. Established Reliability resistors have tighte.
reliability screening, process control, and failure-rate proof
rrom cum ulative life test data, and are therefore recommended for
dormant applications whenever applicable.
" Select a resistor whose resistance value plus tolerance value plus
voltage coefficient plus temperature coefficient plus drift with
time are within the resistance requirements for the required ap-
plication. For dormant applications, especially in extreme en-
vironmental conditions, the designer should consider designing for
cn end of life toleranc with a "buffer tolerance" added to the
purchase tolerance to insure long life reliability.
" Various initial (purchase) tolerances are available depending on

the type of resistor. The designer should realize that the ini-
tial accuracies become meaningless if the inherent stability of
the resistor is not consistent with the basic accuracy require-
ments of the application.
" During dormancy, as well as during operation, any characteristic

(i.e., resistance, power rating, size, etc.) of an, resistor may
change value as d result of stresses caused by the environmental
effects of temperaLure, humidity, pressure, vibration, etc.
Changes caused by envirounental stresses may be linear or non-
linear, reversible or non-reversible (permanent) or combinations
thereof.
4.- I
* Excessive temperature rise due to inadequate heat dissipation is

the predominant cause of failure for operating resistors. Al-
though, a dormant resistor does not experience the same amount of
heat as an operating resistor, the same heat dissipation should be
adhered to, because the dormant resistor will eventually be re-
quired to operate.
* Resistors which are closely mounted together in circuits can cause

moisture and dirt traps. Moisture and dirt traps eventually form
corrosive materials which can deteriorate resistors and other
electronic parts. Moisture can accumulate around dirt even in an
atmosphere of normal humidity. Normal humidity in dormant en-
vironment typically ranges from 40 to 70% relative humidity.
Moisture (humidity, rain, snow, etc.) is the greatest enemy of
dormant resistors. If dormant equipment is subjected to humid
conditions it must be protected with adequate moisture protection,
such as coatings. The designer should design dormant equipment
with hermetically sealed components and conformally coated circuit
boards in order to protect t)e equipment from moisture induced
degradation. A more detailed examination of coatings, preserva-
tives and packaging of dormant equipment is presented in section
4.5.
4.1.1.1 General Characteristics of Fixed Resistors
Fixed resistors are among the most stable electronic components in

dormant or operational configurations. No major problems have been
identified for fixed resistors in a dormant mode. The following sec-
tions present guidelines and general application information for fixed
resistors in dormant modes. These subsections are subdivided by re-
sistor construction type.
Table 4.1.1.1-1 siummarizes environmental factors and effects, and pre-

sents guidelines for fixed resistors in a dormant mode; the table is
categorized by the construction type of the resistors.
4.1.1.1.1 Fixed Composition Resistors
Fixed composition resistors consist of a mixture of finely divided

carbon and a binder, either in the form of a slug or a heavy coating,
on a glass tube. Specially formed wire leads are embedded in the
resistance element. An insulating case, usually phenolic, is molded
around the resistor forming a one-piece enclosure to support the leads
and provide a moisture seal.
Like all plastics, the phenolic case is not completely impervious to

moisture (Ref. 14). After long periods of dormancy in high humidity
4. 1-13
M 41 Al .
.w 4) WA :
43 44 - 4. 4U . 0
m ~ 43 .A 43W4L 4-4 43 .
-1 tm 0..( +1 0. 0
11 0. 0 M0 0
1) 0 0
C = (A "oLA CA =A C C3 2 C M
v 3 0 C >043 C) Q L 4) -
)0 06 0 4)L0 0
43 u4 Q - 3 0
- C c. C Z I3
40
I-
c4 V 0D 4 . 4 _r4
>- £0 = - J £ m3 0 V
F
C .'0a).. -C 4
(A 0
430 *-.-Q 06 zA (A
(A~~ Lc(AJ 0 - ( L d ~ 0 u(L CL
*0 43 43
V4 I- u . 7- .
V4~~ m- C~ C-c (-. C (
CD0 4.'a 0 4
m 4 0 0 o- U ' - .- u 43
0 3 ai LA +14 42. 0m 16 C
0
04 0. 41 0 'J . . C 0
c 4 3 0 4A c) -.
0
00
.
m
0) L0
'-(
W 0
3 1 b-10 A 0 .. C ~ .
0) (NI c- 140'- 4
I4 A ( 3 1. C -
u.. I 0
t-
0 .' V-
o Z (A 0 UL CL 0. 0.
430 43Q (A . 0) 43-1 E 4340 a
COcA
43 ~ ~ I.
U . 40 V 0!
-,
j
(A
Q' CO
4 .
C
4
c
4
mUL
4
.
. 4
CaC £0 .0 O.' £0 > .' z 0200.
xCD CL 0 4- C " 0.- 0..> .

IV ) U CV . C1 4r U C0 C 0 U .00m 4o
IxC C
rx CL 0 43 .' C 0 I3 ix C'.
.3 .0 >0. 2 ' J C
(A 0a)
(A - w .0.P) 4 M--
. 010 (A 0 . 3.
c. A A 0 0e mA .. CL4
(A (A L >-( ,
u4-4
UZ
43c. ~ 434
0 ( - 0.
(
4 0
L
C
.
+1 3.-f
a, -
3 0
L3-AL-
C
)t.-
4 C
t- Z
0
2
4
>
0-
- c>
4 4 3d1 14 4 -3
-. '-- m 0 a. 40 4i w £
>3 . 'a. IV 1.
C
2~~~ 4 . 3-4 C4 3.
03
C 0. 0. 4r00 -
0d 41 LA LQ ' I ) aCO)00 a
CC Q c) -0 0)
L1u.43 -d 3i u
4 M. 4-C
(A ( OL 0- 4j c C (a
>* 0) 0u. +1 a£ c'4
.0 0 0 ( CV >0 'A4 t3.-- 0)
U0 U 43, 0) 0 '- -IV, 3 0
"). Um0
043_. C M ) .- . X C- C Ln(1
3 A 2) C'. 0)
z4-4 .C r 0 . 0 0 U £0 E0 I' (A L 4 L
03 U) C- .. U
:3 ( £ > .
*+ C
(A C
0~
U
3 ~~~~~~~
C--'
.
L.. 4.
- 0
U
3 a%
2
0
.: a-
43
0U
C-
0
4
. 4
.0 ~ ~ C >.,- > 0. C '

0 V O ( 1' 4 C 5 1 A5 () ( 4 ( -3 (
.0 0 40-' .-.- 4300 06.- o
.--
0
(A~~~~~~~ ( ... ~ C 3
-
A
4-.
1
EC)
-.-
A ( 3
(Xx L
OC
0 ~~ ~ ~ ~ ~ 'C4 1-1-( 4.~L 3 4 04
.4
c
m)0 c ~ C '
4~ 10 AM
a CL . a;4 .4' .
C) 'U 0V
(7 0A c cm
U'; _.4) CU 4) m
0) 'A 'U ) ) '
4) C. wA , 0A I"'
z
5 4
4)D4 ) W CL.4 10 It
) m 4 ).m m C4) LM . a , M C
W-.-- 0 .4 00' 0 C X L 0)a
td 'A 0. U'- 0-
M) L. w4
c L M a. "V 0 W Cie 'a- '.0 C
4). W 'U C)4.
' 4)0
w.' L 3- L 4)
C) A 'G C. - M IL- L4. 4 I . . )

L~~~ 4I.4
'U ~ =A 'U 4 ) 4
C~~ 4)0Wl CC4) -) Q.- L4
0~ ~~~~~ -C nA 0- ' I 0
'A 3I-~~ ''A /) 0 0. Q * 03. 4)

0
C 4)4) 0 >. 4
cc
4)' ~ 4) 4)
'
.
'-V)
4
0
.
C .
' -
'A 00 m) 0) 'k' 0U 5
0.4 06 4) C4 L ca) .0
4) -. - c W.-.1"'U
4) C . 445* 'A - C0 ) '
L- .00 L 4 W)U' LA 0-: 4
-L WO. C ap Ut. CL I" L.
QU
0) U
aU
L-
)a - 3- a,'
0)4. I-
wU 00
c' ':3 rm
x M4. 0 p 4 ) 4)0 U 4- =
L U 0 4 . 4)
C f4-- ' -'U 4 , 'A
0 M) M.. .1 a4 4'
cU MC aI ) CL.
4-0. . Ci ) C 4)
w A
0. 4- C. I" m 4do - 'U x
' C - 64 m U 4
L- h m4 4
Q,' -41 4- w)
2) 4)
ai 'a 10 0 4) Cie C .4.0 0 0.
4)-C
cm a, 4-)- 0 C .4 a# ) 0 L .4IsU 9
C L ' 'v L- ' 4) "
- 'V L.4 'AL LO w)
'A ~4) ). .0
*. ~ - C
L.
O U'
A 'A
L. 5. U ,
' ~ ' 0'
4
-0
U'
4
L.
-
w) E 0 U'4 'A 4).. _). 4) 'A ' 'A 4D 'U 0CL C 4)
:3 L&
7- x
LsJ
U4".
00
0
0,.
= '
A
4)
' L
0 1-
U
L-..
0
v 00
~ f-
CL
04-4)9-T
.4
O
CL
X
w
o - 4) 0 4w~C
4) C, L~4 v -- 4)m
40 ' - 0 /
r=)
4) 0 4) '- ' 44 =
0 4-
c
0U
I-
'U
0
-04
(D
U'
)
06
-AU
.
~ 'U
4
U
''
m~
~ 3
0)
0
C4
-o
Qt
c
MUC
E'54
U'-# .0 C LLU C _x
C)CU' 'A 4) CL -V. )tO.
4) 04) -0 z ) c W)-. A C W4- L W. '
L- C. 'U 3I- 4 00' 00 '6
I&--- C 0
C 'DW.4 4)- 4) CL C 'U m) C .
4. 0 'U4. W4C --0 4-
L- -V 4) U. 'U
IV
,- . 06 0 $- M - 4) 00 IV *- 0
L- ) 4- L.,
1 0.3 IU .M X 'L 04D ) C 4)0 0 0 L
L. c' - - 4 ) 0 c ) 'A 4)4-' -4'U 4)j 'A>
ap
U'0 .4 'U C- - Lw. . L0 . 4)
>. uA i5 'U
E0 m. 4) )0 C . 'U 0
C
0 .40 0- L C3 4) CA 0 4- a' 0. 4 0 0) L U
0
C- CC
.4a50
I ~ ~ 4 ~ L
4-
)
4--
W.
-- ~ ' '0U
4 A
WU
~ ~ 4-0 w .44
4-m
r ) c --C
..
vi0)
4 L
w M).. a
.4
0
)
>
01 _r ) It 'A 21 Lo
'A 'U 'A
CL . 4) .
.00
.4
D 0 0
Ci V)-
C.~4 1--15
0) .40 >c-'
'D ; 4) - *-00 C
I 4 > 0D 0D L. 4) 3 4.-
o o CL a CO0 4- - .4
0 E4 w )
ai m + +1 CL M.4 u > C
a)
4-.c 0 C 4- - It v'
C 4) 0-.- 0a4 - 4, )C) 4)
OC~
0.C Ce 00 .'E.' ) 4 )
-4- 4 L- C . >
V~~~~ .0 4 ) - 4) 0 . C
4) L 0).10 . m~ Cl - 0 C.
En .- C
C-- -- 4) 0 - 000 0
m 4- 40
-~~ E 4-
> u D : 4) 4)
W )
CV
c. C) 44 M- 0
u~ 0 u4' L- >,'
C Z) - 0- m) Cc C ' .. M4, C 4) .0- M)
0 0 4)0. 6 C- d U
L4 to M L.4 4 . C C
x "- mL w4 - 04 4) M) _r 4- 0 1) 0 '0
It
:lA w 0E
CP 40 Z 4) cz ( 4j - )4w
It Z m 4
'A 4)>
0t 44t
'A 0 Q,4 14)0CD ( 'A 0 I 'A
Q) F4 4w0 , It CL
> -4-'-
A +1 +1 C 'A w +1 .6 4)
00 0v
Cw 0 0L 0 - 0 xA
C a C 0. c C 4 -4) 4 W) 0
4 . .- 0 0 0 W44
0- -
0 C M) s0 C C
a, -D 04) - .
0 C- 1 - -e 0 0 .j- c E0
4-> C 4-4-c-) IV kAVC4
..- oC '0 -- Z) 4M a4) 4) 4) .0 a)
S4)0.4 C 4- 'A 4 r-d 'Lv 3

CL 4)
OC 40 U' U' 041-C - ' . 4
La L-4
- 04 a) 4) 4r Cc
m~ - ) 0 4
0 0 4)
4-4 4)>- 0t--.-

o-0 c L- c- 00 C
4> +1 C 0) +1 V
4)w'A 40. 4*1 4) 0 4o
CC 50
4)0 4 m 0 0 4- z- CL
2 C 4)0
o 0 c 4-- 4) 0.34
0' 10 -- 'A 0* 0M44 0
-- '
4-ru .4 ) u) 4-'
:) 4)- 4) 4-C
a' 4C 4)
1- 3 0)
u4 ) 0 0
uA 'U
-0)..
CL -)r W C C>m m4
C- ) - 0 000 4 4
0C, > - - wC M r_ r4- ) 2
d)
f-
.
L0
)
0.-E
.
)
'
C
4
CX
4 ~ 'AL- -o
'
4>,
4 4'
C
4 -
4)0.'
- 4)
U' 0 4 -.--. .4- -' 4 C -' 4.- -'
(NJ
4))
4L
C -' C 0
4) Co >,)
LU ) (/) > 0. V (AI-- U
4.-6
C C 'AC 4- C
(m 'V C 4i 4-0 aC CA
'0 C 0 c4
L z 0 zR U A- 4 ' ) C '
4C 0a). 0 Cn
a- C) . C R.'
a0
O C
.0.. 0
*-J
0~
>
'A
'A4
C
010 4w 0
~ x 00 kA
DC-.
w
S1 M. 41 EC 0 R4 0 a>
c4L '0 -> 4 I*
-- z kA 'a &-.
- ~ .0 C..- .00. 1- - 1 41~. 00. C u
0 >. D' 0 - CU
'X .- - C 4- D j CR00
VCJ4 C 0 "0 0 WCDVR
C3 M u CA. 4-.a. . t
'V 'V C C'Vw CV
w W- DC C ' 4- c
C) c00E >, C.) C) .X 0L - c. Q'. w
~ (D 0 C-' * Li-= 4- (D 0C 0-V
=0 0 w' w aC
09 N.' u 'j - 01 z 4# L^ CA 4z LL .0 Q-4
CR
1.. C-C. C 2 4- 0 'V ~ 1-0. C C O'A
C C C
10 >.. 44- ' 4w L- L

4'A. m = : C0
CD 'V.1 c C
.6 '-3 C- 0 r-4 06. 8. 0
M 0 C 0041C ;
to M V 4 0C C7 C. -0
01~4 CR . GA CO W M~V ' W-
ow 4n C - ..- 1 LCR CR 16
ac - .0 a, L -f C-- 0. L 0 m E w
0o
'0
.. ' >
C*
.c ~ Cu
>
A c.
4wC~00
La C 'A W0
C
C''i. CIL- 01 C 'C
4- IT 1- 4- .0 00 . C
0 1"-a# 4)
ZV
CR-- CL
CR
. 4
CV
a) 10
C)
W
LO
~ - 1- 4 A
=.. .
V~ LC -
41 C
0
0 a-4-
C
~ a) 0 CD
C...Cr-
:3
4
=0 - ()
0-
w
0;
)M
3')Ir'
0
0)
x
0..
(L C--
W-
0
:
V) CR0.
'
-4
-- '-
OW '
. ~CR."- C.C..
4
.
C
.0
A ..
0-.- 0.V
'
0
- C
'V
. - R
0
1 4
~~~
LA. OCJ0 L..2 Z> 0 C C A EuJ.
41 X ~ ~ . CRw
i0 IV a, CI CR
0 1-> .0 0 ,do
R .0w C
4..- CD- "4' x
'E- . 0
C.. 4 - 0 CleL C) 0 M
LA3 0.0 cR :-b CR. 4 C
'1. -A2 C
39 m 'o 0- C1 01 "D--0C
-~~ CC 30 0 M- C 0
4 0- . L. 01 C 1 L- m
0 0~~~~u
-W'C- 0 '.- ~ C Z C R~
2m LALO '--LJ 0- oR( 0 0.
C 0--- CI C0
0 30 .0- 'V .0 0 C
.
~ ' 0
0
C
CL I- "DA. 40 -.. aCR 0

a, C>
(D -L 00 .- W LW a- 1- c- L 4
T .0 C R..-n.I .00 4 a, a, . 4 0 a) 0. Cw 0wr n
LA d u- C' X1- 4' W .0 .0 C) U ,j 1-
W, 0V4 0 - - CR - C 4 L ) E C 0
z- - >
cu > C .- 2 L C -'
-
- ~~
CR 0 'V
~ C'V
~ >
~
4'01
CQ
~
CQ-40 G
10 .-
2
O R
-
C ~.
1
V
0
4
0
I L)..- - '-C 4 44 .~ Q C 2 L V 4 - C 0
..w'. > -- cc0 1. -
-)~ C0 .
M)U
U~~ U 4 cL 4 E
ccu).'-4 4) : L'-w4) .0 L- c 0A
C a). 0p -4t -. O 31 3C 04 ..- ) m~ 8> m

- 44 - L CZ 0 L).- L.
*0 C
a; ).. 4.. cA G ~ . L
a) - c mG C7 Lo 4)
. a) c.-.4.
'D u C L c )4). 4 'o
0 a 4- 4 0- ) W U -. %A 0 L3
'A 0 L)
4w 'A 4- 1- ). 4 .
k^U " 04- C 0 L 0
0 0C L W4 4 M M, ). . ' 0 0 . CI 0 c
1 41.- - Ga0 ) 1-0 ~ - U m ) - 44

4).C 4 C U . U 4 C -4 4-04 ). L- 4 u.
c
0
:3
0 4
40.
)40 .1
M44
WC- ~ M
)
M
-E
L
.
)
o )
. L.- L. 0
44Z
~ :
0
04))
~ C
U U V C- rC LU CLC L V L'J 3 m4 x
'I CO -C- Z W4-~ m 4 W) c 06V~
m- 4D a
4) 0 C) > - f Q V. L-LQ U 4 4) ) 4 w)
*. - C z -0 x 0 M4 X~.. m 4#C 0 m L/I CC
c- S)
U) ) 0 mC-Q
Q) 4).. It u. IA wCIC
4- 0.- C M . 4w
GL- c U 0. 04-
4)0.- 0 4- -X CL It r. 0 M*4 z4. 4
r u C) U 4) D 4)..L. cu 4 L ) cu 4
M 'A
M 4)** > 0)
ai 4Ga 4 E * 0 m~ .
L4.~ L 04) U 4 *- o.CL- 4 4 4 ) ' C 0 0. - C
4)4 4-04 41 U ~ . L0 L L 0 -
~V 4
4),4.
4 ) L
a
C
cC
4)) >u
0 4-
0 4-
4)
C
CC W
e 17
) 0X
U 4e
0
- )4 L- 0 0.. 4) C
M A v 'A 4)U W0 4) 0.1
0U 9. 1 4)4 L 0
IA4 L 4.) .0 '.- C U) 0 C
Cf U a; "41. 4) ML
4)4_ 4w).QU - 43 C 4- 0 4)0 4. c
L. w4 'a04 d)) C W 4z L- . 44)
o) C U
-04)41- CL LV Cl .) L UV > L- ) a
4)~ 4- a0 44 4, Li .- L .- L 0 ). m)~
C 4 C M-E ' () L4 A 4,
L.4) w). . u =
W04 a) C Q 4)ItM . w.L %- x )4 L
L IV ar 0L 'A c)- 4
L-4 D CL >~ 0i
W n. L
C0 4? .. :) w 0
L a# c 04 CL.- 0 ) ) a%4
0>,
0- C" .0 U "IU Lo.- 1. 0 4' 4) E
m)
L
)00
C.-
) m~
x
La -0
0 :34-
L OU
'a..Z..
0 w . V 6
) M
.J 0.
3 ~40 %A.
4A00
C 0~
LV)
.
44
0
XE
44.)4U
1.
L
W) 4- I u)
U4) C 0 - C) - t 3 -- 0 M
4)- U Q) a, L U) C
4-j 4)- 4) -- ) .C Ict
o) 10. c C X 1; 04 .. L- 4) .
E) . 0.a 0 - - - M .3 .C0 L. - . >4
c4 4) = c )-. 4) 4)1 L 0 -c m. C 4.
0. L- 4)-L4>
"O G) .
qu - U U L -' ..-.. ) 4 .- CJ4
CoJ 0 .4 U 4) a, a 4 L -. 1- 0)n 06 . U)L
4) 41 IV C 4) 3.,-.K .) 'CC 4) 4 )4)- C .
4, U) L-C C 4 >U0- C U 4A C 4
0 Lp 4) 4) 0~ LIsU - 0 ) ) )
CUQ L .J U) - z)4. 4 C c A 4) c~ 4)4 G

W 4 S.)) .)u 4 W U). C-
4 )C) C. E -
*~~~ ~~ -
% w-.
U)4 )4C 44 >
- .c -
QP
C
4) 4 4' L- Ec 4m L .1
)4L 4 >4) .L U0 > U )L-.
E aC 4 ) ) 4 4w2 39 4.- 4)D E

X 0 C
7 U)
* '-
4C 4 0.
0L6U L "a .> Wi 04.) -
0 a 0 C C) F) .X L MO 044 CL 10 )4 E &A
E o 0> cm
v L LCa X: m. . 4 )' 4
I-a M).- 4
M)0.0
0 LO
t-L1 w- 0.4'-&A C 0.-. a-L
c G)
0 IV
U UT
4.4. 1.-4
0'' -V C d#>4.-
L. cO 'A . c 0. C
0)
aj 3d. ) 0 0 C-
CA 4
Mt ) .1.. A C kA
.4-0)= 3 L
00 0- 2.
ou.> E 40 CA-Z
C. 3.. .- 40 A L.
0
C.-0wo 0 0 ' E--- L.Z
0 c 9 = ~ 0 )'
0) m0.. - 4 0) C 4. 46.-
0) r- 0) .~ O .. 0
, -0. m P ' 0 'CL CL 0 0
0-
C~
r- '04 0 CLX & 4J . L E
V. -1 CL 0).. 40) a% 20 0
a U A 44 ..
fa u 0* .0Q
L.0) 0 -C
n Ca 0. , 4- L. ci c) a m 0 64 - 4-
10
L cC 61 0LO. O 0.0W44 0)0
02 41> 0)
0) 'A 2 C
t~ xC aD
,A 0*C4 .AC
k:C- A 2 . ))~C C
a; 44 A CD L. 4V0) .0 0)
wC4 4 ' A C C 4).. v)
0)' E L x 92 " CM 0. C
'C CM) (D m)0
CCC - ) c E42. - U rL
0' %D*- L- 0 IV. 4 0.> L.0
- L. E) -2 0.L 0 CA
44C .; 0 a4..2 0
0~c '. )4 )0)0)4
C c4 .1 C C-.._ -I !
O ~
C-
I
2
CC-4-
0a-
0
(V
00
.0-a
L.
c
CD4-
.
m40
I ---.
(1
0)
>,4
v
-
(Z
0 44%o
C
)0
C
CA 44 C0- 4 ).- 4 C -CU
> 0
CD..- D 1~.71 0 CL~ 0) C
~ iL 0)0
Lm 0 04
CL
C'.
w.)4
C CL00' 4#
0)i
+- .0C~..-C
C L.
wA W 0
0o 0)I 0 0 , le cm
4~~r 4 CA 0) C0~44 0 )C
cC L U- -44 CD 0 Cf 4) 06- C
0 - C4. 4- >C a - w) f.>a
L L4d
LQ C
4
L.40 E C aL
C
4
0 0 coC m
0) CA 4- 4.. >4 -0 4
C ) L. C M CX0) L 00
0~~~ L
0.i L.
0 0 L - 4 1-)
A
w.
M)
0
x . L
0)0 CL~
L
M --
4
1'0'.
C 0
)'
-0
0 0
)c ~ M
u W) -A> 0 1C. -40-. 41 20 )

)C -. 1 CA C:A M 0) >CD4C
m04 0 M) .- C'Q- .4 C-- ); - -''4.' 4
2--
- 0.- 44G:3
C
041
C
c
0C
- ta'
0)
0 0 0 )z-
E
.1
E~ r=
x
0- ~(.- . f
0z
E CL C,
0) m)0 *; 2; 0 ) A -V
17 0 Z: -0 mC -C - 0D c 41 .- M cA O. A) 1 40u-
i i 00 0 -0 wL CO m00.0 4-.
10 0~' C
3 C -
0 ) .
c C 0. .
EAC 06
0C 4)0 0 0
00CC
C ~'4 C C4
the resistance value can change appreciably from absorbed moisture.

Moisture may have two effects on the resistance value: surface mois-
ture may result in leakage paths which will lower the resistance value
or absorption of moisture into the element may increase the resist-
ance. This is a reversible process which can be counteracted by heat
from periodic operation or from high ambient temperature. A reverse
effect is also observable. Composition resistors can lose moisture
resulting in increased resistance after long periods of storage at low
humidity (Ref. 18).
Although composition resistors are characterized by large resistance

cha-ges with time, moisture, and temperature, open and short circuit
failure modes seldom occur (Ref. 14).
Generally, carbon composition resistors are the least stable resistor

type in dormant applications, but with adequate environmental protec-
tion no dormancy induced problems are anticipated. Detailed environ-
mental factors and guidelines are presented in table 4.1.1.1-1.
4.1.1.1.2 Fixed Film Resistors
Fixed film resistors usually consist of resistive material, carbon or

metal, deposited on the inside or outside of g1ass or refractory tubes
and spirally-cut to achieve specific resistance. Leads in the end of
the tubes and various types of end caps provide connection to the re-
sistance element. As with fixed composition resistors, a molded plas-
tic case provides physical strength and a moisture seal (Ref. 14).
The effect of moistuce on film resistors varies according to the manu-

facturing technique. Properties of materials used in the r sistive
film can be controlled to provide low thermal coefficients (on the
order of 50 ppm/deg C) and good moisture protection (Ref. 18).
Corrosion or electrolytic action involving impurities or surface con--

taminants can be a cause of open circuits in the film ot between the
film and end cap connections. Reduced resistance from this effect
prior to malfunction can be hard to detect because of the common
localized nature of the effect. Moisture absorbed during storage does
not cause serious trouble until after a period of operation with volt-
age applied to stimulate electrolysis.
Manufacturing defects have been found to be the primary cause of fail-

ures for nonoperating film resistors. Increased resistance, decline
in insulation resistance, and eventual opens from defective seals,
swelling of binder from moisture, corrosion of terminal connections
and oxidation of metals have been found to be the most common failure
modes for film resistors in the dormant state. Tighter screening and
4.1-20
process control can be used to identify latent manufacturing defects

which may induce these failures in dormant fixed film resistors.
MIL-R-55342 are fixed film chip established reliability resistors.

They are primarily intended for incorporation into hybrid micro-
circuits. They are uncore, leadless chip devices which exhibit high
stability with respect to time even under severe environmental condi-
tions. These resistors are essentially unaffected by moisture. Unoer
low humidity conditions these resistors may be prone to sudden siani-
ficant changes in resistance (usually reduction in value), and signi-
ficant changes in temperature coefficient of resistance as a result of
static discharge during handling, packaging, or shipment. Substitu-
tion of more suitable implements and materials can help minimize this
problem. For example, use of cotton gloves, static eliminator de-
vices, humidifiers, and operator and work bench grounding systems can
reduce static buildup during handling. Means of alleviating static
problems during shipment include elimination of loose packaging of
resistors and use of metal foil and anti-static (partly conducting)
plastic packaging materials. A primary failure mode for chip resis-
tors is package fatigue from temperature cycling. The designer must
be sure to match the temperature coefficient of chip resistors with
the substrate to preclude stress fatigue failures from temperature
cycling during dormant storage.
Film resistors are more stable than composition resistors and approach
may of the good characteristics of the larger and more expensive
wire-wound types. Because of their stability, even under extreme
environmental conditions, fixed film resistors are recommended for
dormant applications. Environmental factors and guidelines for film
resistors are presented in greater detail in table 4.1.1.1-i.
4.1.1.1.3 Fixed Wire-Wound Resistors
Wire-wound resistors are formed by winding a special alloy resistance

wire on a ceramic form having expansion coefficients matched to that
of the wire. By selecting and matching the resistance wire, almost
any temperature coefficient of resistance can be obtained. Precision
wire-wound resistors provide closer tolerance and better resistance
stability that any other type of resistor. These resistors are gener-
ally well sealed in molded cases for use in high humidity environments.
The reliability characteristics of precision wire-wound resistors vary

considerably with the resistance range. Low values of resistance can
be formed with single layer windings for minimum inductance and other
optimized characteristics. High resistance values require multiple
layers with resultant problems of shorts between layers, hiqher mutual
inductance, and physical disadvantages such as size and weight. Very
4.1-21
low resistance values can also cause reliability problems because of

the practical problems of close tolerances on short lengths of resist-
ance wire. Specification limits on tolerance and resistance should be
carefully observed.
Power wire-wound resistors are similar in construction to precision

wire-wound types but less attention is given to close tolerances and
noninduct~ve winding. Greater attention is given to the means of
mounting for the extraction of heat. Special silicone coatings are
designed for maximum heat conduction and radiation. As with other
wire-wound resistors, reliability is related to resistance range.
Used to obtain high resistance, multiple layers of wire generate hot
spots which are difficult to cool.
Moisture in wire-wound resistors is frequently a cause for leakage

between turns and between layers which ultimately results in insula-
tion breakdown and shorts. Corrosion and electrolytic action results
in openings between resistor wire and end cap terminations (Ref. 15).
Opens can be caused in wire-wound resistors by defective wire and/or

terminations. Opens can also be caused by corrosion resulting from
moisture entering the resistor through poor seals or porous coatings.
(Ref. 20)
Higher value wire-wound resistors are subject to mechanical damage

resulting from vibration, shock, and pressure. This is because of the
construction of the wire-wound and its size. Wire-wounds are typi-
cally constructed with a Plastic or ceramic bobbiu with multiple
layers of wire in the larger sizes. The fragile bobbin and extra
weight may result in damage due to the mechanical factors.
Generally, wire-wounds are the most stable resistor with respect to

environments and time. Failures in the dormant mode are usually due
to manufacturing defects. Tighter screening and process control can
be used to identify latent manufacturing defects which may induce
failures in dormant fixed wire-wound resistors. Because of their
stability, even under extreme environmental conditions, fixed wire-
wound resistors are preferred for dormant applications. Environmental
factors and guidelines for wire-wound resistors are presented in
greater detail in table 4.1.1.1-1.
4.1.1.2 Vatiable Resistors
Generally, variable resistors are highly unstable devices, and they

are not recommended for dormant applications. Tolerances of variable
resistors reflect their instability. Periodic inspection is required
.1-22
to ensure that variable resistors are maintained within their resist-

ance values. This will impact the maintenance concept selected for
the dormant equipment. Refer to section 4.3 (Maintainability Design)
of this text for a more detailed discussion of maintainability design
guidelines for dormant equipment.
Environmental degradation is a significant failure mode for variable

resistors. Variable resistors are highly susceptible to moisture,
temperature, shock, vibration, and contaminant induced failures.
Variable resistors cannot be sealed in a complete encapsulated
jacket. Even when the resistor element is encased in a plastic or
vitreous case there must be a portion of each turn exposed for contact
with the wiper arm. This provides many possible points for the en-
trance of moisture and contaminants.
Variable resistors have movable shafts which protrude through the case
and possibly through the front panel of equipment. This opens the
interior of the variable resistor and Possibly the interior of equip-
ment to the environment exterior. Various types of shaft seals, such
as elastomer '0' rings can be used to protect the resistor interior,
but they are imperfect seals (Ref. 18).
Dormant variable resistors should not be lubricated with a liquid

lubricant, because the presence of oil lubricants can collect dust and
other potentially corrosive and degradative contaminants.
When the use of variable resistors is required, they should be main-

tained in an enclosed unit to protect the interior of the resistor
mechanisms from degrading environmental influences. The following
subsections present design guidelines for variable resistors when they
are absolutely required in dormant equipment. The subsections have
been divided by precision type of the resistor.
Table 4.1.1.2-1 summarizes environmental foctors and effects, and

presents guidelines for variable resistors in a dormant mode.
4.1.1.2.1 Low Precision Variable Composition Resistors
These resistors are constructed in a similar fashion to fixed composi-

tion resistors. Many tapers can be obtained, but nonlinear conformity
and stability characteristics are the least stable of all variable
resistors with respect to time. These resistors have been known to
have an average resistance change of +20 percent per year under ambi-
ent storage conditions. (Ref 13) The designer must insure that a
design is capable of tolerating this significant resistance change if
the design will spend significant periods in storage.
4."-23
lea >61 f" a

L - #V. M 0
4.T. U,,- v ah :3 L.- 0
0.) 4 0 CL 0~ - u
CLQ 6. 0 VA =04 (D
60 w 40.641.z 0 LA ) Im 0
+1
0%
.' ~ 0
--
0
3 C
0
+1-
0%l
. C
0
'.- C ,'f L6 4-C C.-
0., 0 0%7 A 0.- 6 L 4)'
L) cm I- I =0 CL"
m 0. 4-
o
wa
~~ 4
0~
Ua U,
4
6
c
m~ Q CL
a- 0 0
O
~ Q
L.
*.- 4 0OL - 0.0 o 06 4) 0 -. j 0

U,
C
6- ~ C
4
.-
CD~
0% 4) C -'
-)C.-ac "
.6- aC M
0~6
) -
4)
C,
U
C
cm
c *..
a). z
G4 L..
4)
M
c.6
o 0
3~ ~ 0 . 4a'' E .CC c .~ .n0 3 ,
4)
~
U
(2)
~ ~Ud) L
04)
C
O..'
I
~CLE-
0
-
1*'OU.0
4.-
w-U
.1.6.
M
.. 0
--0
4 U,
.
C)
U
a
)
L 0
0
C6U'
4)
41.
04) xd C)
0: Dc 6) 0 0 00 CL0 L or I a- 'aiLw
.0 4 )0
;> 6d
- IV1 cu
6w *- %&. I6 .
3A 06 0 M0 O v Q
m 0 E6.0 4- - 4 00,
+1 a,4) C .- 0 .- c
0 0 m% 0 +1 1; 41-
.-
C3 C
c4 -U'0 . 4-C U
0..- .4 0 U 0. 41- W)
r-
Ci
.6.
.41
Lt--
~0 r-
L) C
u,
.0-
0.
0.
0
4).
0%6
C
06-)
d0
0
40 L- 0 0)6 I
r-0 c).- ~ Itd
6- . 0 6-
.6- C 0 - U' 4. C1 " 3 c0C

- .- 40 W. -W . 4)0
'd
.0 U Ce~ -M 0 U C*- 10 0
w 40 E d C' 4 -0 M2 %-.- 4
4) w . L- 2 n 0 V x
u, E. 3L 0 M EL 06
o ~ (~E *t CL 'o . . o0 0.0 0 .0 4)u .

c- m) 0) 07 X a 0%
4- 0) -, -) a , n
4~_ 0 .- d. )c
.41
4 )4) -10 - -r.6 44) 1"C 0

C3 3a 0. 0L m
C~'44-5264) -O
0 4 4 0 ) IV) 0 c6 I -0
4).- 0a 0 000
C~4 0. U, ,0 AP06 4 4
0% W) c T r I*E
U,
w a 40 >~4 . 0 40 . -. 6-
t'.j. 4)6 .C 0 64 -. 0 14 - L.- 4
6OL I U> 6 C UGO
-04 CD 0 U 4 0 C LA CL3.
0 w)> m) 0.0 4-: 0 0
V -
4 46 C CL C0 04 4)C4))) L.-
0 U') L C 6- 4. IV',6.) . U
4) 0 CL, M )40 ' , C A , 0 0 U
I. i GO IE 0. O . .0 . .t )-
0 o
C cu
4.1 2
0)d 000
0
T) - C >~C
0) C L0
1- ' M- U 0
'A 4,
CL
G)' W0-I~
a, C M) 0.
01' -
M- .0
a,3 UD
4 - A A IA
C qu '
'0 '~ 'A 4
0 C W) 'A a0 Ai . AC0 ) a) - E
C 0, *- 1
' a01- 6-.'' 1- 1. 0
4 -4
cu0
4) C 0) w, - 0.'A.' E '
C C U ' j0.-
'0 4. 0
(.. 0 A 1-0
'A~~ W. C m)-
Z CL--4
'0 0 .0 -
-0 '0 0A
1C 3W
,-- 0 00'
0
., 0 0
4) CL -
L., C. mA
> 0
12 '0 %, 0). 0 C W
6. A 0
'A
06 CP
'0
C
0.
'0U0L4'C ~ '00
004)
U4-
'00.
'0'
W
0)~~ - ~ -
0
4
4-
4 '-
C)
'
0 ,44
.0
0
-I C Go
'Ac' .0). : .u .0e

t- J- al
ex 4'A
'0~'0 )4 0,
0 w0C
4- 4 ) 4-- C 0 ,-0
- 0' -- 0
0,0 'A
A CA
CL CL 'o o0 C ) -
) zz
o M-)'
0 C >,
4 ai ' 0
' o -. - 'A A
WC
6.,) 1 '
0)'~
-)~
~~
'' m'A'A- ~ '4 ~
0- '~ .
L '0 C '0 A - *0. DC
0 - ) ' C0
E ) CL 0 n,'
C) -w)0
.4 A' .-
4C '05C>
C .0
0 ~ ~~
10) 'A 4)4
A - 4' A)L-0 CA .. L
'0 ) 4A )
Q' U) %- 'A M 0U
1
0
C
0 ~ 4)
-
- ='- =A '0)
0. 0)C ,0
.. ,04)
A , 0 C '0E~' 1 2
C Ck) a) A '0a;..'
t6U4 0 3It'
*
3
) - )
-.
C w) 4)-0
'01) ) ~C o' A t.4'
1 0 0
4 3 0 -- 5 w04'
4-
CQ)
(%J~~~~~
W0 'A
0)-
1- :' '0
'A ' 0 ~~~
4 ' C' -.
cl .0 00' - ~ ' A - 0-' 0 c ~ etU
4
0)
c Q-'
.0
1-4)
'A 4-.C t
0- M- A -n r- . C
%- 4- C~ 0-c
0*~~ *4 -- LD- w 1
CGD W- cG D a.
10 0 0 )
41 - 1- 0y 0..
-~~-C) GDG *4 G L.
CLG 4. . D.' G uD 1-.0 G
'U
4' .
-vE~ ~ 0'
)t c. to 'U
>D 0. C
a '1-
IV
CEA
0U
%-
GD w CL- 4.
oD
a
0
'A U
~ .
'U
D
10~.. QE .
4)
-
>.0'
- E.1
r'
*4u
C
xU
0
0'A4D
4 ~ ap
D
W*~.
o
GDI 16G
G D G
u*
D
4)
G
0
D'
- >~
0 V. '.'C C ' D 0.' 0
1-Z G D G U C4 A cm- D C
I C
- ~ ~~
0 -. ~ .... ~ 4' ~ 3
.. ~0 ~ ~ ~ . ~ a l~G
0
4.'d
'A 4.' 0 C I c
4D 4- -W C 0'A 0 -
W0. C-...'
0 GD 4v 00 C
aJ kA U CL.-
G It'A*
-= 4D 0 - CoU~ -
.0 - 0.. GD IV 4 A I- 4 DI
'4- mE-* le uD GD
Oj
'a 1 0)
CL r_ WA . 4- 0
4) GD GD
a. ' km 0. 0
It
~Cc
'A. 41 .CA- GD+1 41 G + ~41-
*0 *- 00 qv 1c - 0 UC.
0' Wv
WD GD .0 C
O
O
GD 0
' '
_
Cr
0-.*
' 4w.
-1 U
+1
C .
+1 u9'
. j- D
0 +1 ~ >I
C >
w 0. *
CA. GD
_ c 0 -*4 U -4-4 0
- 4 >4- 3.. 4 - 'd 4 UUC
c 4-G . 0U -C- C3 GD G>-GD L G D
0.. 1-* Ci C -
t.~- 0 GD 0.' 0
L m0 L 0 -. 4m' (M- D0 % GD 4AC~
C3 C CL 41- 'U U c U A - CL ItU0
'A GD
M 1-G vn GD .-I" .- L 4 G -' :I
- 'A 0'* GD '- 4 C4
A 'r- C * GP A 4.
GDi *.- O x GD ' D0 0 - 0 0 0U C C " G
'Ai 4.)0
*4 4 CA C C '
U~~ 40C *
4-~ 'Alo--U G
1- %..
GD o- 1-CO4 Ci UU z
E Cc-'.
-
L~J -' - 0 - - .4. GD -
0D
~ .
U
C
~~ .
G
L..
-1
GD C 0 Cw- 4 m.
4 CO G - G
2 j -') L'- >- 1- In1 U C .
C.~ 4 C G -GD D C4.1-'A6

~> 6I
%.-
'A
W
M4
.C
-'
0u u
0
Ca
06 0 C C . - >
do-C .'A 4 C -. ' 0 4 0. 4
30 >
0.
MA M C- C 0 C aW Q1 4- -- 0LJ L- *L
C5 C: L 00
on CD .0 C
0 7;
~ K0
.4 '0 . 0 .-
o -'
3'C- V 4 to z -C CO 'D W00 A ' D

0 1= I 't -. L- WO)'O 0 C ' L- 0Co C
0 e L- 0'. .0 - W I" V4 0A C >,'. 1. M44
4)4-L U0 IT4 0
ow L- -C 4 L. M,- W
Vo ao 1 .1. 00. > > L- 41 C. C M E0 L -
0 "a IZ L0 0o C
M., W~ EZ 0
'a 0j M- m 4 E 0 0
C -0 > C L- 0G A - v c.4
c C m. mL IU X4 _u CL C 2, LV '-)> O wA
0 03 E.20 0 0 0 *-L' -- 0
Q 0.
L) 0. 0 0 .- ' 4 4'a C 0 3-- LW LV CA 0l
a)- 0 w
.3
.2
L 'a
I
4-
1A
I
~ ~ *.- 0
~0 L
-0
~ 0d) '
V
C C
'VI.
0--
- 44-'-
0.V
0
00* w
A
1--'
0
'A
'
0 u
'a40
C
o
cO
LV
E"
'D C
L
Od
a) Cu 4- 0
w 0 4) ix. OCL uV 41 L- 0 0 >
0 It000 0 - 0 ao 04 x OZ 00 C rA 'a-
3 . OV 1 .2 5 z 0I I V) (A~ ac UJ SLL'.0 UL) Q'- 0 O0-.'.
a).
'A >eA
0 - a C eA4 0 C U U
4... M c' W- c. 0 0.- 0
4) C-' a 0 C -.- "0 .
0 0-C 41 : UA CA A O..4 Q). 4)
1- 0, 0 C 'a
m - r A.C4C 0 'A 00 44..
L. - . w 0 L L a4 ;11 0
4) Ln (LI
A . 4 0 wO L- V. 0- - M - 0 0op-
4 o ''0 L- LL 0. Z c 'a 0- 1- >
03 'A0 L 0 0 C. c' 31 w.0. m w4
0 C a o 4"
w' -0 L 00 -U- Ccn a) C u.0
u L- .0 -j 0o 'a
'A 44 I. CW W
00 0 t. "1- L8 : c. L 4,- 4A
E0 O .00 A. u L 00C m L~0L.
00 04.. L-C) W, CU Q E w. wA
1 Vo
z4O V-4o0 0) -i C 0- WA 0- 'a O
C 0-- *-e'
M~4 C- CA 0. t- ,m. 06 0 0
00
w cc44
C C
W L4
C
WA
L,
L.
0
(,- c.
40 ~ u
46
L-->
C.. qDC C4
0A-CO 0
60)0 (n D D . 0 )t x 0 X ( L.- C.- 'o40
-0 :lc u4 0.- 4) (A0V 3. L W L 0 0M - 0 '
0~~ - ~~~~~~~~Ca'4 '.' - - 0 0 0 V 4 'A. 0 O -

.4- - 00 0-CLO 0 C CL 0C 06
0 0 0 0c4 0o 0 04 O 00 (U0 C 'AL C -
V - CjC 5Z0 EA . WA L-60 LL~-. E
'- 10. 4
mA a0 .- 'A
0 cI 4u 0--
4-' 14 m c a)CV C
'A4 0 004 P I - Q
Oi 0 K 1:- 1
0 z fCCC 'aLa 0) 0
4-) 0- 4 'A 'AD 4 C O> C- W4a 0.-
0.4 --? C (7 ' 00 0C- 44
CL
n 4 L -w' OL '
-4 .. I L- C I m0. Z A
0.. 0
4 .0 3~ao '06 >C C 0 C 0 V. 0-- 0(w La.
0
> 0 4) C 0J - Z W C
C uAA -- c- c 0. J '0 L..CW-CW
C CCI M.0 M L CL 01 1 ' LC C

0 ,- 0 6- 0''L). 0 'A 44 0(144 .-
40 -0
'a 4AC V- V3 - C, 4u 4)
LI 2 AC O 0 C
-&. n 4 0 L 0 L. &A CL.' M 0
*- -C L) I 44 .j 0-C 2 LV G- >w 'A0- Oc-
-j 60 >. 0)
w.. .' a vC L- 0 >v.
IP
' -0 C4 :5 'A 0 -'. LWax LW 0.- C0
0 0600D
'Aj w~ _r_~. '- z V ( V) CE 0.-V
La 0- Q 0) 0W MV
L--
0t
'A 4..
t- CP 0
>-
C11 V'Y-0
'-'k: V) w
0 040 1-60
t4- m
qu rA I IE C
)tL )I 'A c
0 1 31 1 C 04 4ez-
-A W U 6U d)
o ~~ L-
3. ~ L)
~ a-
U' 4 -4 ~ xC .3 ~ > ~ ~
c
0 ~ C 0M- 4- C o ~ U 4
co0
C1 CA oLr .. 4 C~ M1 LU w CD041'
d.) GD *- 0 L--.I ' WIA 4). 4- ) It..
' 1
4-0U
'
'.-C ~
4- I
~ >0.-
*-
~ 0 L16
.C
1
W4
' '
0U- I L-
0 E
'U a 4 0
c a ie >% -
%A LD- 4- 4-I C C- 0
C-0LU~ 0 C) IA 0 n 4 -0
C D - - 0
10L (C
0) 4)4. 'U
4-:
EA
GD
G 0
IA'~
4
GD
E .
G4
2) U
~GC0C V' Gi. M -
M'
U
E0
..
CUC E-
) xAL
C L L C ' ) LJ CLV >0 CA CL. a)-CCCL L W. C~ 0
0.~
G-D C
I C '
IA
)
L ~
D G U G
0
C
16 -;' ))
4-
4. A- I- IA
A
MU
0
L
.
dL 0004 - 0 4-G 4- 06 -.- 0 c ) I 0
>0
IA X IAW'Ac
GD IA LL - 4-
-0 WC W CU 00li
3o ~ ~ ~ 4- ' Iu 0.-C)- -
GD~~4c
C- 'U IA- )L C 0. 4
-cU cu -v a 004 - - 0.0M
L. IA 41....
z) L3 _r_ IA - CE U I 0) - '
00 w U j - IAc ~ 0 4 a .
4-C C4e 'U4 > 4 CL

0 'U
> -
4-~~ Cm . 4 0 a.D 'o 0
0)-.-,--
IA 0 .A4- w -
X: M
A0 I 'U 0 4 0 10Ct 4). m. '
0j 'D L. 0 C WA, C IA GD r4 uj -00
's'
0a 00) 4- A L E - ' D 1 '
IAJ IA4 GDc- - 4 3 ' . ) - 0 M)
2
4-*. 004 M .6JU -- 0 -V
IA UI/ 'U ME. 00
00m
Ido - r- cr4 -
-0 0 W Q
A 0) 0. IA
M 0) :3 'o I' " c w

30 -
CD 4- M
07 a# .. 41
3 :0 . 0 'U- > CL ) 0 4
4-~~~~L
3,--Z'
00)~~~
4 0
4
'U
U0
C0D)D.
n L-
0)
~ I
''
M44
U
C, 4-1U 4 A "- 0 4 L) : 0 m c) 41-C

L
GDJ C0 0; C 4 GD C 41 > 0
C c04 0)-. IA L 0 .I
t C 16
410 1- 'DU 01 C2 0) LUS C ." ) U
C A 0) -
L I. 4- C
-04.
IC
0 -..
E '
) IA)0
zU D-3 C -4.- 400 'o-00. 00 '
C A- 4.c 1' )- - U> 0C
0 10(-7A L U 4 L - G U '
c L L C ' 1) .L I 1 A L -U '
4-' U ' -I - L. 0 -L C>
I I --4-GDCE -0413'4128IO 'U -4

- I I
0 k (- 0 C
w W
0 4 .C, 0 CD W
U.
3.4 00 0 "0-
E -r- .1- 00-.0 0 U

w 0 L- I .
C:, 2M -
$- 2 4L U . L- CL a
41 Om CL . a 0 0.
Eal 04 C0 40
L L. 4 0) 4- M
.
-,
4C47.~
~ 0. Lii
CL % w
W4.
c . v.-
CD5 z 01 0
c cM V4-L
0 al4 U.
j.
0 W
>4
~ l 0---
tLm1
-.. 41 LC 4W0 .0D4- a C 4 -L

Dc L. 4 0 L-
CL- CL 0 10 - >
a.) 0 4 1 Ai0. O -G
I ~ .4.1 U
'A. Lu e
UW
C-
a 01-0
0)
4) .- i~ 4
0>. -
U. - ~ C
vi
GL )0 C
'a E1
C
m.
-. Lii
m.
L
~ C-3 O 5- 004 4C GI 10
0 D C1
06 W
16 C C ULc.
>) .1 0, 04.
01 - 0. 4.~ 0
4 ~ ~0
..
U. CC 4
10
U 0 L 4
).
7 0 E
0.-
w~ .0
0 20 C 0E CDWC
4 ~ 1 ~~~~
0.1L0
LO
L. 0E
4)C..1
m
L 0
L x
(on 0 x 0. 0
01 ~ ~~ Wu 04,
a
C4
4). L .14-
* .. C0 U .0-vL
0 m 4) + )-
E
0. 4 0 C
c0 . 1 U0%0
c 7 01
0 ZEU0
IV W- C 04
M1 > U 0 W
0 0 U. L . S
0
C 0 0'U 0C U
4)2 )
W L4 U. 01 C 06 f0 1 4
1 U. 4) -. U
4> 0..- 4L ) 0a.
A ap 'A 0 'A 1 C ~ 0
U. 4m
w. c. 441 > I D
C L- w ) M-. 12 00> 5
41 0 _r0
C U.001w05
r 5
C C cu 0'- L- 04% 4)
0 c U.
5 U. 0 L.c a
.' 4 .
U) 0- 04 L w.J 0) . .U
Lii~~~~~L*CL.~ .. 4) 01 ..-
0 ~ ~~~~ .0.
44 ~0 M1C00
U. x 4 0 .110
fu (x. U.0
E 21 ). -. .0 CU4 U
* 004..) -4)1) O 2404.4.1-29-

The maximum operating temperature for these devices is 120 deg. C, and
the maximum storage temperature is 70 degrees C.
Purchase tolerance is typically 10% for these devices.
4.1.1.2.2 S-emi-Precision Variable Resistors
These are wire-wound, medium power variable resistors. The bodies and
cores of RA style devices are constructed of phenolic or other plastic
which may char or distort at high heat or crack under thermal shock.
The RA style is moderately hygroscopic and thus more susceptible to
moisture than either the RR or RP styles.
Purchase tolerance is typically 10% for these devices. These devices

are more stable than the low precision variable resistors, but they
are not recommended for application in a dormant environment, especi-
ally moist environments.
4.1.1.2.3 Precision Variable Resistors
There are two types of precision variable resistors available. A

wire-wound, which has better stability and a non-wire-wound type.
Both types offer a reasonable stability with respect to time for vari-
able resistors. The wire-wound is typically available with a purchase
tolerance of 5%, while the non-wire-wound is 10%. The wire-wound
offers greater stability with respect to time in a dormant mode, but
both types are the best suited variable resistors for a dormant
environment.
4.1.1.2.4 Variable Non-Wire-Wound Trimmer Resistors
These trimmer resistors come in several different non-wire-wound con-

struction types and styles. The stability with respect to time is
comparable to corresponding wire-wound trimmers, but the non-wire-
wounds are generally less expensive than the wirewounds. These
trimmers are typically available with a purchase tolerance of 10%.
4.1.1.2.5 Variable Wire-Wound Trimmer Resistors
Wire-wound trimmer resistors offer good stability with respect to time

in the dormant mode. These trimmers are very similar to non-wirewound
trimmer resistors, but they are more expensive. They are typically
available with a purchase tolerance of 10%.
4.il-]0
4.1.1.3 Special Resistors
Two types of special resistors are considered in this section: Fixed

Film Network resistors and thermistors. The fixed film network rcsis-
tors are similar in construction to fixed film resistors, except that
they are in a network configuration and arc Zackaged in a dual-in-line
package or flat pack configuration. Thermistors are resistors whose
resistive elements are sensitive to temperature.
4.1.1-.3.1 Fixed Film Network Resistors
These resistors are in a resistor network configuration having a film

resistance element and are packaged in a dual-in-line package or flat
pack configuration. These resistors are stable with respect to time,
temperature, and humidity. Maximum ambient temperature for these de-
vices in storage is 125 degrees C. Network resistors typically offer
a very tight purchase tolerance of + .5%. Temperatures above the
rated full load and storage maximums will cause temporary changes in
resistance.
These resistors are particularly applicable to dormant environments,

but only if extreme temperature changes above or below the rated maxi-
mums are not anticipated. Network resistors offer good stability,
long life, and accuracy. They are particularly desirable where minia-
t-UiZaLiC; 5 iimportant, and where a number of the saime resistor
values are used in a circuit. Using networks in dormant circuits,
insteid of a configuration of several of the same value resistors,
will reduce the possibility of moisture and contaminant buildup be-
tween resistors.
Network film resistcrs are similar in construction to fixed film re-

sistors, and they experience the same environmental degradations. See
table 4.1.1.1-1 for environmental effects/guidelines for fixed film
resistors.
4.1.1.3.2 Thermistors
Thermistors are a mixture of metal oxides fired at high temperatures.

Thermistors are manufactured in three distinct forms: beads, disks or
washers and rods. The metal oxides are mixed in proportions to pro-
vide the required specific resistance and temperature coefficient for
the particular application. The final product is a hard ceraric mate-
rial which can be mounted in a variety of ways depending upon the
mechanical, thermal, and electrical requirements.
4.1-31
Measurements of electrical parameters which describe thermal charac-

teristics of the thermistor are difficult to perform accurately be-
cause of the element's complex construction, physical shape, sensiti-
vities to ambient temperature, and direct heating effects of measure-
ment currents. Such parameters as current-voltage relations as a
function of temperature, electrical resistance at a fixed temperature,
presence of hysterisis during temperature cycling, semiconductor
indications, and changes in material constants are frequently measured
to point out weaknesses caused by some adverse envirormental condition.
4.1.2 _ tgrs
Capacitors are functionally classified as fixed or variable. These

functional classifications are further subdivided by construction type
as follows:
Fixed
Paper and/or plastic

Mica or glass
Ceramic
Electrolytic
Vriabl__l
Ceramic
Air
These basic types differ from each other in size, cost, capacitance,
and general characteristics. Some are better than others f r a parti-
cular application while no one type has all of the best characteris-
tics. The choice among them, therefore, depends on the requirements
(both initial and long term), the environment in which they must
exist, and other factors. The designer should refer to MIL-STD-198E
(Ref 22) and the applicable military specifications for a more de-
tailed description of the various types of capacitors and their appro-
priate uses. The following sections present a more detailed examina-
tion of dormancy effects and guidelines for capacitors in various en-
vironments. Section 4.1.2.1 and ubsections present a more detailed
examination of fixed capacitor types, and section 4.1.2.2 presents
variable capacitor types.
Generally, capacitors are stable devices in dormant applications, with

several notable exceptions. The most notable exception is variable
capacitors. Variable capacitors are highly susceptible to environ-
mental induced degradations, and are therefore not recommended for
dormant applications. Electrolytic capacitors, both liquid and solid
4.1-32
types experience definite dormancy induced degradations, and use of

these devices in dormant applications should be controlled.
Foi other capacitor types, dormancy is not a significant degradation

factor, however, dormant capacitors may degrade due to environmental
factors experienced during the dormant period.
As with resistors, temperature and moisture are the primary degrading

environments for capacitors. Typically, a dormant capacitor is not
subjected to the high temperatures experiE-ced by an operating capaci-
tor. Low temperatare can result in significant capacitance changes
for capacitors, whether they are operating or dormant. For low and
high temperature, the designer should consult MIL-STD-198E (Ref 22)
ava the applicable MIL specs to determine capacitance deviations.
Capacitors are susceptible to moisture induced degra'ations. Even, in

hermetically sealed units, moisture present during manufacture can
lead to deterioration of insulation or dielectric materials. This can
be a more serious consileration In ceitain lower grad devices. The
entrance of moisture through cracks in the seals can be minimized i
several ways. Capacitors wi'h seal cracks prior to installation in
equipment should be screened out and removed from manufacturing
stock. Cracks developed during assembly into equipment can be pre-
vented by careful process control and can be screened out by final
assembly inspection. Cracks which develop during use in later life of
the equipment can be traced to low-quality seals or stresses placed on
the leads during equipment manufacture. Certain seal cracks are
treceable to a combination of these causes plus 3tress resulting from
shock and vibrat'on, pressure, and other environments. Environmental
effects and guidelines for dormant capacitors are presented in greater
detail in tablt 4.1.2.1-1.
Listed below are several general capacitor selection,control guide-

lines that are applicable to both operaLional and dormant capacitors.
Subsequent subsections piesent mo-e detailed pait selection/control
guidelines for che various capacitor types in dormant applications.
" SElect a capacitor for each application from the list of standard
types and values. Established Reliability capacitors have tighter
reliability screening, procLss control, and failure proof from
cumulative life test data, and are therefore recommended for dor-
mant applications whenever applicable.
" Various initial (purchase' tolerances are available depending of

the type of capacitor. The desicner should realize that the ini-
tial accuracies become meaningless if the inhererc stability of
4. 1- 13
the capacitor is not consistent with the basic accuracy require-

ments of the application. The capacitance tolerance that the
designer uses in order to design a circuit which will operate
satisfactorily for the required time requires consideration of:
(1) acceptable tolerances according to specification; (2) capaci-
tance-temperature characteristics; (3) capacitance-voltage charac-
teristics; (4) retrace characteristics; (5) capacitance-frequency
characteristics; (6) dielectric absorption; (7) capacitance as a
function of pressure, vibration, and shock; and (8) capacitor
aging in the circuit during storage.
o During dormancy, as weil as during operation, any characteristic

(i.e., capacitance, capacitance-frequency characteristics, size,
etc.) of any capacitor may change value as a result of stresses
caused by environmental effects. Changes caused by environmental
stresses may bc iinear or non-linear, reversible or non-reversible
(permanent) or combinations thereof. For dormant applications,
especially iL extreme environmental conditions, the designer
should consider designing for an end of life tolerance with a
"buffer tolerance" added to the purchase tolerance to insure long
life reliability.
* The use of the self-healing properties of certain types of capaci-

tors may not be desirable in circuits where interwittent failures
and noise would be troublesome. Self-healing characteristics
should be closely monitored in order to insure long life
reliability.
e Oil or liquid filled capacitors should not be subjected to severe

mechanical stresses (i.e., shock and vibration, low pressure,
etc.). Leakage of the fluid can destroy the capacitor js well as
adjacent components.
o Moisture in the capacitcr dielectric will decrease the dielectric

strength, life, and in.,ulation resistance, and increase the power
factor of the capacitor. In general, capacitors which operate in
high humidities should be hermetically sealed. The effect of
moisture on pressure contacts which are not gas tight may result
in a high resistance or open contact.
e Capacitors which are closely mounted together in circuits can

cause moisture and dirt traps. Moisture and dirt traps eventually
form corrosive materials which can deteriorate capacitors and
other electronic parts. Moisture can accumulate around dirt even
in an atmosphere of normal humidity. Normal humidity in a dormant
environment typically ranges from 40 to 70% relative humidity.
Moisture (humidity, rain, snow, etc.) is the greatest enemy of
4.1-34
dormant capacitors. If dormant equipment is suhiected to humid

conditions it must be protected with adequate moisture protection,
such as coatings. The designer should design dormant equipment
with hermetically sealed components and conformally coated circuit
boards in order to protect the equipment from moisture induced
degradation. A more detailed examination of coatings, preserva-
tives and packaging of dormant equipment is presented in section
4.5.
4.1.2.1 General Characteristics of Fixed Capacitors
Generally, fixud capacitors are very stable components in either dor-

mant or operational configurations. The only exceptions are electro-
lytic capacitors, which have been known to experience definite dor-
mancy induced degradations. The self-healing properties of electro-
lytic devices will compensate for dormancy induced degradations during
short dormancy periods (( 4 years). However, even during short non-
operational periods these devices require periodic monitoring to in-
sure reliable operation.
The following sections present a more detailed examination of the

various types of fixed capacitors. Table 4.1.2.1-1 s'mmarizes en-
vironmental factors and effects, and presents guidelines for fixed
capacitors in a dormant mode.
4.1.2.1.1 Fixed Paper and Plastic Film Dielectric Capacitors
Capacitors with paper, paper/plastic, or plastic dielectrics are made

by interleaving thin films of dielectric material with metallic foils
which serve as electrodes. The resulting four-layer wedge is spiral
wound into a tight cylindrical roll. Leads are attached to this capa-
citor sectinn by soldering or welding. This capacitor type includes a
wide variety of dielectric systems, styles, voltage ratings and temp-
erature characteristics. Continuing developments in dielectric mate-
rials, construction techniques and manufacturing processes have re-
sulted in a highly versatile family of devices. As a group they have
high insulation resistance, good stability, low dielectric absorption
and low loss factor over wide temperature ranges.
Metallized paper capacitors have low insulation resistance and are

prone to dielectric breakdown. Plastic dielectric capacitors have
superior moisture zharacteristics in that they are non-absorbent. All
units should be hermetically sealed. Small amounts of moisture can
increase the rate of chemical reactions within the caPacitor materials.
For metallized paper and plastic capacitors, where the conducting

plates have thicknesses in the micrometer or submicrometer range, a
4.1-35
eu4) - a
I I
UA )
Q)+ 4) c -
4 41.... C) U
%A QJ
4) 4) 0M 0 -6- M
'
L 6 . CL- 4).. 0
0 M
0'U .0.4) r! 41-61 4)6f- -
0 =6 (A ''- %-
U) 0c 'a
4 C 4)6 6- ;f>U 4.) m L4- 4 0 u
to a 4)4*M 4)4) m.-.'
4)) A) ' - D'.j . E 4- L- CL 0+
LU LUt 04 . CU g24)
.6. +6 -a> C-4. 4 a) 0 C~4

046 C 6- CU L- 2
40 4Q & 0 C' u 0 -- 4M4to '.'
C06 M0 3c4) m . 3 4- 4..
m Q a CL L an 0) 1. 4C -0..
Q 2 u 0 '-0 L.6.' 4)M L0). 66C. 6-CO

0 ) 4. a, >.+ L. 4j L- :3 .. :
4 *v 04 C 0 4) 0 U) CO a j-
E 14L -M f +6 > to.-- &-.- M 4j
O 6
C
t.)
16Ln Ln
Oa)
,-
Li
L
6 ~ 4)
x
0.4
2
) C-
m
0.
4
'-
4
10
L
C
A
U)
%4--
do0C
= C C
4) Vi
a)-.-
d) U)A
4)>
L.. W. I-- +~ co +' 0 .
W U 00 'a -0 o - L,4e0 3 "t)
L 4) Z6 + . 0
4)
4" c.
U) a) 4 W
00 4)T 2D. -C z - Ci 4) '- 4 U)-3
U) C U 6 6- C v 41 4 L w)4 0 c- U)
> -6- 4- L C 0 3
.- .- L-. 2 c .- )
(D 4) w ' ) C C4 .6IV 4 - --. L 4j L v- W.3, -

>
Q6. o L - Lv It0"" M. 4.60 I- 464
CC M) - 'If..)0 -0 w). C 4) 4)
'a .. > - -j 'o ' L+6 w. 4) 4 0 IV LW*' W

L 04 L - C 3 4 CL-0 U 3x .C
L t L 4,)-
CE I ..- L. - tr U > 6
M I CL
6 >0 06)4 46 C> 4)0 n -M4) C 'A ) 4)
L 4 C 6u
L6-6 z 06 4#
-4
ca.
j
C) u4-
U)6.04 o .> '4 4 4 4 u
4)0 (U 6.6 'o

(. 4. C44) C 4) " M )
L. 6-
(V IG .6 4D U) 00 000gD4) W Q -6
oo6 4) M- )' v E U) m- 46 V I -a 4)
4) 40). 00' 4 4 .- -- 4) C+ )
IU M m).a
m- a) 4) I- m) . 2j C
1) Q) 4j u C'- CO )'
1.. U E0 L. 4- U) " 6-
U64 +6 +64e)0.Mz
'V E-t .-. 45 +6 U0) C£4 00m 4 C ),4
-L6 -4 L+Q.. 06'
a) L) >4c 4' ' C) U)
M ) 4 C6 4) a)+
04 4- IT. 4.A' .- 34. L A 0. -+6 4)'
C L 0 (V 6 C E - -0 4' u. . U) 4 L, U)z
c 4+6+
On C>0 m6 dj
6' C>
w)u)--w 0 )) 4)= C;
O ~ 6 ). u6' 4 - - .- 'V4 C
1.-. -0 m+6
~
6-. -. 4)
4 ~~~~~~ U) 4)004)
0.
> '6.'4) IA
+
444
>LtU C + 4C-
)
U0.+ 06n. ~ '+
U
C *4 4) .C 4- 44 -- -- 0 0C 4) C C t- C )
06 a -o m' 4)
a) U) '-V " - 4 C
4) 04 j do 04
00 E-J U) z66 C 4j 46 V6 0 U ) ) I6'
- a,
C
D +6 4v ) 00 06 0 4 6-+
L) 4
- 4)o U) Ct +IV 46 ) m-). C a 4) A
et) E >-'
00' M- 4 4 C- 00. W o! ) (
Z) C) 4)
Z a) A . kA 0 CL 04 06C- oo U.
a; Us 4- U CI- - 6- +6 4)
0 1 +64)0
W.
C0 Cl 14f ... u .j a, dE
Cu 4) 0 -3 C 6 4C 4)-
- + 044 0 0 0u +06 U E-)
a,6 4 0 -6 ..- ,. 4 .
0'
+6 L)- 4 ) 0 ) - -.- + '~64+ 64£
U) DX - - 0 3 )4- 3C U
C ! - .- !4L ) )6-> 6 6L
o
a)6 > 6 4- >. !44 )L )0)C44.-4
CL6 4 6 6 C 4 U) + -+ )U . ) 2 . .- 6 £
+66c). 4 +- 2 40 L. 6+
6- ~ ~ ~ ~ ~ ~ 4
~ 04)1-360)> ' 4444 404CU
40
40 41- 'A
cic L,04' > 'A
41'
0.
'A
C
.
~ ~ 0.
14
LW
-
>'C I
'A '
4 4 .
w 4 3 0 ol co .1 4 O- 4L Ec
C~~~4 4e 0 V ' -- c a)U4
4k. C- - 4 I. 41 ' 4- 1 C4 44 1 'A
4-u~~~I 1 00u C44* 4- 01-
CC-
'A . - ' L C 4 ' 1> 4
'A 41 C 4) 4. 4 C CL
'aL C C L 'AG
v.
z 1 C0 0r apC O. I" m M1.4 C-.- -o Q C W1
C a 41 U 0 - A L- u . 'A a- mA C . U .- C k A
a;-L 6 - W3 C. c ap. 4 1 -
>.
z4 4
>j - c
C L
4n
'A
4-
3..)41 C C) o n 4L4.-a. I.
'A 0V !Z~ M U" C 0. w. m 4
0 A ' 0'
-~ ~~ ~~~~~ A 4 CI C ~ C C 4
r
1 z 'c
C~~1 0p 41 41.. U, *- 0. cx0~C

0Z -0 41C C*(U. ~C- -- jU L)41 4Q 1 1 z
0 4n EL0
Go. 0 u
00 444 'A 0
a. 'A 0 m L.- . 4
cL. 414 m 4 06
0.0 40 V 4.. 1C
0~ML 0 4A . 0
LI.. 1 0 0 c C - 4
4- Cu~L~ 4
C 0 41 I40 U 4 E
U 4 40-- '
00 4 'A 4k 4V > 01
o~~~~C .t-. .-. u '
L- wI ID) C (Us
u .
Z C4- (
'A 'A 'A

41IA I11-
0 C
leV "0 1 0 -- 4-4

C( 4.) 'A 1 k.v
00 L 41 .L 4
>U 0 I .0 0.
4 041 0 4 .
C~~~4 1-37A I >1

0~0 c 4A
u j4w 0 m'4 Q 4)
4) M 'IA 1 4j-V 4) vI .- 0
mA4 i r (i M4 ui m c
.Cu C 41 1 41 C A 4 ' .
z1L
IA CL w w1 4-'
I 4 .- 1 . Q.- a4 C 0.
CU E' '! 4) 04.4 0 1 L
4. 414 C M L a CL' G11'
. w1 C C =i L- It - L.
w1 4-U
0L .. '4 LirA~4*- M14 wA
GOC 01L'. 41 !:--4-'-~

. 4.14
41 .- .0 41 4 . " 4 C0 VA '4w I
Q m V L. IA 4)
IA w 0m.1 -; 4. 4J 4
0 -.- - L. '4 M C 0 M4~
LI. .14 0 M C 1"4 I A1'
'e do I 4 1 m4 L- w1 '4 1 1' '4a

0.L
m4
41
I..'
L-
Li
..
z- :1K
0..
a,
a.
Li
41E
w'
C 4.
IA
A '' '
.
A ~' IU4
f-
'MAL
.~41 C 41 I 0 W1 C W1 2u Is4- I-4'
j4 41 0. 6. . C
L. 41 W - 4 '4
''4
IA0 0 .- m - WA42'4
C '414 C L I C. '4' m1041.
CM LI Li CL IA Q . 0 'L . 4 I.-.-
Ld w1 IV L '. -- 31I- a.0 .1 L.% 1

--
M1 0 LA w1 '4 j ' 01 '4 C; C1 0 ' 0 Ir-'
0
IA 0 0 -
41 . 1 '. '4 L. mA-..
0 4I'; .4 0 '44 LI 41M560
C~~ ~~ 4)V..
. 1V S
di i "I 0.' IA ovL. '
.C 4 4c l 0 4! Cj
6 1 '4L '4D '
.. 1L -4wt LI 41 4 ?414. .
4- .1 L. '4 z 4 X> M1 r4-.-. 1

4- I C. C L. >.3
4414 .' 4 . V
'406L C, 41. to - L. 4,4 0
4-' IA '. C .f
4' 1.' 41 0.' 1
41~=
LI04' I L L 4'' w1 4-' I 4
2 '' -IA 4p1 . . 0 4 C 0 '4J OD
fa u
IA C'. -A (D0.4 L. at' atZ
>00
M C IA UA Z1'
0 IA dC 4e 4 . 04 4'
41 .- 4IA? .1 14 L C 4 A I- IA 0 ' >1 '-A0~
C- .u .4V. ;'34 cu 4 L- -0'-
04 I 4 4 W. A 02.. Li '4w '44

m EL.
I~- 2 IL I '' ' IV - a.'' z'
LA 1 41 0- c4-i. 1"-V 0..

.a -' Gi cA . 0. 4 ) Li' '4 0 )LI-' LL'.A[ I
C .1 ? 4 C'-''4. L '- -- L') ' If. L. w'V
41 IA C '4T4
V w C r- B. C. 4.
w- '.'i '0 1 W.'
L. '- 0 IA C .1 114 .C L 0L' L 'a ''V 4
0.~
IA. C . - I 0. - L - Z 2 'I . '4L - . '4 L 0 .
41
Q1 '4 -. C C
C aj(D0
4.1-38
n v)40 0 Mn 4
4)0 U ; .- 4 I C . IM
u 0 I L 4 4 , C
1- - C
0 C. 0A 0 to 10. - Z4 a-)440 U,
o C 4 'a
-- 0)- . 4
4 C =~0.
U -, U, 5 4 0 U L U, i- I".
44) ~ M
.V0
40 04 0
at 0
.--
10
..
L.J-..
O .')0. -
)
4* C > 0 L. 4) C - do4 U,# AC ) ) ) 0

4 4-4)U 1004 a'0 - L 04 C.- U 4 u- L0 ,
0 44 0.5 40 C*-*- 0 0.. 05 C c0 4 )C
X) '-0 444 44 u C LU L 4
4
cC a-
) 4).- C a44 a, - 4D do IA V4) 4.
4- 4 U
L .o 'D.. 44.. Cm Ua 0. 4 * 40 4) l-
44 -- L. C~. , UI 4m4C- , ) UC04 4,
L '4A 4) 4) MU 4-A M 0 r_ a. "0 a I*- w "
0 4) IA4 4)4 &.- 4 0 0 C
. Z) . 4 w
L0
0j - L0M 4 04) 0to- CZ4 C40 04 4 04
UE 4 =- 44 4
0 0L4.C 0- C U
a4) Li w 41 3 L. 44) %A4.. 4. C)
U, .CC0..C
L L -- - L-
4W CL c C0
u 0 AP U, w w)40 w,
L-. X M Mu %- k. 04 U '0 .- .0 C L.
44)Go 'D0 0) d?~. 0 4)0
w,
V 0,44 L = 50 0 u
4) MV
4 C- >1 3 4 ..
4)4 40 .:.4 -A C- 04 0 0' 504
44 4)U, C0r- -
:: m, 40
4D C
cu a444 4m 4) ).. to
u4
~ 4 0 Q-4 - 'D U C U W .
CV '10 Q4 m m.L- 4k Cl4
0) W)4 Li.. C 4 M~ ; 4) C 44)
M C 4)
L .40 4 04 . 4)0 Q - 4).. j CL -V
o 4L 4-4) L U, W 0m. w) w)4 to C4
O. L L 10 w4 'D)4 '& 4 UL
Ix4 V4 44 0 0A 4C4--
%A04 0). 4 4) >1)~ )
4p
00
0. .J a. .. ,v , 0 ~ )
5 L.
4--
10 -04 m4 40 4)jImk
X) CL E.j Q I 4A w44u
C 0
06 L0 C 0. - C" U :
M 4) 4) 04 L-4 4) U,4. 41J- 4)
4)~a
U
0)6
L.
a; ,
C ~0 .-
0.
~)). IA
0
L
*0
M--.
4
U,
~ k
a, .
4-4
u
-
40
=
U
z44
0 D 4 44 -.
(U U, L C44 41 CD
44
W 0.
Co
" UM S ) Z; 4 4) U, 4
41 (PI . 1 0 D 4 410. - 41 to
G4 0 4 x C4- L (b, cuz 0 41 "4 4.
e> C
c . 4 C 104 CL CL 94) 4 0 caC . 4).- 4)4
4 X 0 444 U - "0.-
-4
4
44
L.
... ..-
4
00
E
0.-04.
C
r4U-
CJ
C
4)
Ui
L
4)
W4
4)4
U 4) C,
C 4)..
44)
~
U,4 c
1-4
W' 4)
C- 4; 4) LL. 4)M 06 4).U, 0'.. ; 4)4 C
U3 U,
M) 44 44 )4 0 Li 44 'A4 0. m4 M
04- - 4 )t - 04 W) 04&) .1 M) . (I CL'
CI CL L.0.0U
W. Q0 E w ,W.i.0O AW U0 Pc1
'D 0 w ,Q mI. L -d 4) L
4
fI
~ .1
~ V
0L4fC
-0)
C
44
4 L.-
0
J. a)
C
4)
>-
44
U
-
F
A
U0M 4)
a 40k
C
a,
L
L01-
p
0c
(v
-Z ' 0 C.-4 04 0 4- L. Cc U .--
4_r_4
C aU, 0 CM . U,
C- 4) 40
0).. U, ..' - 44 L
0.
tv 44
:D C Li aiU4)C 4 ) U, 4) 4) C0 0'
L 44 CL U, L
-W- 044 . aU~
CL o 44 C# - ,
*~~~4
5
x4
0
~~0~
4f C-
.- *
4 )
LL
0 .- 4)4
.
0
0 CU
4
U,
4
m
4.
4)0.
apE
.
C
0 4 .
L- 4
)
U,-
C
40C
4.. 4)0
u ~
* U 44) L w. U, C
D4 L a - V50 Cal &4 , 4) -w 4)W
CL . 44)-.. Um. rU, 0au 0 4 4" C m- w..
C Li4 .
Id 04C w - o d 4)z 44>
w .C-o4.-
w o 44 . L 4x 4
- 4L
L C V
44 ' Ix."04)0.. C3 CC C- .0 u.4 4)4 )
4 0. ~~~~~
.
~~~.
)4
-
U.
4
CC4 ,4 4) ,4 4.. L
0-'. - 4 , 0 U ).- - 4 4 4 0 u
0. 0 4 LI0 0. . , 0 U O .). 0 U 4 4 . .
Cl LA Cr-jC
04.1-3
L C (- A
(.d A0 '
0-u> 4- 1-C - j l 4 36
1- GD 1-
> cu zD = 0
m~ *1 0 "Da# uD .j W a'A
*1 W. #0 -0 4j. >.W j0 a0 t44 #4
u L- CM Ci Q 0 1 IA0 UC .
C) W- -- ~ =* j x 1.C-W Z
Ca 401 GD .. op- a) 0 -
.. GD 0
Q. D4jV C C 0 4).A :
1-i 24 te 16~. W 0-A1V-
GD 0
*.-.- 0 e
LA M AO L- GD 'a M> W W 0
'0 4 0.1- 1- 4 0 . GD GDu 0 's fo
.1f GD a, 4C4. M C N . 4
o - 4 J c C GD
- .1 G6 * L.DW > GD
Q
.C6 GD
L-4 C* W GD. 0 .CM'4M 0 > 40 W AP
L.I* cc 401 1C 0.00 > A 0 .0 G 'a j.D
G
V&
GD
GD
4~
4-)
~~-4A
4W
Wa1
' L. >.
4
LA
-
4W414
L D
V
GD-
- 0W
'.
.0
.
.
..
iC
G
'
D
4
4 -
GD
04 c a GD0 . . L C G ~
a.' WO0 G 0- O
a-'
C ~ ~ LA.
~ 1- ~ 4
m~
CL' 41
4 .
-
.C
..- A C- .
C G C G
L.U~~~
GD CC
u4 ~ 4.- A
G
-
L--
>.C wD -4
.I C. 4.0
40ap
L.
.-
. 3 -
0 4
I.o 0, JeL 4
w- w- 4-L ; - S L
o0 m -> 4- # -
CL c GD G.0D
5- QV 10 0 t 41
4- LrD G D G
LA AL-DD L .
4 14.11-4
* E G
C) - U.-2 w
M-- V w4 ~ 4.D
>- C C
m 0- 00.-
3~G 0D4 W (D0 40.(V 4
>
'4' GD' DC a,
4-) GD D ~ G m CL
0
C '6-
0 GD
-Mr>
U.
G.-
GD
4.
4-
GD '!GD
'. ~ v.
*. 0
6G
,A
MD 4.4
'4- S L :
U -4j 4 . 0
0: 4) 4- GD G
ap
6- GD a U4 GD 0
GDz 0 C
6 04 GD 0
0
-. 00 0
M. 0 U CA E to
U .- 4-C m '- E
0) ..- M 0 GD GD
' 6-
>i 6 4- ( 4
'441 b 4-L 0' GD
GD
w 4GD a- m >
04 v 6-4 4- 0* CL w GD
) 06d
0
.- I 1 0
C 4) 4- GD 'A C
L&J I 41..- -A- 6- a :3 40
0 C N ' a C~ GDW U OG GD
0
'
* ~~CL 00 Z U
..
-C
G
0 ~4> G..
'A
~- E
;>4
> L.4 D. GD 06 GD0 "-.- 6

_ 6- 0 U4) M 4~6 0 W 6- .0 *-
2
*
O..-.
0 >
0.C0jU
"A * 6- .-
6- C
GD D
U
G)
I-
- ~ 6G
4. 6
L- CL-
C w -0 (7 0 GDO '4- a
.' 0- a 4 >46 A. 41 - . GD GD
r= .
GDE (3.- ' G0 >40 4L GD GD GD DA0)6
.0 c GDL(
4) 0 4w GD C0 6-C 10' C.
D- 'a m GD GD C .-- -GD C 3 >-- D.' -G

6- 06-0 U U.C
UA 04- U 'A 0,G 0, 1
GD -C ' C -.- 6GD D >. G
c-uG
4' f 6 .- >4
GD >.C- -F 2L Cl 2 %4 0 C - U'4- GD 0
0. 4 4 ' .0 U > GD 'A- CL.4- z m 4 GD DfD...
C zA
GD C 0
LA >0 C) -
4.1i-41
> .
In 'V
0o 0In C 01 >.jC
u 4- ItC = 0 '0 %A
. 4- 4= 1.0
%A 'AT CIV - c 0
m IV c4 .~ 0 C. 0 0u .C. x .61 0.
3a14
c 4
C
C0 x
0
.o C1
0D 4 c L w4 a;.
Lc 1 ..
..
M- Mn.FL
LJ.
~cm
L.x - I- i C>. '&A- 401 %A- 4J l: In; CL LLAJ
w u '04 3 I 1:) 0 V L 'V u 0 Z L. 42 0
01 z- 01 CD C' 06 010 0.'
W - 4 J. 0. %A (
3. J-- 'A C > Im3 E
04.. cn. w~ IV. V.C
0 >1- M~ 0 4
It f . Um U 0e 0. %AL
0 0~ 40w0 a wU 0 4 10 0 qu 3.w
W0cc l0Inw0
> L. =4
In 4w
C .- In . 10
=. 4) 04D
W. C 0 1 0'
0 0.4 = C c0 4 0E~'V
>10 cc
I
Z CL m n 3. ' z Z
M E() .41 c1 c * a .-
L i It 0' x 0. zC -3 0 CtA
m- 0 I#. C.- AA C 010> .c
0. a 41. . MA L u M.
4 n I 4- I 01 I In *. 4 X' C 4 C
(~~~C n >1 -W). 41 4. 1 0
L. u
In~~~ 01 In0 Cf C- 1 0 4 0
'V .0 4 6C 9 CI In.. W0 01 L L .*- F,
3.C x1 C- 01. 06 4W LL .04# _x cI .1L
c ai >..0
:3. . 4= 0 U *- In -4. LL
.' .Ew .'.. 01 01 ,-
Inw- 4. cr 1
03 4.1 0 a W (L 0 C' 01 w ci 1 4, aC0 C

ai 0 0 L.4 0 L- 1.4 0 a) 0 1m - L
t-L0 4 -1. 0' 0.4 M 0 L6.)0 V - InjL-
t
.c c
C -- CV' C0 4 'C L 01
-C *IIC 3 4-42 m . .-Z "a4
01U' 0. 4 . L z1 . I wa.
'001 -20 L)(1 0 0 C C n s-41 0
U. n
AA LO. o0 .C.- 0 0 Z ~ .0' In'
'A m %)-A - 4 'A J -
%A1 4- 4 U 4c
zn m 4- *- 01 LV 0i 4M C
C 01 - C,4. a.. %= - C- L- L0">L~.

U - Q %.... 3..- , 41. I- .. *- D m' C
Fk , l
&A Go3EL4 Z. :)C' o1 41 UD
c 01
a) 3. In 1 c
In
I- C.
1
c 4c .
at1
.-.- 6 4 m m A4 x' wV C1 C re .' 530'-
ap 0 E 13 CD
v 1 010 d u. . 4 L- C
4 010 u 20U CD L 0U UL 0 . S " L
I- .C 4 .-01 "V1. 0 0 3.- 1 4I)
ai 0) 00 C) 0 10 C- 01
a) CC
01' EC 0
I .- 0 >1 0'o LU, Ln m 0 0 C w1 0 d 0 In 04 M nO
I Z
Z '0 064 X.. m' LI.. z 14 (A IA 1.4C
V L..L ra
0
C~~ - I0 - 41-> 1.0 C
0.. >, 0# -D~ c C - 'V >43K

(06 'a Ln m L 0 C E Lw
-. 01 -E 0 4-1 C cC m-4. 0
u I '4. 300.4 L C 0 0 mC 0'D n
~n
m 0'7~
cn
I '-0
4
>n
k1 4
mn
L--
.
11* j'31 '0
) -
4n
40
C M
v 4
C
01I
C.1
* ~4m a 0
ao-C
InC
.0 C
0
I
m
00
m L-
0
0
w~II.-
4-
.
M0
0
1
..
ICL
*.-
11.1
* CL
'A 7
U -- .- C >..-0- C) 40 -I.-
4-. IV -L C 1L .j
4' . 4 '. L~ It 4-I 1 0 1 . 3 3U In -0 C' 01
C 4 -3 0 w 1. c 01 'V 01 1'42 0
I 1 . C1 01
L.' C' 0101 c X CL 4 W 0 C,
-b 010 0' EQ00 0. 1.u 'a 0 . L -1
4 C. 400 (U 00 c:: 4 W006 0 X.. 10 0'I 4A 1
1 .
1
C
.- 4-I L -.
-0
;-..
I C .C' n. >. . C 04 01 W V14'
. 'D 0 d) V) .J m00
4
m1 0. M0 3 1 a,0 a) M- . - ' V I
0. 41' 010 C 401010 C ' 4' 01 01 0' C CC 01 C
m 0 03 > DU 01' .L 4) 0 C 01 0D 0 14 In 04 M I 0
C. Z4'0 j (D.. .4 g 00 Ll. E 24 le A Ln VI)C cc .. 1. E M
cC
0I0
C D. L 1 .a,.
0 In .4.1-420
r-1 0 LIC . I 30c- CL

u C 0) 41 CL wC0-L 0 . an M..00 C -' 40 1U
M '
-
c . m1 4
CA o
IA 4 C4
41 3.-V0 o .. 41 - 0 w1 41.- 0) IA 0 L.1 414
41I 0 04v . LA 4 .0 ( u0
%A - 0 C L-
1- 0 40 V I .-- 4 41I C "A* " A, 44- M144)
W-
4 04. . J 4L 0 0 1-I >141 CL *
41 41C e0 1 -4 . L..I C 4. C u A 0 41
4104 1--441 I- 0 . V 4 L1 o4 c - C LO 0 4 44 a)- 410
- jI. 4u L 0 LI 44 04 IA 0C 001 - =..4-
41D -V 41 IMC. 4 *- 3 10l01- 0 -0 0

0 1- LI 4)
IA- 4 Cw I-I 1d 0 L, 3 r4
3t- 4~.
LI CL 4: re '1-L m1 4~ m Q. 41 -1 a1 L-X oA o
C 41 -'A u C M-I a)11 CL 4
3t1 'C C- U- 4
L.4---->L. 0
o c M- 0. .- 40 L) c0 C 44
W . C )tC 0. 6- V (AJ
u- 41.C
a)4 . 4). 4J - 4Q J l4 L L . 44 0 - A - 4C - W1 U . 0W-.-
- I IL CL 41 -1 41
M 01 M
1- .l 1- 4.' - LII Q, -.-. z- '004

0c o
IA 3 1 L., I 1 -
L- -'1 0> - r=
X1 cCW I 41 r- 0 0 3 - 0 .0 014 o0 1-A C 4-I 0
w.. 0- E>I V 00 41 0.1 4DI3 C L- .- 4MA aC
LI~~'
C 1 . . -. . A- a) IA 'D M4
4.. . -- 4 C 44-C %.- 'o :; C 42. CI -.
0.>0 1 31V 0 >41 4 - 40 1D 41 1 0 L D 44
1L. 0 1 0- IA.. 44 1-C A . 4 301 1 M
WA.
4 01 0.C. 4 00 al-I >41 C I- IA D- C
1-1..1 3LIA I " Q m44 M - ;4 W Mv04 CIA M
at a) -- OC 800411- 41.1 L-.-C -91 2r IL I 141
c. u IA 1&.1-40 Co u -u0M4- CC CL. L .-
M.00U
4104ao 4- A 414 0"'- 1- C D1141
x 3 4 1 wX 0
4) '3 610 C-.- IM4-M 0. Z 8. -X~ U1 W "0LI 4
C
.1~~
t1a400
41
V
0
C ~a-0
V
MZ
C
CLC
V
4-1
4.
M
044
CL
L
L
IA
-u
.
4-.61
u001 .(
W~
-
410. a)
1 0C
01-0.0
0)*
41--
0
4 I A 1 L-I w Q441
. 01.- L-. -40 L. 0 - 4- X
.- .0 C M40- V mII. C CO M0 3
IA L- 44- . IA .44
L 4- I-440- aC 0 W>- z1 IAu 4 404 1-
16 m C ( 0 1 0 1 0 1 41 = I
4-4
I 6v C 06 C C1I0
IA I 4 1 4) 0o to.C. u a)
x1 I0A ao. - 0 0 it ~- - 4 041
C -L-
IA 41 .Cj C.-
C LI 0 0 1 0- 1IA
C. C .0
41~ 410. M 1IA 4~ U. 0-. C. 4
L..
>1o 4- 30 ' >14
1V 41 01-4 41 IA, 0)1CL.
')4D 1
- 1I 04j 4 IA L- IA f^ u fa MA 3( 9 M --
w 0 u CL0
a) w c 8 0 Z0(U1c-V 4. cA. IA 4A-A.
0W 1 W11
1 .
c4-.1 1 m .-- L4 - u =1 ... >1 CL IA.- C CLC
(D1-w. CL 1 IaL
z0 I 1. 441- 4 4 I CA IA4-0
(u4.
IA 0 'G 4 06 CO -03 -C 404- = m - 0.0.01
I o4-a V f;! -0 4V 4. C ' -w 4. 1- t-.

30C 4) 1-
IV
-. -.
-V
.1
41
4DI
4
0
. 440A 0 .-
4. 3 4
r_. 03001- w.0 1-0 W -0 - >)-
44) C - 01 .- V-
IA- C 4 C C 1 IA -0 3 j - -4- IA m0
C . 41. Cj 41-to UJ0. 1 ~ 06 VM CL W-1
rI614 4
0
I IC . 4) I 0 1.
41
al 0 0 C 1I 0
(3 C -- C
L04 crc CL).-
IA W .1-1)1 :If V
"- 'D - E
xC IA W0
41 3 -C 0. .0014M 0 1-I C.0
04J0- >.-V W1 o4o 1 3Z CL IA -4.- 46 C-.
-~4 CL .0c 4- A. 4 A'. MV4 =.
- :3. - 0a) C 4V 4- 1 -4 4-' C 42.1m CIA - 0.
I 31 'D. 0 1ft W1 -40 W 41
(V C) IAV 01L.41(414
-O n 0441 IA 1- I 41 f (
Z0 IA - u0 CL'
-
4
1-1. ~~~~
- LN 0 . .
IA
4v
L In ~4.00 A0
416.-I
4 4 - -1 r_23
W-4 31
41
IA-.
L A
o C
41 40 414 V 41 M. 1- C14 L C L- 44 (36' 0

CL
10(3 a.)1.- 0 4V0 CL. 3 r=.. . -W w1 OWLh 0. LI 4-
41 I - 0 V 0 4 0410 ) IA0 C 001 L - 4
- -0 a, V C x IV Q1. C 0 1- -.- .. fi- 1D= 0 C
)t1-D --
V le C- I"
4r .- 3 ap :.C 01 "10-- 00 0
4 0J
L IA 44 m- 1
w4I x IA 340 M - U1-- 1- 0 ..- w 4-X
41 0 U 1L M1 C 41 CL Z
IA4 41 - 00041
41 41
u 2 41 -4 . 41. C 14.1 x-~ w1 W1-4 u 4) 14
0) 4-ap^E 1 > 1w 4 41 -C >4 0.1E - IV.1C W 1-14
C 04.. 4 W 41
0.
4 -o
4.C
~ ~ .4
C0
-I
404
A4-
-
4 C
a.
X:
WVJ
0>
w0)D
N
1 4.
IV
3C1
5' 4.M 0 C4 41. . V 4/
06-
I- I mi do4
C 0 0I
41Q 1 .
E 0
.c C, CL
C 00a
4.1-43
4~- %A 1 -0 r_ I ty, I I I W1
C V .0 0 wIL- 41 *. '. - 4 c
41
wn C C .0 >. w. "D~m. a'44M
.1 .- >,4 w4 m 41 ' 'C,~0 - 0, 0
C
c*1
~I '.
.-1x
u
*-41
3
0..-a
u4
CL
41w > .0 4
-
3 In
U
u
C
'4U
M
-'4
0u
CL
.-
In 41L 0 00 1 C. 0c
on In 4-I c1 0.~
41 0 In C C ' 00U M. C U CU 0 .
L) '4. Q.C. Cn1 41 GIL - 4a) 4-I .11

a U
.C wt "D '41. >4 In~ .0. 3 MC -v aj4 C
c1 C. *-. -1. 4 1 1 0.''
C w1 L 4
ID 4U- 0 m CL. c. "aC' 0 -. ' E 41
Q 41... 0
CL. 412 W1. 41
V' >4
>CN 4
0
.1 '41.1-.I m LL. m- r- a~C > 41 O''. Il 44
41 ~- ~ 10 C.-. 41 4 O 3 C .IC 0.0.12 44
CL L1411DL 'a2 * 41 0.-o 'e U W -41 C C m4
M. n -4
- ~ U ~ U
. Inm
1 m~. 0 In 1; >4-
O.
I..
. .
0.-1
0 '4
In
41) Q;
3 2 n-
LI 141 SL 341 U u 41 P. 4j L 1 1 041w~ 1 m4 L1 .0 c'4
L .I I 4
O. n CL..-.>
tj
c- '4
I" 41
W c 0C
CL44 Q# cC -W
Q. Is 4 w1 In
C ~ 4-
~ M4V1
-0 41
41.. .. U - -C
0.1- 0 U0 041 0'.
41..U-
s0 m c L 0) .. c' LI
u1 CL In.- m4 C.- 2
U 4'oM44(D10 OVE'4>U
~'0 0M m1.
0.0 4 C
4-C -- L 4' 41
0. C 0C. 114 .- f
a) 0 - ,1
C0 IUmn
In~~ 30411I n ' 4 414
a, 0.1 0 C L L In 41(
a)
> m 1
CL 10 m4. 06 C . 'a I -
In 41m 4- 4141
rC In m1 . 0.. I CL
~~~W0CE~u4
4141
' UD '4j
'4
Q.I
o
4
'4
'4 .0In
01
ID-
4-
n
4, In
44
'
C)
4
411C
!
u4
14:
0. ' '441m.M -- f 0.0 41 w c'4 - 4
'aC 00.. un '4 ... C L >.1 - 41
In
In4 Ui 04 M-. C v
o G 0 4D 4*. a- W 4 -. 0 L v14 C v4
M ' .' m1 0 4-. u40 CX
U n a
4- In '4 - 4)..C 41 an C U 4 u
- U 41.-. I0 C ..- L
01 '4) '4C d
C~
'4 In) --- In 4- 4 4 . - '
41 UL '4 '41 U Ina U
CL' 11 U c1 '0 In~
41 m w~ c4 .- L 414 C . 4-. .Z 4 1 0 2I (X
C (3L CL :3 '' a-.. C3 a' 4 L u1 m w m.
0u
I 41 LA :n 2,4
41 I1 45 In E L , L. u c 0
Cl I . .- C 0 Cz Cm- 4 .
44. .- '4A In' C 06 .... 1 4 U . t- I. 41
w- 0. m CW- W. 'A. .-I3 0.0 L Lc 4)
_r_ .0 LI CL -0- MLL
U >4
1 un o44.' w0m
'4 t- Is 4- u0. - m
w-4 C C U U . 1 '
GI
.0 U 414 %A4 Up U
w0 ..
In 41 ~ . 4 - 1
In
(D1 1 C.
n U
In
0.C~'4
-.C 0
4 m'.
C ' 0 -W '4 a) C U 0C' 4- 0
UL 'V. 4144> 0 u 4)~ 0. 41 C.D ' n '4U U 1 .w 4 0

,a 10 ld 0 .- .1- w L- C mn~ 0. 41 CD 4140 In
(A. ix 414 In '4 6 L CL 'D4 C .- 4
'4Cc.- L 4 0 0 ' 0 4 - C 4 1 . 1 . 0 . . 4
4). 4 0 .- C1 U 4. 0 *.- 4 0 I1 1 4 0
0.
IF 00 C ' .4' C I~- ' 4 - 4 4 - U 4 . .
L- ow M -- ~ ' . . 4. 4. n > 1.4 . 40 U '

L M4
I U 0 O 4 n 4 L 4 1J ' ' ' 1 0(
41 .. 4 1 L '.- 4 0 2 .- 4 ' .- 'J 4- I n L
0. '0 0> C~3 0 U '-41 14' .1 ' - 4) '4 1

00
o CL0 -
c0 I
.0 0 .
CL0.
1 WE w
0 uL
0 M J (1 A0
00 >
tr W
a0 0 a
EL
- 01w
- Co
4) U
4 0
o m
01 00I
L.0 CL m
u1 0' C04
C m1
U
0 m
L
-' w
UU
0) CA.0
0 C a
> u 0
S 1
w L3 0
0. 0-4
puncture of the dielectric can cause a relatively harmless vaporiza-

tion of a small area of the plates (known as "clearings") with the
etfc-t of a noise spike and a small reduction of total capacitance.
These plienomenA aie not considered failures of the capacitor, until
enough of them occur to cause significant reduction in capacitance.
However, metallized plastic capacitors should not be used in timing or
memory (storage) circuits or anywhere a momentary breakdown in the
dielectric cannot be tolerated. They should not be used in high im-
pedance or low energy circuits where the fault will not clear.
4.1.2.1.2 Fixed Glass and Mica Dielectric Capacitors
These capacitors have non-flexible dielectric materials. To obtain

the higher capacitance units, thin layers of the dielectric are
stacked between multiple electrodes. Alternate electrodes are con-
nected in parallel. The electrodes can be either metallic foil or a
metallic film painted directly on the dielectric. The assembled stack
of electrodes and dielectrics is held in close contact by clamps or by
the capacitor encasement.
Mica is one of the very few natural materials directly adaptable for
use as a capacitor dielectric. Its physical properties, plus its rare
characteristics of nearly perfect cleavage make it probably the best
known capacitor dielectric. It is inherently stable, both dimension-
ally and electrically, therefore, mica capacitors exhibit excellent
temperature coefficient characteristics and very low aging with
operation.
Glass dielectric capacitor electrical characteristics are very similar

to those of mica capacitors: they have excellent long term stability,
a low temperature coefficient and a history of good reliability.
4.1.2.1.3 Fixed Ceramic Dielectric Capacitors
These capacitors are available either as tubular, flat disc, or flat

plate designs. Tubular designs consist of a ceramic type with silver
bands (electrodes) fired on the inside and outside surfaces. Capaci-
tance is formed between the silver bands with the ceramic as the di-
electric. Leads are wrapped around each end soldered to the bands,
and exit radially from the tube and are parallel. The assembly is
encapsulated in resin which is subsequently vacuum-impregnated with a
high melting point wax.
Disc designs consist of a disc with a thin coating of metallic paint

fired on each fz-e. Parallel leads are soldered to the metallic elec-
trodes. The assembly is encapsulated in resin and impregnated with a
high melting point wax.
4.1-t6
Flat plate capacitors consist of a monolithic stack in a molded case.

The internal stack consists of multiple films of a noble metal spaced
with thin films of ceramic. This assembly is fired to give a mono-
lithic construction.
In dormant applications, capacitors with wax impregnated contormal

coating will collect dust and other contaminants. Under these condi-
tions, the insulation resistance will decrease under prolonged high
humidity conditions.
During dormancy, excessive moisture should be avoided since the encap-

sulation material may absorb moisture and silver ion migration may
occur when the capacitors are later put into service. Silver ion
migration can cause short circuits in short periods of time in humid
conditions.
4.1.2.1.4 Fixed Electrolytic Dielectric Capacitors
There are two types of electrolytic capacitors: aluminum and tanta-

lum. Tantalum electrolytic capacitors are be further subdivided as
solid or non-solid dielectric types. Non-solid tantalums are further
divided into foil and sintered slug types. The construction of each
of these types is summarized below:
Alinuw- Electrolytic Capacitors
Al,-minum electrolytic capacitors consist of an aluminum foil

rolled unto a porous spacer. The spacer is impregnated with an
electrolyte and separates the anode and cathode. The electrolyte
is usually an aqueous solution of ammonium borate, boric acid, and
glycol.
Solid Tantalum Dielectric Electrolytic Capacitors
Solid tantalum dielectric electrolytic capacitors consist of a

porous tantalum pellet or wire which serves as the anode or elec-
trode. The surfaces of the anode are electrochemically converted
to an oxide of tantalum which serves as the dielectric. These
surfaces are coated with an oxide semiconductor which is the work-
ing electrolyte in solid form. This oxide semiconductor estab-
lishes contact with all of the complex surfaces of the anodized
pellet and is capable of healing imperfections of the tantalum
oxide dielectric film.
Solid tantalum chip capacitors consist of a porous tantalum slab

which serves as the anode. The surfaces of the anode are electro-
chemically treated to form a tantalum oxide dielectric. These
4.1-47
surfaces are coated with an oxide semiconductor which is the work-

ing electrolyte in solid form.
Non-Solid Tantalum Dielectric Electrolytic Caiacitors
Non-solid foil types consist of a tantalum foil, as the anode.

This anode is electrochemically treated to form a layer of tanta-
lum oide dielectric. Porous spacer material is used to form a
conventional cylindrical capacitor section with axial tantalum
wires on either end. The section is impregnated with a suitable
electrolyte (usually a weak acid or base) and then sealed in a
suitable container. Solderable leads are welded to the tantalum
leads.
Sintered-slug types consist of a sintered slug of a tantalum as

the anode. This anode is electrochemically treated to form a
layer of tantalum oxide. Porous spacer material is impregnated
with a suitable electrolyte and then sealed in a suitable
container.
Electrolytic capacitors have experienced problems in storage. Table

4.1.2.1.4-1 summarizes the predominant failure mechanisms associated
with solid tantalum types. Table 4.1.2.1.4-2 summarizes those for wet
tantalum types. Electrolyte leakage in wet tantalum capacitors has
hepn the major source of problems, while impurities in solid types has
been one of the predominant problems. Most of the failure mechanisms
associated with these capacitors are accelerated to failure by a temp-
erature cycling environment. (Ref 18)
Aluminum electrolytic capacitors experience deterioration of the oxide

film when placed in a dormant mode. The ox-ide film deforms at noor
low-voltage and the capacitor is destroyed by application of full
rated voltage. These devices are not recommended for dormant applica-
tion, howaver, if these devices are used in dormant equipment, MIL-
STD-1131B (Ref 25) Presents guidelines for the storage, inspection,
and reformation of these devices. If aluminum electrolytic capacitors
have been in storage for longer than 5 years, it is recommended that
the capacitors be checked for leakage.
Although electrolytic capacitors are the most volumetric and cost

efficient, they are not recommended for dormant applications without
application of the following guidelines to insure long life reliabil-
ity in dormant applications. (Ref 24)
e Use only hermetically sealed capacitors to provide maximum stabil-

ity and reliability, and prevent outgassing during low pressure
operations.
4. 1-48
CO 0 0 0
06CU 0. Q
cZ CL CL C C
k 0 0.
C01
m
0 C
~ 0 -
.
u '.
L, .00. . . 0
C M 0 04 .4 4x cc
-c 0 c. .10 1 0
40~
4. 4 -00 0 0
*0 0 0
0 0
c 41 -,.
d-. 0 do
L C 10 C" c 10 M C
0) L- .-- M L
0OW fu 440-,
L4Z 0~ 0 0
C CL C
C) C ) C A
Q 4 -0 :.. E-
C44 0 0- 4 0 0 CL
L 0. It 0 m. (w
0 x4 0'A u
42 4 2 0 .4 0 c4-.
4/1 m- > 0 0
'Ac.-C- L-.
0. ..- 0 06. 0).a)
-~ uI-- 41> E m~
ea 0 W 0 0
M~0C 0~
4L 0' cu (1 u/ 'A L 'A >
C evWc IC
W 44 0
&- 0A o t(O 41
a)1 D n- .A C1 01) 0 0 4
oC "a L > c. .
C~a 0. 0 c4. C. - 0 - 4 )
4m 10 "a ap 0 AC
>A - j4
04 4) 0> L. Cw be 0 z
0.. 4- 0) 0U .- 0 - MV c
0 .- 0 L C 4-'A C . 0 '
wV ai 0 0A 30) a) WAL
i-E C 0 C 0.
cr- "D 10 .C 0 '0
4 -4 4A
47 0- *- ;.: Id I u )
ao w A. mA 0) 00 Co (v
: 0 j1 . IU c 0 14 ) V- 0
> 0 V w
D 4 CL 04 CL V >1 )
C=0. ; x4 u0 CL- 'A
00 4 C 0C . 0 40 V 0M
M 4 'A
0 U'- m Q.0 1 __ A Q4'
S 1C * 1 0 4 4 4 .
- 4 C 4 A C V 0C 0.C .- * 0 0X
I 0 0 * S0> C . A'
u 4 L . 0 . - U C
C 0
*.-£~~C
~ 4 A' .4 L
-
> 0
3 4
04' 44
L-A54'
.-
-
44 0V 4 04.149. 1 C' 4
o LI
IA 0)
o L I Go
2 C
4i 4
0n C -.
L. 'A D L. 06
L u m- 0
1~e l - j a. 0:
0 CL 10 0~ L)
4. , 4#n
0- c0. a r- a I CL
I-- C CCc
C .
E C; C C;
C .. L. M. .
4) 0 c
00 (36L .. C I C
> UL L
4 >4
E~ 01.
4 1!- L.
W :) C 0 4
~0. LI
0n 0 LI
x D I
.00
U.V 4 . I MV.1
(U G4 cI -
*C0 * - CL
<NI~~ 0 . - 0
IA 4 4# 0. r
m _ LID
> C
0 C 0 > . Xn
InL LI0
MM aI. u' M
-j do0 L 4'0 4 LI 4r
0 0
4 0 C
C 0 4-1C5C
* Solid tantalum electrolyte capacitors exhibit excellent storage

life characteristics, however these capacitors should be applied
with a limiting series resistance of 3 ohms per volt minimum to
avoid potentially catastrophic breakdown.
e For foil and wet slug capacitors, the leads should be welded to
anodized tantalum risers external to the hermetic seat so as to
provide a completely insulated structure within the seal.
* The tantalum riser wires should not extend more than 1/4 inch be-
yond the seal to minimize the possibility of bending and therefore
damaging the riser and its oxide, particularly at the seal area.
* The cathode lead of wet slug capacitors in silver cans should be

of a material which is weldable to silver without causing crystal-
lization or punch-through. For this purpose, silver or oxygen
free copper is superior to other material such as nickel.
* Gelled electrolyte is preferred over liquid sulfuric acid electro-

lyte in wet slug capacitors because of lesser mobility.
* The proper high temperature solder must be used in fabrication to

prevent reflow during assembly operations.
* The attachinent of slugs into solid electrolyte capacitor cases

must utilize silver bearing solder to prevent silver paint dis-
solving into the solder.
e Plain foil capacitors are preferred over the etched foil, since
the etched foil is somewhat more prone to manufacturing process
errors.
* Seal test should be performed on 100% of the devices to verify

every seal. In addition, if acid electrolyte is used a litmus
paper or thymol blue test should be added to the usual leak test.
* When capacitors are used in banks, they should he assembled in

easily removable modules to facilitate replacement and test.
* The largest possible case size should be used for a given capaci-
tor voltage rating as this provides thicker oxide dielectric,
lower ESR, lower dissipation factor, better heat dissipation, and
greater capacitance stability.
• For long life high ruliability usage, the Peak voltage including
surges and transients should be limited to 50% of the manufac-
turer's derated ratings fo. all tantalum capacitors.
4.1-5]
" For foil and solid electrolyte capacitors, the allowable ripple
current should be derated to 70% of the manufacturer's derated
rating for high reliability.
" The temperature of solid electrolyte capacitors should be limited

to 50 deg C including internal temperature rise. For long life,
the foil and wet slug types should be held to 70 degrees.
4.1.2.2 General Characteristics of Variable Capacitors
Generally, variable capacitors are highly unstable devices, and they

are not recommended for dormant applications. Tolerances of variable
capacitors reflect their instability. Periodic inspection is required
to insure that variable capacitors are maintained within their capaci-
tance values. This will impact the maintenance cnncept selected for
the dormant equipment. Refer to section 4.3 (Maintainability Design)
of this text for a more detailed discussion of maintainability design
guidelines for dormant equipment.
Environmental degradation is a significant failure mode for variable

capacitors. Variahle capacitors are highly susceptible to moisture,
temperature, shock, vibration, and contaminant induced failures.
Variable capacitors cannot be sealed in a complete encapsulated jacket.
V'ariable capacitors have movable shafts which protrude through the

case and possibly through the front panel of equipment. This opens
the interior of the variable capacitor and possibly the interior of
equipment to the environment exterior. Various types of shaft seals.
such as elastomer '0' rings can be used to protect the capacitor in-
terior, but they are imperfect seals.
Dormant variable capacitors should not be lubricated with a liquid

lubricant, because the presence of oil lubricants can collect dust and
other potentially corrosive and degradative contaminants.
When the use of variable capacitors is required, they should be main-

tained in an enclosed unit to Protect the interior of the capacitor
mechanisms from degrading environmental influences.
The use of variable -apacitors in dormant applicatiors should be

avoided, however when a variable capacitor is necessary, air trimmers
should be used because they offer the highest degree of stability of
variable capacitor types.
4.1-52
4.1.2.2.1 Variable Ceramic DieleQric Capacitors
These variable capacitors consist of a single stator and a single

rotor for each section, made of ceramic material impregnated with
transformer or silicone oil. Pure silver is fired and burnished on
the top of the base of the stator in a half moon-pattern. The rotor,
usually of titanium dioxide, has pure silver contact points. The con-
tact surfaces of both the stator and the rotor are ground and lapped
flat, thus eliminating air space variations with temperature.
These capacitors are unstUble with temperature changes and at high and
low temperature ratings. Changes in capacitance from the nominal
value measured at 25 degrees C may vary from -4.5 to .14% at -55 de-
grees C or -10 to .2% at +85 degrees C. Temperature sensitivity in
these units is nonlinear over the capacitance range and varies greatly
between the units. These capacitors should not be designed into cir-
cuits as temperature compensating units.
4.1.2.2.2 Variable Capacitors Tubular Type
These capacitors are constructed of a series of concentric circular

metal bands with inter leaf and are variable by adjustment of the re-
lated depth of the interface. All other style capacitors are con-
structed of glass, quartz, sapphire, or alumina dielectric cylinders
and metal tuning pistons. A portion of the cylinder is plated with
metal to form the stator and the metal piston, controlled by a tuning
screw, acts as the rotor for these variable capacitors. The overlap
of the stator and the rotor determines the capacitance.
4.1.2.2.3 Air Dielectric Capacitors
Air dielectric capacitors consist of multiple rotors and stators, each

having a half moon shape. The overlap of rotors an stators determine
the capacitance. The rotors can be rotated continuously and full
capacitance change occurs during each 360 degree rotation.
4.1.3 Microcircuits
Microcircuit devices are broadly classified by the way in which they

are packaged (hermetic vs. non-hermetic) and by the type of testing
and screening they are subjected to during processing. The screening
specifications for microcircuits determines the quality level to which
the devices are classified. For military and aerospace applications,
the highest quality, most thoroughly screened microcircuits are pack-
aged in hermetic packages and are procured per MIL-M-3S510. This
specification establishes the design, inspection, testing, and their
reliability assurance requirements for monolithic, multichip, and
hybrid microcircuits.
4.1-53
Microcircuits are precision composite structures composed of materials

which have been chosen to provide desired electronic functions. Their
small geometries and the intricate nature of the composite materials
employed make detailed chemical and geometrical stability highly
critical to device reliability and performance. Conditions which
alter the intended physical arrangement of materials in minute detail
can have catastroghic consequences on device performance.
It is generally true that the materials employed in most microcircuit

structures will change very slowly when protected from degrading en-
vironmental factors. Properly fabricated microcircuits might well
remain stable for decades if stored in a perfectly dry environment at
a low and constant temperature. (Ref. 3) Unfortunately, these condi-
tions are seldom achieved for most military electronic equipment.
Table 4.1.3-1 summarizes.the effects of various environmental factors
microcircuits during non-operating periods.
In his study of the failure mechanisms and materials degradation proc-

ess which will cause the failure of microcircuits during long-term
non-operating periods (Ref. 3), Livesay concluded that the most im-
portant environmental stresses during non-operating periods are
mechanical, chemical, and low thermal. Mechanical stresses occur due
to thermal mechanical interactions and residual stresses resulting
from device fabrication and processing. Chemical stresses result form
contaminants such as residual process chemicals and environmental
gases which are introduced through improper or failed seals in her-
metic packages. Although purely thermal stresses are much less im-
portant during non-operating periods as compared with operating condi-
tions, certain low temperature reaction rates and diffusion processes
are temperature dependent and will proceed at reduced rates over the
non-operating temperature range. Of critical importance is the syner-
gism of the three primary non-operating stresses. While any one of
the three acting alone may not have a deleterious effect, the combined
effect of two or three working together can cause device failures.
The failure mechanisms of greatest importance during non-operating

periods are those related to manufacturing defects, corrosion proc-
esses, and mechanical fracture. Since a non-operating envitonment is
normally free of electrical stresses, electrical and potential current
induced degradation processes are not important in the nonoperating
environment except when the environment itself provides the electrical
stresses (e.g. See EMP effects in Table 4.1.3-1).
Although the operating and non-operating environments for microcir-

cuits may be quite different, experience (Refs. 5-10) has shown that
most failures that occur during dormancy are of the same basic kind as
those found in the operating mode and are the result of the further
4.1-54
ICL
CL-
W W
Q. ~ I
S4C -
0.0 o E V 0 w0
0 0 C CE e 0- .
1 0 O C 0 wA-.
C CL CO= AO 1l 4)
-o
-r a w-. W L
U~ 41 EL ~ CU -CL 0
&JJ CL4
0~~ - wO
c! -. o.
IA G 0 CLcm
0 0 'A Q.L
Q100*Q1 WCO IA 0.- 4T
16 Cw
v0. 4,. I .Z 4
0I
CL C
0 4
10 c L1
Ln M 0) 4, AA
00 O0A 04 w a.
0. C u 0 oC 4 n . U a
IA 0V 00 .. 00-44 Q..
0 I0. 0. I a). L CL- 0 -
x CO 'D > ,. CA m4 L.
-
0~~
C .- L
~ 0'
~
cl
0
- ~I -
C 1
1C
00 W
4..)
a,0
CL~IO*L~
I
J
0:: 2; 0[ o\1
L C 1
%A . 1
t/ 4.04 S w 4U
IA . .10 .0 Ln 0 064 W
o~QEL 00 4C . 0
O0 %A 0. 0' 'A. . C0
>- m0 0 4 C1
X Z UQ 0
EO U .0 CL faE 4) -
WC. CL o 0. 000.) M L
0 W vIA 0 IA>. 01
01 0 0
r >-.1 12 - .C .1 *'. 0 u IA
C 0
EL =
Q
4 IA
--
IA
M 4)
0
CO
L
V I
0) C CA
-0
E
Go CW. IA1. U, 00 u u0 V 0V'
C. 04 CU 4 U0 cr . 0 j 04.'A
0. W~ 0 a)U C - 0 0. w 1 0 z.
- L 4 0 9 0
C- L U C'C-' A .
0. C 0
* ~~~0 S.~(.4IA
-
0
LU c
It...
1 0cL
20o
Q -
*,
M..
U 4
SC 4) IA 44rI--0.
C4 - 0C I 0 0
IA 4- 10 M.1~ .1 0 CO 01. CL 0 L 40.
0CL4 IA 4 *- L 04 .0 c 0 1 . 40.4)A
LV - IA 00 . 0 1.I 400 n0 -
D
- - )>l C A C ~ L0 0-. .. 0 a) .- IA
0W 0 IA 0
.J - 0 C A -0.. C n a0 1 00 C 0 w
~.1 L 0 . L1-50
CW
c0 C.,
0' L i.' - L.'
01 0 11, 0)_
w. T0 t -0(
0~~~ a)L
01 .. 4-I .0
0 D CO.(4 fj C 0- I-
c M..U
. ml 0 0- O . 04
2 I r 'a .0 .- 4.j .-A" 0

WVd) D
0 ~ L C M. 4 V ')
4j,~
3
ID 3: C
0
C14
140
. C .
C~
0
0.
CL... .1
M
L
0 M ~ 4w'
r- 30I-
-U
c-
04
C- CO 01 4 0. -0 0 >---C >. E
w4 0 u OVC a ao LO.44. L. .0 u
0E 01A.l a, ... =- C L- c03D w a >
u 0) a-4 0 j I 0 o z-.'- I 0 L- V
S A WA 4) tA o0.- -- 04- . L-r
IA ~ M' C~.. 4)
01
4
-.
C
4
f^-
V~ 4- 0
> 0
£ L
0IU 0-
C -' c..
IA 0.Q
4-' 41 044. (n 04a1. 44 .1jfa40 00
It ~C~ n>2- 0- LV 2 E
L~ ~ ~ s220.
) 0M a, u
%0
I 03 L,
L
0j
C
0in
4£ .
C) *. 01 C 4- EV 03 f4- 0 .CO
0~~ I A
E~. 0. l A-0 L C .
c1 >1.. 0
(11 ~ Z' C
w 0 9 2
L 6 W0 w u
0 .0 0. >4
L- 610I0
- 2
w. 4 - c*
000 0 .0 0L 0.W
w-
0 C CL' >. C 0 4)
C C~A 0 CL ) CL
Cx u-4 0 04 0
a#1 01 -C 1 >
L~I o a) - z 0 5 -4, > 4j- ; ;
E;
1 4-I 1 0 m "L0 0x
0'Q
uj ~ L0)0
V
.
a) C
13 a)0 0D
0 -0
w('
Q) 00w
0X- au4'0
4j.
C3
w'
u0
0 .
& -
m4 o
0u r=a ao 4 :
2D 06V L-
m CC0.I 0 1 L 0
0£ m- 000. 'D. C
LI t- v-1C IC D wA.-
4- 0. 1 C
m.
0 .
> M.- a)-0
00 . )
0 -v- CLI
4Q L 4 -
Cz 0 0 L
m 10 *30
C 4- ..- L C)
t 2- - 04-.QC. 0 I
0 w0.0 L. L- 00 L- 00 £4lo~0
LL 0S 04 r_ z o . - a. m 0 I mA.. -
I 0. 4- o o1 eC
C 4-
C> ~~~~~0
4' m .
C
CJ u. W0
0?
L0
34
> >
4'44-. -
40
~ 4s
V
US C 0l E u C 06M" 4, C4L A u2> C0
3.. 4) 0.) $- a, - U o--
L 0. X. x LL £4'
-0 E L (100 .0 4. I v a)- 04 I. E- Co
L ~ -0
L IA I U, V .- 4- 4-' 0 0
W 0C ,0 4.-C - . £ 0 0
- 400
4' 0) LL -
-
~ ' 0. 00>
-Oc4
- 0 0 0
4
L
ao ..a O0 4 I ' A 0 A 0 - -
0 L 0 44 0V 0
E
0.
E1~ 0.
CO A
2 ~ C ~ ~L 0
0.
C
4 I .A 4
A
Cl CU 4 V C . . 4 V 0 0 2
C I:
4.~ 1-5
I!, C Ua) Downloaded from http://www.everyspec.com
-If 0 0 w 0 .- -
- .L- 'V' w 4-'
I43
C 4w CL a) 'A 0
cc' C 43 C. 0:
l CL 'a 5 A toL> -
0.J'
-~~
.
C
.'
4. I (-A
~
LV
CD; aC ;
4
-
0. 0 Co:f4
.0.435
Lg .
w
0
CL''V
I
0 wU Im I 43> m
0 Go -A C 3 I
IM 0 43 43 ' - 1 ,.- . Q
- 4) 4 .4A m .* m U
V 10 - 4A4 0 w- V-0 c43 'U1
Z 0C .8- ;V 4, U30- a 4, 0(-.0 w m. o
C C C L a334 (M L.' 1 > 1 'U
m 0
-
-V L- 0 ,- c-u 'D a
E 04 ao - 43 CC
G.. 43 4 'U CCm C U 0
-- I c , 4' a). -, .- IM -u . i,• >u M ,-
" .. 43CC C14 A V Q0 'U C 4. * L 3 - uU. m.

0.- - C
C(A ,A A- w ap 40 0
A C . (U C 43 U m..- 'U A30 0N( C

0 a. C C 4u,
.. .- .(4( 4. 43.-
43... (A 0~ CL(3 ( C" U .1 -
%A m 1w. 0 'U C L '
0 o0 C 2- • .- C oU
10 4) 4 w C c w 4= C Q.
L -. (A. € q I .. C >U -'V C4.' .C 0 1 .. 0.
4.A I 'V W30 1 (AC.'
"0,0
43 4 CCC
01 -A 3 U C
LC1 Z.. M .- 7
oV( 16L. >, f406 0.C = - '0 40.430W 3-
CC1-- 'A *'z . C C 43 4.14 0. ( - C

-. C m. C CL - E 4 0 ;Z
-0. -(
0~A to 00 cU.- 'UV>( 4'rC '
' C. - IA. A 4 - 3 (..
- 43 v 4 (A *-
- ~ 3 U
43'
43
do 0 .'
4
0U24 w4.
vC 4
3
'U *'.
C
4.'C C a,1 . C >'U 4 3 C 0
A4kC' (1A41 C0 1P 43 3 - w3 . CX
Cw I~ U.
CI mO 0C.- 'U
Zi'I0. U U (A ..- ' > m 'U E24'-
01
U43
CLl
4. -5
Downloaded
I
from http://www.everyspec.com
it
0. 0
CU
0 l0
(U 0
=~~~ . . OG 1
0 MD4 0 a
= GD.- -ac-.-
ai 0 (w -SC
IV GL IV 40
u 4 0D~ - u L
u01 0.. L.
0in' 4
01 Qi 40 in - G M
u ZU 0 t. 0 -C C.
o
L 4- 4 1 4) d)Id
Li C41 a a) Go
WD 0
* 414 C 01
4GD
0i 1 -
C'
0V ~ *. 1 a0pG4-
c 0D
o 1-
0
V4 1-
LLa
C)4 0 0
GD Q# 1.UD 4
Q#M 'C0
C , 4-
00 It M0w
c4. 4j C 0M
GDM 40 1 GlDM I-L , m V

U jo0 1 jM -
-6
E0 al I-- X4 4-
r
'a- d) GD
W0 1
zD C . 4. G 04-
4- 'a1 41 - 'o GD C" ~ - L >1

L-3, 40 od 0 G -:3OL.
0.
3 1
00 C6 I%.
GD
lw-i
-
CD G i C OL
i ~z~- i
0 G . in 4 C 4# G n~.J G
x M- in CL GD GD GD inX 1
GDc 1 - n n 4 G 0
'0~ ~- ~~ LG C ' '

4-~ ~~~~~
0 CZ C l . D I 4 I -
C C I
0 0 -:
L- GD
C
0 10 1;
1- U.'-4.1-581
U c
0 0
0' 0 a
o o
a, cm,
• .- x u 0 > 4
0 do tol 0
- "C . '0. ..
"
. Q
-1
L-
IA b-- A C
00
MGo 4u L- o
IA- .. - 6 A 6
4A)
%A
01c
.C . .~I
6". 00 - ,-- '-- "C
41 1
cc
0 *.u-s € u+,. 0.-
, . 41 E 41'.. ,4 .
3 'A
V,, ..- M
IA0 M$ 0--
UU .00
04 0 -L a3M> M
'4m leO L- >41 4 9 - '
I.AJ .. i' 41 a41Ec CL. c .1 C &
0.1 o-4 tAA

a
'A- 0
0
'#A
1Q 1m )
)
.
06 041-C
.E E 1W IVWI -A I- , 4 W
01 n 4 %-- (P w1> *-
C 41 - k 0 41 -c 1 . A 41 1.
C 0) 4*.C 41E1 4 -
~~~
C~~~ 4
V1. D~ CM
4
C0 C& &A
A 4'J.
Q.
~A1
>I. v,4
-C
*
I4)~.C
-4V.
D..
0.
4- 4- '..0. CL .j1.-'
A 41. -0
41c 034 4. IA a 41. 4 M-4'5 4 1.'

o *O cA. L0 0A4. 6) 0 . 0- 0. 41 1. Uw 1- . C a3ll
m IA 5u C - u. 1 0 - 4. 4.) V14. A LI 44
x 0 4' . 0 0- )C . 0.1. ~ U .
.' ~~ ~ ~ . ~ 1-~ 4 C~.

1-59 1 U 4. A
C.-;
.
01.
UU IA C z.- I b
41 '- 4 - , v ' 4 4 '. ) .
C,1 at 4j -. 41 L'0
41. 4 W .A .0 M L 4 0 4 41-c m1. MI) IA
41 .0.4- 0- ' W1410 4
10 UC %- A -4 0 4V U
C0 4- 0 4
wA
o - U-.- ) 40 .0 m- c
- ~~ ap0'
~ A.
~ 04
~ ~ ~ f ~ m1
4-
IU..
0
I..-
m
'4'. I.-
Z1--LU
* ~~ ~ U 4 4-
) -14 6
OW4 A L 4 U 4 2'
C0 UL A '. * .l 4 U CI44
u 4wU w-. 4U 4
>A r- M>1 40 0A.
CA IA Un11 - L0 M1- 4 c AI.4
U ell 1. CL CLI 0.u' 4 U-4
I
00
44
06 C 01 A - r
0 C 4 uc
CVC
U
%0
4" C
P.- QLI > . > I
->z a)- M0Lz c a
C IU C1 a1 4-
Cj 4wIC 5 4
>- 0 0 .' 0 ~U 4 - - 4 4 .
w
-C-..
V)1
Ui 4
~05 A
0 C
4'-
4
1-AC
AC LC-
I'
C
.
C ~C U 'V 04-'.- 4 .1-CI60U 4-C 44 L

i V 4 .00
£ an~ -an 0 0.0 -W

41 m C- m Vc 04 w Co
f^ at 0 u 0 0 0.a 0 0
A3 C>, u c L-. =adL

'a OA 40 f^ 41. 3- 4 01 L C
W cn a.
414. L.4' (10a
4314an *4 M f L4
Q? .1.
'-3 0. 0 0'
A j C-
44 V m~ W) Li
0W >0 C3
3; "a m, - >. &A.
.a 43 0A
"V. I- - .j C
w-' 43 43 C 0 V4(
0z o 0 --
u. 0 . 0 Li
L 141 06 CK10 UK0a
wi -i 41 4 .6C 430- - 3 z
L
IA
L
4 .
4)
>04
K
)0
~4L0
c0
e
mi
a C 43
0i CC- 43
WaC .1 414C L. -4w
j 43 L M0 43 an 0. M0
0.. /
4A~. 41 'a. ~ i 0 9a
%a
0 )
a.c C l*
4 aM
w 4A
(/1 9
£~~ 4-c-- 4'
0 -CK
*0_ 01 0K ' 43 4
143~~a Ua a 33 3 do
4- CD vi
9 0 a
4w L 4.44 '
c- di do!'VUCl
c
0 M .0
v0
m
40 14
4
--
a0
£C
~ 44 a4
4.4C an > 40
v,4 -. an 9
06. 1* 0. .a- 434- . 43 0 9 4
C an 43 M an a -.- 3.4 0 a > L- E'- Li
uW W Li C C i..i 0 n'v gLia)-
anaA0 4 C L MU 0n40
*~4 4 n'' a C -L 0 430 3 -
- t mi * OZ *-.1 a- a0'4- "0 C 3
E ap-3-I W3 m
a-.-.0
1.. 40 443 -V L33- an 0' 43
*~~ 0 Vt --
IV kA. w a.-L
VV
W .4>. W
*a 4 43£ > .. - u3.-mC_
43 an
9.-41 Mn 00 a- .4. C Gal . n 0 - 4
4w L- 04 = -
Q-. w.. C 4. 4)0 c
.C0 Mn C- u- i ' 0. CvLi 4
>J 414'
0I
- .4
00~L
4432
433.-.)
0...-- 4W 9.'*
0~.-. ~ KCna
£Cc c VV
VnV C j. 33 ~ -(. A Q
Li -06-
0 m~
c~ C3
O a-p>
C 0X
0a -
4. 1-61
4^ I 40
.
CL. C
411
44- 0 u
wA - 4 -4W=1
-, LI
060 c 0 a r pu -
W 4AC >apCW u :
Cl.~~I ' 4) W
I 41"
1 0 .1 %A C I u '
W C 01 0. 4 L
1 M~ a# j- %A .0
-40. 1. L .
.0 44L 1- 1 a) C WA
413 IA LI. 0.1 C41r 41 1.
4J
C - >L4r Ce A41o LO) ' L
MW U--. LM.
%A g -;;4Z414E4141 4.. CA4 0..- ; =
,A
.'
a
U - -:; 0
3 .1
1" -Z 5
o w~ ,C
! 0-$ L J--
4 4
4
.. I ~01 -
0
M C ..- E~A4
414 41 41. 0
I..) 1 4.' 41.. - 1.W 1.. w 4 0 05 L. *1 3
r_ IW L 1 1 )> 1.g -
41, 1 414 0
W
IA
~C*mI.u
~ 04LI.0~~0
LI
~ -0)
~ 4A ~ M. 'C
C 0. . C
Q
L.
- L
U Cp LU M1 4 M cm W- 1> LL EL C 1
1 C CL 41
aL0IA >1.4 0 0 E063
LI a, 'n>0 I A 41 M o - I L I" *. 0 n14. Ok I

1.j 41 LI
"a 0.j U . %j >. O a U4w 4 1EA
C3 U 411 L >. 1 - LI 4'.
44
41.
Cll L u
,A CL I)
41 414 06 01
C ~ ~ 4 1
CL..-
041 0 'a
0
04
~~ 4 0 0
0.m4.uq.-)
)3 E 0>12
1P wCc m AQA 0D
41 A0
44 CA N. 00uA0j 1
... C- L. CL. 0 C
14.; J CI0 w. m
VIA 41 M 4
C'1)'z 4 '0. 0, C k D .. A( .
41C 4p ) : I.W"
41 4D I1 LI 4m1 mA -. u' 41 u C
m t
E z 44 C u' CL .01
40'~- -1.. L 4-Ec
XC I- 1. MA
IA 41 - C 4..- f 410
1.
0 41
IA.'
41C
U
L-
IA-
43
40
CE
IA UI
I.
IA.
~ 01
a0
m1
~
CD I M
IA 40 - - - L J 0 .' 0 40 31
41..u A ' - 4 .' 4 4 A C4 0 Y
1.Id ~ AX 0 -. C -r 3. ~ >
E A '' 1' . - 0 4 c 3
4 U 41* 41~ 4 1 A2A IA 1 0 4 0 0 4 .

4; 1 1 0 *-
-0U- CI Z 00 -. 0
> uL- L.-
.0 0
4)> w. Wn 4 . C
.L
C~~.
MCLc 06
4)
'a0e
. IV
%-
C
0
. 0
I.. 4
~ 0J
0~ 000c
40.
'A M L .
fi.c-~
c - 0
0 c .uc
0n
0 0 . V)
-. 0 4) 4
; a) L- C,
0 C 0
0
>.. c
Z m M - _C Ia c C
4) 40 0 0 00
0L- 0J 31 0
n4) 41 0
1 Jc Q?
~
0 doua
4
~ 0) ~ 4- . )M
)I
U
v a
r- I*4. d0o
44
00.
IV0
00. cc 4 0 4
M w0 3
41*A w 0 41 C
0) > 41 4)- 4
C,4 4- 4Mu0)
a 401
in IOk c
0# "0 W.C/ 'AUm

C V C %04 ( W I
40 M Z~ U4 c
- 0 *- 1W IA
-4 4 0 U
.00 L4 L 4) '
- U *. IV 'An
.L Cf > 0 o
in41 00 - u o 0
C O*.-
> CL IA
0 i 40 0 0
4-~ 3.
AA LnL.0
> w) 16.0.
0 u 0.. 04
W4 0
L- ,
Zi> D C/ 4> >/ IVC "0
C 4)0 0 kA' 'L C m .4- 0)

0)z 04 c zn mO c 1..- 0 4w UU C
T u
. 0 Cr- 4 in C C
0 a
C
L
4) W
O
a; V-
m
in
V)inC
IA ., I*CU)
3A I
>
m
....
.. t- 4.'r
)0
~ . 0a
I- >0 .cu~
0 4- CO
A w 0 L-440
004
inI0
IV).
4A 04
(- 0£'a
(A5 (n iz
>i-
40 0 Z 0 .- ~ *-i.
- .' 0 n
n
40 icA
m ~ o
-l .00 L. L 9. 0
.' .' CO -Li f O
C > in U*.L-
j - S 4i)L L C*-*-
44 c CM u. Cc> 0. 0 -.- 00 i 4
4~~~~~~ 0 . L 0 )'*
.' ~ ~ ~ in 4.
~ ~ ~ 4 ~ inL.V . . n.'r- C- 4 L . ) 4
4)
.' £ .
1-630 CC 4 C- 0 . 4
c
C.. C
-) -
0
I o CL
0 0
Ul
C v
L641
41 -
L&oJ
C.) 4A C
00 'a00 (j
U m m C 3 CL v4
> *1- M 00o4 j oc'
4J4= 41 >, a UM 0 a
ui %A4 0414) M l ol l 6
-C tI( 0> OlX
000
41 41&A '
0. kA n h > . '
-C U:, LI j -j4 0 L
Go c UL ) W M v M C '
L, I- apI4o t0 U 4- w E 4
U) aC - -04
c1 c M 4 00.U pW 41- -; . 04
m- r3C 1 M VC V %
0 -oj 4 j4
Li- 4 06U- CU04 5 U

41 w Is - ap 4 u ..- 4 C I 4T L. 41. W 0 'A 4
o C '4 ( C 4 ~ C >4 0 164. A - ~ 4

6..~4- -. ~. 44 ~ 4 A
0
1
~LLJ ~ 4
~ 41 C ~
14141~4.
C -
4.
L Z C OC~ 6-i413
IL 4
CL- C4-l 3 A 4 C (A>X0- 0
0~~~~~L~4 1-6C4 0 C A 0 40 L 14
0 aC. .0 0 N%
0
c C 4 c C C
0 c 0 0
W.. 0 4 4'' 4' L L-. W
z c 4- cCi C 41 41 C
40 0
U A 3 33
o 0 0 C 0 - -
-
0
L- - 0 -2 -
UW 0a) 4 ' 4' It3 4#
4' ~ .0 .0
4- '
(CNL .0 M0 0 0 .0
m 44 4 h - -
4' 4 ' ' 4
4' -4S - 0 .
0 S. S
4c
L.-
0 4'
41
4A
c 44
-c C
4'o
C 4'O *a
4j V4
UA C so 0 I'
4- *014..
4' >
C
> ~ u C
:I >-A
4- .- Q.. -4 W (A
WL 4.. 40 4'le~ (
.0 0 W4 do 0.
do
4'.
c '4- ~ >r Cr 4..)M V)
4-
E ' ~~~t
' 4 ~ IV44
4 M'4#L '4
4) C Cr;4' C
C w' w o U.*0LL ..- 4' L.) C- M4
4.C4'
m~
V4 c-4
0'(
C
04 04 4 ~ 0 '
4-. 10 4A
0p C c4
440 04'
L.00 Id a 04 Co 0-
3S L. 6. . 4A.
- 4- (# 0 4'-
M'e)4A ' .
Lc
44 VUC4 4
44-a 04 4w C 0 D
4-0 40 ai
4 4 0' 04
C z. m ' 0' ~ 40
4. 1-65
degradation of latent manufacturing defects which are not detected

during device screening. Even the corrosion and mechanical fracture
failure mechanisms can usually be traced to a latent manufacturing
defect.
The overwhelming area of concern for long-term non-operating reliabil-

ity of microcircuit devices is the influence of chemical contaminants
introduced either from Lie environment of else during fabrication. As
stated in section 4.1, moisture, in particular, is very critical be-
cause of the corrosion processes it induces directly and because the
presence of moisture activates such contaminants as residual chlorine
ions. It may be the single most important factor to long-term non-
operating reliability of microelectronic devices. Because the amount
of water required to degrade the materials within a device package is
very small, as little as one molecular layer according to some sources
(Refs. 3,4), all possible steps should be taken to eliminate it within
microelectronic devices. The use of hermetically sealed devices,
therefore, is highly recommended. Plastic encapsulated devices are
non-hermetic and, as a general rule, should not be used for dormant
designs.
The following sections provide further details to assist in the selec-

tion of microcircuit devices for equipment that will spend a signifi-
cant part of its service life in a non-operating state. Sections
4.1.3.1 and 4.1.3.2 cover hermetic and non-hermetic microcircuits
respectively. Section 4.1.3.3 covers the application of VHSIC/VLSI
technology. Finally, microcircuit stress screening is described in
section 4.1.3.4.
4.1.3.1 Characteristics of Hermetic Microcircuits
The primary concern for long-term non-operating reliability is the

sensitivity of electronic components, bonds, metallization, etc. to
the presence of water vapor in the device package. Sources of water
vapor include the sealed-in atmosphere, epoxies used for die attach-
ment, polymer conformal coatings, leakage through polymer seals. leak-
age through minimal leaks for devices that Pass the screen tests, and
leakage due to seal failures resulting from the operating and non-
operating environments (e.g. thermal cycling).
In an effort to protect microcircuit devices from the effects of am-

bient atmospheric moisture and other contaminants, microcircuit de-
vices are encapsulated in glass, ceramic, or metal hermetic packages.
While hermeticity implies a perfect package such that the electronic
devices are protected from the detrimental effects of atmospheric con-
stituents, a perfect hermetic seal with a zero leak rate is impossible
to achieve.
4.1-66
Accepting a finite leak rate implies that there will be a gaseous ex-
change process taking place during non-operating periods until atrno-
spheric equilibrium is reached. The time it takes to reach equili-
brium is dependent upon the leak rate. For example, a 1 cm3 package
with a leak rate of 5xlO - 6 atm cm 3 sec -1 will reach equilibrium
with the atmosphere within days. (Ref. 3) Such a leak rate is intoler-
able for long-term non-operating microcircuits in an uncontrolled,
humid environment. The consequences of moisture within a microcircuit
package are a number of failure mechanisms including corrosion, ionic
drift, reactions with phosphorous-doped passiv.tion glasses, and the
formation of migrated gold resistive shorts.
Even if the hermetic seal is perfect (i.e. zero leak rate), moisture
may still be present within the device package as a result of en-
trapped water vapor during sealing or the outgassing of water vapor
from the materials enclosed within the device package (e.g. epoxies,
polymer conformal coats, etc.). To minimize the amount of moisture
introduced during fabrication, Livesay (Ref. 3) recommends that all
hermetically packaged microcircuits be vacuum baked at 150 degrees C
for at least 4 hours and sealed in dry nitrogen without ever being
exposed to moisture containing gases such as air. The moisture con-
tent of the nitrogen sealing chamber must also be less than 100 ppm.
For a more detailed discussion of hermeticity and the impact of mois-

ture on microcircuit device reliability, the interested reader is re-
ferred to reference 3. The following sections provide additional part
selection information concerning hermetic monolithic microcircuits
(Section 4.1.3.1.1) and hermetic hybrid microcircuits (Section
4.1.3.1.2).
4.1.3.1.1 Hermetic Monolithic Microcircuits
Monolithic refers to a device fabricated on a single chip of semicon-

ductor material. The two primary technologies for device construction
are bipolar and MOS (metal oxide semiconductor). The term bipolar
refers to the fact that carriers of both polarities, + (holes) and -
(electrons), are necessary for device operation. In contrast, MOS
devices are unipolar since only one type of carrier is used. For P
channel MOS, the carriers are holes. For N channel MOS, the carriers
are electrons.
Another distinction arises from the differing location of active

regions. Bipolar devices are considered "bulk" devices because the
active region, the base, is located several microns beneath the sur-
face between the emitter and collector. MOS devices, ou the other
hand, are "surface effect" devices because the active region consists
4.1-67
of a channel that is induced at the silicon/silicon dioxide inter-

face. This makes MOS device more sensitive to surface contamination
than bipolar devices.
References 18 and 26 break down device construction intn seven major

areas: Bulk materiel and diffusion, oxide, metallization, glassiva-
tion, die bonding, chip connections, and Packaging characteristics.
For a more detailed discussion of each of these construction areas the
reader should refer to these references.
The failure. mechanisms affecting semiconductors are generally the same

regardless of the device type. They are also the same regardless of
whether the device is in an operating or dormant state. The differ-
ence is the frequency at which individual failure mechanisms occur.
In general, the mechanisms can be grouped into three categories (Ref.
18):
1. Mechanisms for which failure occurrence is independent of the

application environment.
2. Mechanisms for which failure occurrence is dependent on the

application environment.
3. Mechanisms for which the failure occurrence is time-related and

environment dependent.
The mechanisms in group 1 are undetected defects, such as improper

diffusions, oxide pinholes, etc., which passed through device screen-
ing. The rate of occurrence of these mechanisms is the same regard-
less of whether the device is applied in an operating or non-operating
environment. The only difference is the time at which the defect is
detected.
The mechanisms in group 2 are defects which do not fail the device
immediately but require an environmental factor to trigger the fail-
ure. Examples of such defects are bond and metallization defects
which progress to failure as a result of temperature or mechanical
stresses.
The mechanisms in group 3 are similar to group 2 except that they are
more time dependent. Examples of such defects are metal migration,
intermetallic compound formations, corrosion, etc.
Since the mechanisms in grouPs 2 and 3 are dependent upon the environ-
ment, they will occur at different rates depending upon whether the
de.rice is operatiug or dormant. In most cases the non-operating en-
vironment will be more benign than the oPerating environment and the
rate of occurrence of these mechanisms will be much lower.
4.1-68
In considering operating and non-operating failure rates, the complex-

ity of the device is also important. A greater number of circuits in
a given die area increases the temperature to which the devices are
subjected and also requires greater process control in production.
Diffusions, metallization patterns, and interconnections are very
critical in a high density device.
Table 4.1.3.1.1-1 presents a summary of microcircuit failure mecha-

nisms. This table identifies the points in the fabrication process at
which a defect is introduced, the resulting failure mechanisms and
modes, and the failure detection methods useful for screening the
defects.
Since the failure mechanisms for microcircuit devices are the same
regardless of whether they are operating or non-operating, the same
part selection and control guidelines apply when selecting microcir-
cuits for both operating and dormant designs. Optimum part selection
should be aimed at assuring zero latent defects Present in the parts
selected. It is impossible, however, to obtain parts with zero latent
defects and tradeoffs will have to be made between cost, reliability,
and ather factors. Parts most thoroughly screened for defects are
those chosen from a MIL-M-38510 QPL vendor. Therefore, whenever pos-
sible, parts for use in military and aerospace systems that will ex-
perience significant periods of dormancy should be chosen from a
MIL-M-38510 QPL vendor.
Considering that compromises and tradeoffs are inevitable, Parts may

sometimes have to be chosen from lower classifications. For example,
there are many microcircuits that are procured to MIL-STD-S83 Class S
or Class B screening that are not qualified to a MIL-M-38SIO QPL be-
cause their manufacturing process did not conform to MIL-M-38510 in-
process controls. These devices could be selected as second alterna-
tives to QPL parts. There also exist various lower quality grades of
devices such as "vendor classes" and commercial grades. Since such
lower grade devices frequently contain contamination and latent de-
fects, they should not be used for dormant designs without additional
screening to eliminate these defects (See Section 4.1.3.4).
The need to provide the most capability in an equipment design often

necessitates the use of new devices and technologies. Where possible,
devices that have demonstrated successful age histories and inherently
good storage characteristics should be used. Where there is a lack of
data concerning non-operating failure modes, mechanisms, and rates,
devices must be carefully evaluated for potential non-operating fail-
ure modes and mechanisms prior to use. This evaluation should include
accelerated aging tests (See Section 5) and material compatibility
studies.
4.1-69
Table 4.1.3.1.1-1. Microcircuit Defects/Screens (Ref. 1)
Point at Which a
Reliability-Influencing Failure Mechanism
I Failure Mode Failure Detection
Variable is Introduced .. _ _ Method
Dislocations and stacking Degradation of junction Initial electrical test;

faults characteristics operational-life tests
Nonuniform resistivity Unpredictable component Initial electrical tec.;

values Joperational-life tests
Irregular surface Improper electrical Initial electrical test;

Slice performance and/or operational-life tests
Preparation shorts, opens. etc.
Cracks, chio,, scratches Opens. possible snorts Initial electrical tests;

(general handling damage) in subsequent metalli- visual (pre-cap); thermal
zation cycling
Contamination Degradation of junction Visual (pre-cap); thermal

characteristics cycling; high temperature
i storage; reverse bias
Cracks and pin holes Electrical breakdown in High-temperature storage;
oxide layer between i thermal cycling; high-
metallizatlon and sub- voltage test; operating-
strate; shorts caused by life test; visual (pro-
Passivation faulty oxide diffusion cap)
mask
Nonuniform thickness Low breakdown and in- High-temperature storage;

creased leakage in the thermal cycling; high-
oxide layer voltage test; operating-
life test; visual (pre-
cap)
Scratches, nicks, blemish- Opens and/or shorts iVisual (pre-cap); initial

es in the photo mask electrical test
MasKing Misalignment Opens and/or shorts Visual (pre-cap); initial

Ielectrical test
Irregularities in photo- Performance degradation Visual (pre-cap); initial

resist patterns (line caused by parameter !electrical test
Iwidths, spaces, pinholes) drift, opens, or shorts i
Improper removal of oxide Opens and/or shorts or i Visual (pre-cap); initial

intermittents electrical test; opera-
I tional-life test
Etchin; ,I L
Undercutting Shorts and/or opens in Visual (pro-cap); initial
metallization electrical test
4.1-70
Table 4.1.3.1.1-1. Microcircuit Defects/Screens (Ref. 1) (Continued)
Point at Which a
Reliability-Influencing Failure Mechanism Failure Mode Failure Detection
Variable is Introduced Method
Spotting (etch splash) Potential shorts Visual (pre-cap); thermal

Ftching cycling; high-temperature
storage; operational-life
test
Contamination (photo- Low breakdown; increased Visual (pre-cap); initial

resist, chemical residue) leakage electrical test; thermal
cycling; high-temperature
storage; operational-life
test
Diffusions Improper control of doping Performance degradation High-temperature storage;

profiles resulting from unstable thermal cycling; opera-
and faulty passive and tional life test; initial
active components electrical test
Scratched or smeared Opens, near opens, Visual, (pre-cap);

metallization (handling shorts, near shorts thermal cycling; opera-
damage) tional-life test
Thin metallizaton to Opens and/or high- Initial electrical test;

insufficient deposition resistance intracon- operational-life test;
or oxide steps nections thermal cycling
Corrosion (chemical Opens in metallization Visual (pro-cap); high-

residue) temperature storage;
Metal 1izati on thermal cycling; opera-
tional-life test
Misalignment and contam-- High contact resistance Visual (pre-cap); initial

inated contact areas or opens electrical test; high-
temperature storage;
thermal cycling; opera-
tiona1-life test
Improper alloying Open metallization, Initial electrical test,

temperature or time poor adhesion, or high-temperature storage;
shorts thermal cycling; opera-
tional life tests
Improper die separation Opens and potential Visual (pre-cap); thermal

Oie resulting in cracked or opens cycling; vibration;
Separation chipped dice mechanical shock; thermal
shock
4.1-71
Point at Which a
Voids between header and Performance degradation X-ray; operational-life;-
die caused by overheating acceleration, mechanical

shock; vibration
Overspreading and/or Shorts or intermittent Visual (pre-cap); X-ray;

Die loose particles of shorts monitored vibration;
Bonding eutectic solder monitored shock
Poor die-to-header bond Cracked or lifted die Visual (pre-cap); ac-

shorts celeration; hock.
vibration
Material mismatch Lifted or cracked die Thermal cycling; high-

temperature storage;
acceleration
Overbonding and under- Wire weakened and breaks Acceleration; shock;

bonding or is intermittent; vibration
lifted bond; open
Material incompatibility Lifted lead bond Thermal cycling; high-

or contaminated bonding temperature storage;
Wire pad acceleration, shock,
Bonding vibration
Plague formation Open bonds High-temperature storage;

thermal cycling; accel-
oration, shock. vibration
Insufficient bonding pad Opens or shorted bonds Operational-life test;

area or spacings acceleration. shock,
vibration; visual (pre-

cap)
Improper bonding procedure Opens, s!'irts, or in- Visual (pre-cap); initial I

or control termittent operation electrical test; ac-
celoration, shock, vibra-
tion
Improper bond alignment Opens and/or shorts Visual (pre-cap); initial

electrical test
Cracked or chipped die Open Visual (pre-cap); high-

tmoerature storage;
ther-nal cycling; acceler-
at--. shock vibration
4.1-72
Point at Which a
Excessive loops, sags or Shorts to case, sub- Visual (pro-cap); X-ray;

lead length strate, or other leads acceleration, shock,
vibration
Wire Nicks, cuts, and abrasions Broken leads causing Visual (pro-cap); ac-
Bonding on leads opens or shorts coloration shock vibra-
(Conti nued) tion
Unremoved pigtails Shorts or intermittent Visual (pro-cap) acceler--

shorts ation, shock, vibration;
X-ray
Poor hermetic seal Performance degradation; Leak tests

shorts or opens caused
by chemical corrosion
or moisture
Incorrect atmosphere Performance degradation Operational-life test;

sealed in package caused by inversion and reverse bias, high-
chanroling temperature storage,
thermal cycling
Broken or bent external Open circuit Visual; lead fatigue

leads tests
Final Cracks, voids in kovar-to- Shorts and/or opens in Leak test; electrical
Seal glass seals the metallization caused test; high-temperature
by a leak storage, thermal cycling;
high-voltage test
Electrolytic growth of Intermittent shorts Low-vol tage shorts

metals or metallic com-
pounds across glass seals
between leads and metal
case
Loose conducting particles Intermittent shorts Acceleration; monitored

in package vibration; X-ray; moni-
tored shock
Improper marking Completely inoperative Electrical tests
4.1-73
The following presents some additional lessons learned in the use of

microcircuits in dormant designs:
" Do not use microcircuits that contain nichrome-deposited resis-

tors. They are susceptable to electrolysis with trace amounts of
moisture and they are subject to deterioration and failure at con-
tact interfaces with aluminum conductors due to interdiffusion.
" Do not use microcircuits with gold-aluminum systems that have not
been determined to be free of "purple plague" problems. During
bonding and subsequent packaging operations, gold-aluminum wire
bonds form intermetallic phases. Reliable gold-aluminum bonds can
be made. It is necessary to minimize the total mass of aluminum
available for diffusion and to keeP to a minimum the cumulative
time-temperature product experienced by the device in both fabri-
cation and use. Design limits for using gold-aluminum bonds can
be found in reference 27.
" Analysis indicates that a single metal should be used for the con-
tact metallization and interconnection interface. The all-alumi-
num system shows a definitely more reliable non-operating capabil-
ity than the aluminum metallization/gold wire system. (Ref. 26)
* Do not seal chlorine or other halogen-conttining materials within

microcircuit devices.
" Polymers used within microcircuits should be simple hydrocarbons

or compounds of carbon, hydrogen, and oxygen. Nitrogen containing
polymers should be considered with skepticism. Ammonia and amines
from curing epoxy resins can react with ions to form complexes and
with acids or water to form cations. (Ref. 3)
4.1.3.1.2 Hermetic Hybrid Microcircuits
A hybrid microcircuit consists of a combination of solid state active

circuit components (integrated circuits or discrete), thin or thick
film-deposited passive elements, and other discrete components, inter-
connected by film patterns on one or more substrates in a single pack-
age to perform one or more circuit functions. Hybrid microcircuits
are most commonly classified as either thin or thick film. In a thin
film circuit, the mean free path of the current carriers (usually
electrons) is comparable in length to the thickness of the film,
usually in the range of a few thousands Angstroms. The film circuit
is either vapor-deposited, vacuum-evaporated, or sputtered. In a
thick film circuit, the film circuit is dePosited by screen printing
(or spraying) with subsequent air drying and high temperature firing
steps. Thick film thickness overlays the range of thin film thickness
and extends to approximately 2.5 mils. (Ref. 18)
4.1-74
A hybrid microcircuit has the characteristics of both a monolithic

microcircuit and a printed circuit board. It is analogous to a
printed circuit board in that the hybrid is a functional assembly
consisting of a variety of solid state and passive devices connected
on a substrate (analogous to the printed circuit board), with electri-
cal paths defined by conductor Patterns on the substrate. It is simi-
lar to a monolithic microcircuit in that the hybrid assembly is con-
tained in a single package which is similar in size, appearance, and
function to a conventional monolithic microcircuit. In the most
general sense, hybrids are not as much a device type as a packaging
technology with provisions for multiple devices of various types and
the required conductor Patterns ro connect these devices in a func-
tional manner. They are usually specified, procured, and used in a
manner similar to monolithic microcircuits.
Hybrids are often used in specialized applications where off-the-shelf

monolithic microcircuits which meet the design requirements are not
available, and the cost of a custom monolithic microcircuit is prohi-
bitive. Hybrid technology also offers the designer a means of imple-
menting specialized functions involving very tight tolerances, thermal
constraints, or other critical parameters which may not be available
in conventional designs. For example, hybrid fabrication techniques
permit a high degree of thermal coupling between adjacent semiconduc-
tor chips not easily achievable in conventional printed circuit board
designs making thermal compensation circuitry relatively easy in
hybrids. (Ref. 28)
Hybrid failure mechanisms include all those listed for monolithic de-
vices plus those that are unique to the hybrid technology. Hyb-id
devices exhibit problems as a result of the number of different mate-
rials used in one package, the number of interconnections and bonds,
the amount of processing with the chance of error or inclusion of con-
taxinants, and hermetic sealing of a larger package. Materials must
be carefully selected and the processing must be carefully controlled
to prevent thermal mismatches between materials; leaching, diffusion,
and migration of materials; intermetallic compound formations; and
corrosion. (Ref. 18)
Tables 4.1.3.1.2-1 and 4.1.3.1.2-2 summarize the mechanisms unique to

thick and thin film hybrid devices. Most of the mechanisms described
are detectable during processing and screening.
In thick film devices, the faulty substrate bond or cracked substrate

which is not detected during processing will be accelerated to failure
by mechanical vibration and shock. The frequency of this failure,
whether the device is operating or dormant, is dependent upon the fre-
quency and level of mechanical vibraLion and shock experienced by the
4.1-75
LA ': - 0 0 0 0 0 C'
0.
W a) f- L
6 L. C.
1.-
M-4~ 0. Q.. 0. 0. 0L m4(I
>~~~~ 41> v4u "
C~~
S
..
L
4- ~m
aLA.-
44
. - L
L0
L) 0 .LLa CL L LL L wL~ -LL 0
0 a)
Q u
C C
Go4 4 4)
0 M 0m 0 m-
0 04 4 0 0
. 0 0 L 01.. a
-. L 0. 0. 1 0A.
- Q '4-
a) 0) di 0, 0, 0w 0a
0
a64 0. 0. 0. 4- 0. 4- 6 L- a. 06
I.: L . 030 0 0 0 0 0 0 0 0
W
U. W . L
m r' - -
c 064-
L- 0 L1 q?
C . 1 In
40 4 4 4 (1..
>. .ft L. zC a I-
U6 %A U
Ud 4) 0 uC ee
IIU
U 47 0
0'"0 C' M
-
CL 1. C1 00 0 -M
0'F. 0 L *. 0 C- v ..
c C :3 -' 0 rU 0 U 0M 0
'AV (-.- 0 1-- 4,
'A z L 0 Q uL L
L , 1.. . V a, Z- 4
r-. e ) : L m 'L
Lj a Q. LA
W' -- L- 04-.m-E W 0-o
00 L 0 - .- 0 A
0. LALA U LAL>- 0. LV 4 C O
w CD U 06L 0 C -0 u
L * -- 4I '
C'. U -' 4- ~ - 0 -U 0o
C, A 04 L .C 0> U L
u' 230 (1 ' 04 1 ,x 0..

C
'a .-
. C A>
04
L
0L .
C
I
I *.)
A C
A
C
L ~ '4 ~L .
L
Ca, LA ON
L 0 L CO -- 3 U 4 ' ..
0 I- A~A...- OIL A.. . 4-0 0.' .U V0 0
0~~L *.CD A L C 4' 0 C LA '- C 4 ~ ~ 4 OL
LAn
4C 6
u -04- -4 4t 4 41 V
IT 0~ 0~ 02 00000.0
0
41
!-0
(1) 1 M m-
E4 L.r-r
Vi Vi ui u.
mi 01 ) 0) 4) al Q) w
C C 2 : X:
z w 0 M 16 6. 0) ea l8 .
0 42 0 0c~ Ui
6. 06 U
4 4 41-
0 C)
U 0.) I)) .CVII.. 0 ~0) 4.
a) C)
IA Vi 41I
W1 0 W r_ -
IV 1 0 0? C 0o
- 60. ..- 0) 4P0
0L - 0. 1-~ a1) w41 6
0 0l c0 z. C S.
0) -0 L C4- 0. (m
-0~ C 3 Vi ) C E -. a
j :3 .- Vi IC 2 0) 0 U-
L. Q- 'a 0 C.- 0 0 - tU
o 4*. 0Vi.. L4- C_ 0m0
ViM 6! L- 0. 0i1.
C 4. C W1 L- 0 0 01 0. 0
0) 00.. 0, C_ 0 - 0
C~~~.)~ I).- 6
I-. I * . 0
31C It. ..- >0)L C 4) 0 C.
4I- C. C .- C 6 0 00.
u CL0 1
(' 4 r- -- Ai 0.1
W Ai X .0 40 . 0.
Q
0.0 Vi 0 >0 W i a 6.
L. 0- 6.
C 01 0- C -0 LO. 0 3 E '4- .
Vi C 01
.0~~~ % ~ ~- ~ c. c,
01
C
0- 0
Vi 01 C
CC0 0
w. 0
Vi m C 40 :3
vi _rC C C0 10
w k c 61 -n 0JL
4 V7
00
u 413
a)*0 10 a) c (
L~. = w 0 L. 03
4) 0
xA0'- L 0. 0 .
C3L Lx
:3 M, a
03 03
do v. 4m 4#-
03 A d)
o- V) A
ON 4 L. L Q.
4- 0.
'a 00. '
:3 C.
E 03
C r-'
0- W 'Ac07
'3 w It : '
43 C 1" 03 x L
'3 .- L. %-
a3 0 ~ -- z3 03
(I03-
.ez
u8 L
)
(V
L
IL '30 >
'0 .03 -- a) (D c
An 03 '0 4V ' Z. o
LA Is3
V)3 ) VI >0.
- ' 0. 4 030 03
-C -
iC I-£ £ *
I. 4 I
C0 0 0
.-ap
A. 03 - 0 04 An
(i Co CLE CJ C --
.0 ~~ ~ ~ ~ 4 1-A. 7'08 C '

m C C
Ui u1 U L)
Cl CL. UL L .
ai
0 LJ
Z z~
c do
o L
Li L
-) 0
CIL
c c
IA C IA
w)
L
cA E
IIn
~
*C M
U- c
CC
0 .0
4A1
CC
0. CD
L--
M .0
CC
*p I C 0
a,
-,
4. -7
device, usually as a result of the transportation and handling of the

equipment in which the device is used.
The failure mechanisms for thick film resistors include processing

failures not detected during device screening, defects that are ac-
celerated by high temperature or thermal cycling, and corrosion. The
rate of occurrence of the last two groups of defects are dependent
upon factors in the non-operating environment.
The chip element failure mechanisms in thick film devices are the same
as those experienced by monolithic devices except that bonding mate-
rials and processes may be diffezent.
The number of conductors and interconnections in the hybrid device can

lead to shorted conductors, faulty bonds, etc. Defects of this type
are accelerated to failure by thermal/mechanical stresses. Since
silver migration depends uPon a high current density, its rate of
occurrence will be decelerated during non-operating periods.
Aluminum-Silver bonds are highly susceptable to corrosion if moisture

is present. The corrosion is accelerated at high temperature and in
the presence of halogen contaminants such as NaCl, etc. Significant
degradation can occur at room temperature and moderate humidity
levels. Prior thermal aging in the 100 to 200 degree C range in-
creases the susceptability to degradation. The use of an Ns40H-Hzo
ultrasonic bath to remove chlorine and vacuum bakes to minimize mois-
ture and other volatile contaminants before sealing has been shown to
markedly reduce Aluminum-Silver bond degredation. (Ref. 29)
The failure mechanisms for thin film devices are similar to those ex-
perienced by thick film devices. The failure mechanisms that are
unique to thin film devices are those associated with the element
films. The rate of occurrence of these defects is dependent upon the
environment, with many of them being accelerated to failure by thermal
stresses.
4.1.3.2 General Characteristics of Non-Hermetic Microcircuits
As the name implies, non-hermetic microcircuits are devices that are

encapsulated in packages that do not provide a hermetic seal. Common
non-hermetic packaging methods include plastic encapsulation and
polymer sealing of open cavity packages. Because they do not provide
a hermetic seal, non-hermetic packages w5l not protect microcircuit
devices from the effects of atmosPheric moisture and other contami-
nants. As stated in section 4.1.3, the overwhelming area of concern
for long-term non-operating reliability of microcircuit devices is the
influence of chemical contaminants, particularly moisture. Since they
4. 1-%
are not protected from these contaminants, non-hermetic microcircuits

should, as a general rule, be avoided for dormant designs.
There may be situations, however, when the use of non-hermetic devices

are contemplated for dormant designs because of cost or other factors
(e.g. shock resistance) favoring their use. To assist in these deci-
sions the following information is provided. Section 4.1.3.2.1 ad-
dresses plastic encapsulated devices (PEDs) and Section 4.1.3.2.z ad-
dresses polymer-sealed devices.
4.1.3.2.1 Plastic Encapsulated Devices
In contrast to hermetic and polymer-sealed devices which are sealed in

an open cavity, PEDs are completely enclosed (except for the leads) in
a dielectric material by transfer molding or other encapsulating
method. Materials commonly used for encapsulation include epoxies,
silicones, and phenolics. The advantages of plastic encapsulation
include lower cost, mechanical shock and vibration resistance, freedom
from loose particles, and smaller package sizes. While new techniques
and materials continue to be developed, all plastics contain moisture
and are permeable to moisture to some degree. Figure 4.1.3.2.1-1
illustrates the relative effectiveness of various sealant materials as
moisture barriers.
In addition to being transported through the package material, mois-

ture also enters PEDs along imperfect plastic/lead bonds. While
several plastics have extremely low moisture vapor transmission rates
through the bulk material, the interface between the metal and plastic
is an abrupt discontinuity in composition, expansion coefficient, and
structure. The forces holding the surfaces together are weak with
little or no chemical bonding taking place making moisture barrier
tightness largely due to the bulk mass of the plastic compressing the
lead. Cracks can develop as a result of temperature cycling and lead
bending during fabrication and handling prior to fabrication. (Ref. 30)
Passivation coatings are widely used to improve the performance and

reliability of microcircuits in both hermetic and non-hermetic pack-
ages. Passivation coatings in direct contact with the device chip are
classified as primary and passivation coatings that are separated from
the device chip by an underlying dielectric layer are classified as
secondary. The function of the primary passivation layer is to pro-
vide good dielectric properties, low surface recombination velocity,
controlled immobile charge density, and device stability. The func-
tion of the secondary passivation layer is to provide additional sta-
bility in various ambients, in both production and use, and to serve
as getter, impurity barrier, or mechanical shield. As a barrier
4.1-81
PE RMEABI L!TY (g/cm-s-torr)
10610-8 10-10 10, 12 10.14 10.16
SI LICONES
-2
EPOXIES
FLUORO-CARBONS
o) .4
GLASSES
-5 METALS
-61
MIN HR DAY MO YR 10 100
YR YR
TIME FOR PACKAGE INTERIOR TO REACH 50% OF EXTERIOR HUMIDITY
Figure 4.1.3.2.1-1. Effectiveness of Sealant Materials (.Ref. 31)
4. 1-82
against ambient effects, passivation layers provide only partial pro-

tection. For example, while silicon-nitride devices may be protected
against alkali penetration into an underlying silicon dioxide layer,
they may become susceptible to potential buildup at insulator-
insulator interfaces or at the ambient-upper dielectric interface.
For a more detailed discussion of passivation coatings, the interested
reader is referred to reference 33.
The thermal properties of epoxies and silicones are also a limiting

factor in the use of PEDs. While most military systems require device
operating temperatures of 55° C to +125 0 C, most PEDs are only rated for
O° C to 70° C. The limiting thermal characteristics are the glass tran-
sition temperature (e.g. 225° C for silicon and 124*C for epoxy
Novolac) and the thermal coefficient of expansion. If the temperature
of the plastic device is not kept below the glass transition tempera-
ture, severe changes in the nature of the Plastic material will occur
along with sharp increases in the coefficient of expansion (Ref. 31).
Mismatches in the thermal coefficients of expansion between the encap-
sulating material and the other device materials such as the lead
frame and bond wires can lead to bond failures under thermal cycling
conditions.
The sensitivity of PSDs to salt environments is dependent upon the

encapsulating material. Tests performed in Panama by the US Army
(Ref. 32) have showu that Epoxy Novalac devices are relatively unaf-
fected by a salt environment while silicone devices are very suscep-
tible to NaCI penetration resulting in contact metallization deterior-
ation in devices with aluminum metallization and gold dendritic growth
in devices with gold metallization.
One situation where PEDs have shown a definite advantage over hermetic
devices is in applications with high levels of mechanical shock and
vibration (e.g. cannon launch). In an encapsulated device the entire
device (i.e. chip, lead bonds, etc.) is encased in plastic. As a re-
sult of this, only the most severe external mechanical stresses ad-
versely affect them. In addition, because the encapsulating material
is in direct contact with the chip, failures due to particulate con-
tamination, a source of failure in open cavity packages difficult to
detect during screening, is eliminated.
In a 1984 study (Ref. 31) of the main factors governing the relative
reliability and applicability of plastic commercial, hermetic commer-
cial, and JAN-qualified microcircuits, the Reliability Analysis Center
at the Rome Air Development Center, Griffiss AFB, presented the fol-
lowing conclusions and reconendations concerning the application of
PEDs in military equipment:
4.1-83
* Certain families of devices have experienced severe problems in

plastic encapsulation. These include low power dissipation
devices, linear devices, and recent technologies. Bipolar
TTL-type devices have performed best in plastic encapsulation.
" The long-term, high-stress, and non-operating reliability of PEDs

has not been determined aud cannot be assumed.
* Screening programs differ from manufacturer to manufacturer and

should be examined to assure that the procedures address known PED
failure mechanisms, including intermittent internal connections,
chip surface effects, corrosion, contamination and those resulting
from marginal manufacturing anomalies.
" The Purchase cost savings associated with commercial PEDs does not
include any custom or unique Parts. Large cost savings are for
industry-standard part types, most sPecifically SSI/MSI and memory
Parts.
" Special screens above and beyond the manufacturer's standard in-
house screens will upset the normal high-volime flow of production
and delivery cycle and should be expected to cause significant
cost increases.
" When selecting manufacturer's, select those in continuous volume

production of mature products for which they are willing to make a
reliability commitment. Manufacturer's selected should conduct an
in-house screening program and demonstrate high reliability when
subjected to qualification tests.
" The quality of PEDs has been known to vary from manufacturer to
manufacturer and even between lots from the same manufacturer.
Manufacturers' products should be subjected to periodic qualifica-
tion tests for approval.
" When employing PEDs, avoid part types which are known to experi-
ence reliability problems in plastic encapsulation.
" Fox LCt , ie perfc rmaLe, PEDs must be used in an environment

with strict temperature and humidity controls. High atmospheric
salt concentration may also present a problem.
" When employing PEDs it is important to be assured that environ-

mental controls are exercised during all stages of device life
including manufacturer, shipping, storage, and operation.
4.1-84
4.1.3.2.2 Polymer-Sealed Devices
Polymer-sealed devices are similar to hermetically sealed devices ex-

cept that the package is sealed with a polymer-sealing compound in
place of the glass, metal, or ceramic used for hermetic seals. Be-
cause the seals are non-hermetic, these devices are not protected from
the effects of atmospheric contaminants such as moisture. They have
all of the disadvantages of PEDs concerning moisture and atmospheric
contamination sensitivity with none of the PEDs advantages in the
areas of mechanical shock and vibration resistance and freedom from
loose particle contamination.
In an investigation of the reliability of plastic encapsulated and

polymer-sealed hybrid microcircuits (Ref. 34), Mazenko and Hakim made
the following conclusions:
* Polymer sealed, metal packaged hybrids are useful only in low

humidity environments.
* Polymer seals deteriorate badly in moist environments. During

testing the package sealing perimeter deteriorated so badly that
the device lids fell off through routine handling and testing.
e Transfer molded devices survive moist environments better than

polymer-sealed devices.
Based upon these findings, polymer-sealed microcircuits are not recom-

mended for dormant equipment designs.
4.1.3.3 VLSI/VH$IC Microcircuits
VLSI and VHSIC devices are taxing the constraints of existing fabrica-
tion processes and equipments. Geometries are exceeding the limits of
conventional photolithographic techniques requiring the use of new
techniques such as electron beam, ultraviolet, and x-ray lithography.
Also a problem is the chemical etching process used to define the
metallization pattern on the surface of the chip, where process con-
trol has limited line widths and spacings. Limitations in the conven-
tional photolithographic mask generation and registration process has
also resulted in the use of new diffusion techniques.
The multiple layer structure for VHSIC devices presents several addi-
tional concerns and potential failure mechanisms. Step metallization
integrity is aggravated with increasing step count and becomes most
compromised in th uppermost layer. Due to smaller metallization
widths and thicknesses the problem of detection becomes important, yet
visual inspection of all metallization is not feasible in multilayer
4.1-85
structures. Further, MOS devices employing polysilicon in their fab-

rication are subject to a unique mechanism. The integrity of the di-
electric which interfaces with the polysilicon is in question due to
the tendency of polysilicon to nucleate into large grains at the
interface, causing disturbances at this interface (Ref. 28).
Interconnects are an important aspect of reliability in VLSI and VHSIC

devices. As more devices are integrated on a chip, a higher percent-
age of the chips area is occupied by interconnects which are prone to
failure from electromigration, whisker formation, faulty ohmic con-
tacts, corrosion, and masking faults. This makes these devices more
susceptible to moisture related problems.
A technique being considered for electrical connection bonds on VHSIC

devices is solder bumps. This technique must be carefully evaluated
in dormant designs since it has been noted that they are susceptible
to fatigue from mechanical stress. Electrical bonds are a prime reli-
ability concern since not only are the bonds smaller, but there are a
much larger number of them. (Ref. 28)
Contamination and latent defects introduced during the fabrication

process are a primary determinant of long-term non-operating reliabil-
ity in all microcircuits. This is especially true for VLSI and VHSIC
devices. The dry etching process which is to be widely used in VHSIC
fabrication uses a chlorine based gas (Ref. 28). Any residual
chlorine remaining in the device package will have catastrophic
effects on long-term non-operating reliability.
Stray particle contamination during fabrication will also have adverse

consequences as a result of the extreme device scaling. The thin
oxide layers of VLSI and VHSIC devices ace approaching the diameter of
the minimum size particle that can be effectively removed in clean
room air filtration. A high density of 50-100 angstrom particles may
exist in a class 10 clean room (Ref. z8). Such particles can cause
problems such as oxide defects, crystaline structure defects, and
stacking faults.
The use of new materials may also impact device non-operating relia-
bility. The use of polyimide as a dielectric between metallization
layers will improve step coverage and alpha particle absorption, how-
ever, it is hydroscopic and will expand in the presence of water
vapor, setting up mechanical stresses that can lead to device failure.
The chip size of VLSI and VHSIC devices is a factor in long-term non-
operating reliability. Previous studies of power transistor dice
bonded to copper headers have shown that such dice are subject to
cracked die failure mechanisms if the silicon die is made larger than
4.1-86
250 mil square. This is due to the mismatch between the linear expan-
sion coefficients of the die and the copper header during thermal
cycling. VHSIC chips of 400 mil square and larger are currently being
built and bonded to non-copper substrates. The oerformance of these
devices in a thermal cycling environment must be carefully evaluated
prior to using them in dormant applications.
4.1.3.4 Microcircuit Stress Screening
Experience has shown that most failures that occur during dormancy are
of the same basic kind as those found in the operating mode and are
the result of latent manufacturing defects. Screening of devices is,
therefore, one of the most effective ways of improving the reliability
of microcircuits in both the operating and non-operating environment.
Stress screening entails a combination of visual inspections, electri-

cal tests, and the application of stress tests to a device with the
purpose for revealing inherent weaknesses (and thus incipient fail-
ures) of the device without destroying the integrity of the device.
Inferior devices will therefore fail and superior devices will pass.
Table 4.1.3.1.]-l identifies microcircuit failure modes and mechanisms

and the detection methods useful for screening them out. Table
4.1.3.4-1 compares the various screening methods in terms of cost
effectiveness and effectiveness in screening the defects noted.
It is not feasible, however, to construct a comprehensive screening

procedure that is able to remove all types of weak or defective micro-
circuits from a given population of devices. The potential failure
mcchanisms are numerous and device construction details are too di-
verse to realistically search out all weaknesses. There are a limited
number of ways to observe and exercise the delicate composite struc-
ture of a microcircuit without undue damage to good devices. Some
screen tests will rapidly extend a device population into the wear-out
phase if the stress parameters are to severe. Latent defects can also
be created during the screening procedure. This must be avoided.
Some of the defects important to dormant microcircuits are not re-

vealed in reasonable times with current test methods. For example, it
is necessary to destructively puncture a package to accurately deter-
mine if the moisture level is high. Furthermore, moisture i luced
failure modes often require longer time periods to develop than are
practical with current screening stresses.
4.1-87
Table 4.1.3.4-1. Co.pzsrisor, of Screening Methods (Ref. 1)
Screen Defects , e~ivnss Cost Coments

C
Internal visual Lead dress Good Inexpensive This is a mandatory screen for
Metallization to moderate high-reliability devices. Cost
Oxide will depend upon the depth of
Particle the visual inspection.
Die Bond
Wire Bond
Contaminaton
Corrosion
Substrate
Infrared Design (thermal) Very good Expensive For use in design evaluation only.
X-ray Die Bond Excellent Moderate The advantage of this screen is

Lead Dress (gold) Good that the die-to-header bond can
Particle Good be examined and some inspection
Manufacturing Good can be performed after encapsula-
(gross errors) tion. However, some materials are
Seal Good transparent to X-rays (i.e., Al
Package Good and S1) and the cost may be as
Containation Good high as six times that of visual
inspection, depending upon the
complexity of the test system.
High temperature Electrtical (sta- Good Very This is a highly desirable screen.
storage bility) inexpensive
Metal I izati on
Bulk Silicon
Corrosion
Temperature Package Good Very This screen may be one of the most
cycling Seal inexpensive effective for aluminum lead sys-
Die Bond tems.
Wire Bond
Cracked substrate
Thermal mismatch
Thermal shock Package Good Inexpensive This screen is similar to temper-

Seal ature cycling but induces higher
Die Bond stress levels.
Wire Bond
Cracked substrate
Thermal mismatch
Constant acceler- Lead Dress Good Moderate At 20,000 G stress levels, the ef-
ation Die Bond fectiveness of this screen for
Wire Bond aluminum is questionable.
Cracked substrate
4.1-88
Table 4.1.3.4-1. Comparison of Screening Methods (Ref. 1) (Ccn .inued)
Screen Defects Effectiveness Cost ,nen ..
Shock Lead Dress Poor Moderate The drop-shock test is considerea

(unmonitored) inferior to constant acceleration.
However, the pnuemopactor shock
test may be more effective. Shock
tests may be destructive.
Shock Particles Poor Expensive Visual or X-ray inspection is pre-

(monitored) Intermittent short Fair erred for particle detection.
Intermittent open Fair
Vibration Lead Dress Poor Expensive This test may be destructive. Ex-
fatigue Package cept for working hardening, it is
Die Bond %ithout merit.
Wire Bond
Cracked substrate
Vibration van- Package Fair Expensive

able frequency Die Bond
(unmonitored) Wire Bond
Subs trate
Vibration vari- Particles Fair Very The effectiveness of this screen

able frequency Lead Dress Good expensive for detecting particles is part-
(monitored) Intermittent open Good dependent.
Random vibration Package Good Expensive The. % a better screen than VVF
(unmonitored) Die Bond (unmonitored) especially for
Wire Bond space-launch equipment, but it is
Substrate more expensive.
Random vibration Particles Fair Very This is one of the most expensive
(monitored) Lead Dress Good Expensive sc-eans; *hen combined with only
Intermittent Open Good fair effectiveness for particle
detection, it is not recommended
cxcept in very special situations.
Helium leak test Package Good Moderate 'his screen is effective for de-
Seals tecting leaks in the range of
10- 8 to 10 Atm. cc/sec.
Radiflo leak teat Package Good Moderate This screen is effectice for leaks
Seals in the range of 108 to 1012
Attm cc/sec.
Nitrogen bomb test Package Good Inexpensive This test is effective for 9-
Seals tecting leaks between the gross-
and-fine-leak detection ranges.
4.1-89
Table 4.1.3.4-1. Comparison of Screening Methods (Ref. 1) (Continued)
Screen Defects Effectiveness Cost Comments
Gross-leak tst Pacakge Good Inexpensive Effectiveness is volume-dependent.

Seals Detects leaks greater than 10 Atm.
cclsQc.
High-voltage Oxide Good Inexpensive Effectiveness is fabrication

test dependent.
Isolation Lead Dress Fair Inexpensive

resi stance Metal I i zati on
Contamination
Intermittent Metallization Good Expensive Probably no botter than ac oper-

operation life Bulk Silicon ating life.
Oxide
Inversion/
Channel ing
Disign
Parameter Drift
Contamination
Ac operating Metalization Very good Expensive

life Bulk Silicon
Oxide
Inversicn/
Channel i ng
Design
Parameter
Contami nati on
DC operating Essentially the Good Expensive No mechanisms are activated that

life same as intermit- could not be better activated by
tent life. ac life tests
Wih-temperature Same as ac Excellent Very Temperature acts to iccelerate

ac operating operating life expensive failure mechanisms. This is
life probably the most expensive screen
and one of the most effective.
High-temperature Inversion/ Poor Expensive

reverse-bias Channeling
4.1-90
4.1.4 Discrete Semiconductors
The failure modes and mechanisms for discrete semiconductors are the
same as those identified in section 4.1.3 for microcircuits. As was
true for microcircuits, most discrete semiconductor failures that oc-
cur during dormancy are of the same basic kind as those found in the
operating mode and are the result of the further degredation of latent
manufacturing defects which were not detected during device screen-
ing. The discussions and tables in section 4.1.3 regarding monolithic
microcircuits are directly applicable to discrete semiconductors and
will not be repeated here. Only differences between discrete semicon-
ductors and microcircuits will be discussed in the following.
The difference between discrete transistors and microcircuits lies in

the physical size, number, and complexity of manufacturing processes.
Compared to the average integrated circuit, a transistor is a rela-
tively simple device, with fewer number of junctions and leads. The
distances between different parts of the device are larger and the
manufacturing processes are fewer and simpler. Although the failure
mechanisms Are similar to those in microcircuits, the differences be-
tween microcircuits and discrete semiconductors tend to shift their
emphasis. Bulk defects are more common due to the larger blocks of
silicon required thus increasing the probability of crystal defects.
Crystal defects collect mobilized contaminants resulting in breakdown,
leakage, gain failures and, in high power devices, thermal runaway.
Diffusion defects are not as critical due to the lower density of dif-
fusions. Oxide and metailization defects are not as prc :ounced as in
microcircuits because the metallization patterns are mi. ' simpler.
(Ref. 18)
A large percentage of transistor failures are the result of die and

wire bonding defects. Contamination, both ambient aDd within the
mnat'erial, is also a major problem in transistors. (Ref. 18)
The failure mechanisms for diodes are similar to those found in tran-
sistors. In addition to those mechanisms in Table 4.1.3.1.1-1, alloy
bonded and point contact diodes can develop intermetallic compounds at
the junction, however, this has not been identified as a severe prob-
lem. Loss of contact is also a potential problem is spring loaded
contacts. This happens when the contact material loses its compres-
sion strength or by slipping off the contact. For these reasons, the
use of spring loaded diodes with spring loaded contacts is not recom-
mended for dormant applications. (Ref. 18)
4.1--91
4.1.5 Inductive (Magnetic) Devices
Inductive devices refer to a wide variety of components dependent upon

a number of turns of wire designed to oppose a change in current flow
in an electric circuit, to produce magnetic flux or to react mechani-
cally to a changing magnetic flux. There are three categories of in-
ductive devices covered in this section: transformers, coils, and
inductive filters. Standard magnetic devices are listed in MIL-STD-
1286 (Ref. 39) and are selected from among devices covered by appli-
cable MIL specifications. Most transformers, inductors, and coils
used in military aerospace applications are custom designed for mini.
mum size and widti. to fit available space in compact equipment. As a
result, relativel, little standardization of electrical parameters has
occurred. However, a family of standard sizes should be used whenever
possible, even if the electrical characteristics of the,new devices
are unique. (Ref 1)
Inductive devices create a magnetic field derived from the Physical

construction of the device and its electrical input.
A -Ioil is simply several turns of wire around a supporting structure.

SiLce inductive operation depends on the physical arrangements of the
wire, provisions are taken to prevent contact between adjacent wire
turns, This is accomplished by insulating the wire. Potting the
entire device also provides insulation and provides additional
mechanical strength. (Ref 18)
A transformeL is a device consisting of two or more coils coupled

together by magnetic induction. Its main components are input and
output coils and a core around which the coils are wound. As in the
case of the simple coil, the wire turns in the input and output coils
must be insulated from each other. (Ref 18)
An inductive filter is a network designed to selectively block or

allow passage of certain frequencies or bands of frequencies. It is
comprised of several coils in network form mounted n a supporting
structure such as a printed circuit board or any other suitable means.
In its basic form the RF choke can be considered the simplest form of
inductive filter. (Ref 18)
The selection of a transformer, inductor, or coil depends upon many

factors. These factors include: function, construction, circuit ap-
plication, operating temperature, altitude, type of mounting, environ-
mental conditions, size and weight, life expectancy, and reliability.
After the preliminary selection of a transformer, inductor, or coil
has been made, the appropriate specifications and military standards
should be examined to vf-rify that the item is capable of performing
4.1-92
satisfactorily in the intended application and that parameters impor-

tant to the new application are controlled to the degree necessary.
(Ref 35)
Shorts are usually the result of the breakdown of insulation. During

dcrmancy, insulation breakdown is the result of chemical changes and
deterioration accelerated by temperature. humid.ty and reactions with
atmosphere gases. (Ref 1)
Therefore, celection of insulating materials is of paramount impor-

tance. Inductive devices must function satisfactorily under a variety
of environmental conditions to which they are exposed. Life testing
of insulation used for inductive devices has identified that: vibra-
tion is not a significant factor in transformer life, humidity is also
not a major factor (humidity effects reduce the life of transformers
by a slow change in resistivity between windings), thermal cycling
appears to bring about the largest decrease in life expectancy. (Ref
37 )
Thermal aging mechanisms for insulating materials in complex and

varies with material composition and service conditions. Typical
mechanisms of degradation include: (Ref 15)
* Lcss of volatile constituents such as low molecular weight com-

pounds present in the original insulating materials arising from
processes of manufacture or compounding, or formed as products of
the aging process.
* Oxidation, which may result in cross-linking and embrittlement or

the production of volatile materials.
* Continuing molecular polymerization, which may result in an ini-

tial increase in physical and electric strength, but subsequently
causes decreased flexibility, embrittlement, and failure under
mechanical stress.
* Hydrolytic cleavage by reaction with moisture under the influence

of heat and other factors leading to molecular breakdown.
* Chemical breakdown of constituents with formation of degradation

products typified by the liberation of hydrochloric acid which in
turn, may catalyze further decomposition.
The degradative processes classified above may be retarded or effec-

tively suppressed by the addition of stabilizing materials. Oxidative
breakdown may be controlled by addition of antioxidants. Liberation
of hydrogen chloride, which is specific for vinyl chloride polymers
4.1-93
and is reported to exert a catalytic influence on further decomposi-

tion, is minimized by addition of basic acid absorbing materials.
(Ref 15)
Different insulating materials react to a different degree and in dif-

ferent ways to the aging processes. Their thermal performance cannot,
therefore, be reliability predicted from the chemical composition of
the material. The thermal aging process may at first lead to in-
creased strength due to loss of volatiles and additional polymeriza-
tion, followed by loss of the strength and embrittlement. In other
cases, especially in closed systems, the buildup of degradation prod-
ucts may lead to softening. (Ref 15)
The thermal capability of insulating materials is an important factor

in determining the performance of electronic equipment. Interest,
therefore, centers on accelerated life tests to determine appropriate
limiting temperatures and fields of usefulness for various insulating
materials and systems. (Ref 15) The designer should refer to Refer-
ence 15 for a more detailed discussion of aging mechanisms in insula-
ting materials,
The most common insulation system of power traasformers generally con-

sists of cellulosic paper and pressboard impregnated with mineral
oil. The life of such a system is governed by the oxidation and ther-
mal decomposition of the cellulose, oxidation of the oil and caLalytic
effects of the deaoadation products. The replacement of the paper
insulation with more stable materials offers the best hope for im-
piovement in the life of insulation systems. (Ref 36)
There are two basic problems hindering the use of new materials.
First, there is a need for improved correlation between the results of
life tests conducted on the material itself and its performance in the
insulation system of the transformer. Second, there is no reliable
method available for measuring the life of a complete transformer
without testing a full size unit. (Ref 36)
Opens in inductive devices result from the breaking of the fine wind-
ing wire. Unless caused by mechanical shock or stresses, opens are
normally associated with manufacturing problems such as stress in re-
lief loops, wire nicks, and soldering of lead wires to the windings.
(Ref 1 Failure modes for inductive devices are accelerated by en-
vironmen.al conditions. The effects of various environments in opera-
tional or dormant applications is summarized in Table 4.1.5-1.
4.1-94
CL 0.
0'.
11 Lu u 4-' U
4- 0..
GD 4-
0 6 C
I- 0 c.
U 0. 0.
00
*0 -W
LW u
>D 4-
4
"0 ) C 10
co CL 10 0. 0,C C 4
4.' W- O5
.C .- v v
0D 0 &-0
0..
C 04 4- 0
0D 0. -0+-4
V) 4A 04. 0
Ow '4 0 .0 GD
.4 u 0 0-
SW GD n-.- LU U G
z4-0 Q;4
GD 0.
GD~~4 0 1-95.
4.1.6 Relays
There are two basic types of relays: electromechanical and solid-

state. In an electromechanical device, mechanical linkages are actu-
ated by an electrical coil which joins or separates electrically con-
ductive mechanical contacts. Solid-state relays are divided into two
categories: pure solid-state and hybrids. Pure solid-state relays
accomplish their switching functions by means of an arrangement of
semiconductors and passive devices. Hybrid relays are a combination
of solid-state circuitry and electromechanical linkages in a single
package. The solid-state circuit drives the electromechanical relay
to perform the switching function. (Ref 1)
These relay types are further subdivided by functional classifica-

tions, which include: Balanced armature, balanced-force, low-level,
magnetic latching, and polarized types. Balanced armature type relays
have the armature pivoted at the center of the mass, which balances
the armaLure with respect to static and dynamic external forces. This
typp works better than other types in shock and vibration environments.
(Ref 35)
4.1.6.1 Snlid-St,.te Relays
This section addresses pure solid-state relays (i.e., relays with no

mechanical components). Solid-state relays are primarily semiconduc-
tor devices, and are particularly well suited for dormant applica-
tions. These devices are generally completely sealed units and they
undergo minimal environmental degradation.
Sections 4.1.3 ands 4.1.4 of this text are applicable to solid-state

relays.
4.1.6.2 Electromechanical Relays
This section addresses electromechanical devices, it includes any type

of relay with mechanical components. The storage life of hermetically
sealed relays, with proper materials and processes employed to elimi-
nate outgassing, is in excess of 10 years. (Ref 24) Designers should
consult Reference 11 to identify potential material outgassing effects.
The chief problem in electromechanical relays, whether in a dormant or

operational configuration is contamination. Relay contamination can
occur if the construction of the relay is not matched to its applied
environment.
Corrosion of metal parts used in electromechanical relays will occur

in humid environments. Dust, sand, or other contaminants will damage,
4.1-96
block, or degrade relay contacts if contacts are not properly pro-

tected. Fungus growth on coils will degrade performance.
Mechanical shock generally causes failure in the form of the armature

failing to hold during the shock.
An electromechanical relay should not be used when a solid-state relay

can satisfy the application, or when the application can be tailored
to enable the use of a solid-state relay. When it is necessary to
employ electromechanical relays, redundancy should be employed when-
ever high reliabilty is required. (Ref. 24)
The analysis conducted to define the redundancy configuration should

not only consider the circuit application, but also the characteris-
tics and failure mechanisms of the specific candidate relay. Failure
mechanisms should be explored very thoroughly as the different types
of mechanical arrangements combined with manufacturing variability can
produce unexpected modes such is partial, intermittent, closing of the
relay. The parallel redundant configuration is the most commonly used
approach since it protects against the open failure mode resulting
from particle contamination. Quad redundancy may be necessary if the
potential metallic contamination is large compared with the contact
gap, as has been experienced with some relays. One form of quad re-
dundancy consists of two coils (two relays) with two contacts within
each relay.
When the use of electromechanical relays is unavoidable, the guide-

lines presented herein will increase the reliability of these devices
in dormant applications. Section 4.1.6.2.1 presents design guidelines
and section 4.1.6.2.2 presents process control guidelines to electro-
mechanical relays.
4.1.6.2.1 DesignGuidelines
The design guidelines presented are both application independent and

application dependent. The guidelines are subdivided under these two
groupings. (Rif 24)
Application Independent
e Relays should be hermetically sealed for minimum leakage. Elec-

tron beam welding should be used. This type of seal has the least
leakage and introduces practically zero contamination.
* Do not use a getter. If cleaning processes and material selection

require a getter, the relay is not a high reliability relay. In
addition, getters are a source of particulate contamination.
4.1-97
" Use compression glass for pin seal. To minimize susceptibility to

cracking, undercut the front and back side of header. This will
puddle the glass and prevent miniscule creep.
" Do not use built-in devices for ac rectification (coil) or induced

voltage suppression. These additional components decrease relia-
bility through particle contamination and outgassing.
" Use balance clapper armature design. Although other designs ap-
pear to have higher reliability possibilities (suspended, rotary,
diaphragm), each exhibits a weakness which results in less relia-
bility (shock/vibration, particulate susceptibility, hydroc3rbon
outgassing).
Application Dependent
" For dry circuit (i.e., open circuit voltage ! 0.03V, current :
200mA) to intermediate loads, use gold plating on contacts. The
softness and inertness of gold reduces susceptibility to film
contamination, which is the primary problem in this load range.
* For intermediate to high loads, use palladium for contact material.

The hardness, high melting point and relatively inert properties
reduce the susceptibility to material transfer, erosion and carbon
generation.
• The tradeoff involved with backfill gas is: inert gas versus a gas
with some oxygen for lubrication. Unfortunately, oxygen enhances
polymer formation in addition to being an oxidizer. Small-ultra-
clean-single-cavity-relays should use oxygen. Small dual-chamber
relays should use oxygen in the contact chamber. These relays are
more susceptible to sticking from self-adhesion due to the small
motor forces and the noble metals use on contacts. There is some
evidence that oxygen in larger relays is effective in reducing
wear and particle contamination.
4.1.6.2.2 Process Control Guidelines
* Coil wire lubricant is a major cause of hydrocarbon contamina-

tion. Wind the coil, using dry wire. This can be done by immers-
ing, the spool in a bath of trichlorethylene and wiping the strand
during winding.
" Coil wire should be pretensioned during winding operations at just

below the elastic limit. Scramble wound is preferred, as space
factor is just under layer wound, and it is easy.
4.1-98
" All subassembly components (armature, frame, can, contacts,

header, pins, bobbin, springs and pins) should be cleaned prior to
assembly. The recommended cleaning process is ultrasonic tri-
chlorethylene, GN2 drying, ultrasonic alcohol, gaseous nitrogen
(GN2 ) drying, ultrasonic distilled water followed by GN 2
drying.
" All handling/moving of components and subassemblies should be done

with the components sealed in a plastic bag.
" All relays should be subjected to a high temperature, high vacuum

degassing process. The vacuum should be less than 5 mm Hg, and
the temperature at 200 degrees C for 4 hours minimum. At least
one GN 2 purge cycle is recommended. The hydrocarbon outgassing
which condenses on the chamber walls should be cleaned after each
batch processing.
" Plating is a major source of metallic contaminants. Avoid plat-

ing, if possible. The plating material electrolyte, temperature
time, etc., should be very precisely-controlled. In addition,
samples should be analyzed from every plating lot to assure uni-
formity, adhesion and porosity of plating.
4.1.7 Switche.
Switches used in electronic equipment can be grouped into three basic

types: rotary, non-rotary, and sensing.
Guidelines for switches in dormant applications are presented in sec-

tion 4.1.7.1. Environmental effects and guidelines are presented in
section 4.1.7.2.
4.1.7.1 Guidelines
A major problem with switches is contamination, both particles and

contaminant films on contacts. The storage life of switches exceeds
10 years when the devices are hermetically sealed and internal non-
metallic materials are eliminated, or if very stable materials are
used. The following guidelines divided into design guidelines, proc-
ess control guidelines, test guidelines, and application guidelines
will assist tl.e designer in selecting and designing switch applica-
tions for dormancy. (Ref 24)
The reader should also consult MIL-STD-1132, Selectio" and Use of

Switches, (Reference 67) and the applicable MIL specs for detailed
specifications for the various Military Specification switches.
4.1-99
4.1.7.1.1 Design Guidelines
" Use hermetically sealed, metal enclosed switches. Accomplish the

hermetic seal by electron beam welding. Soldering or non-metallic
sealers introduce both particulate and gaseous contaminants.
" Use either a metal bellows or metal diaphragm for the seal around
the activation mechanism. Do not use an elastomer seal. This
metal seal should also be accomplished by electron-beam welding.
" Contacts should be alloys of nickel, silver or palladium for loads

greater than 500 milliamperes. For loads less than 500 milli-
amperes, the contacts should be gold plated to minimize oxide
films and maintain low contact resistance.
" The minimum thickness for gold plating is 0.000100 inch. Thinner
platings may burn through. Thicker platings are neither necessary
nor cost effective.
" External electrical terminals should be sealed with glass with a

matched thermal expansion coefficient for strength, dielectric 2nd
gas seal qualities.
" Solder hook external contacts are required to provide stress re-
lief to the glass seal. Plug-in units are not recommended because
the glass seal may be stressed.
" Getter devices should not be used. They may break or move and
interfere with switch operation. Switches must be clean enough
withoit a getter or they are not reliable parts
" Minimize the use of non-metallic materials because they are prone
to outgas in low pressure applications, and they are a source of
particulate contamination (consult reference 11 for outgassing
effects). Non-metallic materials also have lower mechanical
strength, are more temperature sensitive, and do not have the long
term stability of metals.
4.1.7.1.2 Process Control Guidelines
" Switch assembly should take place in a clean room environment to

preclude particulate contamination; which is a major problem en-
countered with switches in dormant applications.
" The transfer of switch components between assembly stations should

be accomplished in sealed plastic bags to reduce contamination.
.1-100
" Vacuum bake-out of switches should be conducted at 200 degrees C,

1 mm pressure for a minimum of four hours to prevent subsequent
contaminating films of non-metallic materials. Some components
may require as much as 16 hours, two cycles. One purge cycle with
gaseous nitrogen (GN2) is recommended as a minimum.
" The out-gassing hole in the switch enclosure should be sealed in

the backfill chamber to prevent contamination.
" The back-fill and bake out chamber should be cleaned after each
operation to remove hydrocarbon (oil) condensate from chamber
walls.
" During assembly, each mechanism piece should be checked against

the installation print for proper installation. A go/no-go test
should be made for every clearance specified, with no hindrance of
motion.
" Plating of piece parts should be avoided. Where unavoidable, the

plating process (temperature, raw material purity, bath cleanli-
ness, voltage stability, etc.) should be continually monitored and
controlled. In addition, destructive parts analyses should be
conducted on 5% of the piece parts to verify plating adherence,
absence of cracks, and the achievement of a fine-grained structure.
* Cleaning baths should be constantly filtered and recirculated with

approximately a 20% replacement after each cleaning process.
" Incoming material inspection should be on a sample basis as a

minimum and should verify materials, dimensions and tolerances as
well as workmanship. Materials should be stored in contamination
protected containers or bags.
4.1.7.1.3 Test Guidelines
The reader should consult section 4.5 of this document, "Testability

Design", for testability design guidelines for dormant systems. The
following design guidelines are presented to insure reliability in
switches used in dormant applications.
" Qualification and acceptance (screening) testing should be accomp-

lished using MIL-S-8805 as a minimum. Additional program peculiar
requirements should be added as required.
" For acceptance testing of production hardware, the above specifi-

cations are inadequate in certain areas and, therefore, must be
supplemented with the following inspections and tests.
4.1-101
* Radiographic inspection on 100% of the devices should be accomp-

lished in accordance with more comprehensive inspection criteria,
such as MSFC-STD-355 to detect contamination, bent parts, and mis-
aligned parts.
* For switches, an operational run-in (2000 cycles of operation at

-65 degrees C, 2000 cycles at +125 degrees C and 500 cycles at +25
degrees C) should be conducted as a 100% screen test. The voltage
drop across the contacts should be monitored during cycling and
the electrical characteristics after the test.
* Seal tests should be performed on 100% of the devices to verify

the seal integrity.
* A sinewave vibration test of 30 g's 52000 Hz should be performed

as a 100% screening test with contact resistances monitored during
the test to insure the absence of contact chatter and transfer.
During vibration, switches should be monitored for a short to case
condition.
9 Accelerated testing is basically accomplished by the run-in test-

ing described above. However, one variant which should be con-
sidered for selective application is the exposure of the nonoper-
ating switch to high temperature followed by operation under a
very light load to establish that outgassing of inorganic material
has not produced a contamination film on the contacts. This ac-
celerated test approach is aimed at a final verification of the
adequacy of the bakeout processes. The reader should consult sec-
tion 5.0, "Evaluation Methods", to identify more comprehensive
evaluation guidelines for dormant systems.
4.1.7.2 Environmental Effects/Guidelines
Many types of enclosures are available to protect switches from exter-

nal conditions, particularly moisture and contaminants. Such classi-
fications include the following: open, sealed, enclosed, environment-
ally sealed (resilient), and hermetically sealed. With the open con-
strtction switch, no effort is made to protect the switch or its parts
fron the external environment. This type of switch is not recommended
for application in dormant. equipment. Degradation of contacts due to
storage in severe environments may cause malfunction of equipment when
it is required to operated. The enclosed switch is one in which the
contacts are enclosed in a closed case made of plastic or metal and
plastic. Plastic and metal cores do not create an effective hermetic
seal and are subject to degradation from moisture and low pressure,
therefore, plastic and metal sealed switches are not recommended for
4.1-102
application in dormant equipment that may be subjected to these en--

vironments. The environmentally sealed (resilient) switch is -n a
completely sealed case where any portion of the seal is a resilient
material such as a gasket or a seal in the bushing of a panel mounted
switch. This type of switch is recommended for application in moist,
or sand and dust environments, but the gaskets may be subject to
degradation in low pressure applications. The hermetically sealed
switch is made airtight by a sealing process which involves fusing or
soldering and does not use a gasket. The hermetically sealed enclo-
sure offers the greatest protection against such elements as moisture
and other degradative contaminants, and avoids potential outgassing of
plastic materials in low pressure applications.
The following lists the major environmental considerations that the

designer must consider when selecting a switch for dormant
applications:
High Temperature
o Chemical action to which switches are subjected to is accelerated

by high temperatures. Insulation resistance between the switch
contacts and ground decreases as the temperature increases. In
high resistance circuits with 3 megaohms in parallel with a cir-
cuit impedance of 1 megaohm, the circuit impedance will change to
d point where operational tailure of the equipment may occur.
High temperature may also affect the insulation from the stand-
point of voltage breakdown due to a change in dielectric
strength. Also, the increased speed of corrosion of contacts and
switching mechanism is affected by high temperature.
Low Temperature
* Exposure to low temperature may cause certain materials of a

switch to contract, thus causing cracking, permitting moisture or
other foreign contaminants to enter the switch causing potential
short circuits, voltage breakdown, or corona.
Moisture
* Moisture in the dielectric will decrease the dielectric strength,

life, and insulation resistance and cause corrosion by increasing
the galvanic action between dissimilar metals. In general,
switches should be hermetically sealed. and/or the use of ac-
cessories such as boots, "0" rings, or diaphragms placed over
sw;itch openings, is recommended to eliminate moisture entry.
4.1-103
Low Pressure
* With a decrease of pressure, the spacings required to prevent

flash-over increase substantially. Small switches, because of
their very close contact spacings, are partially susceptible to
malfunction at high altitudes. To compensate for increased arcing
at high altitudes, users must derate the current rating given by
the manufacturer. To compensate for increased flash-over, the
user must derate voltage ratings. No problems are anticipated for
dormant switches, other than possible outgassing. To mitigate the
potential effects of outgassing, the user should avoid switches
wiLh non-metallic (organic) components. For more detailed out-
gassing information, the reader should consult Reference II.
Shock, Vibration, Acceleration, and Gun Fire
9 The primary degradative aspects of acceleration and gun fire for

dormant switches is shock and vibration experienced during these
environments. Switches should be selected which meet shock and
vibration requirements (see applicable MIL specifications). Those
siich-s with specific contact chatter requirements are suitable
for application in low frequency and shock environments. High
frequency of vibration will determine the effects of fatigue and
resonance on the mechanical construction of the switch contact
elements. When contact bounce at the time of closing of switch
contacts is important, this requirement should be a consideration
to be specified.
Tempetature Cycling
* Variations of temperature must be considered, as moisture conden-

sation within the switch could develop. In choosing a switch fo:
a wide range of temperature, the entire temperature range must be
carefully considered rather than one extreme. To mitigate the
effects of moisture condensation switches should be hermetically
sealed. If the contacts are in a sealed enclosure, the fill gas
should include oxygen. (Ref. 67)
Sand and Dust
* A combination of sand and dust and small amounts of moisture will

materially increase the possibility of voltage breakdown of the
insulation between closely spaced terminals.
Nuclear Radiation
e Nuclear radiation affects switches primarily by damage to organic

insulating materials. Some microswitches suffer damage to the
4.1-104
plastic cases and actuators at gamma-ray exposures as low as 4 to

6 x 108 ergs g-1 (C) or integrated neutron fluxes at 1015 n cm-2
(0.7 Mev). (Ref 16)
" A secondary problem, induced radioactivity, arises if handling and

maintenance of the units are required. This induced radioactivity
may constitute a radiation hazard to personnel. (Ref 16)
* Contacts of gold or silver should be avoided, if possible, because

they have high neutron-absorption cross sections as compared witn
aluminum or some tin alloys.
" The major radiation resistance factor for switches is insulation

materials. The designer should consult reference 16, Effects of
Radiatior on Materials and Components, for a detailed examination
of insu'. on materials and radiation effects.
4.1.8 CQR'_0 to S
Connectors come in a wide variety of shapes, sizes, densities, support

hardware, electrical and mechanical characteristics, and environment
withstanding properties. The reader should refer to MIL-STD-1353 (Ref
70) and the applicable MIL specifications fcc detailed application and
specification information. This section will present guidelines for
the selecticn and control of connectors for dormant applications.
The proner determination by the user of which of the numerous connec-

tor types and associated hardware is most suitabJe for his application
is one of the most important contributions to equipment reliability
tnat the designer can make. (Ref 1)
The primary factors affecting the failure rate of connectors are in-
sert material, contact current, number of active coLtacts, mate/unmate
cycling, and the environment in which it is applied. Contact current
and active contacts are not considerations during dormant periods,
howe*,er these aspects must be addressed by the designer to insure re-
liable operation during the operating period. Designers should con-
sult MIL-STD-1353 and applicable MIL specs to determine operational
ratirgs for contact cirrent and the maximum number of active c( :s
[or the particular application. The primary failure mechanism
connectors during dormai.t periods arise from mate/urL 3te cycling and
environelxt31 degradation of insulating and contact materials. The
remainder of this section presents these failure mechanisms and guide-
lines to increase the reliability of connectors during these phases.
Achieving jood electrical contact in a connector is a function of

mechanical degradation, contact surface Lilins (oxides and sulfides),
4. 1-I1,'
surface roughness, contact area, plastic deformation of the containing

materials and load applied. The performance of the connector is de-
pendent upon the chemical, thermal, and the mecha ical behavior at
these points.
Mate/unmate cycling is the primary mechanical degradation of cofnector

contacts. The most obvious malfunctions due to cycling are broken
contact elements and ineffective contact of these elements. Many con-
nectors, particularly of the cable type that are repeatedly plugged
and unplugged during maintenance and testing, continuously expose the
contacts to possible corrosive contaminants. These cycling effects
also crerte the problem of physical wear on the connecting inter-
faces. The result is increased interface resistance, higher contact
temperature during operating periods, and degradation of the electri-
cal connection. As a result, mate/unmate cycling increases the fail-
ure rate of electrical connectors. Control of maintenance practices
will assist in the control of these failure mechanisms. Detailed
maintenance guidelines for dormant equipment are presented in section
4.5, "Maintainability Design" of this text.
Environmental factors are the next primary failure mechanisms for dor-
mant connect)rs. Section 4.1.8.1 presents gener'al design and applica-
tion guidelines for selection and control of connectors for dormant
applications. Section 4.1.8.2 presents environmental effects and
guidelines for dormant connectors.
4.1.8.1 Guidelines
To protect unmated dormant connectors from degrading envi:onmental

factors, all unmated connectors should be kept covered with moisture
proof or vapor proof caps specified by military specifications or
military standards and designed for mating with specific connectors.
Where such protective caps are not available, disposable plastic or
metallic caps designed for this purpose should be used. Plastic and
metallic caps, however, are subject to degradation, and should be
chosen with these degradations in mind. Metallic caps should be
chosen to mitigate the corrosive effects of moisture on metals, and
plastic caps are subject to outgassing in low pressure applications.
Plastic caps should also be chosen to match the thermal requirements
of the connector.
The following are general design guidelines or connectors for

dormant applications:
SWhet ver applicable, use gold platin' on contacts. The softness

and inertness of gold reduces susceptibility to film contamina-
tion, which is a major degradation mechanism in dormant connectors.
4. 1 - t
" The minimum thickness for gold plating is 0.000100 inch. Thinner
platings may burn through. Thicker platings are neither necessary
nor cost effective.
" If the anticipated operational load is greater than 500 milli-

amperes, than the contacts should be alloys of nickel, silver, or
palladium. Otherwise, gold contacts are recommended.
" Plating processes should be closely monitored. Plating is a major

source of particulate contamination.
" Minimize the use of non-metallic materials because they are prone
to outgas, are a source of particulate contamination, have a nor-
mally lower mechanical strength, are more temperature sensitive,
and do not have the the long term stability of metals (instability
arises in metallic components, see the low temperature guidelines
in the following section).
Many connector contact failures are induced by the growth of films at

points of contact. These films can cause increased contact resistance
or open circuit. Ions in impurities or contamination in the surface
pores of contacts will migrate to the points of highest potential,
which are f.-equently localized hot spots. Ions interfacing with elec-
trons and other constituents at the poinLs of high chemical activity
usually gene-ate nonconducting films. There is also a continuous sup-
ply of material for the growth of insulating films from environments
where there are corrosive elements such as hydrogen sulfide, moisture,
oxygen, ozone, hydrocarbons, and various dusts.
The following presents environmental effects and guidelines to assist

the designer in controlling connectors to maintain reliability during
dormant periods.
High Temperature
* High temperature can cause failure of connectors by breakdown of

insulation or by breakdown in the conductivity of the conductors.
Either malfunction can be partial or complete. A typical break-
down caused by excessive temperature occurs progressively. As
temperature increases, insulation tends to become -Dre conductive,
and, simultaneously, the resistance of conductors incrc.ases. Com-
plete failure will occur if the temperature reachL'3 th.! point
where che conductor melts, breaking electrical conductivity, cr
where the insulation fails, causing a short.
4.1-107
Low Temperature
* Metals and nonmetals tend to become brittle and shrink at differ-

ent rates. How important each characteristic is depends on the
application. Designers should consult the applicable MIL specifi-
cation for the chosen ;onnector to identify low temperature rat-
ings. The coefficient of expansion of plastics and elastomers are
so different from those of the metals used in structural members
that they will shrink enough at extremely low temperatures to open
seals. An open seal may not cause a malfunction unless moisture
and contaminants enter through the opening. If a seal opens after
the temperature of a connector falls below the freezing point of
the contaminants present and then seals itself before the melting
point of the contaminants is reached, foreign matter will never
enter. However, if a connector seal opens at a temperature where
liquid or gaseous contaminants have not been frozen they can enter
and contaminate the connector.
Moisture
e Moisture is a primary degradation mechanism for dormant connec-

tors. Moisture is a corrosion inducing element, and also serves
as an adhesion element for other corrosive contiminants.
e Unmated connectors should be protected from moisture by protective

metal or plastic end caps. Whenever possible additional moisture
protection methods should be applied. For example connectors
should be maintained in sealed, moisture free containers or addi-
tional protection applied to the external portions of the cap and
connectors junction, such as an additional wax type sealant.
* Mated connectors should be contained in sealed moisture free con-

tainers to mitigate moisture entry to contact elementr.
* Extreme caution should be observed during maintenance and testing

to preclude tho entrance of moisture during connector mating/un-
mating cycles tj reduce failure mechanisms identified in the pre-
vious section.
Low Pressure
* The primary failure mechanism for dormant connectors in low pres-

sure applications is outgassing of organic (nonmetallic) mate-
rials. For low pressure applications the connector should be con-
structed of non-organic (metallic) components to reduce the pos-
sibility of outgassing. For more detailed outgassing effects, the
designer should consult reference 11.
4.1--08
* During operational periods, connectors intended for use in low

pressure applications should be derated to account for the ten-
dency of voltage flashover. This derating is specified in the
military specifications for conductors intended for use at other
than sea level barometric pressure.
" The primary degradative aspects of acceleration and gun fire for
dormant connectors is shock and vibration experienced during these
environments.
" The primary failure mechanism introduced by shock or vibration is

ineffective electrical contact at the connector contacts. For
shock and vibration environments, connectors should have positive
screw coupling mechanisms and adequate support for the cable or
wire bundle.
Temperature Cycling
o Variations of temperature must be considered as moisture condensa-

tion within the connector could develop. (See Moisture)
Saud and Dust
e A combination of sand and dust and small amounts of moisture will

materially increase the possibility of voltage bieakdown of the
insulation between closely spaced contacts.
Nuclear Radiation
" Materials frequently used in the manufacture of electrical connec-

tors include: plated copper or bronze alloy for contact pins and
receptacles, dielectric materials such as plastics, ceramics, or
glass, and outer shells or shields which are usually steel, brass,
or aluminum. Since metals are known to undergo relatively small
changes in electrical characteristics when exposed to nuclear
radiation, limited concern is necessary for metallic connector
components. (Ref 16)
* The effects of radiation on the dielectrics of connectors is the

main concern. Two types of degradation are encountered, and both
deal with the dielectric characteristics of the insulating in-
serts. Degradations which change the physical characteristic of
dielectric materials may also result in mechanical weakness in pin
supports and are indicated by the appearance of brittleness in
4.1-109
organic type materials. Permanent and/or temporary loss of insu-

lation resistance between contacts or in the outer shell is
another form of degradation. (Ref 16) The reader should consult
reference 16, Effects of Radiation on Materials an Components, for
a more comprehensive presentation of the degradation of insulating
materials in radiation environments.
4.1.9 Cables/Wires
Wires and cables are divided into several types based on performance
characteristics and on construction details. The most common types
are: (Ref. 14)
Wires
* Single conductor, shielded.
o Single conductor, unshielded, high voltage leads.
Cables
o Single conductor wires twisted together, shielded and unshielded.
o Coaxial cables.
e Multiple-conductor round cables.
o Multiple-conductor flat cables.
The reader should consult the applicable MIL specifications for a more
comprehensive description of the various wire and cable types. The
following sections present the primary degradation mechanisms and
guidelines for cables and wires in dormant applications.
Wires and cables are generally subjected to three different primary

mechanical stresses depending upon their usage: installed and never
moved or flexed, flexed during installation and occasional servicing,
and repeated or constant flexing. Cables in dormant applications may
be subjected to any or all of these stresses during the equipment
life-cycle. Flexed installations require proper stress relief at
connectors to prevent breakage. Installations requiring repeated or
constant flexing requires engineered cable and harness flexibility,
including balanced assembly (or twisting) of wires in the harness to
eliminate short lengths and stress relief of all conductors at
termination.
4.1-110
There are two predominant failure mechanisms for dormant cables and
wires: insulation degradation and conductor breakage. Insulation
degradation and possible breakdown can be induced or accelerated by
several different environments during the dormant period. It is
beyond the scope of this document to present all of the various types
of insulation and corresponding failure mechanisms. The following
subsection presents some general part selection/control guidelines for
cables and wires in dormant applications. The reader should refer to
Reference 15. A Review of Equipment Aging Theory and Technology, for a
more detailed discussion of the aging mechanisms of specific insula-
tion materials.
Generally conductor breakage is due to mechanical stresses as de-

scribed above or through additional induced mechanical factors experi-
enced during the equipment life cycle. These environmental factors
include shock, vibration, gun fire, and acceleration. Conductor
breakage failures can most often be attributed to poor workmanship and
poor assembly procedures. Furthermore, the mechanical weakness of
conductors sized #24 AWG copper and smaller increases the problem.
wire breakage has been a major problem with small conductors.
Overcoming this problem, for both small and large conductors, involves
providing proper stress relief of all wires at connectors. (Ref. 14)
The following presents environmental effects and guidelines to

assist the designer in controlling cables/wires to maintain
reliability during dormant periods.
High Temperature
o As temperature increase, insulation softens and loses its cut-

through resistance. If a thin edge is pressing against it, flow
can occur at high temperature until the conductor is exposed and
shorting results. (Ref. 14)
o Temperature in excess of the insulation rating will result in

insulation degradation (shortening of life) and possible excessive
detericration by flow or carbonization. Insulation aged and
hardened by excessive heat may be subject to cracking if flexed.
(Ref. 14)
o Avoid the use of polymeric insulators as these tend to deteriorate

more rapidly in high temperature environments.
4.1-111
a Thermosetting plastic insulating materials such as phenolics per-

form better than ABS, polycarbonate, polypropylene, or acetal
resins.
Low Temperature
* Temperatures below the insulation rating may result in fracture if

the wire or cable is given sharp flexure or impact while it is
cold. (Ref. 14)
Moisture
* Moisture can be a primary degradation mechanism for insulation in

cables and wires. Resulting moisture penetration is a corrosion
inducing element for the cable/wire conductor. Wherever appli-
cable, cables should be kept in enclosed moisture-proof enclosures.
" Use molded cables/wires in preference to bundled wire runs to

reduce the area available for moisture and other contaminants.
Low Pressure
* Outgassing of insulation materials (non-organic) is a primary

failure mode for cables/wires in low pressure applications. For
more detailed outgassing effects the designer should consult
reference 11.
* Cables/wires with polyvinyl chloiide insulation should not be used

in low pressure applications. (Ref. 71)
9 The primary failure mechanism introduced by shock or vibration is

ineffective electrical contact at the cable/wire connector con--
tacts. For shock and vibration environments, connectors should
have positive screw coupling mechanisms and adequate support f'-)
the cable or wire bundle.
Temperature Cycling
* Temperature variations within the ratings of the cable/wire rat-

ings will generally not affect performance. Vr ijations in temp-
erature must be considered at the connector cor:acts.
Condensation within the connector contact could cause corrosion of
the contacts and cable/wire conductor.
4.1-112
Sand and Dust
* A combination of sand and dust and small amounts of moisture

entering into the insulation will materially increase the possi-
bility of voltage breakdown of the insulation.
* In humid environments the designer should use molded cables and

not bundled wire runs to reduce the area available for moisture,
and sand and dust absorption.
Nuclear Radiation
e Metallic conductor materials undergo relatively small changes in

electrical characteristics when exposed to nuclear radiation,
therefore limited concern is necessary for metallic conductors in
cables and wires.
o The effects of radiation of the insulating (nonorganic) materials

is a primary concern. Two types of degradation are encountered.
Degradations which change the physical characteristics of the in-
sulating meterials, and permanent and/or temporary loss of insula-
tion resistance. The reader should consult reference 16, Effects
of Radiation on Materials and Components, for a more comprehensive
presentation of the degradation of insulating materials in radia-
tion environments.
4.1.10 Batteries
Batteries are divided into two basic groups: primary (non-recharge-

able) and secondary (rechargeable). Primary batteries are not elec-
trically recharged easily and are usually discarded once they are dis-
charged. The general advantages of primary batteries are good shelf
life, high energy density at low to mcderate discharge rates, little
or no maintenance, and ease of use. (Ref. 76) Secondary batteries
are easily recharged after discharge. Secondary batteries generally
have high power density, high discharge rate, flat discharge curves
and good low temperature performance. The energy density of secondary
batteries is generally lower than the energy density of primary bat-
teries. The charge retention of secondary batteries is poorer than the
charge retention of primary batteries, but when the charge of the se-
condary battery is properly maintained by recharging, it will retain
its charge longer than the primary. Generally, for dormant systems
the choice between primary or secondary battery systems is the choice
between the low maintenance, good shelf life characteristic of the
primary battery or the recharge maintenance, superior overall shelf
life characteristics of the secondary battery.
4.1-113
Reserve batteries are a type of primary batteries from which an active

component is removed from the rest of the battery prior to activa-
tion. The active component is usually the electrolyte. Thermal bat-
teries are a type of of reserve battery with a solid electrolyte that
melts upon heating. Reserve batteries are intended for long term
storage, but the missing component must be added before they can be
activated. Generally, reserve batteries do not experience any signi-
ficant degradation during dormant periods, and they are recommended
for dormant applications.
Generally, batteries are a perishable product and deteriorate as a

result of chemical actions that proceed during storage. The type of
cell design, electrochemical system, temperature, and length of stor-
age are the most significant factors which affect the shelf life of
the battery. Low temperature extends the shelf-life of most bat-
teries. However, most refrigerated batteries should be warmed before
discharge to obtain maximum capacity.
The primary failure mode for batteries during the dormant period is
self discharge. The rate of self discharge of most batteries is in-
creased by high temperature environments. The rate of self discharge
for most batteries is reduced in low temperatures. Most batteries
should be kept in a cool, dry environment to increase the storage
lifetime.
Lithium and Magnesium cell types develop protective coatings on the

active material during storage. These films can improve the shelf
life of the battery, however, when the battery is placed on discharge
after storage, the initial voltage may be low until the film is worn
off (Ref. 76)
Figure 4.1.10-1 compares the shelf life characteristics of some of the

more popular primary and secondary battery systems. A closer examina-
tion of the shelf life characteristics of the major primary battery
systems are plotted in Figure 4.1.10-2. These figures present per-.
centage capacity loss per year from 20 to 70 degrees C. The relation-
ship is approximately linear when the log of capacity loss is plotted
against the log/temperature (Kelvin). The plots assume that the rate
of capacity loss remains constant throughout the storage period, which
is not necessarily the case for most battery systems. For most bat-
tery systems, the rate of capacity loss tapers off as the storage
period is extended, but for purposes of comparison of the shelf life
of these battery systems, these figures present a reasonable approxi-
mation. (Ref. 76) The reader should refer to reference 76, Handbook
of Batteries and Fuel Cells. for a more detailed discussion of these
battery systems.
4.1-114
S10
Q.
(n 20
0
*~30 -Z/g
50
70 manganese dioxide
La 100
Zinc-carbon
200
300
500 -Seated
nickel-cadmium Sealed lead-acid
700
1000 1 1L -
20 30 40 50 60
Temperature. IC
4.1fi~i. Sh~ff i!e Charac-_eristi-s of Vari2ous Batterg Sy;s-ems (Pe'. 7,3
4.1-115
10
20 Li/SC 2
0
S30
50
70
100
Zn/HgO
200 Zn/Ag 2 0
300 - Alkaline - MnO 2
Zn/air (seated)
500 -
700
1000 Zn-carbon
20 30 40 50 60 70
Temperature, IC
Figure 4.1.10-2. Shelf Life CharacteristIcsof Prinary Battery Systems Re:. 76)
4.1-116
Table 4.1.10-1 presents the characteristics of the most common primary

battery systems, maximum storage temperatures, as well as, advantages,
disadvantages, and dormancy guidelines for the battery type. Table
4.1.10-2 presents corresponding information for the major secondary
battery types.
The following subsections present general guidelines to aid the de-

signer in insuring long term reliability during the dormant period.
The guidelines are divided into design, process control, and test
guidelines. (Ref. 24)
4.1.10.1 Design Guidelines
" Design excess capacity into the battery to reduce the percent
depth of discharge and compensate for capacity decrease with
usage. The tradeoff is cost and watt-hours/pound.
" Hermetically seal batteries to avoid degradation as a result of

possible electrolyte leakage during the dormant periods.
" For battery types where explosion is a possibility, employ a

pressure relief valve to prevent personnel injury and equipment
damage as a result of overpressure.
4.1.10.2 Process Control Gvidelines
o Employ clean areas during processing and manufacturing to reduce

the amount of harmful contaminants. Also, use clean lint-free
cotton gloves when handling components. Store components in clean
plastic bags when not being processed.
* Employ clean processes, remove the carbonates and keep the ni-
trates content down to prevent gas pockets thrt pop off active
material.
* Flush and brush plates prior to installation to remove

contaminants.
o Control the brazing temperature-time relationship to pievent ex-

cess dwell during brazing operations that can cause active mate-
rial penetration of ceramic seals.
o Avoid rapid cooling after brazing to prevent cracked ceramics and

brazing voids.
o Require process and test controls for each active element:

plates, separators, and electrolyte to reduce end product
variability.
4.1-i17
Table 4.1.10-1. Characteristics of Primary Batteries
Battery System,
Applicable MIL
Numbers, Maximum
Storage Temps. Advantages Disadvantages Guidelines
Carbon-Zinc -Low cost -Low energy density Not recommended for long term
MIL-B-18 -Comparatively low shelf life dormant storage especially at
-40 to 45*C -Inoperative below -20*C high temperatures
-Rapid deterioration at 50° C
Magres i um -Good capacity -Loses its excellent storage -Not recommended for long term
MIL-B-55252 retention characteristics after being intermittent use
-4C t: 70*C even under partially discharged -Can be stored for 5 years at
high tempera- 20° C with only 10-20% capacity
ture storage loss
Alkaline - -Good shelf life Moderate cost -90 to 85% capacity retentio,,
Magnese-Dioxide -Operates at after 1 year of storage at 20*C
MIL-8-55252 temperatures -10 to 20% capacity losses after
-40 to 50° C as low as -40*C 3 years of storage at 20° C
-High capacity -20% capacity loss
retention
Mercuric Oxide -Excellent stor- Expensive Storability also depends on the

MIL-8-18 age character- discharge load placed on the
-20 to 550C istics even cell. Failure in storage is
under high usually due to the breakdown of
temperature cellulosic compounds within the
storage con- cell, which at first results in
ditions a reduction of the limiting
-Capacity loss of current density at the anode.
10 to 20% after Further breakdown produces low
2 years at 209C drain internal electr'cal loss
-Capacity loss c paths and loss of capacity due
20% after 1 yr. due to self discharge. Eventu-
at 45'C ally, complete self discharge
can occur, but these processes
at 209C and below take many
years.
-After long periods of storage

the cell electrolyte tends to
seep out of the seal and is
evidenced by a white carbonate
deposit at the seal insulation.
4. 1-I16
Table 41.10-1. Characteristics of Primary Batteries (Continued)
Battery System,
Applicable MIL
Numbers, Maximum
Advantages Disadvantages Guidelines
Storage Temps.
-Cells should be adequately

packaged and not stacked up to
high to prevent damage, because
Mercury cells are heavy.
-Should be stored in a relative

humidity less than 65%.
-Severe shock and vibration

could damage the cell.
Silver-Oxide -High energy Use limited to button and
MIL-8-82117 density miniature cells because of
-40 to 609C high cost

-High tempera-
tures acceler-
ate capacity
deterioration
-Good low tempera-

ture performance
-Periodic inspection and main-

-High energy -Periodic maintenance required
Zinc/Air
tenance of water level to
-20 to 40° C density achieve proper electrolyte
level. The use of trans-
-Long storage -Susceptible to drying out of
cells parent cases allows visual
life if sealed
inspection and simplified
maintenance.
-During storage the air access

holes of the cell should be
sealed to prevent gas transfer
decay (the access holes must
be open when in operation,
oxygen from the cell's ex-
terior is one of th,
reactants)-
4.1-119
Table 4.1.10-1. Characteristics of Primary Batteries (Continued)
Battery System,
Applicable MIL
Numbers, Maximum
-Elevated temperatures dramati-

cally increase self discharge
rates.
-Storage life at high tempera-

tures can be optimized through
trado-offs between other per-
formance parameters and choice
of cell design components.
Lithium Cells -High energy -Voltage delays after storage -Design should be able to with-
-55 to 701C density stand short voltage delays
after activation of long terr
-Best low tempera- storage
ture performance
-Can be stored
for long periods
even at elevated
temperatu res
-Effective stor-
age for 10
years at 20"C
is projected
-1 year at 70"C
with little
capacity loss
-Hermetically
sealed, and
protected dur-
ing storage by
the formation
of film on the
anode. This
film may create
voltage delays
after long
storage.
4.1-120
Table 4.1.10-1. Characteristics of Primary Batteries (Continued)
Battery System,
Applicable MIL
Numbers, Maximum
Solid Electro- -No loss of Ca- Only available in small sizes

lyte Lithium pacity after 4
-40 to 600 C years of stor- -High cost
age at 20*C
-No loss of ca-

pacity after 1
year at 60*C
-Good low temp-

erature per-
formarce
4.1-121
Table 4.1.10-2 Characteristics of Secondary Batteries
Applicable Maximum
Battery System MIL Numbers Temps. Advantages Disadvantages Guidelines
Lead-Acid MIL-0-11188 -Good high and -Long term storage in -Charge must be
-SLI MIL-B-15072 -40 to 55"C low temper- discharged condition maintained in
-Traction MIL-8-:76g -20 to 401C ature opera- can lead to irre- long term stor-
-Stationary W-8-134 -10 to 409C tions versible polariza- age
-Portable oD-8-15072 -40 to 600C tion of electrodes
-Performance
-At 70*C this cell can characteristics
only be stored for can be greatly
30 days without - r rovcd :!
charge stabilized by
removing the
electrolyte
-Each cell should

not be allowed
to discharge
below 1.8V
Nickel-Cadmium MIL-STD-1578 -Can be stored -Low energy density -Best practice is

-Vented-Pocket MIL-8-81757 -20 to 456C for long peri- to store in an
Plate MIL--49436 ods of time upright posi-
-Vented Sin- MIL-8-23272 -40 to 50"C without re- tion at a tem-
tered Plate DoD-8-8565 charge perature range
MIL-8-55130 -44 to 45*C from 00 to 300C
-Sealed M!L-8-26022 -Can be stored
over a broad
temperature
range
Nickel Zinc -20 to 600 -High energy -Poor cycle life -Should be stored
density in a well ven-
tilated area to
-Good low temper- prevent the ac-
ature perform cumulation of
ance potentially ex-
plosive Hydro-
gen
Zinc/Silver MIL-8-82117 -20 to 60*C -High energy -Short cycle !ife

Oxide MIL-8-29361 density
MIL-B-11814 -Can provide extremely
DoO-8-24506 -Low self dis- tremely high cur-
000-8-24531 charge rents if shorted
4. 1-122
Table 4.1.10-2 Characteristics of Secondary Batteries (Continued)
Applicable Maximum
Battery System MIL Numbers Temps. Advantages Disadvantages Guidelines
-Electrolyte is ex-
tremely caustic
solution of Potas-
sium Hydroxide
Nickel-Hydrogen 0 to 500 C -Long cycle life, -Self discharge in
even cn deep 100 hours at am-
d*;charging bient temperature
-High energy
density
Rechargeable -Good shelf life -Useful capacity -Available

"Primary Types" -Good capacity is only 1/3 of energy
(Magnesium, -20C to 40*C retention primary versions rapidly with
Silver Oxide) -Completely -Small sizes only cycling
sealed and
maintenance
free
4.1-123
4.1.10.3 TeSt Guidelines
* Helium leak check test the assembled cells.
z Wherever applicable, subject battery during acceptance test to a

minimum of three charge/discharge cycles, high impedance short
test, and leakage tests. These tests should provide assurance
that the basic operating characteristics and construction are
satisfactory.
" X-ray along three axes to find gross battery defects.
" Conduct a minimum of 30 charge/discharge cycles on assembled cells

to minimize infant mortality and to confirm the matching of indi-
vidual cells. Resident inspection should observe and confirm
these tests.
4.1.11 Fiber Optic Components
This section addresses the reliability of fiber optic cables in dor-

mant applicaLions. Fiber optic connectors are similar in function to
standard connectors, and the guidelines presented in section 4.1.9 are
also applicable to dorma-it fiber optic connectors.
Fiber optics is a relatively new technology and the amount of failurc

data available for dormant tiber optic devices is limited. The fol-
lowing section presents general guidelines for the selection and con-
trol of fiber optic devices in dormant applications. This section
addresses glass optical fibers, both single and multimode. Fiber op-
tic connectors, splices, taps, and pigtails are also addressed.
Glass in its various forms and material compositions was developed

thousands of years ago. It has proven to be one of the most durable
and strongest substances known. It has been stated that glass can
last forever if it remains mechanically unstressed and in the absence
of moisture. (Ref. 80)
Optical fibers are dielectric wave guides relying on total internal

reflections in which light is reflected from a refractive index dis-
continuity within the fiber. Failures in optical waveguides, connec-
tors, and splices can be inherent or induced. Induced failures may be
due to externally applied shock, vibration, temperature extremes or
cycJing, radiation or other stimuli resulting from accidents or main-
tenance action. The durability and strength of glass suggest that
most fiber failures are induced. (Ref. 80)
4.1-.24
The absence of yield causes glass to be brittle and unable to with-

stand heavy shocks. However, glass also can develop inherent fail-
ures. An example is crack growth, where dominant flaws can increase
in size under induced stress until fracture occurs. (Ref. 80)
A variety of materials are used for optical fibers. The most commonly
used is ultra-pure fuse,! silica (Si02). This is produced by a chemi-
cal vapor deposition process, usually through the collection of Si02
soot obtained by the combustion of silane (SiH4) or silicon tetra-
chloride (SiC14). Selected doping elements or compounds are added to
the process to control index. The resulting fibers combine extremely
high physical strength with optical transparency. The absence or
scarcity of impurities reduces the probability of heterogeneous grain
boundaries where flaws and weaknesses are likely to occur, and reduces
Rayleigh scattering in the optical path. (Ref. 80)
Typical dopants used are metal oxides, sulfides, or selenides. Be-

cause they are fused mixtures, rather than fixed compounds with crys-
talline structures, their compositions are infinitely variable and
large numbers of different mixes can be manufactured. The need for
low optical losses, i.e., high chemical purity has led to the develop-
ment of glasses which can be prepared directly from the vapor phase.
Small amounts of dopants are introduced into the gas stream to modify
the refractive index of the material so produced. (Ref. 80)
Since the main and almost sole catastrophic failure mode for fibers is
fracture, strength is surely the most important factor affecting long
term optical cable reliability. The maximum tensile strength for
ultra-pure fused silica fibers, as used in fiber optical applications,
may be as high as 2 x 106 lb/in 2 making this commodity one of the
strongest materials known. However, brittleness, or lack of yield,
has made this material vulnerable to shock, vibration, rough handling
and other such applied stresses. To overcome some of the difficulty,
it has become necessary to coat the fibers individually to protect
them from mechanical damage and to preserve their high inherent
strength. Coating used may be epoxy--acrylate or urethane-acrylate,
both of which are extremely strong plastics. (Ref. 80)
Besides brittlenesses, glass exhibits delayed failure or "static

fatigue", and a wide variability in fracture strength. Static fatigue
is a property which causes a fiber to break when exposed to constant
stress over long time periods. The fracture strength variability
greatly contributes to difficulties encountered when attempting to
predict reliability or useful life of fibers and cables. (Ref. 80)
4.1-125
Connectors, installed at cdble ends, facilitate quick removal and re-

placement of fiber optics .omponents, azsemblies, or units. Occasion-
ally, connectors must be uncoupled for service, maintenance, inter-
change of equipment or for reasons of mobility. During an uncoupling-
recoupling cycle, the fiber ends within a connector are exposed to the
immediate surroundings. If these include dust, sand, dirt, or mois-
ture, connector failures are possible. The failure may be catastro-
phic if fracture has occurred or recoverable if effective cleaning is
possible. (Ref. 80)
The reliability of splices depends upon chemical and physical stabil-

ity of the materials involved. Reliability also depends upon the care
taken during splicing procedures. In addition, light transmission
losses occur as a result of misalignment or differences in core diam-
eters and geometry. (Ref. 80)
A fiber may be tapped for beam splitting or other reasons. These are
usually accomplisheO by grinding two fibers at designated spots until
the two cores are exposed. Then, they are cementad together or
welded. Precautions are necessary similar to those for splices.
(Ref. 80)
Pigtails are attached to fiber optic components such as light sources

or photosensors to allow input or output from these components to be
applied to connectors and fibers. Protection is mainly provided by
the de-ice enclosures. However, when these devices are opened for
servicing or other reasons, the pigtails then become vulnerable to
damage. This has proved to be one of the major failure modes of fiber
optics systems.
Generally, fiber optic components are very reliable devices in dormant

applications. Optical fibers are designed so that inherent mechanical
strength is sufficient to withstand heavy vehicles running over them
without causing damage. This is true if the ground surface is paved
and reasonably smooth. It is not true if the ground is uneven or
soft. (Ref. 80)
Some difficulties have been experienced with dust and other contami-
nants accumulating inside connector assemblies. Cleaning will restore
the system to full operational conditicn. (Ref. 80)
The following presents some guidelines for the protection of dormant

fiber optic components from the primary degrading environmental fac-
tors. Other than the induced factors described above, the following
are the primary degradative environments.
4. I-12-
Temperature
Glass optical fibers are generally not particularly sensitive to

temperature variations. However, due to differences in linear
thermal expansion coefficients among the various materials making
up fiber optic cables, temperature variations and cycling can
cause compressive and tensile strain to the fibers. Aside from
the probability of fracture, losses in signal can also occur due
to microbending, that is, microscopic rippling of the fiber, where
the signal is partially reflected and some of it is lost in the
cladding and fiber coating. These losses do not always constitute
permanent failure modes. A return to "normal" temperatures can
often effect complete or nearly complete recovery; that is, unless
fracture has occurred. (Ref. 80)
Extended exposures may cause gradual degradations or irreversible

change in system insertion loss with time and with succeeding
temperature cycles. The losses, thus induced, are a function of
microscopic roughness in the core-cladding interface. (Ref. 80)
A technique for screening fiber optic cable for fracture strength

has been proposed by C. Veraa (Ref. 80). According to this tech-
nique examination of sample fiber optical surfaces, using an elec-
tron microscope, can aid in determining the size and nature of
flaws responsible for the fracture failure mechanism. The rela-
tion between flaw size and tensile stress required for fracture
has been mathematically demonstrated by Olshansky and Maurer
(Ref. 81). By screening samples of fiber optic cable with an
electron microscope and determining average flaw size, one coula
theoretically determine the tensile stress due to thermal cycling
or thermal shock at which the fiber optic cable would fracture.
Using this technique as a quality control tool, however, would be
difficult and costly.
In practice when selecting fiber cable for a dormant application

in an environment where thermal cycling and/or thermal shock are
expected factors the designer should use simpler techniques to
assure the product quality. In particular samples of the fiber
optic cable procured should be tested to Methods No. 4010 and 4020
of DoD-STD-1678. (Ref. 81)
Moisture
It is known that water has an adverse effect upon the strength of

glass. The presence of moisture reduces the surface energ'y so as
to accelerate strength corrosion or fatigue with time. With no
moisture present glass would not fatigue and will theoretically
4.1-127
last forever. In practice, however, some moisture is always pres-

ent. Selecting fiber coating materials which resists moisture is
necessary to protect the fibers from this source of deteriora-
tion. It is also recommended that molded cables be used in pref-
erence to bundled wire runs. This reduces the area available for
moisture and other contaminant absorption. (Ref. 80)
Shock and Vibration
Severe vibration or mechanical shock can cause fiber fracture.

Fiber fracture can cause dispersion of the energy being trans-
mitted through the fiber, and this can result in loss of function
of the fiber.
Vibration and acoustical energy exposure can induce bending losses

in fibers. This effect is intensified in the case where fibers
are used in a curved position. (Ref. 80)
There are no types of optical fiber which are known to be less

susceptible to vibration effects than other types, and part selec-
tions can not be used as a means to lessen vibration susceptibil-
ity. It is recommended that when designing with fiber optic de-
vices for use in high vibration environments, the system should be
designed to avoid sharp bends in the fiber cables. Any form of
isolating/vibration absorbing packing placed around the fiber op-
tic cables would also be helpful in lessening vibration damages.
Nuclear Radiation
Neutron and total dose ioni3ing gamma radiation can effect the
performance of fiber optic cable. The effects are primarily rrani-
fested by an increase in attenuation. The neutron radiation ac-
tually damages the glass structure causing flaws and thus attenua-
tion. The total dose gamma radiation reduces the transmitivity of
the optic fiber and produces attenuation.
In using optic fibers in applications where a nuclear radiation

environment is expected to be a dominant factor it is recommended
that the system be over-designed since shielding the fiber from
neutrons and gamma rays is impractical. The cable should be se-
lected to allow for an 80% attenuation loss of system function.
4.1.12 Quartz Crystals
Crystal units are generally made up of a quartz crystal mounted in a

metallic holder. The size of the crystal is inversely proportional to
the frequency of operation. The crystal unit may also contain i
4.1-128
heeing element which will help stabilize the crystal temperature.

(Ref. 35)
The five important parameters of the crystal unit are: frequency

range, mode of oscillation, temperature range, load capacitance, and
driving level. The principal stress parameters are the driving power
and temperature. However, dormant crystals are not degraded signifi-
cantly by them.
The frequency and frequency stability of a crystal is sensitive to

high levels of moisture and temperature. Crystals should be chosen so
that its operating temperatures are well within the specified operat-
ing temperature ranges. Proper selection for operating temperatures
will also insure proper temperature ranges for dormant periods. In
high humidity, metallic cases of the crystal units are subject to cor-
rosion and should be suitably protected. Metallic crystal holders and
cases are susceptible to corrosion in a salt-fog environment and
should be maintained and suitably protected within sealed enclosures.
Crystal cases should be glass sealed to protect the crystal unit in a
humid or salt fog environment. Frequency shifts will result from the
degredation induced by these environments.
Environments of shock and vibration can damage the fragile crystal

unit and lower the frequency of operation of the larger types.
(Ref. 86) Consult the appropriate crystal specifications for the
recommended safe limits of shock and vibration. (Ref. 35)
Crystal units are also susceptible to frequency shift in an accelera-

tion environment. Designs that contain crystal units that experience
acceleration should allow for a frequency shift as per MIL specifica-
tion requirements.
Nuclear radiation may effect the frequency and frequency stability of

crystal units. Nuclear radiation effects can be reduced by selection
of the proper crystal types based on manufacturing techniques. Sus-
ceptibility of crystal performance to nuclear radiation is ranked from
least to most susceptible as follows:
" Swept synthetic crystal oscillators
* Synthetic quartz crystals
" Natural quartz crystals.
Natural quartz crystals are not recommended for application in nuclear

radiation environments. High frequency crystal units of any type are
more susceptible to nuclear radiation than low frequency units.
4.1-129
While not normally considered electrostatic sensitive devices, most

crystal units are susceptible to electrostatic discharges greater than
4000 'volts. This level of electrostatic voltage can be induced by
improper handling. The damage often results in degradation rather
than failure. Utilization of suitable handling implements and mate-
rials can help minimize this problem. For example, use of cotton
gloves, static eliminator devices, humidifiers, and operator and work
bench grounding systems can reduce static buildup during handling.
Means of alleviating static problems during shipment include elimita-
tion of loose packaging of crystal units and use of metal foil and
anti-static (party conducting) plastic packaging materials. The de-
signer should consult DoD-STD-1686 and DoD-RDBK-263 for more detailed
guidelines for the handling of devices to mitigate damage from elec-
trostatic discharge.
4.1.13 Electromechanical Devices
Electromechanical devices addressed in this section are motors,

blowers, fans, and solenoids. Electromechanical devices, generally
produce a mechanical output for an electrical input. Motors, blowers,
and fans are electromechanical devices that rotate in resionse to an
electrical input.
The principal stress factors for electromechanical devices are

mechanical load, winding temperature, and ambient temperature of
operation. Mechanical load is a factor encountered during operational
periods, and it is not a degradation mechanism during the dormant
periods. Winding temperature is generally higher during operational
periods due to temperature increases induced by electrical and
mechanical resistance.
Temperature is considered to be the primary factor for determining

the life and efficiency of electromechanical devices. Although, high
temperatures induced during operation are the primary degradation
mechanism for electromechanical devices, surrounding natural environ-
mental temperatures need to be addressed during periods of operation
as well as during dormancy. During operation as well as dormancy,
high temperature will cause the winding insulation to fail. Low temp-
eratures will cause the bearings to fail, and low temperature will
also cause slowed mechanical reaction to applied electrical stimulus.
Readers should consult applicable Military specifications for tsmpera-
ture ratings.
As with other electrical devices presented in this section, moisture

and contaminants are the primary degradative environmental factors for
electromechanical devices during the dormant period. Moisture induced
failure mechanisms include: corrosion, insulation breakdown, and low
4.1-130
resistance to electrical leakage. Moisture and contaminants are par-

ticularly degradative to electromechanical devices, because these de-
vices are generally not hermetically sealed. In order to mitigate the
effects of these environments during the dormant as well as subsequent
operational periods, electromechanical devices should be maintained in
enclosed, protected enclosures. The designer should refer to section
4.5 of this text for more comprehensive packaging guidelines to reduce
moisture entry.
Temperature cycling, during dormant periods may produce moisture con-

densation on sensitive portions of electromechanical devices. Sand
and dust, and other contaminants can cause reduced mechanical and
electrical efficiency in electromechanical devices. These contami-
nants also serve as moisture traps, and in conjunction with the mois-
ture, they may promote accelerated corrosion. Dormant devices should
be protected from these environments, it is recommended that hermeti-
cally sealed units be used whenever possible. Non-hermetic units
should be sealed in in a protective enclosure that still allows suf-
ficient heat dissipation when the device is called to operation.
Interior and exterior portions of electromechanical devices are prone

to accelerated degradation and corrosion in salt fog environments.
Electromechanical devices in salt fog environments should be main-
tained in sealed enclosures and they should also be plated or painted
with a protective coating. These coatings should be periodically in-
spected and maintained to insure a protective seal. Plating, paint-
ing, and packaging guidelines that will inhibit corrosion are pre-
sented in section 4.5 of this tert.
Some other failure mechanisms that are of primary concern during the
dormant period are: lubrication (drying and oxidation are chief con-
cerns, low tempeLature annealing of rotating surfaces, and corrosion
of surfaces). Other failure mechanisms include swelling, cracking and
general material degradation of O-rings, packing and insulators. Cor-
rosion of bearing, contacts, switch parts, gear assemblies and motor
are also possible. (Ref 93)
4.1-131/(.1-]2 blank)
4.2 DERATING
Derating for electrical, mechanical, and electromechanical parts was

appropriately defined by Brummett, Cross, Davis, and Towns (Ref. 35)
an "the practice of reducing electrical, mechanical, or environmental
stresses below the maximum levels that the part is capable of sustain-
ing in order to increase application reliability." In a more general
definition, R. T. Anderson stated that (Ref. 2) "Derating can be de-
fined as the operation of a part at less severe stresses than those
for which it is rated. In practice, derating can be accomplished by
either reducing stresses or by increasing the strength of the part.
Selecting a part of greater strength is usual'- the most practical
approach."
Derating cannot be used to compensate for using parts of a lower qual-

ity than necessary to meet usage reliability requirements. The qual-
ity level of a part has a direct effect on thp predicted failure
rate. (Ref 92)
Electrical testing of all parts in a lot is not guaranteed for commer-

cial or JAN level military parts. For high reliability applications,
only fully tested and screened parts/(including burn-in) should be
used in additicn to applying the appropriate derating levels. (Ref.
92) The designer should refer to sections 4.4 and 5.0 for more com-
prehensive test, screening, and evaluation guidelines for dormant
systems.
In this study dormant systems have been defined %s systems that ex-
perience a significant portion of their service life with zero elec-
trical stress. From this definition of dormancy, it is apparent that
reducing electrical stress will not directly increase the reliability
of dormant systems during the nonoperating period. However, derating
of electrical stress will indirectly increase the reliability of dor-
mant systems by reducing operating failures and thus increasing the
operational readiness and availability of dormant systems. Although
dormant systems spend a significant portion of their life cycle in
non-operating states, it is recommended that electrical stress levels
be derated by operating load criteria in order to preclude over stress
failure during operation periods.
It is beyond the scope and purpose of this document to derive or pres-

ent operational derating criteria for electrical stress, especially
when several excellent references are available. For more detailed
operational stress derating guidelines the reader should consult
Reference 35, "Reliability Parts Derating Guidelines", or Reference
92, "Reliability Derating Procedures".
4.2-1
In a similar adaptation, dormant systems generally experience less

severe mechanical stresses than those encountered by operating sys-
tems. The two primary mechanical stresses encountered during non-
operating periods (inertial forces and thermal-mechanical interac-
tions, see section 4.1) are generally the result of induced environ-
mental stresses. For example, inertial forces are usually introduced
by shock and vibration encountered during transportation and handl-
ing. As a result, it is concluded that the primary stress encountered
during dormant periods is environmental stress.
Several previous studies have concluded that the failure modes en-
countered during dormant (non-operating) periods are generally the
same as those encountered during operating periods. A Martin-Marietta
study (Refs. 7 & 20) provided the following breakout of electronic
part non-operating failure modes:
Bonding/Welding 21.5%
Photcetching 17.2%
Transportation and Handling 12.9%
Seal Aging 12.9%
Expansion Coefficient 12.9%
Conductive Cement 8.5%
Defective Hermetic Seals 4.3%
Plating 4.3%
Soldering 4.3%
These failure modes are directly attributable to inadequate manufac-

turing processes, improper part selection and control, and ineffective
part screening. For some components the failure mechanisms are inde-
pendent of the application environment and for others, it may be
time-related and environment dependent. Reduction of environmental
stresses is a means of mitigating or decelerating many of the above
failure mechanisms.
The rate of deterioration of dormant electronic systems is, for the

most part, dependent upon environmental conditions, either natural or
induced. Section 3 of this text presents information and general
guidelines for the most significant environments encountered during
the dormant period.
Each environmental factor makes its own contribution to the de-

terioration of dormant systems, *-nd the designer must realize that
these environments rarely occur singly. Often it is the combina-
tion of factors which determine the severity and rate of deterior-
ation. Many materials experience small to moderate deterioration
when subjected to only a single environmental factor, but these
same materials may be seriously affected by combinations of
4.2-2
environmental influences. From this discussion it is apparent that

reducing or derating environmental stresses is a means of preventing
failure during the dormant period.
Reference 50, "Deterioration of Materials and Recommended Storage Con-

ditions", presents guidelines for environmental factors encountered
during storage. This text also presents recommended storage condi-
tions. The recommended storage conditions provide for a reduction in
environmental stresses, however, derating factors for these environ-
mental stress are not presented or discussed.
Controlling or reducing the environmental stresses for dormant systems

can amount to a significant cost, even for the control of a single
environmental stress. This presents a cost tradeoff decision of
choosing reduced dormant reliability , with less stringent environ-
mental control, and less cost, or increased dormant reliability, en-
vironmental control measures, and respective increased cost. Packag-
ing, and Transportation and Handling guidelines for dormant systems
are presented in sections 4.5 and 4.6 of this test, respectively.
Part Selection and Control guidelines are presented in section 4.1.
These guidelines are presented to aid the designer in designing a
reliable design for dormant applications. For derating purposes, %he
guidelines can be expanded allow to for an even greater reduction of
environmental stresses during the dormant period.
An alternative solution to environmental stress derating is suggested

by Anderson's definition of derating, which could also be defined as a
policy of overrating. In his definition Anderson suggests that se-
lecting a part of greater strength than required is usually the most
practical means of establishing reduced stresses. This derating cri-
teria presents a cost-effective and practical approach to establishing
an over'll derating policy for dormant devices.
An overall approach for establishing a derating policy for dormant

systems should include operation derating guidelines, environmental
stress derating guidelines, and overrating policy for part selection
for anticipated stresses. A combination of these derating criteria,
in conjunction with an effective part selection/control, packaging,
and transportation and handling program based upon anticipated dor-
mancy environments should provide a cost effective reliability program
for dormant design.
4.2-3/(4.2-4 blank)
4.3 MAINTAINABILITY DESIGN
All equipment should be subject to an appropriate maintainability

program in accordance with MIL-STD-470 (Ref. 73). Equipment
maintenance is a prime driver of equipment life cycle costs (LLC) and
a major determinant of equipment availability. All equipment,
regardless of the amount of its service life that will be spent in a
non-operating state, must be designed to an appropriate maintenance
philosophy that achieves reliability and availability requirements and
minimizes life cycle costs. The following sections provide
maintainability design guidance for systems that will spend a
significant part of their service lives in non-operating states.
Maintenance concepts for dormant equipment are first discussed in
section 4.3.1. Maintainability design considerations for these
maintenance concepts are then discussed in section 4.3.2.
4.3.1 Maintenance Concepts for Dormant Eguinment
The selection of a maintenance concept for a given system design

depends upon the nature and characteristics of the given system. For
some systems, (e.g. orbiting satellites) maintenance is difficult or
impractical. For other systems (e.g. missile systems) maintenance
concepts specifying regular, periodic servicing periods are often
prescribed.
The development of the maintenance concept is the central activity of

maintenance support planning. The maintenance concept defines
criteria for maintenance activities and resources allowable at each of
the specified maintenance levels. It is derived from the operational
and ILS concepts of the system and from the policy statements which
forma the constraints and boundaries of the support system as expressed
in requirements documents. The maintenance concept serves two
purposes: (Ref. 74)
(1) It provides the basis for the establishment of maintainability

design requirements.
(2) It provides the basis for the establishment of maintenance

support requirements in terms of tasks to be performed,
frequency of maintenance, preventive and corrective maintenance
downtimes, personnel numbers and skill levels, test and support
equipment, tools, repair parts, facilities, and Information.
The maintenance concept must be both rea3istic and sufficiently

definitive to meet the needs of the system/design engineers and the
requirements of logistic support planners. Since the primary purpose
for which a system is acquired is intimately related to some set of
4.3-1
missions, analysis of the implications of maintenance policies on

system design starts logically with mission and operational
requirements. The maintenance concept is concerned with policies and
goals pertaining to: (Ref. 74)
(1) Operational states of the system.
(2) Maintenance activities.
(3) Maintenance and support resources.
(4) System effectiveness.
These categories are further subdivided as shown in Table 4.3.1-1.
Over the years, two maintenance concepts have developed for equipment
that will spend a significant part of its service life in a
non-operating state. They are the no-maintenance or "wooden round"
maintenance concept and the periodic monitoring/repair maintenance
concept. These two concepts are discussed in sections 4.3.1.1 and
4.3.1.2 respectively.
4.3.1.1 "Woode Round" Maintenance Concegt
The "wooden round" maintenance concept is essentially maintenance-free

storage of equipment through the elimination of organizational level
preventive maintenance throughout the storage period. There are two
types of wooden round maintenance concepts:
" Total wooden round which eliminates the need for organizational
level maintenance and uses maintenance-free storage for the
storage life of the system.
" Quasiwooden round which uses limited on-condition maintenance at

long intervals (5 years or more), carried out by contractor
personnel using commercial test equipment, special tooling,
commercial documentation, and spare parts (if required).
Systems that are used intermittently or only once (such as a missile

or munition), or are not economically repairable once they fail (such
as inexpensive hand-held calculators) are good candidates for the
wooden round maintenance concept. Design considerations for the
wooden round maintenance concept are discussed in section 4.3.2.1.
4.3-2
Table 4.3.1-1 Classification of Maintenance Policies and Goals
A. OPERATIONAL STATES
1. Non-operating (Dormant) Period

2. Scheduled Downtime Period
3. Operational Demand Period
a. Standby
b. Alert
c. Reaction
d. Mission
e. Deactivation
B. MAINTENANCE ACTIVITIES
1. Preventive Maintenance
a. Service
b. Inspection/Test
2. Corrective Maintenance
a. Detection
b. Diagnosis
c. Correction
d. Verification
3. Maintenance Level
C. RESOURCES
1. Personnel
a. Operators
b. Maintenance Technicians
2. Equipment
a. Prime
b. Support
3. Facilities
4. Repair Parts and Supplies
5. Information (Publications and Data)
D. EFFECTIVENESS
1. Downtime
a. Detection Time
b. Diagnostic Time
c. Correction Time
d. Verification Time
2. Reliability
3. Availability or Operational Readiness
4. Dependability
5. Mission Completion Success Probability (MCSP)
4 .3-3
4.3.1.2 Periodic Monitoring/Repair Maintenance Concept
The periodic monitoring/repair maintenance concept is shown in figure

4.3.1.2-1. It consists of fixed calender inspection limits, operating
times, checks, and overhaul of equipment. To carry out these
measures, it is necessary to remove equipment and assemblies from the
dormant systems and subject them to preventive and corrective
maintenance procedures in order to maintain their reliability and
operational readiness states.
This maintenance concept was often employed on first-generation

missile systems. This required these systems to be designed to meet
test enviroirents (that is, they had to be designed to be capable of
withstanding testing for hours); whereas, the real operating time was
often only a few minutes. The frequent handling and testing of these
systems was found to cause many failures. According to a Redstone
Arsenal study (Ref. 18), as many as 50 percent of confidence test
rejected missiles were erroneously faulted or were found to contain
test-induced failures. These first generation missiles were often
tested to failure and overhauled to death.
As a result of these experiences and findings, inspection intervals

for equipment have been enlarged and often multiplied without any
reduction in reliability. In fact, by drastically reducing the
testing, reliability is seen to improve, and, in many cases, it has
been possible to entirely eliminate the periodic checks. (Refs. 7, 75)
Design considerations for the periodic monitoring/repair maintenance

concept are discussed in section 4.3.2.2.
4.3.2 Maintainability Design Considerations for Dormant Systems
The following sections provide maintainability design considerations

for the two maintenance concepts previously described for dormant
systems. Design considerations for the "wooden round" maintenance
concept are discussed in section 4.3.2.1 and design considerations for
the periodic monitoring/repair maintenance concept, including guidance
in selecting an optimum inspection/test frequency, are discussed in
section 4.3.2.2,
4.3.2.1 "Wooden Round" Design Considerations
The objective of the wooden round, maintenance-free storage concept is

to design a system and its storage package for the elimination of the
need for inspection (except for periodic inspections of the package
container to assure its environmental integrity), maintenance, and
repair. The design for maintenance-free storage must therefore
address the following factors: (Ref. 75)
4.3-4
z
uJ
~ui
0
ClC
LL0
zC
L4-I
uJ
Lr
Z0 a
< < 5
m
u'-
a: -0C
0 w 2 cr
a..
4.3-5
* Use of 100-percent burn-in, screening and preconditioning of parts

or subsystems, on both electronic and mechanical items.
* Avoidance of storage-time-critical components and materials (for

example, electrolytic capacitors, open contacts, natural rubber,
and mineral oils. See section 4.1 for part selection and control
guidelines.).
" The system should be designed for simplicity using a minimum of

parts. This minimizes the possibility of failures occurring in a
numerical sense.
" Use of an inert atmosphere (such as nitrogen).
" Use of transport and storage containers as tactical packaging to

mitigate the effects of long-term exposure to life cycle
environments (See sections 4.5 and 4.6).
" Interconnection plugs should only be used if no other design is

available (cost reduction and increase in reliability).
" Reduction to a minimum of built-in test points for external

testing as they are unnecessary and may provide paths for moisture
and contamination intrusion.
* Use of a compact structure, combining PC boards and sealing the

system.
" Use of digital electronics with no field adjustment.
" Manufacturing and quality assurance methods to increase

reliability, through increased usage of autom;,tic test equipment
during manufacture.
" Complete documentation of manufacturing parameters and test

results.
" Replacement of parts which, due to their present technology, are

time change items (lifetime limited, for example, magnetrons).
This should be accomplished without any disassembly or test, and
the system should remain in its container. Replacement of modules
will be limited to the smallest possible individual element.
4.3-6
In order for the wooden round concept to be effectively applied, the

reliability of the candidate equipment must exceed the required
reliability over the designed storage life of the equipment as shown
in figure 4.3.2.1-1. If this is not possible, a cost tradeoff should
be made considering the use of redundant equipment to increase the
non-operating reliability above the required level vs. a change in the
maintenance policy to a periodic monitoring/repair policy (See section
4.3.2.2).
4.3.2.2 Rgriodic Monitoring/Repair Design Considerations
The objective of the periodic monitoring/repair maintenance concept is

to improve the reliability of non-operating equipment by periodically
monitoring (testing) the equipment to detect non-operating failures so
that they can be repaired before the equipment is called upon to
perform a mission. This is illustrated in figure 4.3.2.2-1. As shown
in the figure, the wooden round concept reliability does not meet
system requirements over the planned non-operating life of the
system. However, with the introduction of periodic monitoring and
repair, non-operating failures are repaired and the system is able to
achieve the desired reliability over the planned non-operating life.
In implementing the periodic monitoring/repair concept, it is very

important that the monitoring and repair operations are carefully
controlled in order to avoid the introduction of contaminants that
will cause long-term non-operating reliability problems, and to avoid
failures induced by the monitoring and repair operations, some of
which may go undetected. Because of this, it is not recommended that
the monitoring and repair operations be performed under field
conditions. In order to maintain the highest reliability and
availability, the monitoring and repair operations should only be
performed be highly trained personnel under environmentally controlled
conditions. Handling, teardown exposure times, and test times should
be kept to the absolute minimum.
The establishment of the optimum monitoring (test) interval is of

primary importance. An interval that is too short will result in
needless tests and inspections, lower availability, and an increase in
the number of repairs due to maintenance induced errors. An interval
that is too long will result in reduced reliability. Two methods for
determining the optimum test frequency for periodic testing of dormant
systems are presented in the following sections. Section 4.3.2.2.1
provides a method for determining the test frequency which minimizes
dormant system unavailability considering the impact of test induced
failures. A similar method for determining the optimum test interval
for achieving a dormant system end of storage life reliability
requirement is presented in section 4.3.2.2.2.
4.3-7
1.00
0 99
0 98
Uj
0.97 G
0 GOAL
I- -
z 096-
"NO TEST" Sb
F 095 - WOODEN
5 ROUND
rr (n94
z
a. 093
LU
0.92
0.91
0.90 I 1 I I I I I I I
0 1 2 3 4 5 6 7 8 9 10
TIME(YEARS)
Figure 4. 3. 2.1-1. "Wooden Round" Reliaoility Curve
4.3-8
1 00-T
I I I I
0 99
OA98
cc 097
0 GA
S096
-J NOTEST' *1
co 0 9r WOODEN
ROUND
09"1
z
0.93 ~um
0 92
091
0 1 2 3 4 5 6 7 8 9 10
TIME (YEARS)
Fi:r e 4. 3. 2.2- 1. Periodic n:-.eiR e11aDilI.ty Curve
4.3-9
4.3.2.2.1 Qtimum Test Intervals To Minimize Dormant SyJ.jrn

Unavailability
S.H. Sim (Ref. 79) has developed a model for determining the
optimum test interval to minimize dormant system unavailability
under two test policies:
I The system is tested every T hours.
II System tests are conducted W hours after the completion of the

last test or after completion of a repair, whichever comes later.
This model is based upon the following assumptions:
" The system is in steady state. Only one system is tested at a

given time.
* The system time to failure in the dormant state is exponentially

distributed.
* Testing can cause failures in the system that are unrelated to

failures in the dormant state.
" All failures are detected during the test.
" The system undergoes repairs immediately on completion of the test

after it has failed.
Test Policy I
For Test Policy I, Sim shows that system unavailability U is given by

the following:
1
U = X PiE[Dli]/T
i=o
where,
P = the probability that the system is in state i just before test.
E[DIi] = the expected downtime between the starts of tests n-1 and n,
given that the system is in state i just before test n-l.
T = the fixed test interval.
i = system state, i=O refers to a good state, and i=l refers to a failed
(bad) state.
4.3-10
Formulae and definitions for the derivation of these parameters are

given in Figure 4.3.2.2.1-1.
The optimal test policy is found by minimizing U with respect to T.

J.K. Vaurio (Ref. 77) has developed an approximation of this optimum
test interval as follows:
TOPT = (2(- +-/A)iX]I,2
Examples illustrating the optimum test interval for various conditions

are presented in Tables 4.3.2.2.1-1 and 4.3.2.2.1-2.
Test Policy II
For Test Policy II, Sim shows that system unavailability U is given by
the following:
1
U = PiE(Dli]/E(Lli]
i=o
where E[LIi] is the expected time interval b-tween the start of

tests n-l and n, given that the system is in state i just before test
n-l. Formulae and definitions for the derivation of E[DIi] are
given in Figure 4.3.2.2.1-1. All other parameters are as previously
defined.
The optimal test policy is found by minimizing U with respect to W.

Examples illustrating the optimum test interval for various conait ons
are presented in Tables 4.3.2.2.1-1 and 4.3.2.2.1-2.
4.3.2.2.2 Optimum Test Interval To Achieve Minim~um Dormant Reliability

Requirement
As shown in figure 4.3.2.2-1, periodic testing of dormant systems can

be employed to achieve a dormant reliability requirement over the
planned service life of a system when the "wooden round" maintenance
concept fails to do so. The objective in determining the optimum test
interval is to make the test interval as large as possible but not
allow the reliability to drop below the minimum reliability
requirement over the planned service life.
E.C. Martinez (Ref. 78) has shown that the minimum reliability just
prior to the N-th periodic test is given by the following:
RI(MIN) = e-((N-l)(l-a))IDTD]e-XDTD
and the reliability after the N-th periodic test is given by:
4.3-11
Figure 4.3.2.2.1-1 Parameters and Definitions for Optimal Test

Policies Minimizing Dormant System
Unavailability (Ref. 79)
Definitions
i = system state, i=O refers to a good state, and i=l refers to a

failed (bad) state.
Pi = probability that the system is in state i just before test.
qij = probability that the system is in state j just before test

n, given that it is in state i just before test n-l.
= test downtime.
X = constant failure rate for system.
(GA) = moment generating function of system repair time.
= mean repair rate.
= test-induced failure probability.
U = steady-state system unavailability.
E[DIi] = expected downtime between the starts of tests n-i and n,

given that the system is in state i just before test n-l.
E[LIi] = expected time interval between the starts of tests n-I and
n, given that the system is in state i just before test n-l.
Test Policy I (System Tested Every T Hours)
q00 = (1-x)e-XT + c ()e-X(T-T)
qi0 = 4(A)e-A(T-T)
00( = e"I/ for fixed constant repair time.
= - for exponential
repair times normalized to a specified time interval, T-T
4.3-12
Figure 4.3.2.2.1-1 Parameters and Definitions for Optimal Test

Policies Minimizing Dormant System
Unavailability (Ref. 79)(Continued)
Test Policy I (System Tested Every T Hours) (Continued)
P 0 = ql0/[l+ql0-q00]
P1 = [1-qo0]/[l+qlo-qo0]
E[DliO] = T-[l-(1-o + (x
E[D l i= l ] = T---(l-e - X( T - T ) ) , ( ) I),
1
U = JPiE[DIi]/T
i=o
Test Policy II (System Tested W Hours After Last Test/Repair Is

Com i e ted)
P0 = e-)W
P1 = 1-P0
E[mli=0] = T+ -i + W_(l-e-X W)/A
E[Dli=l] = t + +w-(-e-XW)/x
E(Lli=0] = + W + =I/
E[Li=--l] = + W
+-I
1
U lPiEDIi]/E[LIii]
i=o
4.3-13
Table 4.3.2.2.1-1. Optimal Test Policies for Component in Dormant
(Standby) State (Ref. 79)
(No Test-Induced Failure)
= 0.5 hour, P-1 = 6.0 hours
Test Policy I Test Policy II

Dormant
Failure
Rate X T* U* Po W* U* Po
(per hour) (Days) (%) ) (Days)ixi _%)_
- 5
0.5 x 10 18.6 0.23 99.6 18.6 0.23 99.6
- 5
1.0 , 10 13.2 0.32 99.3 13.2 0.32 99.3
4
0.5 x 10- 5.9 0.73 96.5 5.9 0.73 96.5
- 4
1.0 x 10 4.2 1.05 93.1 4.1 1.05 93.1
- 3
0.5 x 10 1.9 2.47 69.8 1.8 2.46 69.8
1.0 x 10- 3 1.3 3.62 48.8 1.3 3.57 48.7
4.3-14
Table 4.3.2.2.1-2. Optimal Test Policies for Component in Dormant

(Standby) State (Ref 79)
(Test-Induced Failure Probability = = 0.04)

1
= 0.5 hour, p- = 6.0 hours
Test Policy I Teat Policy II

Dormant
Failure
Rate X T* U* Po W* U* Po
(per hour', (Days) ill (Days) L1 %
- 5
0.5 x 10 22.7 0.27 99.6 22.7 0.27 99.6
1.0 x 10 - 5
16.1 0.39 99.3 16.0 0.39 99.3
4
0.5 x 10- 7.2 0.88 96.5 7.2 0.88 96.5
1.0 x 10 - 4
5.1 1.26 93.1 5.0 1.26 93.1
- 3
0.5 x 10 2.3 2.92 69.8 2.2 2.92 69.8
1.0 x 10 - 3
1.6 4.24 48.8 1.6 4.22 48.7
4.3-15
RN(MAX) = e-[N(I-a)-ADTD]
where,
N = the number of test periods.

a = the effectiveness of the periodic test in detecting failures.
XD = the failure rate for the period between tests.
TD = the time period between tests (See fig. 4.3.2.2-1)
Based upon these equations, the optimum test interval (TD) is found
by setting the minimum reliability at the last test interval
(RN(MIN)) to equal the reliability requirement (RR) of the system:
RR = e-[(N-l)(1-a)XDTD] e-XDTD
From this it is seen that:
TD = lnRR/((N(a-)-a),D]
where lnRR is the natural logarithm of RR.
The optimum test interval is then found by maximizing TD such that

the following conditions are met:
TD X N < TSL
and,
TD X (N+) > TSL
where TSL is the planned service life of the system.
As an initial test of whether it is possible to achieve a reliability

requirement for a given set of conditions, the limit of RN(MIN)
should be calculated as TD goes to zero. This yields the following:
RN(MIN)(TD=O) = e-(l-a))DTSL
If RN(MIN)(TD=0) exceeds the minimum reliability requirement, than

an optimum periodic test policy exists. If it is equal to the minimum
reliability requirement, then the optimum periodic test policy is to
continuously test the system throughout its service life (i.e.
TD=O). If it is less than the minimum reliability requirement, then
no amount of periodic testing will allow the attainment of the minimum
4.3-16
reliability requirement over the planned storage life without an

increase in the effectiveness of the periodic test.
The previous discussion does not include the impact of test induced
failures. If test induced failures are to be considered, then: (Ref.
78)
RN(MIN) = e-[(N-I)(I-a)(XDTD + FT)] e-)DTD
and,
RN(MAX) = e-[N(l-a)(XDTD + FT)]
where FT is the number of test induced failures due to on-off power

cycling and powered operation per test period.
The value of TD that must be maximized to obtain the optimum test

interval is then:
TD = lnRR/[(N(a-l)-a)XD] + [(N-l)(l-a)FT]/[N(a-I)-a)D]
As before, the optimum test interval is now found by maximizing TD such

that:
(TD+TT) x N < TSL
and,
(TD+TT) x (N+I) > TSL
where T T is the total power-on test time per test period.
4.3-17/(4.3-18 blank)
4.4 Testability Design
Testability addresses the extent to which a system or unit supports

fault detection and fault isolation in a confident, timely, and cost-
effective manner.(Ref. 82) Designing a successful testability pro-
gram involves the accomplishment of the following:
* Preparation of a Testability Program Plan.
* Establishment of sufficient, achievable, and affordable testabil-

ity, built-in, and off-line test requirements.
* Integration of testability into equipments and systems during the

design process in coordination with the maintainability design
process.
* Evaluation of the extent to which the design meets testability

requirements.
* Inclusion of testability in the program review process.
These tasks are the same regardless of how much time the system will
spend in a non-operating state throughout its service life. They are
fully described in MIL-STD-2165 (Ref. 82) and will not be repeated
here. RADC-TR-82-189 (Ref. 83) is another good testability design
reference. Whereas the MIL-STD-2165 treatment of testability pays
more attention to the managerial and organizational aspects of the
subject, RADC-TR-82-189 is more technically oriented and provides use-
ful guidance for designing testability features into a system. The
reader is encouraged to reference both of these documents in estab-
lishing a testability program.
Section 4.4.1 discusses testability concepts and section 4.4.2 ad-

dresses testability design considerations for dormant systems. This
information can then be used in conjunction with the recommended
references to develop a successful testability program for these
Fystems.
4.4.1 Testability Concepts
Testability is divided into three concepts: Manual Test, Automatic

Test, and Built-In-Test (BIT). Any of these test concepts can be
applied at system, subsystem, equipment, card, or component level to
aid in fault detection and isolation within that level. All three
concepts have been employed in dormant system designs.
4.4-1
In the manual test concept, the test equipment designed during the
integration of testability into the design of a system is intended to
be manually operated. Manual testing generally relies completely on
manual operation, operator decision, and evaluation of results. The
manual test equipment may consist of standard "off the shelf" oscillo-
scopes, voltmeters, signal generators, etc. which are manually at-
tached to specially designated test points external to the system.
However, this manual equipment may also be specially designed in it-
self. The essential characteristics of the manual test concept are
that:
a) The test equipment is not an integral part of the system. The

test equipment can be removed from the system.
b) Test personnel must manually connect the test equipment to the

existing external test points of the system.
c) The operation of the test equipment must be manually initiated

by test personnel.
d) Any switchover to redundant equipment in order to maintain sys-

tem availability must be manually performed.
In the "Automatic Test" concept, the test equipment designed during

the integration of testability into the system is designed to be auto-
matically operated. Under this concept, the performance assessment,
fault detection, diagnosis, isolation, and prognosis is performed with
a minimum of reliance on human intervention. According to MIL-STD-
1309 (Ref. 84) definitions, automatic test may include Built-In-Test
(BIT), however, throughout the following discussions automatic test
equipment is discussed separate from BIT. The automatic test equip-
ment definition in the following refers to a removable stand-alone
piece of equipment that is separate from the system. It should also
be noted that Automatic Test Equipment (ATE) may sometimes be manually
connected to the system and monitored by personnel. However, the
definition of such equipment specifies only "minimum reliance on human
intervention", not zero reliance on human intervention.
In the "Built-in-Test" (BIT) test concept, the test equipment designed

during the integration of testability into a system is designed to be
built into the system. BIT refers to an integral capability of the
mission equipment which provides an on-board, automated test capabil-
ity to detect, diagnose, and isolate system failures. The fault de-
tection and isolation capability is used for periodic or continuous
monitoring of a system's operational health, and for observation and,
possibly, diagnosis as a prelude to maintenance action.
4.4-2
BIT is further subdivided. There is "Initiated BIT" which is a type

of BIT which is executed only after the occurrence of an external
event such as an action by an operator. There is "Passive BIT" which
is non-disruptive and non-interferring to the prime system. There is
"Periodic BIT" which is & type of BIT which is initiated at some fre-
quency. An example is BIT software executing during planned processor
idle time. There is also "Turn-On" BIT, a specific type of initiated
BIT which is exercised each time power is applied to the unit or
system.
The essential characteristic of BIT equipment is that it is part of

the system itself. Current BIT systems typically use 1% to 10% of the
system hardware to test 90% to 100% of the system functions (Ref.
85). For additional information regarding BIT the reader should refer
to Reference 85.
The selection of an appropriate testability concept for a system en-

tails cost tradeoffs between the three basic concepts in order to
arrive at the most cost effective testability approach that achieves
reliability, availability, and supportability requirements at the low-
est life cycle cost for the system. The optimum approach may contain
a mix of the three basic concepts, or, in the case of some dormant
systems, it may contain no testability features. The blind applica-
tion of an approach such as BIT will not ensure that this objective is
met. As previously stated, the steps to be followed in selecting a
testability approach for electronic systems and equipments are de-
scribed in MIL-STD-2165.
4.4.2 Testability Design Considerations for Dormant Systems
The following sections provide testability design considerations for

dormant systems. The maintenance concept inpact on the selection of a
testability approach is discussed in section 4.4.2.1 and the impact of
test effectiveness is discussed in section 4.4.2.2. Section 4.4.2.3
addresses "Go/No-Go" testing versus "Parameter" testing for dormant
systems. Finally, testability design guidelines for dormant systems
are presented in section 4.4.2.4.
4.4.2.1 Maintenance Concept Impact on Testability
The maintenance concept chosen for a system will have a significant

impact on the degree to which testability is designed into a system.
In section 4.3, two broad maintenance concepts were identified for
dormant systems: "Wooden Round" and Periodic Monitoring/Repair. The
"Wooden Round" maintenance concept is essentially maintenance-free
storage of equipment through the elimination of organizational level
preventive maintenance throughout the storage period. In contrast,
4.4-3
the Periodic Monitoring/Repair maintenance concept consists of fixed

calendar inspections, operating times, checks, and overhaul of systems
throughout the storage period to maintain required reliability and
operational readiness states.
The "Wooden Round" maintenance concept objective of maintenance free

storage of equipment throughout the storage period implies that system
testability is not required. This implication is also reinforced by
the common objective that "wooden round" systems require no checkout
before use. The attainment of these objectives requires that the
system be designed for simplicity using a minimum of parts to minimize
the possibility of failures occurring. Since typical BIT systems com-
prise up to 10% of the system hardware, a significant decrease in the
failure rate can be achieved by the elimination of BIT hardware.
Built-in test points for automatic or manual external testing may pro-
vide paths for moisture and contamination intrusion and they contri-
bute to the overall system failure rate. They are unnecessary and
should also be avoided.
The Periodic Monitoring/Repair maintenance concept, on the other hand,

implies the incorporation of a high degree of testability in the sys-
tem. The optimum mix of the three basic concepts of manual, auto-
matic, and built-in-test that should be employed will rely on cost
tradeoffs to determine the mix that meets system reliability, avail-
ability, and supportability objectives at the lowest life cycle cost.
Regardless of the maintenance concept chosen for a system, the system

and its subparts must be designed for testability during fabrication,
in-process testing, acceptance testing, and, where appropriate, re-
pair. It is important that this is not forgotten, especially for
"wooden round" systems which require no field testability once de-
ployed. The guidelines provided in Reference 83 are applicable to the
incorporation of testability at this level of system design.
4.4.2.2 Test Effectiveness Impact on Dormant Systems
Test set fault detection capability has a marked effect upon the
feasibility of implementing periodic confidence (go/no-go) testing for
dormant systems. When confidence tests are considered, a reduction in
fault detecting effectiveness of the test set requires an increase in
system production requirements in order to achieve desired mission
reliability goals. The level of fault detection that can be achieved
is, in turn, dependent upon the configuration of the dormant system
and the dormant failure modes.
4.4-4
An analysis of missile system failure rates in a non-operating en-

vironment (Ref. 88) suggests that such failure rates are n.At con-
stant. In fact, the number of missiles failing confidence tests was
found to be virtually the same regardless of the dormant storage
times. This is attributed to the following: (Ref. 88)
1. Compounding of storage failures with factory recommended stress

screen escapes. The number of confidence testing rejects at-
tributed to this item is directly related to the environmental
stress screen (ESS) strength and the quality assurance levels
applied to vendor piece part screening as well as sub/prime cnn-
tractors in-house quality control programs.
2. Confidence test set operator errors: technical judgement re-

lated to accept/reject criteria and the technical level of the
operator.
3. Confidence test set calibration.
4. Confidence test set induced failures and false alarms.
5. Non-operating hours attributed to handling and shipment which

are usually equal for dormant storage controlled test.
6. Non-operating hours spent in dormant storage which are directly

related to the non-operating environment.
Item (1) is expected to appear as a decreasing hazard rate during dor-

mancy, but at a much slower rate than that observed during operating
environments. Items 2-5 are independent of time and should result in
a fixed number of failures each time that the confidence test is per-
formed. Fail1ures attributed to items 2-5 can be substantial. Accord-
ing to a Redstone Arsenal study (Ref. 18), as many as 50 percent of
confidence test rejected missiles were erroneously faulted or were
found to contain test-induced failures. Item ,5) lz the rczaL of the
hazard rate distribution for dormant storage and is assumed to be con-
stant during non-operating periods (i.e. the number of failures will
increase as a function of non-operating time.).
It was determined in Reference 88 that regardless of the overall dis-

tribution , non-operating failures exert themselves in dormant systems
as a direct result of the number of parts used in the system design
(i.e. the higher the complexity of the design the higher the fallout
during testing). While the level of test rejects related to items 2-5
can be minimized by test set design changes. operator instructions,
and BIT enhancement, the factors attributed to items 2-5 will over-
shadow that expected from item (6) for systems using a small number of
4.4-5
parts (e.g. 2000 electrical parts). For such systems, a confidence

test would be counter productive. In general, the advantage of imple-
menting confidence testing increases as an inverse function of non-
operating reliability.
Mathematical models demonstrating the impact of test effectiveness and

test-induced failures on dormant system reliability and availability
are presented in section 4.3.2.2.
4.4.2.3 "Go/No-Go" Versus "Parameter" Testing
The time interval between tests for dormant systems is typically quite
long, often exceeding 12 months. While the failure mechanisms that
occur during non-operating periods are of the same type that occur
during operating periods, they occur at much slower rates. Critical
to maintaining the reliability and availability of dormant systems is
the ability to not only detect failures that have occurred since the
last test, but also to determine the likelihood of failures occurring
before the next scheduled test or system demand.
Many BIT systems are designed for "go/no-go" testin- for the purpose
of monitoring the general well-being of the system, informing the
operator of any malfunction, and aiding in the location of failed com-
ponents. Simple "go/no-go" test results, however, are not able to
indicate if a system has degraded from the last test but is still
operational, or if a failure is imminent. Often an impending failure
can be anticipated due to early signs appearing in the main or BIT
systems. They can take the form of noise, voltage level changes, ex-
cessive current drains, or missing data bits.
Early recognition of these signs can be of great benefit since it

could prevent a mission failure. In cases where components are known
to degrade over their lifespan and these characteristics can be meas-
ured, a test point should be installed. Key parameters should also be
monitored and stored for an early failure prognostic by a computer or
logic circuit. The removal of a questionable unit and rtplacement
with one that is not suspected of impending breakdown can considerably
increase the confidence factor in a mission.
A non-volatile memory such as an EPROM may be used as a maintenance

log. It can be incorporated into the BIT test system to store the
history of the parameters tested and record component failures. This
could then be made available during test periods to determine trends
and provide more efficient fault diagnostics. If automatic test
equipment is used the equipment can perform this same function. Even
if manual testing is employed, the actual parameter values should be
recorded and analyzed for trends from test to test rather than just
recording whether or not the item passed or failed the test.
4.4-6
4.4.2.4 Titability Design Guidelines for Dormant Systems
As previously stated, the tasks involved in implementing a successful

testability program are the same regardless of how much time the sys-
tem will spend in a non-operating state throughout its service life.
Testability program tasks are fully described in MIL-STD-2165 (Ref.
82) and technical design guidance for incorporating testability fea-
tures are covered in RADC-TR-82-189 (Ref. 83) and the Joint DARCOM/
NMC/AFLC/AFSC Built-In-Test Design Guide (Ref. 85). The following
guidelines are provided to supplement these sources in designing for
dormant system testability:
0 The selection of data monitoring points for R & D tests should

also be harmonious with those to be required in subsequent
stockpile testing. Careful selection should be made to maximize
the requirements for the full life cycle of the program while
minimizing cost. One way is to design data monitoring points,
telemetry counterweights, antenna cutouts, etc., into the final
technical data package. In this way, dormant systems randomly
selected for testing can easily be modified to accommodate
telemetry components, etc. Further, since data monitoring
points will be identical to those collected during R & D, re-
sults of periodic storage tests can be readily compared to base-
line data. (Ref. 89)
" Technical manuals should stress the proper care and procedureF to
be followed so that unnecessary damage will be avoided. Where
extensive handling, mating, and re-assembly of hardware will be
required as part of the test, the most used hardware should re-
ceive special design attention to make it more rugged and less
susceptible to damage.
" Removable hatches (knock-outs) should be incorporated in missile

skins to add telemetry antennae for missile flight tests. In
addition, standard telemetry connectors should be used. (Ref. 90)
" Components that are designed to respond to environmental stimulae

should be designed to be easily disconnected so that they can be
connected to simulators for individual testing. (Ref. 90)
" It is erroneous to assume that simply because a mechanical device

has been operationally checked that it is as-good-as-new. (Ref. 91)
" Since operating failure rates are higher than non-operating fail-
ure rates, equipment "on" time during testing should be minimized
to avoid degrading reliability.
* Tests should be carefully analyzed to ensure that they will not

induce failures or degrade equipment reliability.
4.4-7,' 4.4-8 blank)

4.5 PACKAGING
The basic purpose of packaging is to ensure that items are fit to

perform their intended function when the time comes for them to be
used. It must protect an item from the time of production, through
transport and storage, to operation and use. The selection of
adequate packaging protection must be based on the application of
packaging techniques and materials to protect an item against known or
anticipated logistics environment conditions. In order to provide
this protection, three factcrs must be known to the designer: (Ref. 46)
(1) The physical and mechanical properties of the item.
(2) The conditions of the environment to protect against.
(3) The technical capabilities of the packaging design to protect

the item.
Factors (1) and (2) are covered in Sections 3 and 4.1 of this
document. Section 3 discusses dormancy environments and provides
general guidelines for protecting equipment from the effects of the
natural and induced environmental factors experienced in thqse
environments. Section 4.1 presents part selection and control
information and provides specific information concerning the impact of
these enviroramnental factors on the various piece parts that make up
electronic equipment designs. The third factor is the subject of this
section.
Proper procection against damage and deterioration of components and

equipment during shipment and storage involves the evaluation of a
large number of interactive factors and the use of tradeoff analysis
to arrive at a cost-effective combination of protective controls.
MIL-P-116 (Ref. 47) describes six basic methods of preservation to
protect equipment against corrosion, physical and mechanical damage,
and other forms of deterioration. These methods differ in the amount
of protection afforded against environmental factors. The six methods
are:
I Preservative coating applied, wrapper not sealed. Water as

liquid or vapor and corrosive atmosphere has relatively free
contact with the equipment.
IA Preservative coating applied as required, water- vaporproof

barrier sealed. Only traces of water-vapor penetration to
equipment is possible.
IB Equipment, wrapped or unwrapped, is enclosed within a

coating of strippable compound. Nc penetration of liquid
4.5-1
and only traces of water-vapor can penetrate to the

equipment.
IC Preservative coating is applied as required, waterproof or

water-resistant barrier is sealed. Only water-vapor can
penetrate to the equipment.
II Preservative coating is applied as required; waterproof,

water-vaporproof barrier is sealed with dessicant inside.
Only traces of water-vapor can penetrate to the equipment
and this is absorbed by dessicant.
III No preservative is applied to the equipment. The equipment

is packaged for physical and mechanical protection only.
Liquid and water-vapor has relatively free access to the
equipment.
The amount of protection required is dependent upon the envizonment to

which the equipment will be subjected and the susceptibility of the
equipment to damage by this environment. If electronic equipment is
to be stored in a dry, well-ventilated warehouse, where the
temperature of the air surrounding the equipment can be regulated so
that it does not fall to dewpoint values, then relatively little
protection is required such as afforded by Methods I and III.
Generally, however, this is not the case for inost military equipment
and more protection will be required such as afforded by Methods IA,
IB, IC, and II.
The amount of protection to apply should be based upon the worse case
conditions to which the equipment is likely to be exposed. This is
especially true for dormant equipment designs since any degradation
that does occur during non-operating periods may go unnoticed for long
periods of time, possibly resulting in equipment failure and loss of
mission. As stated in Section 4.1, the most important environmental
stresses during non-operating periods are mechanical, chemical and low
thermal. The minimal packaging for dormant equipment designs should
provide protection from these stresses.
'.detailed discussion of packaging designs is beyond the scope of this

document. Detailed guidelines for packaging and protecting equipment
are presented in references 48 and 49. Reference 48 is a design
handbook for packaging and pack engineering and reference 49 presents
design guidelines for the prevention and control of corrosion. The
following sections provide additional information applicable to the
packaging and protection of dormant electronic equipment. Section
4.5.1 presents a discussion of material degradation. Section 4.5.2
discusses cleaning methods and 4.5.3 discusses the dielectric
4.5-2
embedding of electrical and electronic assemblies. Assembly packaging

issues are discussed in section 4.5.4 and shipping and storage
container design is covered in section 4.5.5. Finally, section 4.5.6
presents guidelines for the development of in-storage cyclic
inspections.
4.5.1 Material DekrioratiQn
This section describes the fundamental factors causing deterioration

of materials, the effect of climatic conditions on materials, and the
recommended storage conditions for different types of materials.
Materials are classified into two basic groups: metal and organic
materials. It can generally be stated that metals deteriorate
primarily by electrochemical processes while organic materials
generally undergo deterioration due to chemical reactions. Many of
the factors which are influential in the process of deterioration are
present everywhere in the atmosphere. With an understanding of the
deterioration of these fundamental material types the design engineer
will be mote effectively prepared to design proper preservation and
packaging for the dormant environment.
4.5.1.1 Metal Deterioration
The principle deterioration mechanism for metals is corrosion.

Corrosion is a process that occurs when metals are placed in contact
with water or a chemical solution. There is a tendency for
electrically charged metal particles (ions) to go into solution.
Because the solution must remain electrically neutral an equivalent
number of positive ions of another element must be displaced. In the
case of a metal such as iron in water, hydrogen is plated out on the
surface of the metal as a thin film, known as corrosion (or rust).
The most important effects of temperature on the corrosion of metals

are its relation to solubility and chemical reactions. An increase in
temperature generally increases the rate of chemical reactions and
reduces the solubility of gases in solution. Changes in temperature
may also affect the solubility of reaction products in such a manner
as to change the nature of the corrosion products. (Ref. 50)
In order to prevent corrosion of metals, control must be extended over

the influencing factors. The prevention of metal corrosion during
storage relies on the use of various coatings, inhibitors and
passivators, and dehumidification.
4.5-3
There are two types of protective metallic coatings that can be used
to inhibit corrosion. Cathodic coatings coat the metal surface with a
more noble metal resulting in reduced corrosion. Pinholes and
discontinuities in the coating accelerate localized corrosion where
the base metal is exposed. Care must therefore be used in the
application of cathodic coatings and in some cases it may be advisable
to apply an outer insulating coating. Anodic coatings, being more
anodic than the base metal are preferentially attached in the
electrolytic process. The anodic protective properties will however,
eventually degrade. The anodic coatinq will still provide a
protective coating even if the coating is likely to be broken or
contain pinholes. (Ref. 50)
Organic coatings are widely used in the protection of metal surfaces

from atmospheric corrosion. They act as mechanical barriers
preventing corrosive mediums from reaching the metal surface. The
performance of an organic compound is therefore largely dependent on
its permeability and its ability to adhere to the metallic surface and
in many cases it is necessary to apply more than one coat to obtain
satisfactory results.
An inhibitor may be defined as any chemical substance or mixture that

effectively decreases the corrosion rate when added in small amounts
to a corrosive environment. The type of material to be protected and
the environment to which it is subjected must be thoroughly understood
before the selection of the inhibitor can be made. An inhibitor may
effectively decrease the corrosion of a metal in one environment and
actually promote corrosion of the same metal in a different
environment. In some cases, an inhibitor may decrease and at the same
time promote localized pitting. (Ref. 50)
Passivators are inhibitors which change the electrode potential to a

more noble value. The tendency of an anodic inhibitor to act as a
passivator is greater than that of a cathodic inhibitor. The ability
of an inhibitor to act as a passivator as well as the amount of
inhibitor required are dependent upon the type of metal, the
environment, and the temperature. (Ref. 50)
Dehumidification is a method of preventing metal corrosion by reducing

the amount of water vapor which is in the air. By proper humidity
control it is possible to keep the moisture content of air at a level
low enough so that corrosion is not promoted. The term humidity
control applies equally to both air conditioning processes of
humidification and dehumidification. These two processes are
differentiated by the fact that one adds moisture and one removes
moisture from the air.
4. 5-4
For a detailed discussion of the factors influencing corrosion, forms

of corrosive attack including stress-corrosion cracking and
hydrogen-stress cracking (hydrogen embrittlement), corrosion
characteristics of metals, general methods of protecting metals and
alloys from corrosion, and information on corrosion testing, the
reader should refer to MIL-HDBK-729.
4.5.1.2 Organic Deterioration
The basic unit of composition of many organic materials is either

cellulose or hydrocarbon derivatives and they can be affected by
numerous chemical factors. Upon exposure to undesirable conditions, a
chemical change takes place which results in changes in the physical
properties of the materials. The changes brought about by the factors
are not common to all materials and likewise the degree of exposure to
such factors produces varied results. The chemistry involved in the
chemical changes taking place in the deterioration of organic
materials is extremely complicated and cannot be discussed within the
scope of this text. However, it is intended that this discussion will
cover the highlights of most organic materials and the factors which
are primary in causing deterioration. This section gives a brief
overview of the material found in Reference 51.
The deterioration of plastic and rubber materials is difficult to

describe without '-ecoming involved in lengthy discussions of each
individual compound. Plastics and rubber are usually considered in
the same light because they basically have the same molecular
structure. They are referred to as polymers of high molecular
weight. Chemical deterioration of plastics results in cracking,
reduced strength, warping, and loss of transparency. Chemical changes
in polymers depend to a large extent upon the basic design of the
polymer. Polymers are classified in two main groups according to the
molecular structure - (1) linear or chain polymers and, (2) branched
network polymers. The linear polymers are often referred to as
thermoplastics and the branched polymers as thermosetting polymers.
Electrical equipment in general consists of a combination of organic

materials and metals assembled together. It follows that the basic
materials of construction are affected by the same physical and
chemical agents. This type of equipment when exposed to moisture not
only undergoes metal corrosion and insulation deterioration, but
electrical properties and performance are also affected.
The deterioration of materials of organic origin is to a large extent

dictated by the end use of the material and the degree of exposure to
which the material is subjected. By anticipation of the end use of
the material it is possible to add or provide resistance to exposure
and thereby increase the shelf life of the material. Many organic
4.5-5
materials cannot be treated in a manner similarly used for metals by

protective coatings and surface treatments to improve resistance to
various undesirable elements of exposure. The general procedure in
this case is to either condition the environment to remove the
environmental factor causing the deterioration or select a material
that will not be adversely affected by the environment. For example,
to give added resistance to materials which are affected by exposure
to the physical elements such as heat, sunlight and moisture,
additives are added which tend to retard the chemical reactions which
take place when these materials are exposed.
Great care must be exercised in using organic materials in dormant

designs, especially in low pressure environments and in sealed
assemblies, because they may outgas harmful contaminants. Reference
1 should be consulted for outgassing data on organic materials.
4.5.2 Cleaning
The overwhelming area of concern for long-term non-operating

reliability is the influence of chemical contaminants. They can be
introduced at any point in the fabrication process and include
fingerprints, perspiration, inorganic residues, organic residues,
water, and other liquids. To obtain the maximum benefits from the
various preservation and packaging methods, items must be perfectly
clean prior to preservation and packaging in order to prevent any
chemical action that might result in corrosion or other forms of
deterioration resulting from these contaminants. Basic cleaning
requirements are established in MIL-P-116. To maximize the
reliability of long-term non-operating equipment, the methods and
procedures for precision cleaning described in MIL-HDBK-407 should be
followed.
4.5.3 Dielectric Embedding
To isolate circuit components from the degrading effects of

environmental factors such as oxygen, moisture, heat and cold, and
mechanical shock and vibration, components and assemblies can be
coated, buried, or encased in dielectric materials. The earliest
substances used for such purpose were materials such as waxes and
asphaltic materials. These are now used to a limited extent.
Synthetic polymers are currently most widely used for
embedding. (Ref. 52)
The materials most employei are epoxy resins. Other materials

commonly used are the polyurethanes and the silicones. Additional
types finding special uses are vapor-deposited polyxyly]enes.
Thermosetting hydrocarbons, thermosetting acrylics, polyesters, and
polysulfide resins are used less often. (Ref. 52)
4.5-6
Embedding does not provide hermetic sealing, however, it increases the

reliability of any given assembly by sealing it against moisture,
dirt, fungi, and other contaminants. Also, components are fixed in
position resulting in the mechanical strength of the embedded assembly
being greatly enhanced against vibration and shock.
There are some limitations in the use of embedded electrical and

electronic assemblies. Many are hard to repair. Although flexible
and rubberlike polymers can be repaired, any repair can present
difficulties. Additionally, the weight of an assembly is increased by
embedding since in most cases the amount of additional mechanical
structure for protection without embedding can be designed to be
relatively light. Embedding resins have higher dielectric constants
and loss tangents than air which limits their use in applications
where very low electrical loss is a desirable factor. Offsetting this
is the fact that voltage breakdown between two potential points is
improved. The advantages and disadvantages of embedding
elt :tronic/electrical components are summarized in Table 4.5.3-1. For
further details concerning the embedding of electronic/electrical
assemblies including the selection and application of coatings and
embedments which will minimize failures due to moisture, corrosion,
and biological degredation during dormant periods the reader should
refer to reference 52.
4.5.4 Assembly Packagina
This section provides guidelines for assembly level packaging of

equipment to mitigate the effects of environmental factors relevant to
dormant periods. Section 4.5.4.1 covers moisture protection measures
and section 4.5.4.2 covers sand and dust protection measures.
Mechanical shock and atmospheric pressure protection measures are
covered in sections 4.5.4.3 and 4.5.4.4 respectively. Solar radiation
is addressed in section 4.5.4.5 and electromagnetic radiation is
addressed in section 4.5.4.6. Finally, section 4.5.4.7 addresses
nuclear radiation.
4.5.4.1 Moisture Protection
Moisture is present in various environments, such as: humidity, rain,

snow, and fog. Humidity closely follows temperature in importance as
an euvironmental factor. (Ref 7) Moisture is a chemical and is
probably the most important chemical deteriorative factor of all.
Moisture is not simply H20, but usually is a solution of many
impurities. In addition to its chemical effects, such as the
corrosion of many metals, condensed moisture also acts as a physical
agent. An example of the physical effects of moisture is the damage
done in the locking together of mating parts when moisture condenses
4.5-7
4
Table .5.3 --. Advantages and Disadvantages of Embedding
Electronic/Electrical Components (Ref. 52)
Advantages
a Use Reliabilityi
,1) Sealing (not fully hermetic) against fungi, water vapor, and gross moisture, dirt, gases; assemblies are fixed in
resin of known mechanical and dielectric characteristics.
,2) Packaging strength (shockproofing, antivibration re~ponse) increased
b Improved Design:
1) Air spaces are eliminated.
(2) Components are held in compact three-dimensional form.
S3) Wider application of module construction, miniaturization, and plug-in units is permitted
(4) Selection of resins allows upgrading of electrical performance (e g , low-loss response of high frequencies. in.
creased thermal resistance, and/or heat dissipation)
t5) Colored resins may be used for identification of circuit components.
6) Electrical noise in high-gain amplifier devices is reduced
c Economy
:1) .Most or all mounting hardware which may add up to 25-30% weight to an assembly is eliminated
2) Need for auxiliary protection for the components is reduced or removed since the resin matrix now serves this pur-
pose.
3) Less skilled personnel can remove and replace embedded units.
k4) Circuit assembly is more rapid since use of point-to-point wiring can be made (e.g , in place of circuit boardsl
2 I)isadvantages
a l)ifficult Repairs:
I'] Embedded assemblies are not easily accessible for making minor repairs.
'2) Solvent soaking procedures are difficult
-%3,Hui-dliliiig (with tianspacert matrices) is expensise and time-consuming.
,41 Embedded circuit must be treated as an expendable unit (though costly, embedding can be shown to increase re-
liability and prevent iampering)
h Lowered Heat Dissipation
,I) Thermal dissipation in resins is lower than in air -- temperature derating may be required.
,2' Heat-sink and other sophisticated design variants may be required to control heat buildup.
F'hermal Limits
I Most resin stability is limited above 200 0 C (certain silicones can surpass this temperature)
,2; (ertain high-temperature rated components are required in various systems and require special packaging (but
in many uses moderate temperature limits are satisfactory)
3' With low temperatures, sharp and irregular parts of components can possibly cause resin cracking. Filled or lexi-
hle resins improve lo,, temperature performance but at the general sacrifice of electrical properties (however, sili-
(one elastomeric resins can be used with sharp-edge assemblies)
d Weight Increase
i1 ' Certain applications can add excess weight to an assembly
2. l)esign techniques may be required in tertain instances ke g , air-borne or space components) to reduce weight;
:e q . use of (onformal coating rather than potting or casting)
,1, Certain foamed resins and low density ,hollow beady compounds can be used to reduce .eight
r \dverse l)ielectric Properties
I - Components can increase circuit capacitance by having dirlectric constants close to that of the embedding resin
2' \%th high frequenty output, electrical losses can be increased. however, design methods can be used to com-
pensate for known dcielectri(c properties, uniform for given conditions. in the circuit
fVariable Stresses in Cured Matrix
Si. Shrinkage occurs (luring resin (uring
21 Difference in :oeffit-ents of thermal expansion ,resin/metal/glasses.'other materials) is a source of problems -
,e. breakage. irushing, other component damagze but effects are lessened with use of flexibilized resin, or
elastomer coatings. e g . sili, ones
4.5-8
on them and then freezes. Similarly many materials that are normally
pliable at low temperatures will become hard and perhaps brittle if
moisture has been absorbed and subsequently freezes.
Moisture in conjunction with other environmental factors, creates

difficulties that may not be characteristic of the factors acting
alone. For exam ple, abrasive dust and grit, which would otherwise
escape, are trapped by moisture. The permeability (to water vapor) of
some plastics is related directly to their temperature. The growth of
fungus is enhanced by moisture, as is the galvanic corrosion between
dissimilar metals. (Ref 1)
Salt fog environments can be a particularly severe source of moisture

degradation. It severely promotes corrosion effects in metallic
components and can foster the creation of galvanic cells, particularly
when dissimilar metals are in contact. Another deleterious effect of
salt fog is the formation of surface films on nonmetallic parts which
cause leakage paths and degrade the insulation and dielectric
properties of these materials. Absorption of the salt solution by
insulating materials can cause a significant increase in volume
conductivity, and dissipation factor of materials so affected.
Some design techniques that can be used singly or combined to

counteract the effects of moisture and to minimize moisture intrusion
are: (Refs. 1, 49)
(1) eliminate moisture traps by providing drainage or air

circulation,
(2) use dessicant devices with humidity indicators visible from
outside the equipment to remove moisture when air
circulation or drainage is not possible, and change the
dessicant when the indicator registers greater than 20%
relative humidity,
(3) use parazylylene or other conformal coating on printed
wiring boards (see section 4.5.3),
(4) use clear conformal coatings to coat exposed surfaces,
(5) provide rounded edges to allow uniform coating of protective
material,
(6) use materials resistant to moisture effects, fungus,
corrosion, etc.,
(7) use hermetically sealing components, gaskets and other
sealing devices,
(8) use impregnating or encapsulating materials with moisture
resistant waxes, plastics, or varnishes,
(9) separate dissimilar metals, or materials that might combine
or react in the presence of moisture, or of components that
might damage protective coatings,
4--9
(10) use "0" rings for sealing around all control shafts that
must penetrate into the enclosure,
(11) avoid the use of knurled wheels that intrude into the
enclosure. Effective sealing around such wheels is
extremely difficult,
(12) access plates utilizing a gasket for sealing are less
subject to leakagE when mounted on a vertical surface,
(13) mount electrical connectors horizontally (through vertical
sides). If it is essential to mount a connector vertically
on the top of a housing, provide a slightly (at least 3/16
of an inch) raised base foL the connector mounting relative
to the surrounding horizontal area,
(14) use an "L" type connector so the wiring enters horizontally
into the upper portion of a vertically mounted connector,
(15) for all multicontact connectors, wiring shoulO lead upward
toward the connector to prevent water from running along
wire into the connector. Strain relief is essential to
ensure no side loads exist on wires entering backshell
grommets or potting.
(16) use solder flux with the lowest possible acid content to
minimize acid induced corrosions,
(17) mount printed wiring boards vertically with the edge
connectors on vertical edge or back of board
(18) mount equipment and components at least 1/2 inch above
potential standing water level.
(19) thoroughly clean all items to remove contaminants such as
residual process chemicals, etc.
The design engineer should design on the assumption that moisture will
eventually get into the equipment. Except for hermetically sealed or
pressurized equipment, this means protection from moisture effects
must be provided inside of the equipment, and provisions must be made
to let the water drain out of the equipment. (Ref. 49)
Ideally, the maximum resistance to moisture is achieved by the

hermetic sealing (solder or glass fusion) of an equipment or
component. The hermetic sealing must include a dry, inert gas
atmosphere within the sealed box.
There remains much controversy over what is an acceptable level of

water vapor within a hermetically sealed package. Livesay (Ref. 3)
suggests that 200 ppm of water vapor is the maximum allowable for high
reliability. For a package with an internal volume of 2200 cm 3 , a
leak rate of 2 x 10 12 atm cm3 sec - 1 would be required to
maintain the moisture content below the 200 ppm level for 10 years.
Such a seal is very difficult to attain. In addition, cooling
requirements, size (rigidity), container penetrations for wiring and
4.5-10
similar considerations also generally preclude hermetic sealing.

(Ref. 49)
The next greatest resistance to moisture is obtained through the use

of a pressurized equipment housing or compartment. The introduction
of pressurized dry air into a semi-sealed (organic seals vice solder
or glass fusion) area qreatly minimizes moisture intrusion while there
is a positive pressure differential over ambient. During periods when
positive pressurization is not available, however, there will be
increased diffusion of ambient air into the housing or compartment.
While this is a very effective means to minimize moisture intrusion,
the additional weight required to achieve the necessary housing
rigidity and pressure tight sealing has largely precluded the use of
this configuration.
For equipment with neither hermetic sealing nor pressurization,

protection is best achieved by applying a sealant to prevent all fluid
intrusion. Two types of sealants that are often used for this type of
application are polysulfide sealants and RTV (silicone) sealants.
Non-acetic acid cure RTV is useful for sealing small areas. Clear
types of RTV allow visual inspection of the sealed surface, however,
these clear RTV may be susceptible to damage by operational fluids.
Acetic acid cure RTV should not be used to seal equipment, because
acetic acid will cause corrosion. Table 4.5.4.1-1 lists those RTVs
(silicone sealants), adhebives, and coatings considered corrosive.
For further information concerning non-hermetically sealed assemblies
and the use of dielectric embedding as a means of protecting against
moisture, dirt, fungi, and other contaminants, the reader should refer
to section 4.5.3 and references 49 and 52.
The designer must also consider possible adverse effects caused by

specific methods of protection. Hermetic sealing, gaskets, protective
coatings, etc., may for example, aggravate moisture difficulties by
sealing moisture inside or contributing to condensation. The gasket
materials must be evaluated carefully for outgassing of corrosive
volatiles or for incompatibility with adjoining surfaces or protective
coatings. Other materials sealed within the assembly or container
should also be evaluated for outgassing of corrosive volatiles. Solid
and liquid propellents in all-up-round (AUR) configurations are of
particular concern to the long-term non-operating reliability of
dormant systems.
4.5.4.2 Sand and Dust Protection
Sand and dust primarily degrade equipment by: (Ref 1)
(1) Abrasion leading to increased wear,
4.5-11
Table 4.5.4.1-1. Corrosive Silicone Se.lants,

Adhesives, and Coatings (Ref. 69)
RTV Q3-6090 RTV 785 RTV 92-018 RTV 103
RTV 140 RTV 891 RTV 92-024 RTV 106
RTV 142 RTV 1890 RTV 92-048 RTV 108
RTV 236 RTV 3144 RTV 92-055 RTV 109
RTV 730 RTV 20-046 RTV 94-002 RTV 112
RTV 731 RTV 20-078 RTV 94-003 RTV 116
RTV 732 RTV 30-079 RTV 94-009 RTV 118
RTV 733 RTV 30-121 RTV 94-034 RTV 154
RTV 734 RTV 90-092 RTV 96-005 RTV 156
RTV 780 RTV 92-00; RTV 96-080 RTV 192
RTV 781 RTV 92-007 RTV 96-0P1 XTV 198
RTV 784 RTV 92-009 RTV 102
4.5-12
(2) Friction causing both increased wear and heat, and

(3) Clogging of filters, small apertures, and delicate equipment.
Equipment having moving parts require special care when designing for
sand and dust protection. Sand and dust will abrade optical surfaces,
either by inipact when carried by air, or by physical abrasion when the
surfaces are improperly wiped during cleaning. Dust accumulations
have an affinity for moisture and, when combined, may lead to
corrosion or the growth of fungus.
In the relatively dry regions, such as deserts, fine particles of dust

and sand readily are agitated into suspension in the air where they
may persist for many hours, sometimes reaching heights of several
thousand feet. Thus, even though there is virtually no wind present,
the speeds of vehicles or vehicle-transported equipment through these
dust clouds can cause surface abrasion by impact, in addition to the
other adverse effects of the sand or dust. Although dust comonly is
considered to be fine, dry particles of earth, it also may include
minute particles of metals, combustion products, solid chemical
contaminants, etc. These other forms may provide direct corrosion or
fungicidal effects on equipment, since this dust may be alkaJine,
acidic, or microbiological.
Since dust is present in almost any environment, the question is not

whether to allow dust to enter, but, rather, how much or what size
dust could be tolerated. The problem becomes one of filtering any
circulating air to remove dust particles above a nominal size. The
nature of filters, however, is such that for a given working filter
area, as the ability of the filter to stop increasingly smaller dust
particles is increased, the flow of air or other fluid through the
filter is decreased. Therefore, the filter surface area either must
be increased, the flow of fluid through the filter decreased, or the
allowable particle size increased, i.e., invariably there must be a
compromise. Sand and dust protection, therefore, must be planned in
conjunction with protective measures against other environmental
factors. It is not practical, for example, to specify a protective
coating against moisture if sand and dust will be present, unless the
coating is carefully chosen to resist abrasion and erosion, or is
self-healing.
4.5.4.3 Vibration and Shock Protection
Electronic equipment is often subjected to environmental shock and

vibration during normal use and testing. Such environments can cause
physical damage to components and structural members when deflecticns
produced cause mechanical stresses which exceed the allowable working
stress of the constituent parts.
4.5-13
The natural frequencies of subsystems comprising the equipment are

important parameters which must be considered in the design process
since a resonant condition can be produced if a natural frequencies
within the vibration frequency range. The resonant condition will
greatly amplify the deflection of the subsystem and may increase
stresses beyond the safe limit. The vibration environment can be
particularly severe for electrical connectors since it may cause
relative motion between members of the connector. This motion can
produce fret corrosion which generates wear debris and causes large
variations in contact resistance.
The most critical mechanical stress applied to a system is that of

shock. Shock occurs when the equipment is subjected to a suddenly
applied force, for example acceleration, gunfire, or when a system
falls freely. Shock damage is caused by acceleration forces developed
during impact. A methodology for calculating the imract force
experience by a syt-am is presented in Reference 46. The underlying
principle behind all mechanical stress protection packaging is that
every acceleuating object possesses kinetic energy which it must
dissipate by decelerating through a given distance and time interval.
More comprehensive guidelines for designing systems to withstand
mechanical stresses are presented in section 4.5.5 (Container
Design).
4.5.4.4 Pressure Protection
The effects of pressure are not sufficiently common to make it one of

the more important environmental factors. Pressure variations due to
meteorological processes range from lows of about 880 mb to an upper
extreme of 1,083 mb. The low pressures are associated with the eyes
of tropical storms while the high pressures are associated with winter
time continental high pressure systems.
Pressure variations also are associated with changes in altitude. As

altitude increases, the mass of the air above a given point decreases
and, consequently, the pressure decreases.
Variations in pressure can cause the rupture of seals, distort

containers, or move objects. Pressure changes cause leakage of fluids
fro,. containers and control systems, the condensation of trapped water
vapor, and equivalent effects. Other effects of pressure include
those phenomenon associated with the availability of a certain amount
of air. Combustion processes are less efficient at lower pressures,
the lubrication capability of oils and greases decreases, electrical
breakdown occurs more frequently, heat transfer is less efficient, and
liquids vapoiize more readily. Design engineers should be alert to
these effects of pressure on materials chosen for application in
pressure changing environments.
4.5-i4
4.5.4.5 Solar Radiation
The effects of solar radiation on material are often similar to those

of high temperature since solar radiation elevates the temperature of
many materials quite rapidly. Nonthermal effects, however, are
important. The high energy short wavelength components of solar
radiation induce reactions in materials that often deteriorate their
functional properties. Textiles, paper, plastics, rubber, and various
surface coatings are susceptible to solar radiation induced changes.
A commonly observed effect is that of bleaching of colors from
textiles through the action of sunlight. This is brought about
through chemical changes in the dyes induced by the ultraviolet
portion of the solar radiation. Solar radiation is an environmental
factor for which special material requirements are established that
may be as costly as the deteriorating effects that solar radiation has
on materials.
4.5.4.6 Elecromagnetic Radiation
Electromagnetic radiation can cause disruption of performance levels

and, in some cases, permanent damage to exposed equipment.
Electromagnetic radiation often produces interference and noise
effects within electronic circuitry which can impair the functional
performance of the system. Sources of these effects include corona
discharges, lightning discharges, sparking and arcing phenomena.
These may be associated with high voltage transmission lines, ignition
system, brush-type motors, and even the equipment itself. (Ref 1)
Protection against the effects of electromagnetic radiation has become

a sophisticated engineering field of electromagnetic compatibility
design. The most direct approach to protection is, in most cases, to
avoid the region in which high levels of radiation levels are found.
When exposure cannot be avoided, shielding and filtering are important
protective measures. In other cases material design changes or
operating procedural changes must be instituted in order to provide
protection. (Ref 1)
4.5.4.7 Nuclear Radiation
Although a natural background level of nuclear radiation exists, the

only nuclear radiation that is of interest is that associated with man
made sources such as reactors, isotope power sources, aznd nuclear
weapons. The most important of these sources is nuclear weapons, the
effects of which can produce both transient and permanent damaging
effects in a variety of materials.
4.5-15
X rays, gamma rays, and neutrons are the types of nuclear radiation of
most concern. As opposed to charged nuclear particle, which also
emanate from nuclear reactions, those forms of radiation listed have
long ranges in the atmosphere; thus, they can irradiate and damage a
variety of materials.
Among the nuclear effects that have been of most concern to operating
equipment are those called "Transient Radiation Effects on
Electronics". They are due to the transient radiation pulse from a
tactical nuclear burst and are often rezerLid to as TREE. These
transient effects are due primarily to the ioni ation and total dose
effects of gamma rays and X rays, as well as inouced permanent lattice
defects due to neutrcn fluence.
The effects of ganmna and X rays on unpowered devices is negligible.

Neutrons, however, damage the lattice structure of active devices
under any bias condition and will result in degraded performance or
failure of the equipment. In addition to these effects, the physical
properties of polymers will be affected by ionizing radiation such as
gcmma rays, X rays, and neutrons. This may lead to a loss of
structural integrity and electrical properties. For a more detailed
discussion of the effects of nuclear radiation on equipment, the
reader should refer to references 16, 43, 44, and 45.
It is impossible to completely protect materials items from nuclear

radiation as can be accomplished for some other environmental
factors. The variety of effects produced by nuclear radiation for
different materials and components makes protective design difficult.
The procedure employed is to define a radiation hardness level in a
given material item and to design and test the item to that level.
(Ref. 1)
4.5.5 Container Desin
The level of mechanical protection required of a container is dictated

by the characteristics of the object to be protected and the
environment in which the container must operate. The previous
sections have described the need for protection of various materials
in various environments. This section will describe container design
guidelines.
Of prime importance to container design is the characteristic of

fragility. This relates to the ability of an object to withstind the
effects of externally applied mechanical forces. The term fragility
pertains to the sensitivity of the object to damage and alludes to its
degree of inherent elasticity. The basis for the measurement of
fragility is the Gm; a dimensionless ratio of the maximum
4.5-16
acceleration that an object can safely withstand to the acceleration

due to gravity: (Ref. 46)
Gm = a/g, g-units
where a = maximum allowable acceleration an object can

2
safely withstand, ft/s
2
g = acceleration due to gravity, 32 ft/s
Essentially, Gm is a measure of the items elasticity, i.e., its

inherent capacity to retain or recover its shape upon or after the
application of a distorting force. If the force is in excess of the
Gm of the item, the elastic limit of the item may be exceeded and
result in permanent distortion. AJn item with a low Gm is considered
fragile; one with a high Gm is considered rugged. The purpose of
the container is to attenuate--to a level equal to or less than the
critical Gm factor of the item to be protected--the forces
transmitted to its contents. The magnitude of these forces and their
duration constitute the hazard, and it is these forces with which the
designer is most concerned.
The most critical of externally applied forces imposed upon the

container is that of shock. Shock occurs when the container is
subjected to a suddenly applied force, such as when the container is
dropped upon a rigid surface. Shock damage is caused by acceleration
forces developed during impact.
The underlying principle is that every object in a free-fall possesses

kinetic energy which it must dissipate by decelerating through a given
distance and time interval. The rate of decceleration is determined
by the maximum number of g's Gm that the item can withstand. The
rate of deceleration of the impacting object becomes a function of the
resistance of the object to crushing or the density of the protective
cushion or elasticity of the suspension system. The density or
elasticity of the suspension material and ability to preclude the
deflection required, determine the amount of shock experienced by the
object subjected to impact. If the suspension does not provide for
sufficient deflection and gradual increase in stiffness, bottoming may
occur with consequent transmission of excessive forces. Reference 25
demonstrates in detail the methods for calculating Gm, and the
identified shock parameters.
A condition peculiar to material in transit is that of vibration

generated by the carrier vehicle. The carrier, functioning as a
fluctuating force, imparts to the container a forced vibration; the
container (assumed to be restrained) is forced to vibrate at the same
frequency as the carrier. Should these vibrations coincide with the
4.5-17
natural frequency of the suspended contents, the associated forces may

become great enough to cause fracture or damage to the contained
item. The following is a general summary of the most common forcing
frequencies that will be encountered by different carriers: (Ref. 46)
(a) Railroad : 2 to 7 Hz
(b) Truck: 5 to 500 Hz
(c) Aircraft: 20 to 60 Hz
(d) Ships: 11 to 100 Hz
The mode of transportation and the applicable frequencies tabulated

provide the required data relating to imposed vibrations.
The natural frequency of a container system is that frequency at which

the system will vibrate if displaced and allowed to vibrate freely.
Any system comprised of a body suspended on cushioning has a natural
frequency at which it will vibrate with greater amplitude than at a
frequency just above or below it. To determine a rough approximation
of the natural frequency of a container system apply the following
formula: (Ref. 46)
fn=2Gm/(h)1 / 2 , Hz
where Gm= G-factor of the container as designed, g-units
h = drop height, inches
fn= approximate natural frequency, Hz.
The amplitude of vibration depends upon the magnitude of the forced

vibration and the ratio of its frequency to the natural frequency of
the object or system being vibrated. When this frequency ratio
becomes equal. to unity, the amplitude of vibration may build up to a
dangerous value. Theoretically, for the effective isolation of
vibratory forces, the ratio between the disturbing frequency and the
natural frequency of the system must always be greater than the square
root of two; and as the ratio assumes values greater than the square
root of two, the system becomes progressively more efficient.
Conversely, as the ratio becomes less than the square root of two,
magnification of the vibratory forces will occur, and this condition
must be avoided. As the ratio approaches unity, the magnitudes of the
vibratory forces are increased until at unity, where resonance occurs,
they become infinite.
The container and its contents, in combination, must provide for all
environments and modes of transportation in addition to the hazards
peculiar to the handling of material in transit. Section 4.6 presents
guidelines for the transportation and handling of dormant electronic
systems.
4.5-18
Reference 46 provides additional details concerning container design

guidelines and container qualification tests.
4.5.6 In-Storage Cyclic Inspection
Figure 4.5.6-1 from reference 1 presents a matrix that can be used to

determine inspection frequency and optimize in-storage inspection
coverage. The matrix includes:
(1) The most deteriorative items to the least deteriorative in

terms of a total ranking factor that accounts for
deterioration, complexity, cost, accessibility, and
criticality.
(2) All combinations of storage and packaging conditions ranging

from the most protective (containerized package and a
controlled humidity environment) to the least protective
(commercial package and an open area).
Application of the matrix to a given materiel item involves assigning

appropriate values to each o2 the weight factors depicted in figure
4.5.6-1 in order to arrive at a total ranking. This ranking
represents a rough measure of the overall deterioration/cost
sensitivity of the item to the storage environment. The ranking is
then inputted at the proper weight column of the matrix to determine
inspection frequency for any desired combination of packaging and
storage protection level.
For new items, the matrix allows broad tradeoffs to be made to arrive
at the optimum balance of packaging, storage, and inspection
requirements. Also, the combining of deterioration with cost and the
other weight factors via the matrix approach allows the specification
of cost-effective inspection periods. This cost-effectiveness is
illustrated by considering two items where one exhibits low
deterioration properties but cost and the other factors are high and
the other exhibits high deterioration properties but the total of the
other factors is low. A relatively low cost or nominal test
inspection frequency may be computed for both items that reflects an
effective balance of all factors, whereas, if only deterioration was
considered in computing the test periods, over-inspection (excessive
cost) of the high deterioration item and under-inspection (low
readiness assurance) of the low deterioration items would most likely
result. For those items where all factors including deterioration and
cost are high, frequent inspection would be required to assure
materiel readiness, and for those items where deterioration and the
other factors are low less frequent inspections would be required.
4.5-19
01
-j i La -
'.id < -J= i -
A.. < Wl K a.
L"
0. -2 -L
IA L" L"0'
-L. L M -j L Z
>~~u Li * 14U-I
o I- Z I- '- U) U
-~~V 0-OL~
L" W i9-< -
-o U- -
H Li ~ U vi LL- ~ .I "j I
Li1 ->-
CD wj w"
0 "ii Con L I--
a:, -'I oi:- -
CI 0 - / w-: 0.< = I -CCI

ui:zm0 ,I Il V *. -o Ir-
ID (D*ILoillX C - WW )V L
__ Z:V))r 'IJ __ _j
I ! K ,.,,,.ILILiJI
I =L
C31~ I
LiIj0 0
0 0 000
*1j c <DICD -
'a. 'o
II:ii
z~ .zI I
C-C,~~~
010101
iI~j ~4.S-20
The matrix approach also provides flexibility for regulating the

number and type of items subjected to cyclic inspections by adjustment
of the weight assigned to the factors that relate the materiel to the
storage environment.
4.5-21/(4.5-22 blank)
4.6 TRANSPORTATION AND HANDLING
Transportation and Handling guidelines for dormant systems are

described in the following sections. Transportation is covered in
section 4.6.1 and Handling is covered in section 4.6.2. Since
virtually all equipment is non-operating during transportation and
handling operations, the same guidelines that apply to the
transportation and handling of normally operational systems are
applicable to dormant systems as well. Where appropriate, these
general guidelines are augmented with additional information relevant
to dormant systems.
4.6.1 Transportation
This section describes various transport types, their capacity, and

their limitations. It describes the transportation environment as it
affects the transportation of dormant systems, and provides guidelines
for protection of dormant systems during transportation. Table
4.6.1-1 summarizes the most significant environmental factors
experienced during transporation for the various transport types.
4.6.1.1 Transportation Equipment. Capacity. and Limitations
There are four major types of transportation: truck transport, rail

transport, air transport, and sea transport. Each transport type
imposes specific limitations to the shipment. These are primarily
restrictive to size and weight of the shipment. In addition, legal
limitations are imposed on shipments by the regulatory agencies of the
individual states, Federal government agencies, and by foreign
governments in international shipments.
4.6.1.1.1 Truck Transport
Regulations restricting the size and weight of loads that can be

legally transported on the highways in the United States are
determined by the individual states. Although these limitations can
be exceeded if permisbion is obtained, it causes a considerable
increase in transit time and transportation cost, and requires special
permits, special routing, etc. The principal limitations imposed when
shipping by highway arise from three sources:
1. Limits imposed on size and weight of loads by various state

governments.
2. Limits imposed on size and wight of loads by foreign governments.
3. Limits imposed by interior dimensions of the vehicles.
4.6-1
0'04 >1 '

0 14 0 53
.14 .4 -4 -4
Li0 01.40 3 '
1w -.4 ON
0o 43 0' $
43I - 1-. -.
'0 00 0 .
03 vi :4 t 0 to0
A.) 414A 0 w
1. 434A3
"., .. 4 tv "., .,q
W044 li 'low V0 0
0 1..
$d 0.
IV.. .. 44
wi 0 FU0 0 U A.)
.4 '- @
000 0 0) t00
> a ~ i- 0 w0Li"4 $4
'0
00 tad )t 0
4J to 04 to .,01
a 00' 0 0) 0I
0. '0 04 0j
t
Ud
to 314 $4 0.-4 W.
61 w'
"4400 4:3 430v 0 0 )
'0 "4L 0 :3 "4 L " i 0 0 a0
04 . 0 0 000 to "
) to 4)
10 c 04 . .
x 2 %0 N. 4~~
L 4.) d
0 (1)L
0 0 0 Li0 44
wii .. 4 0 ."4 Li L
H-- 0 43 434 43 0 00u
00
to U3 to 43) 44 )
0.-o 0i Li0 a
0. 4.)
94-4 .0 0 0
04.4 .. 4 0 " 44 toV
4-1 >044 > .0 C
A.) M0- i 44 -4 4j .. 4 -
to 'V 4.4 10 0o 4.)
w4 0 4 93 ul.
V).0 43 to )' 43
93 U 03 0 .'
.0)0 10 V 0 U)4 3
00 u 0 0 O4i 4 10 >0 (
0 4) 4) 0vI 0 10
*M4.4 Li rl Ini r.00
JJiw 0: U 0 ".40 0 V :5 t4o
0o4) yl "4 > V3 > m4 v -0
,1 0 4.) 0 iv>qi & 0 43 0U
43W 0 0 '-4 43 24.4-4 0 t .9 4304) d
".0 0 '-40 to "i4 03 w
*O 4 V- 44 " 0 0o 0
I~ ~ 430 0 4D
0 40 L
-4 0
10 "4.'"4
0 0 >- 043
4.) 4 r. 'b 0. 4) 0 $
0 (a .0 4 .140 0't .4 0A. 0t
- 0 .10 pq~ 0 4a~ in
0 0 00004 0 m- 4)~
0~ 04
4.)
-4z -
0z
04.6-2
>1A.) 0
0 10 V0
"4s "4j"
to > -- o 4
w 41 4) w LwQ E
o6 to 0U $4 V0 1 1
0 .) 10 4 . m 4-I
1.4 mO uE 0 0
A.) 0U 1i 4.010 E Cw4 0 CIA4
r. 0i 0
u w 4.4 0 u 0 -
0' 4) Ui 00 L ~L
LA 0 L1SS 500
4.)' %64 0.4 0 :3 9 0
0 u 0 'j.J 04. -0~~~ 0.
. 0-4 to0 ''-4 -m IV E
-4-~
wO*.
Li .0z 4- 4 *-4 E S4-1
:5L 4) 4J'4
w. 0 "4L- 0
".4 5j
41 *.4 > 2 0 45.4
0 uUE to ~ it 0 U
C E
oA4V 10
-(12
Z~ot0
-44w -- I
V) u m 4.
11-4 -4 Q004o- 0) 00
OLE4. 4.) .. 4 c,.l

ud4 0 0% : V Li
0 Li
Eu"
0,E u-) 0 . 0 00
C p
>i 000
to2. 0U 0. -40
00EU0O 1 U
w WUEUE
.00 Vo V. ~ 2
ca
00 V
4.6-3~ 04-
0 d
.404 0
J.40
4 0 4 0 t
U-
. 5.)
,4 t5.E
104 0 44 ~ to 41 04
4) wU S.M 4) w
A' M4.) 000 4 : 0 4 C0)
cn
41
) -q
4
4
.. )4
~ l 0 .) ,q
M~ e ul 41 4
ul 4) F O *.4 fu E o
O4 u.14 9:4 O.) L, V0 b
4) tf ). *4)
1) .O )Zd )
4 ) 1.)
') 93. 0). aO 0 3 E
U3)0 0a0Ad -M 14) 0Su
0) 41 *>4a 0 414 E20.
oa aj.1q0 41 0) 0
0) 6 0. 4)a 64V ll, -r- 04 4) 44
0 .,1C 044)0 $4 0 0 0) 1 A 14>0m
0 1 4 U*.w00 0j 4) 0).40
0) 0, '0 (a ~
410 -
sq ESVI4 ul0) )t 5-4) &.

0J
0*- 00j- s0J~ D( 0 m .) -1jJ>0
'0 4d
0' V
0r
00
04
U)
0 9
0j 06 0
Q1.4 u -4.1.
014) 04 0.
13 -44
0)14 0 4
w4 01 41
d)$4 0 0
04 0 0 4.,1-4
a0 .4
w2 w a4 *4 U
00 0 0 0
-4.-I4.6-4
Because of the many different regulations, there is no one unitized

group of restrictions that can be followed in each case. For the
United States alone, there are 51 different jurisdictions, many of
which change their regulations frequently. In addition, shipments
destined for overseas must meet the regulations of the countries
through which they pass. (Ref. 54). For the most current
information, the transportation department of the appropriate
jurisdiction(s) should be contacted.
In addition to the restrictions on weight, the limits placed on the

vehicle dimensions must also be considered. The shipping container
should be designed to permit versatility of transport through any
state without having to obtain special permits, routing, etc. The
transportation department of the appropriate jurisdiction(s) should
also be contacted for the maximum width and height requirements for
over-the-road equipment.
In a similar manner, consideration must be given to the interior

dimensions of the military vehicles that may be used to transport the
material. Dimensions of standard military Oehicles are listed in
TM9-236, Military Tactical Vehicles.
4.6.1.1.2 Rail Transport
The principal limitations imposed on shipments by rail result from

three causes:
1. Limits imposed by dimension of railroad cars.
2. Limits imposed by dimensions of the rignt-of-way.
3. Limits imposed by the weight and distribution of the load.
The dimensions of railroad cars in use throughout the world vary

widely. The Official Railway Register (Ref. 55) lists all railzoad
cars used. When the shipment must also travel in foreign countries:
the more restrictive limitations of these countries musz be taken into
consideration.
The most frequently used railroad car on the railroads of Forth

America is the closed car, in particular the Class X box car.
Material too large to fit into the closed car is carried on open cars,
either a gondola car, a flat car, or a depressed-center flat car.
To promote maximum efficiency of railroad transportaticn, loaded

railroad cars should be within the clearance limitations of the
right-of-way over which the shipment wil) travtl. The types of
4.6-5
physical barriers which determine right-of-way clearance limits are

bridges, tunnels, telephone poles, etc., which are subject to seasonal
variations. If a load exceeds the limitations, a special routing must
be established for the shipment. This adds enormously to the
transportation cost and time. A complete listing of clearances for
most world railways is given in World Pailways, Sampsop lows,
Publisher, London, England.
The total load that may be carried by a railroad car is restricted by

the weight limit of the car, and by the load limits of the rail bed,
bridges, etc., over which the shipment will trave]. When the I7ad
approaches the specified limits, the local freight agent shouald be
consulted to ensure that the shipment can be efficiently transported
by the carrier.
4.6.1.1.3 Air Transoort
Shipments by air transport are subject to the limitations imposed oy

the characteristics of the aircraft used for the shipment. The
principal limitations are as follows:
1. Limits imposed by maximum aircraft payload.
2. Limits imposed by size, location, aud configuration of door

openings, int-luding limitations caused by the use of ramps for
loading.
3. Limits imposed by the size and configuration cf cargo

compartments including limiting factors that may prevent full
utilization of available space.
4. Limits imposed by quantity, location, and strength of tiedown

fittings.
6. Limits imposed by the aircraft center of gravity limits forward

and aft.
7. Limits imposed by altitude and temperature changes, especially

in unpressurized cargo compartments.
8. Limits imposed by restricted cargo.
Maximirm ,.mensions for trvnsportabiil.L- iA US Air Force aircraft are

defined in reference 54.
1.6-6
4.6. 1.1.4 SeTranspo rt
Sea Transport is capable of carrying any shipment that can be

transported to dockside and loaded on board ship. Generally, on-deck
stowage is limited only by the capacity of the deck, and maximum
weight limitations. Below-deck stowage adds a,.ditional limitations.
While special handling equipment can be made to load any given weight
a ship can carry, the weight of individual containers should be
limi ted to the boom capacity of the vessel to permit loading and
unloadinig without the iuse of shore equipment. Since all cargo ships
are eqcipped with at least a 5-long-ton (2240 pounds) boom, an
ll,20C-pound load can be loaded and unloaded anywhere by any type of
vesse'. If the container ° includes an explosive material, such as a
copiete ,missile with a high-explosive warhead and a solid or liquid
prpe lant rotor, a Coast Guard weight limitation cf 7467 pounds must
be observed if the material is to be capable of being loaded and
unloaded axrywhere by any type of vessel.
The limit.ations imposed by each ship varies widely; however, for easy
stowage in most vessels. the container dimensions should not exceed 35
feet in length, 20 feet in width, and 11 feet, 4 inches in height. A
n.igt~t of 6 feet 1--1/2 inches has been determined as bleing optimum.
4.6...5 Combina___tion Transport
When weapon systems are shipped for long distances, and to overseas
dest.inations, a combination of different types of transpoct is
required, For shipments using several types of transport, three
nethods of combination transport were developed.
These method are:
1. Roll-0n/Roll-Off systems for truck-sea combination.
2. Trailer-On-Flat-Car system for truck-rail combination.
3. Trailer-Containers-on-Fiat-rar system for truck-rail-sea

-omb i.-nat ion.
Comb at on tra:nsport provides the economy of a cost-effective use of

ti t Lty-;es,
df .,oYo end elimirnates the cost and the inherent risks in
E 3x 1Vt) IT atarIa] handlino,. This is especially important for dormant
systerrs since falittres which occur as a result of transportation and
handiinc; may not detecte d for long periods of time. However, the
most restrict.ive lirnitatiors imposed by any of the transport types
used in cobinarinn tran .i-jct will apply, and must be coitsidered as a
ta(:.or in the .-syezn und coitaincr dtign.
4.6-7
4.6.1.2 Transportation Environment
The transportation environment consists of the conditions and forces

to which material are subjected during transportation. This includes
both normal operations and accident conditions, and consist,
essentially, of the following elements:
1. Shock and Vibration.
2. Terminal handling for truck, rail, air,and sea transportation.
3. Varying atmospheric pressures normal to each mode of

transportation.
4. Accidents.
5. Temperature and humidity variations normal to each mode of

transportation because of equipment or operating environment.
6. The worlu-wide weather environment in terms of temperature,

atmospheric pressure, and humidity.
7. Critical physical and geometric limitations.
6. Forces causing puncture and abrasion.
The elements of shock and vibration experienced dur'j ransportation

are described 3n the following paragraphs. Other eni.-,nmental
factors are discussed in section 3 of this report.
T!-' most damaqing element encountered by a shipment is shock and

.ation. The shock and vibration to be expected depends on the mode
or transportation, and is further dependent upon the manner in which
the materizl is packaged and stowed. The techniques used for
protecting packaged material against the harmful effects of shock and
vibration include blocking and bracing, use of cushioning material,
an6 use of shock and vibration isolation systems.
Although shock and vibration are often treated as separate phenomena,

the distinction between the two is not clear cut. The difference
between transient shock motion and periodic vibration is fairly
obvious; but the existence of any basic differences between shock and
random vibration, which is not periodic, is much less obvious.
However, shock may be con,idered as intermittent excitation and
vibration as sustained excitation.
4.6-8
Vibrations and shocks will impose forces on, and deform, any flexible
or elastic structure. The severity of the deformation depends upon
the nature of the imposed force, and the geometrical configuration,
total mass, internal mass distribution, stiffness distribution, and
damping of the item or equipment.
Vibration is an oscillation wherein the quantity is a parameter that

defines the motion of a mechanical system, Vibration has also been
described as the variation, usually with time ,of the magnitude of a
quantity with respect to a specified reference, when the magnitude is
alternately greater and smaller than the reference. Vibration may be
periodic in nature, or it may be non-periodic.
The simplest form of periodic vibration is simple harmonic motion,

which is motion that varies sinusoidally. Simple harmonic motion can
be identified by any two of the four parameters: frequency, amplitude
of excursion, velocity, and acceleration. Any periodic motion can be
considered as consisting of motions at one or more frequencies, with
the motion at each frequency being harmonic. A periodic, or
steady-state, vibration can be completely defined by designation of
the frequency or frequencies, the maximum value of the harmonic
variable at each frequency, and the phase relationships that exist
between the component harmonic motions. The harmonic variable may be
expressed in terms of displacement, velocity, or acceleration.
There are two types of non-periodic vibration: random and

white-noise. Random vibration differs from periodic vibration in that
the amplitudes at the component frequencies vary randomly with respect
to time, and therefore cannot be predicted. White-noise vibration has
no defined component frequencies, and both its frequencies and
amplitudes may vary randomly with time.
The response of a structure to shock and vibration is determined

largely by the excitation frequency and the resonance characteristics
of the structure. Resonance affects the magnitude of of the applied
load and its transmission characteristics. Any shock or vibration at
the resoaant frequency is amplified in force, resulting in an
increased chance for damage.
2 e ratio of the output vibration amplitude to the applied vibration

is the transmissibility. Transmissibility can be considered a
magnification factor, and is greatest at resonance. It decreases down
to unity below resonance, and can become less than unity above
resonance.
Saock connotes impact, collisicn, or blow, usualiy by physical

contact. It denotes a rapid change of load, or a rapid change of
4.6-9
acceleration with a resultant change of load. A shock motion cannot

be defined by assigning numerical values to established parameters; it
can only be defined by describing the history of a significant
parameter such as acceleration, velocity, or displacement. The time
duration of a shock pulse is important, since il helps in determining
the way which an object will react to that pulse.
Actual shock and vibration likely to be encountered varies for the

four major types of transport. Shock and vibrations in truck
transport are generated through normal highway operation, starting and
stopping movements, acceleration and deceleration, coupling and
uncoupling, docking, and additionally such minor accidents as
sideswipes, rear-end collisions, etc. Vibration frequencies in motor
trucks are dependent upon the natural frequency of the unsprung mass
on the tires, the natural frequency of the spring system, and the
natural frequencies of the body structure. The vibration amplitudes
are dependent upon road conditions and the speod of travel.
Intermittent road shocks of high magnitude can occur, with resultaac
extreme truck-body displacements. These large displacements may
result in a severe shock environment for unlashed cargo as it bounces
about the truck floor. Vibrations caused by the truck engine are
relatively insignificant in the cargo area. The predominant natural
frequencies of various commercial vehicles can be obtained in
reference 56.
Shock and vibration effects in rail transport are generated through

normal operational conditions including end shock caused by switching
and humping operations, longitudinal shock caused by slack conditions
in over the road operations, and vertical and lateral vibrations also
caused by over-the-road operations. Vibrations in moving freight cars
arise from track and wheel irregularities, and occur principally in
the lateral and vertical directions. The exciting frequencies caused
by rail joints and wheel imbalance vary from 0 to 13 Hertz per
second. Railroad car structural-frame frequency is usually in the
range of 55 to 60 Hertz. Shock and transient vibrations during
coupling and during starting and stopping are generally considered to
be the most damaging phases of rail shipment. The mean speed of
impact is about 7 miles per hour, which is well above the approximate
5 miles per hour limit for which the switching gear provides
cushioning protection.
Shock and vibration effects in air transport are generated through

normal operational conditions in flight, and during take-off and
landing. In-flight vibrations are of two types: mechanical and
acoustical. Acceleration stresses from sharp-edged qusts and boosts
tend to supplement vibration stresses. The largest single source of
vibration appears to be the propul:;ion system. Self-excited vibration
4.6-10
is caused primarily by flutter, the vibration of lifting, and

transitory loads to the aircraft. This is most evident in taxiing,
take-off and landing, and in abrupt maneuvers. Forced mechanical
vibrations result primarily from the mechanical forces. The in-flight
shock and vibration environment is generally not too severe. The
loadings which are important are the dynamic loadings that occur
during flight in rough air. These are differentiated from shock
loadings in that they consist of fairly high magnitude accelerations
imposed for a prolonged period of time. These accelerations can be as
high as 2 to 3 g's during normal operation oi large transport aircraft.
In ship transport, the principal excitation forces for shipboard

vibrations result frcm the ship structure interfering with the flow of
water, from the propellers, and from imbalance or misalignment of the
propeller shaft system. The frequency range of the vibrations is
about 5 to 25 Hertz, with attendant acceleration reaching a maximum of
amount 1g. During normal service, ship cargo do not experience shock
loads of any significant magnitude.
Detailed specifications for and information on shock and vibration are

given in MIL-STD-810, MIL-P.-116, MIL-STD-648, FED-STD-101, MIL-S-901,
and MIL-STD-1670.
Detailed specificatior. for and information on package cushion design

to mitigate the effects of shock and vibration in equipment design can
be found in MIL-HDBK-304.
4.6.1.3 Protection During Transportation
Systems can be protected during transportation by the following

methods:
1. Design of the system.
2. Design of the transport vehicle.
3. Packaging of the system.
4. Shoring, bracing, and tie-down of the system.
5. A combination of these methods.
Design guidelines to protect a system from various dormant

environments are found in sections 3, 4.1, and 4.5 of this report. The
following engineering design parameters for transportability of
military material were developed by the Department of the Army, and
are provided as design guidelines. (Ref. 48)
4.6-11
1. Truck Transport - For truck transport, the cargo and its

restraining system should be capable of sustaining the loadings
incident to a 1,000 mile road trip over a paved highway. For
all shocks and vibrations, the stresses in the restraining
system should not exceed the yield strength of the material, nor
should they exceed one-h lf the yvild strength of the material
under static load conditions.
Envelopes of maximum values recorded during Transportation Corps

field studies for both shock and vibration are provided in
MIL-HDBK-409. It is recommended that the vibration time for
design purposes be consistent with a 5,000 mile trip and that
design safety factors, if any, be applied by increasing the time
of vibration. For design purposes, it is recommended that the
restraining items be designed to withstand 30 shock applications.
2. Fail Transport - For rail transport, the cargo ane its

restraining systems should be capable of withstanding a
simulated transportation shock environment as described in
MIL-STD-810.
The stresses in the restraining members should be less than

one-half the yield strength of the material in the static, or
restrained condition. The combined static and dynamic stresses
must not exceed the static yield strength of the material in any
restraining system component during the dynamic portion of the
impact loading. Additional margins of safety may be required
during design of the restraining systems because of the cargo's
peculiar nature, train safety considerations, or accident
effects considerations.
The carqo and its restraining system should be able to withstand

without failure or impending failure, a transportation vibration
environment equivalent to one produced by over-the-road movement
in a 150-car train. The car transportiLg the cargo should have
standard freight car suspension and draft gear, and should be
considered for the end of the train car position. Vibrations,
both intermittent and continuous, should be of a duration
equivalent to the input from 3,000 miles of Class I railroads
containing at least 50 percent long maximum grades.
3. Air Transport - Induced shock and vibration environments for air

transport are normally considered the least severe as regards
loading of* the cargo and its restraining system. Factors of
aircraft safety, cost of cargo, and military value of the cargo
dictate the highest degree of reliability for the strength of
the cargo and its restraining systems. Many strength and safety
4.6-i2
factors are employed both in design and operation for

restraining systems involving air transport, with consequent
multiplication factors applied to the basic environmental data.
The basic data should be especially accurate in order to
minimize cumulative error on the inaccurate portion of the data
that is proportioned or multiplied for safety or design reasons.
For air transport, the cargo and its restraining system should
be capable of withstanding all the aircraft vibrations
occasioned from a time period consistent with the maximum range
of the aircraft. It is considered important that the amplitude
and frequency of the vibration be accurately duplicated and that
safety factors be applied in terms of the length of the
vibration. it is recommended that the restraining system be
designed to sustain the vibration for a period three times as
long as would be anticipated based on the mission of the
aircraft.
T1'e shock acceleration normally occasioned by landing should be

based on a veiocity at touchdown for the aircraft of 10 feet per
second. Again, any safety factors should be applied by
increasing the number of shocks rather than the severity. It is
recommended that the restraining system be capable of
withstanding 20 landing shocks with no signs of failure or
impending failure to any of the components.
4. Sea Transport - For sea transport, cargo and its restraining

system should be capable of sustaining an environment occasioned
by a seaway induced loading of a transport ship consequent to 20
days of Beaufort Sea State Condition 12. During this condition,
the components of the restraining system should not exhibit a
combined static and dynamic stress in excess of 80 percent of
the static yield strength of the materials. The static stresses
occasioned by normal tiedown procedures should not exceed 50
percent of the static yield strength of the materials.
Particular emphasis must be placed on the effects of stacking

cargo for shipboard transport. Stacking may subject the cargo,
the cargo container, or the restraining system to severe loading
conditions. As the dynamic and static loads are resisted by
each succeeding lower unit of cargo in the stack, the cumulative
effect on the bottom units must be considered in design. The
same consideration as regards stacking and dynamic loading must
be given in the horizontal plane since longitude accelerations
will also cause a load buildup on the end unit unless load
dividing measures are taken. Effects of cargo stacking are most
troublesome in sea transport because of relatively large cargo
holds which accommodate excessive stacking.
4.6-13
Whi'.e most operationally ready pieces of equipment require some sort

of preparation for shipment, most dormant equipment are kept in
shipping and storage containers and require relatively little
preparation for shipment. Depending on the item, however, certain
subassemblies may require folding, removal, or reorientation. Exposed
machined metal surfaces will require a protective coating to prevent
corrosion. Dissimilar metals will set up a galvanic action, and must
be insulated from each other in the original design or as part of the
preparation for shipment. Fuel systems may have to be drained,
purged, and reserved. These features, and other similar ones should
be considered when the item is designed to reduce the time required to
accomplish these tasks, ease their performance, and reduce the
possibility of damage. (Refs. 54 and 57)
The design of some items of material/equipment limits the points at

which attachments may be placed for lifting and tieing down. Certain
tiedown practices may cause extensive damage to the item. Therefore,
consideration should be given to the design, location, and marking of
attac'ments for lifting, ground handling, and restraining for surface
and air transportation. MIL-STD-209 provides guidance for the design
and location of lifting and tiedown provisions. All devices used as
hoist or lift equipment for hazardous materials should have an
allowable design stress below one-fifth the yield stress of the
material. (Ref. 54)
Restraint criteria should be met for all applicable modes of

shipment. MIL-STD-209, MIL-STD-814, MIL-A-8421, and Air Force System
Command (AFSC) Design Handbook (DH) 1-11, Air Transportability,
provide appropriate guidance for restraint of cargo. However, the
designer should consider also the tiedowns and any other appurtenances
or possible attachment points, from the viewpoint of the person
charged with moving the item, to determine how an inexperienced person
would restrain it. Parts of the item that could be damaged if they
are used as restraining points should be marked to preclude improper
use, and suitable parts should be marked for use as restraining
points. Typical lifting and tiedown provisions for various military
vehicles are preseuted in MIL-HDBK-151.
4.6.2 Handling
This section presents guidelines for providing protection for systems

during handling. It presents various types of material handling
equipment and systems and their limitations. Handling methods and
hiuan factors considerations are also presented. Handling of systems
occurs throughout all phases of transportation and storage. Zhe
designer should consider the limitations of existing handling
equipment concurrently with the development of the material and its
4.6-14
containers. If required, special handling equipment should also be

developed concurrently, although the use of specialized equipment
should be kept to a minimum.
4.6.2.1 Handling Equipment, Capacity. and Limitations
The design and specification of handling methods is an essential part

of planning for the protection of all systems during handling. This
is especially true for dormant systems since many of the failures
reported for these systems are induced by operator or maintenance
handling and such failures may go undetected for long periods of
time. Generally, specific instructions are required for each system.
The extent of detail required depends primarily on the size and
complexity of the system, and the quantity of special equipment used
in material handling.
The various types of material handling equipment that may be used at

various cargo handling and storage installations include the following:
1. Powered self-propelled equipment consisting of forklift trucks,

wheeled warehouse tractors, warehouse cranes, hand-lift pallet
trucks, fixed platform trucks, straddle-carry trucks, overhead
and gantry cranes, and hoists.
2. Nonpowered mobile equipment consisting of platform warehouse

trailers, hand trucks, and dolly trucks.
3. Conveyors, including portable belt-types, roller gravity and

wheel gravity types.
4. Pallets made of wood or metal.
5. Hoisting slings and spreaders, hoisting beams, and hoisting bars.
The design engineer should ensure that the packaged material can be
handled by standard handling equipment. If necessary, several
equipment units may be used to lift extremely long or heavy
containers. The container should be clearly marked showing the
handling locations, such as lifting eyes, rings, handles, brackets,
and center of balance. Lifting eyes should be permanently attached
and large enough to accept rigging cable hooks. They should also be
far enough above the center of gravity of the container to stabilize
it. Handles should be so positioned that they will not catch on other
units, cables, lines, structural members, etc.
If the container is too awkward or heavy for manual lifting, some

method of mechanical handling must be provided. To permit handling by
4.6-15
forklift trucks, the bottom of the container should be raised three

inches above the ground. This is usually accomplished through the use
of skids spaced to allow for the entry of forklift tines. Reinforced
channels should be constructed at the bottom of the container to
prevent the fork times from damaging the container wall. Two fittings
should be located at the end of the container, adjacent to the skids,
for use in moving the container. Various standard types of powered
and no-powered handling equipment and conveyors for use by all
military services are illustrated and described in TM-743-200 (Ref.
58).
MIL-STD-137 (Ref. 59) specifies the standard handling equipment

normally required for specific loading and unloading operations.
Various types of pallets are used in the handling of material,

including general purpose and special design pallets. Generally,
where possible, material should be palletized in a unit load for
shipment and storage. MIL-STD-147 (Ref. 60) provides the practices
and procedures to be followed in palletizing or containerizing unit
loads, which are mandatory for use by all department and agencies of
the Department of Defense.
4.6.2.2 Human Factors Considerations
Human factors considerations should be recognized by the design

engineer as being a necessary part of the planning of the handling
process. Human engineering principles axnd criteria should be applied
together with other design requirements to identify and select the
equipment to be operated, maintained, and controlled by man.
Essentially, this effort should ensure that the human functions and
tasks are organized and sequenced for efficiency, safety, and
reliability. Appropriate training and technical publications should
be provided, as required.
Specifications for and information on human engineering requirements

can be found in MIL-H-46855 (Ref. 61), MIL-STD-1472 (Ref. 62), and
MIL-HDBK-799 (Ref. 63).
4.6-16
4.7 REFERENCES
1. MIL-HDBK-338, Electronic Reliability Design Handbook, 15 October

1984.
2. Reliability Design Handbook, R. T. Anderson, ITT Research Insti-

tute, Report No. RDH-376, Reliability Analysis Center, RADC, Grif-
fiss AFB, March 1976.
3. "Reliability Factors for Electronic Components in a Storage En-

vironment", B. R. Livesay and E. J. Scheibner, Georgia Institute
of Technology, U.S. Army Missile Research and Development Command
Report No. DD14-23, September 1977.
4. "The Reliability of Electronic Devices in Storage Environments",

B. R. Livesay, Georgia Institute of Technology, Solid State Tech-
nology, October 1978, pp. 63-68.
5. "R/M Design for Long Term Dormant Storage", George Kasouf, General
Electric Company, and David A. McGoye, Joint Cruise Missile Pro-
gram Office (DoD), 1984 Proceedings Annual Reliability and Main-
tainability Symposium, pp. 168-176.
6. "Long-Term Storage and System ReliabiliLy", Stanely M. Cherkasky,

Singer-Kearfott, Proceedings 1970 Annual Symposium on Reliability,
February, 1970, pp. 120-127.
7. "An Approach for Assessing Missile System Dormant Reliability", R.

D. Trapp, et al, BDM Corporation, NTIS Report No. AD-A107-519, Air
Force Contract F29601-79-C-0051, 30 January 1981.
8. "Accelhrated Versus Real Time Aging Tests", E. W. Kimball, Martin

Marietta Corporation, P.O. Box 5837, Orlando, Florida, Undated , 4
pages.
9. "Dorrant Storage Reliability Assessments - Data Based", G. T.

Merren, Sandia National Laboratories, IEEE Transactions on Com-
ponents, Hybrids, and Manufacturing Technology, Vol. CHY.T-4, No.
4, December 1981, pp. 446-454.
10. "Tomahawk Program Failure Summary and Analysis Report", J. K.

Stich, et al, General Dynamics Convair Division Report No.
GD/C-SLCML-001 (8) for Joint Cruise Missile Program Office, Wash-
ington, D.C., 1 Septem~ber 1985 through 31 October 1985.
4.7-1
11. "An Outgassing Data Compilation of Spacecraft Materials", William

A. Campbell, Jr., et al, Goddard Space Flight Center, NASA Refer-
ence Publication 1014. January 1978.
12. "A Study of dormant Missile Systems; Vol. 3: Dormancy Design Cri-
teria", General Dynamics - Convair Division AF Systems Command,
Technical Report BSD-TR-66-52, DTIC Report No. AD 478929, February
1966.
13. MIL-STD-1996; Resistors, Selection and Use of; 4 June 1974.
14. "RADC Reliability Notebook Volume II", Clifford M. Ryerson, et al,

Hughes Aircraft Company, RADC-TR-67-108 Vol. II, NTIS Report No.
AD-821640, December 1966.
15. "A Review of Equipment Aging Theory and Technology", S. P. Car-

fagno and R. J. Gibson, Franklin Research Center, Philadelphia,
PA, Electric Power Research Institute Report No. EPRI NP-1558,
Project 890-1, electric Power Research Institute, Palo Alto,
Calif., September 1980.
16. Effects of Radiation on Materials and Components, cdited by J. F.

Kircher and R. E. Bowman, Reinhold Publishing Corporation, New
York, 1964.
17. "Literature Search and Radiation Study on Electronic Parts", W. E.

Home, JPL Contract 952565, NASA Report CR110496, May 1980.
18. "Storage Reliability of Missile Materiel Program - Storage Reli-

ability Analysis Summary Report Volume I - Electrical and Elec-
tronic Devices", Dennis F. Malik, U.S. Army Missile Research and
Development Command, Redstone Arsenal, Alabama, Report No.
LC-78-2, January 1978.
19. "High Pulse Power Failure of Discrete Resistors", H. Domingos and

D. C. Wunsch, IEEE Transactions on Parts, Hybrids, and Packaging,
Vol. PHP-11, No. 3, September 1975, pp. 225-229.
20. "Dormant Operating and Storage Effects on Electronic Equipment and

Part Reliability", D. F. Cottrell, et al, Martin Marietta Corpora-
tion, RADC-TR-67-307, DTIC Report No. AD-821 988, July 1967.
21. MIL-HDBK-978; NASA Parts Application Handbook; 15 December 1980.
22. MIL-STD-198E; Capacitors, Selection and Use of; 29 May 1984.
4. 7-2
23. Proceedings - Reliability of Tantalum Capacitors and Microcir-

cuits; A one-day Seminar at New Jersey Institute of Technology on
Wednesday, September 11, 1985; The New Technological and Scien-
tific Activities Committee of the New York Section, IEEE and the
New Jersey Institute of Technology.
24. "Long-Life Assurance Study for Manned Spacecraft Long-Life Hard-

ware - Volume I - Summary of Long-Life Assurance Guidelines", Mar-
tin Marietta Corporation, NASA Report No. NASA-CR-128905, NTIS
Report No. N73-23853, December 1972.
25. MIL-STD-1131B, Storage Shelf Life and Reforming Procedures for

Aluminum Electrolytic Fixed Capacitors, 16 September 1975.
26. "Storage Reliability of Missile Materiel Program - Monolithic Bi-

polar SSI/MSI Digital & Linear Integrated Circuit Analysis" Dennis
F. Malik, U.S. Army Missile Research and Development Command, Red-
stone Arsenal, Alabama, Report No. LC-78-ICl, January 1978.
27. "Design Limits When Using Gold-Aluminum Bonds", Elliott Philofskv,

Motocola Inc., Proceedings 1971 Reliability Physics Symposium,
March-April, pp. 114-119.
28. "VLSI Device Reliability Models", David Coit et al, ITT Research
Institute, RADC-TR-84-182, DTIC Report No. ADA-153 268, December
1984.
29. "Susceptibility of Microwelds in Hybrid Microcircuits to Corrosion

Degradation", J. L. Jellison, Sandia Laboratories, 13th Annual
Proceedings Reliability Symposium, 1975, pp. 70-79.
30. "Metal Problems in Plastic Encapsulated Integrated Circuits", E.

D. Metz, Motorola, Inc., Proceedings of the IEEE, Vol. 57, No. 9,
September 1969, pp. 1606-1609.
31. "IC Quality Grades: Impact on System Reliability and Life Cycle
Cost", Mary Gossin Priore, ITT Research Institute, Reliability
Analysis Center, Griffiss AFB, N.Y., Report No. SOAR-3, Winter
84/85.
32. "U.S. Army Panama Field Test of Plastic Encapsulated Devices",

Edwdrd B. Hakim, U.S. Army Electronic Research and Development
Command, Microelectronic Reliability, Vol. 17, Pergamon Press
Ltd., Great Britain, 1978, pp. 387-392.
4.7-3
33. "Passivation Coatings on Silicon Devices", G. L. Schnable, en al,

RCA Laboratories, J-urndl of the Electrcrhenical Society: Solid-
State Science and /echnology, Vol. 22 No. 8, August 1975, pp.
1092-1103.
34. "Reliability Investigation of Plastic Encepsulated/Polymr Sealed

Hybrid Microcircuits", J. J. Mazenko, Hughes Aircraft Company, and
E. B. Hakim, U.S. Army Electronics Research and Development Com-
mand, Materials National Association of Corosion Engineers,
Performance, January 1980, pp. 9-15
35. "Reliability Parts Derating Guidelines", S. L. D:urmett, et al,

Boeing Aerospace Company, RADC-TR-82-171, June 1q82.
36. "I!.sulation Aging - A Historical and Critical Review", E. L.

Brancato, Naval Research Laboratory, IEFE Trans. Electr. Insui.,
Vol. EI-13, No. 4, August 1978, pp. 308-317.
37. "Life Testing of Electronic Power Transformers - Part II", Thomas

W. I-akin, et al, IEEE Transactions cn Electrical Insulatitn, Vol.
EI-3, No. 1, February 1978, pp. 13-i.
38. "St-ady of Plastic IC Reliability in the MLRS Fuze", Barry

Scheiner, Harry Diamond Labo - *ories, Project No. 424048, Report
No. R--440-82-1, 17 January 196U.
39. MIL..STD-1286C;Transformers, Inductors, and Ccils, Selection and

Use of; 28 October 1985.
40. MIL-M-38510F; Microcircuits, General Specification for; 31

October 1983.
41. "Storage Reliability of Missile Materiel Program - Relay Analy-

sis", D. F. Malik, U.S. Army Missile Research and Development Cori
mand, Redstone Arsenal, Alabama, Report No. LC-78-EM3, February
1978.
42. MIL-STD-1346B; Relays Selection and Applications; 29 April 1985.
43. "Design Guidelines for Transient Radiation Effects on Tactical

Army Systems", Rose M., Horrity, J., Rudie, N., Rasmussen, K.,
Holpin, J. J., HDL-CR-81-015-l.- July 1981.
44. "Hardness Assurance foi Neutron - Induced Displacement Effects in

Semiconductor Devices", Berger, R. A., HDL-CR-78-135-3. May 1978.
4.7-4
4 . "Pr.inciples and Techniques of Radiation Hardening", Rudie, N.

Western Prriud.ical Co.,. 1Q80.
46,. "Zuoiaeering Design Han.rbook - Rocket and Missile Container Engi-

nec::irq Guide", U.S. Army Materiel Development and Readiness Cou-
mand, LARCOM Pamphlet DARCOM-P 706-298, OTIC Report No. ADA 1YF
264, January 1982.
47. -1L-P-II H; Preservation, Methods of; 11 April 1985.
4F. "Engin--ering Design Handbook - Packaging and Pack Engineering",

U.F. Army Material Command, AMC Pamphlet AMCP 706-121, DTIC Report
No. AD- 01 5.3, March 1972.
49. "Desinr Guidelines for Prevention and Coitrol of Av.onic Corrosion

- Avit niu De-iqn Related to Fleet Eniro nent", Departnent of the
Na,'y., NAVMAT P 4855-2, June 1983.
51). "Oeterioration of Materials in Storage and Recomniended Storage

Conditions", M. L. Erickson, University of Minnesota Institute ct
Technology, Navy Contract 79585, DTIC Report No. AD-600 560,
January 1956.
51. MIL-E-17555H; Electronic and Electrical Equipment, Accessories,

and Prcvisioned items (Repair Parts): Packaging cl- 15 November
le4.
52. 'Enqineerirg Design Handbook - Dielectric Embedding of Electrical

or Electronic Components", U.S. Army Material Development and
Readiness Command, DARCOM Pamphlet DARCOM-P 706-215, DTIC Report
No. ADA 074 139, April l 'i9.
52. Not used.
54. MIL-HDBN-l5lB; Clothing Equipage, and Footwear Components Pa,-ts

and Mauerials; I August 1961.
55. The Official Railway Registe:.. The Railway Equipment and Publica-
tion Cc., 424 West 23rd Street, New York, New Yoik.
56. Shock Pnd Vibration Symposium bulletin, Shock and Vibration .nfor-
mation Center, U.S. Naval Resear 1i Laboratory, Washington D.C.
4.7-5
57. REi-STD-IOIC; Test Procedures for Packaging Materials; 13 March

180.
38. 2M 743-200/NAVSANDA PUB 2RA/AFM 67-3/NAVMC 101/DSAM 4145.1, Stor-

itge and Materials Handling, Department of the Army, the Navy, the
Air Force, the U.S. Air Force, and the Defense Supply Agency, 30
January 1961.
5). MIL-STD-137C; Material Handling Equipment; 25 June 1963.
60. MIL-STD-147C; Palletized Unit Loads; 30 August 1979.
61. MIL-' '6855B; Human Engineering Requirements for Military Systems,

Equi at and Facilities; 3? January 1981.
62. M±L-STD-1472C; Human Engineering Design Criteria for Military Sys-

tems, Equipment and Facilities; 2 May 1981.
63. MIL-HDBK-759A; Human Factors Engineering Design for Army Material;

30 June 1981.
64. AMCP 706-115, Basic Environmental Concepts, U.S. Army Materiel

65. Not used.
66. "Storage Reliability of Missile Materiel Program - Switch Analy-

sis", U.S. Army Missile Research and Development Command, Redstone
Arsenal, Alabama, Report No. LC-78-EM4, February 1978.
67. MIL-STD-1132A; Switches and Associated Hardware, Selection and Use

of; 23 March 1976.
68. DoD-HDBK-26,; Electrostatic Discharge Control Program for

Protection of Electrical and Electronic Parts, Assemblies and
Equipment (EL-luding Electrically Initiated Explos.ive De *-es); 2
May 1980.
69. Not used.
70. MIL-S:D-1353B; Electrical Connectors, Plug-in Sockets, and Asso-

ciated Hardware, Selection and Use of; 2 July 1980.
71. MIL-STD-454K; Standard General Reqairements for Electronic Equip-

ment; 14 Febr.ary 1986.
4. 7-b
72. "Storage Reliability of Missile Materiel Program - Missile System-

Battery Analysis", U.S. Army Missile Research and evelopment Com-
mand, Redstone Arsenal, Alabama, Report No. LC-78- A, February
1978.
73. M7L-STD-470A; Miintainability Program for Systems and Equipment; 3

January 1983.
74. "Engineering Design Handbook - Maintainability Engineering Theory

and Practj.e", U.S. Army Materiel Command, AMC Pamphlet AMCP
706-133, DTIC Repcrt No. ADA 026 006, January 1976.
75. "Simplicity - The Basis for a No-Maintenance Concept", E. L.

Peterson, Rockwell International, and G. Mitchell, Teledyne Ryan
Aeronautical, Circa 1982.
76. Linden, D., Handbook of Batteries and Fuel Cells, McGraw-Hill, New
York, 1984.
77. "Unavailability of Components with Inspe:tion and Repair", J. K.

Vaurio, Argonne National Laboratory, Nuclear Engineering and
Design 54, 1979, pp. 309-324.
78. "Storage Reliability With Periodic Test", E. C. Martinez, Motorola

Inc., 1984 Proceedings Annual Reliability and Maintainability S--
posium, pp. 181-185.
79. "Unavailability Analysis of Periodically Tested Component5 of Dor-

mant Systems", S. H. Sim, Ontario Hydro Research Division,
Toronto, Canada, IEEE Transactions on Reliability, Vol. R-34, No.
1, April 1985, pp. 88-91.
80. "Factors Affecting Fiber Optic Reliability", C. F. Veraa, Vitro

Corporation, 19811 Proceedings Annual Reliability and Maintain-
ability Symposium, pp. 236-241.
81. "Tensile Strength and Fatigue of Optical. Fibers", R. Olshansky and

R. D. Maurer, Joirnal of Applied Physics, Vol. 47. No. 10, October
1976, pp. 4497-4499.
82. MIL-STD-2165; Testability Program for ElectruQc Systems and

Equipments; 23 January
83. "RADC T*! ability Notebook", J. Byron, et .I, Hughes Aircraft Com-
pany, RADC-TR-82-189, june 1982.
84. MIL-STD-1309C; Definitious of Terms for Test, Measurements, and

Diagnostic Equipment; )8 November 1983.
4.7-7
85. "Joint DARCOM/NMC/AFLCiAFSC/Commanders Built-In-Test Design

Guide", Departments of the Army, Navy, and Air Force, NAVMAT2
9405/DARCOM P 34-1/AFLCP 800-30/AFSCP 800-39/NAVMC 2721, 1981.
86. MIL-STD-683F; Crystal Units (Quartz) and Crystal Holders (En-

closures), Selection of; 5 May 1978.
bT. "Storage Reliability of Missile Material Program - Storage Reli-

ability Summary Report - Volume II - Electromechanical Devices",
D. F. Malik, U.S. Army Missile Research and Development Command,
Redstone Arsenal, Alabama, Report No. LC-78-2, February 1978.
88. "RiM Design for Long Term Dormant Storage", G. Kasouf, General
Electric Company, and D. A. McGoye, Joint Cruise Missile Program
Oftice (DoD), 1984 Proceedings Annual Reliability and Maintain-
ability Symposium, pp. 168-175.
89, "AFDC Corporate Lessons Learned", U.S. Army Armament Munitions

Chemical Command, Armament R&D Center, Issue No. 1, April 1985.
90. "Corporate Memory Lessons Learned on Tactical Nuclear Systems", by

Lockheed Electronics Company, Inc., for Nuclear Systems Division,
Product Assurance Directorate, U.S. Army ARRADCOM, Dover, N.J.,
Contrcct DAAK10-82-C-0209, June 1983.
91. "TLCC Versus Confidence Testing During Long-Term Storage", G.

Kasouf, General Electric Company, 1981 Proceedings Annual Reli-
ability and Maintainability Symposium, pp. 101-107.
92. "Reliability Derating Procedures", D. J. Eskin and C. R.

McCanlass, Martin Marietta Aerospace, RADC-TR-84-254, DTIC -eport
No. AD-A153 744, December 1984.
4.7-8
5.0 EVALUATION METHODS
5.1 INTRODUCTION
Military electronic systems must be evaluated throughout their design,

development, production, and deployment to ensure that they meet all
performance requirements, such as functional performance, reliability,
maintainability, availability, useful life, etc. This is especially
true for systems that will spend a significant part of their service
life cycles in non-operating states since such evaluations are often
the only means available for determining the integrity of these syr-
tems. The following sections present evaluation methods that have
been found to be useful in evaluating systems of this type. Section
5.2 discusses in-plant evaluation methods and section 5.3 discusses
evaluation methods useful in a field environment. Special evaluation
methods are covered in section 5.4. Additional information concerning
evaluation methods for dormant systems is available in Reference 6.
5.2 IN-PLANT EVALUATION METHODS
In-plant evaluation methods are comprised of the many analyses and

tests that electronic equipment are subjected to during design, devel-
opment, and production to ensure that all performance requirements are
met once they are fielded. Many of these analyses and tests are the
same regardless of the amount of time that the fielded equipment will
spend in a non-operating state. This section covers several in-plant
evaluation methods that are either significantly affected by equipment
non-operating periods or have been useful in testing and evaluating
systems that will spend a significant amount of time in non-operating
states. Section 5.2.1 covers worst case circuit tolerance analysis
and discusses the need to consider the impact of long-term non-operat-
ing periods. Section 5.2.2 covers a method for performing reliability
assessments considering both the operating and non-operating periods
throughout an equipment's life cycle. Accelerated life testing as a
method for evaluating the impact of long-term operating and non-oper-
ating periods is discussed in section 5.2.3. Finally, environmental
stress screening is discussed in section 5.2.4.
5.2.1 Worst Case Circuit Tolerance Analysis (Ref. 1)
All circuits are reqired to function properly and to the requirements

of the applicable specification under their worst-case conditions.
These worst-case conditions include not only electrical parameters,
but environmental factors, packaging factors, noise, aging, storage,
and other factors as well. Worst case (circuit tolerance) analysis
ensures that the end product will have the capability to satisfy all
5-1
functional requirements, within the required performance and safety

margins, und3r the most unfavorable combination of realizable condi-
tions.
Three types of analysis are performed: d.c., a.c., and transient.

Worst-case d.c. analys.;s is accomplished by calculating the d.c. cur-
rent and voltage levels which exist at each node of a given circuit
under worst-case conditions. The final output as a function of the
input will then yield the d.c. transfer characteristics to be compared
to the design requirements. Worst-case a.c. analysis is accomplished
by calculating the a.c. current and voltage levels which exist at each
node of a given circuit under worst-case steady-state conditions. The
final output as a function of the input yields the a.c. transfer func-
tion which is then evaluated for discrepancies in characteristics such
as magnitude, phase shift, and propagation delay versus frequency.
Worst-case transient analysis is accomplished by calculating the ef-
fects, at various nodes, of worst-case transients resulting from
worst-case transient (step) functions at the various inputs and at
other nodes of the circuits where electromagnetic interference and/or
cross-talk can occur. Worst-case transient analysis is also performed
to determine the effects of worst-case shock, nuclear, and other
environments.
To perform the worst-case analysis, three analytical techniques

have been developed:
1. Absolute Worst Case - Circuit is evaluated to determine if it

meets all performance requirements with all components degraded
to their end-of-life tolerance in the worst direction.
2. Taylor Worst Case - Circuit is evaluated to determine if it

meets all performance requirements with all components at their
worst initial (purchase) tolerance and with any one component
part at its end-of-life tolerance.
3. Modified Taylor Worst Case - Circuit is evaluated to determine

if it meets all performance requirements with all components at
their nominal values and with any one component at its worst
end-of-life tolerance.
As a minimum, the Modified Taylor Worst Case analysis should be ap-

plied to linear analog circuits. For digital circuits and analog
feedback circuits, the Absolute Worst Case or Taylor Worst Case analy-
sis should be applied. In addition, the Taylor Worst Case and Modi-
fied Taylor Worst Case analyses should be applied to each component
whose parameter variations have, as indicated by a sensitivity analy-
sis or engineering judgement, a significant effect on circuit opera-
tion.
5-2
In addition to these analysis techniques, computerized methods such as

ECAP (Electronic Circuit Analysis Program) and SCEPTRE (Systems for
Circuit Evaluation and Prediction of Transient Radiation Effects) can
also be employed. The important consideration for dormant systems is
that parameter drift and device degradation during dormant periods
over the life of the system be considered in the analysis.
When performing the worst-case analysis, the following procedures

should be followed:
1. Define the circuit performance requirements.
2. Generate analytical models for each circuit to be analyzed.
3. Determine the worst-case tolerances of the independent variables

(i.e., part parameters, supply voltages, interface impedances,
electro-magnetic interference, cross-talk, temperature, input
signals, noise, etc., including end-of-life end-point values)
which appear in the analytical models.
4. Determine additional effects to the various independent vari-

ables due to the various environments of nuclear attack, long
term storage, transportation and handling, etc. These environ-
mental considerations should include shock, vibration, humidity,
temperature, gamma rays, x-rays, neutrons, beta rays, etc.
5. Using the worst-care tolerances of the independent variables,

solve the analytical models for the minimum and maximum values
of the circuit dependent variables (i.e., voltages, currents,
dissipations, output requirements, etc.)
6. Compare worst-case minimum and maximum values of dependent vari-

ables with design requirements.
7. Note all circuits which will not meet their performance require-
ments under any set of required worst-case conditions along with
recommendations of dppropriate corrective action. When possi-
ble, an analysis of the probability and/or duration of these
circuit malfunctions should also be made.
The following precautions should be observed when performing the

analyses:
1. Input and output loading at each circuit should be strictly ac-

counted for.
2. Stray effects of special circuit-interfaces developed for ease

of analysis should be delineated.
5-3
3. All assumptions regarding the circuit should be noted in the

analysis.
4. The effects of noise, electro-magnetic interference, and cross-

talk should be considered. Nuclear environments should be con-
sidered, where applicable.
5. Worst-case analyses should be sufficiently generalized to show

that entire circuits and combinations of circuits will perform
their required functions under worst-case conditions.
6. Analyses may be necessary both for the individual circuits and

for various groupings of the same circuits in order to show that
the subsystems will meet design requirements.
7. Temperature ranges assumed for various parts and elements during

operating and non-operating periods should be in agreement with
the results of any thermal stress analyses performed.
8. Traceability must be maintained in the performance of the analy-

sis to permit any reviewer to readily follow the calculations
and rationales, and to allow for ease of updating of the analy-
sis as data or requirements change.
5.2.2 Life Cycle Reliability Assessment
A comprehensive life cycle reliability assessment requires the consid-

eration of all possible states and the time in each state that an
equipment must survive. While most contractual reliability require-
ments are given in operational terms, the performance of electronic
equipment depends on the effects of other life cycle states such as
transportation, testing, equipment power on-off cycling, and non-oper-
ation.
Reference 2 provides an eight part procedure for developing a compre-

hensive life cycle reliability assessment. The eight steps are as
follows: (Ref. 2)
1. Define the mission profile.
2. Determine the use scenario.
3. Determine component operating failure rates.
4. Determine equipment operating failure rates.
5. Determine component non-operating failure rates.
5-4
6. Determine equipment non-operating failure rates.
7. Compute the average service life failure rate.
8. Compute the reliability.
Step 1, the definition of the mission profile, is the initial phase of

any reliability assessment process. During the mission profile defi-
nition process, specific time intervals and stresses that the equip-
ment will experience in every operating environment must be defined.
The use scenario, step 2, considers the mission profile defined in

step 1 plus the non-operating time between missions. In addition to
the operating environment information provided from the mission pro-
file, specific time intervals and stresses encountered during the non-
operating phases of the use scenario are defined.
Steps 3 and 4 are well defined in MIL-STD-785 (Ref. 3) and MIL-HDBK-

217 (Ref. 4). Steps 5 and 6 represent the corresponding reliability
assessment tasks for non-operating periods based upon the use scenario
and are fully covered in References 2 and 5. Steps 3 through 6 yield
two unique parameters characterizing equipment reliability: the in-
herent equipment operating failure rate and the inherent non-operating
failure rate.
In step 7, the average service life failure rate, which is defined as

the number of failures per unit time regardless of operational mode,
is computed. It is dependent upon the operating failure rate, non-
operating failure rate, operating duty cycle, and non-operating duty
cycle and is represented by the following equation for a scenario of
one operating state and one non-operating state: (Ref. 2)
XSL = X 0 (to/(t 0 + tN)) + XN(tN/(to + tN))
where,
XSL = service life failure rate
X0 = equipment operating failure rate
XN = equipment non-operating failure rate
to = one mission operating time
tN = non-operating time between missions
While this equation will become more complex as the number of operat-
ing and non-operating states increases, it will take the same general
5-5
form. The service life failure rate is one of the most informative
comprehensive reliability pc.:ameters. (Ref. 2) It is constant with
time and is therefore useful for comparing or evaluating the total
reliability for equipments with an indeterminate mission duration.
This is extremely important since, as demonstrated by E. Demko (Ref.
7), using only the total number of failures and the total operating
time will give the appearance of a changing failure rate due to the
statistical effect caused by a changing usage profile. The service
life failure rate also provides a good indication of the reliability
of equipments which are exposed to intermediate amounts of both oper-
ating and non-operating periods, and for equipments with a use sce-
nario dominated by non-operating periods, it approximates the dormant
equipment failure rate.
The (t0 /(t 0 + tN)) 'erm represents the operating duty cycle, and
the (tN/(tO + tN)) term represents the non-operting duty cycle.
In cases where there is no definitive service life time interval (to
+ tN), estimates can often be made of the respective duty cycles.
The service life failure rate could then be computed with appropriate
duty cycle values rather than specific values for to and tN .
Given the time intervals defined during the definition of the ube
scenario, the equipment reliability is given as:
R (( oit0i + A NitNi))
where,
R = reliability
XOi = operating failure rate in the iTH operating state
toi= time in the iTH operating state
)Ni = non-operating failure rate in the iTH non-operating state
tNi time in the iTH non-operating state
This model assumes a "wooden round" maintenance concept (i.e. no test

or repair prior to use.). Adjustments to this model that are neces-
sary to account for the impact on reliability of a periodic monitor-
ing/repair maintenance philosophy are described in section 4.3.
When using this procedure, it must be understood that the use of hand-
book derived component failure rates to calculate the service life
failure rate and equipment reliability limits the usefulness of these
parameters to relative comparisons among competing designs. It is
5-6
inappropriate to use these parameters so derived as absolute measures

of equipment performance in an actual use environment. If absolute
measures are required, the assessments should be based upon test data
or actual field experience for the equipment in question. These
topics are covered in succeeding sections. For additional information
on methods for assessing dormant system reliability, the reader should
refer to Reference 6.
5.2.3 Accelerated Life Testing
Accelerated life testing (ALT) is a technique often used to predict

the lifetime of electrical or mechanical components in an intended
usage environment or to obtain failure rate data in a relatively short
period of time. This is accomplished by subjecting items to above
normal stresses over a short time period in order to simulate the ef-
fect of normal stresses over a much longer time period. It requires a
knowledge of the predominant failure mode under rated stress condi-
tions, the environment which excites the failure mode, and a quantita-
tive relationship between the stress level and the failure occurrence
or material deterioration rate. Once this relationship is determined,
it can be used to estimate the failure rate or aging behavior within
the usage environment of interest.
ALT is a very effective method of reducing the time required to accom-

plish the magnitude of testing necessary to provide a statistically
significant measure of reliability on low failure rate items. It is
therefore suitable for evaluating dormant systems since they typically
have very long lifetimes (often 20 or more years) and low failure
rates due to reduced stresses during non-operating periods.
ALT is most effective when applied at the component level rather than
at higher levels of assembly. The foremost problem in applying ALT at
any level is the determination of the acceleration factor. Accelera-
tion factors have been validated at the component level by testing at
various stress levels. At higher levels of assembly, however, the
validation of acceleration factors is very difficult, if not impossi-
ble, to achieve. The excitatiou of failure mechanisms at higher as-
sembly levels is also a problem. Since individual components are
likely to be sensitive to different stresses, it is not reasonable to
expect that a single stress could be chosen to provide an accelerated
test.
The two most common methods of accelerated aging are referred to as

time-compression techniques and over-stress techniques. (Ref. 9) In
time-compression techniques, the rate of cycling of an environmental
variable (e.g. shock, vibration, power on-off cycling, etc.) that is
cycled during the component's lifetime is increased. It is an appro-
priate accelerated aging technique as long as the cycled variable is
5-7
the sole cause of the failure mode and if the failure depends ou the
number of cycles but not on the rate of cycling.
Over-stress techniques involve increasing an environmental variable

above its ambient value in order to accelerate the aging process.
Several models relating aging to stress have become well esrablished
and are used frequently in applying the over-stress technique. These
models are: the Arrhenius Model, the Eyring Model, the Inverse Power
Model, and the 10-Degree Rule.
The Arrhenius model relates the variables of time and tEmperature to

the deterioration of materials and therefore is often used for calcu-
lating accelerated thermal aging parameters. It is usually applied in
the form: (Ref. 8)
E / kT
L = Be
where,
L = time to reach a specified endpoint or lifetime
B = constant (usually determined experimentally)
E = activation energy (eV)
k = Boltzmann's constant (0.8617 x 10- 4 eV/K)
T = absolute temperature (K)
To use the Arrhenius model for the purposes of accelerated thermal

aging, the activation energy must be known. Table 5.2.3-1 lists
activation energies for a number of materials and components. In ap-
plying this model it must be understood that extrapolation of the
model to time periods (temperature) far exceeding the range of experi-
mentally determined data is extremely risky, and large errors may re-
sult. (Ref. 8)
The Eyring model provides a formulation that is more thermodynamically

correct than the other models and is the only one that can include
additional stress terms to allow for synergistic effects. It is most
difficult to apply in practical aging programs, however, because the
model parameters have been determined for very few materials and com-
ponents. The model takes the following general form: (Ref. 8)
K = K 0 enS = aTwe-b/kTeS(c + (d/kT))
where,
5-S
Table 5.2.3-1. Activatio. Energies for Electronic

Components (Ref. 8)
Activ.
Energy
CO ponent 64V) Remarks
Capacitors, chlorinated 1.17 DC life. Stressed at 1000 Volts per

diphenyl. No stabilizers. nil. See Note 1.
Capacitors, chlorinated 1.53 DC life. Stressed at 1000 volts per

diphenyl. 0.5% ioi1. See Note 1.
anthraquinone
Capacitors, chlorinated 2.00 DC life. Stressed at. 100. volts per

diphenyl. 0.5% mil. See Note 1.
azobenzene
Capacitors, chlorinated 0.86 Dielectric stressed with dc poten-

diphenyl Kraft paper tial, 106 V/in. See Note i.
Capacitors, chlorinated 1.50- Dielectric stressed with dc poten-

diphenyl Kraft paper 1.93 tial, ,6 V/in. See Note 1.
with 0.5% azobenzene
Capacitor, dielectric, 2.42 10% capacitance increase. See Note

tubular paper 1.
Capacitors, metalized 1.32 Life defined as time required to

paper regain original -alue of capaci-
tance after initial increase. See
Note 1.
Capacitors, titanilm 0.09 Formed by anodization. Tests with

titanium dioxide, thin- rate of temperature rise approxi-
film. @25-100 deg. C mately 2.5 deg. C/min.
Diodes, Si - reneral 1.13-

2.77
Diodes, Si (-1960) 1.14
Diode, silicon, 1N673 1.8 50% failure. See Note 1.

and 1N696
Diodes, silicon, p-n-p-n 1.41
5-9
Table 5.2.3-1. Activation Energies for Electronic

Components (Ref. 8) (Continued)
Activ.
Energy
Component (eV Remarks
Diodes, silicon, 2.31-

varactors 2.38
Diodes, other 1.13-

2.77
Diodes, varactors 2.31-

2.38
Microcircuits, CMOS 1.0 25h @ 250 deg. C = 50% failure.

type CD4024A See Note 2.


Microcircuits, CMOS
4C07 freak pop. 0.9
main pop. 1.3
Operational Amplifier
741
-freak pop. 0.7
-main pop. 1.6
-mixed pop. 0.8
-freak pop. 0.8
-main pop. (1/2 voltage) 0.9
Printed circuit board 1.05 50% retention of electrical

material (1/32 in.), strength. See Note 1.
NEMA G-10 and FR-4
Printed circuit board 1.49 50% retention of flexural

material (1/32 in.), strength. See Note 1.
NEVA G-10 and FR-4
Semiconductor devices, 0.9- predominant value - 1.1 eV.

silicon 1.4
5-1 1

Activ.
Energy
ComRonent (eV) Remarks
Silicon transistors and 1.1 Testing of transistors and inte-

integrated circuits grated circuits based on Arrhenius
model.
Transistors 0.66
Transistor, Ge alloyed,
OC 1972 (1964) 1.26
(1966) 1.08
Transistor, Ge alloy 1.25

LT123 (1958)
Transistor, bipolar, 1.65

p-n-p-n
Transistors, CMOS 1.18 Eyring model.
Transistor, diffused 0.87 Step-stress tests without moisture

germanium moisture getter. Median life. See
Note 1.
Transistor, diffused 1.24 Constant stress tests with moisture

germaniun moisture getter. Median life. See
Note 1.
Transistor, Ge gettered 1.24
Transistor, Ge mesa, 1.00

AF106 (1969)
Transistor, Ge mesa,
2N559 (1958) 1.17
(1959) 0.95
(1960) 1.14
Transistor, Ge MADT, MADT = Micro alloy diffused tran-

211501 (1958) 1.07 sistor.
Ge MADT, 2N501 (1959) 1.07
5-11

Activ.
Energy
Component (eV) Remarks
Transistor, Ge MAT 1.0 MAT = Micro alloy transistor.

2N292 (1960)
Transistor, Ge MAT 1.0 MAT Micro alloy transistor.

2N393 (1959)
Transistor, Ge ungettered 0.88
Transistors, germanium 0.99-

@ 60 deg. C 1.26
Transistors, germanium 0.17 Near and below room temperature.
Transistors, germanium, 0.88

ungettered
Transistors, germanium, 1.24

gettered with vycor or
molecular sieve
Transistor, Si mesa, 0.38 Conditions unknown.

2N269 (1961) 0.58 Constant stress.
Transistor, Si mesa,
2N560 (1959) 1.12
(1960) 1.50
Transistor, Si mesa,
2N1051 (1960) 1.12
Transistor, modern, 1.4

submarine cable
Transistors, MOS 1.2
Transistors, MOS 1.10 Median life for failure criterion

of 1.0 V shift. See Note 1.
Transistors, MOS 1.10 Median life for failure criterion

of 0.5 V shift. See Note 1.
5-12

Activ.
Energy
Transistor, power, MSC 0.81 Median time to failure. See Note 1.

1330
Transistors, Si main pop.

(1960) 1.02
Transistor, Si planar,
BFY 33 (1969) 1.12
Transistor, Si planar,
4A-2 (1967) 1.18 Step stress.
(1967) 1.50 Cnnst*ant stres4.
(1963) 1.29 Constant stress.
Transistor, Si, p-n-p-n 1.65
Transistors, Si, (All)

-before wearout 1.12
-at wearout 1.46
Transistor, Si, bipolar 1.02 With surface inversion failures.
Transistor, Si, bipolar 1.02- With Au-Al bond failures.

1.04
Transistor, Si, bipolar 1.77 With metal penetration into Si.
Transistor, silicone 2.16 50% failure. See Note 1.

mesa, 2N560
Transistors, Si, typical 0.96 t1 0 lifetime. See Nc-e 1.
Transistors, Si, typical 1.11 t 50 lifetime. See Note 1.
Transistors, submarine 1.30 0.025% failure. See Note 1.

cable
Transistors, submarine 1.24 50% failure. See Note 1.

cable
5-13

Cnmponents (Ref. 8) (Continued)
Activ.
Energy
Transistors, 2N559, 0.89 Me, i life based on failure

vacuum baked. cri. ria of collector breakdown
voltage and reverse current, and
emitter breakdown voltage. See Note
1.
Transistor, Vycor 1.02 50% failure. See Note 1.

gettered germanium,
2N559.
Notes:
1. Calculated from Arrhenius-type plots.
2. Burned in @ 125 deg. C, then powered during life testing @ 250

deg. C. Main failure mode was high leakage currents.
5-14
K0 = reaction rate in the absence of applied stress
K = reaction rate in the presence of applied stress
k = Boltzmann's constant
T = absolute temperature
a, b, c.
d, and w = experimentally determined constants independent of
time, temperature, and stress
S = a function of the applied stress
n = c + (d/kT)
The Inverse Power model has been applied to problems such as the fa-
tigue testing of metals, the dielectric breakdown of capacitors, and
the aging of multicomponent systems. It has the following general
form: (Ref. 8)
L = I/kVn
where,
L = time to reach a specified endpoint
k,n = positive parameters characteristic of the material and

test method
V = stress (e.g. current, voltage, temperature)
The 10-degree rule is an approximate relationship for describing the

rate of a temperature-dependent reaction. It has been empirically
derived and often fits the observed data well. The rule states that
for each 10-degree (C or K) rise in temperiture, the specific reaction
rate doubles. The general form of the equation describing this rela-
tionship is as follows: (Ref. 8)
R = C2t(Tl - T 0 )/10]
where,
R = the specific reaction rate
C = constant determined by experiment
T = temperature, aegiee C or K
5-15
A better match between the model and experimental data can often be
achieved by modifying the 10-degree rule using a value for the con-
stant, usually ranging from 5 degrees to 20 degrees C, which best fits
the data.
The interpretatiun of the results from ALT is the subject of frequent

controversy. According to S. P. Carfagno (Ref. 8), aging that can be
accelerated in ways that yield verifiable correlation between real and
simulated aging is an exception rather than the rule. Several envi-
ronmental and operational stresses may cause degradation which is de-
pendent upon the stress level, time, changes in stress level, and the
sequence of such changes. Since it is very difficult tu predict the
stress history that a component will experience in service, the
specification of ALT parameters to simulate the service life is also
very difficult.
Models relating the causative factors to equipment degrad&tion are

essential to accelerate equipment aging. The models presented here,
despitp the considerable mathematical sophistication of some, are only
rather simple approximations of the real-life problem and sufficient
correlation with observed real-time aging to substantiate their valid-
ity for long periods of time is not available. While the models re-
quire experimental determination of key parameters, this has only been
accomolished Eor very fe-i components, and then for comparatively sim-
rle stress histories.
In conclusion, equipment aging on a rigorous scientific basis appears

to be beyond the present state of technology. While it is inappropri-
ate to use the results of presently available accelerated aging tech-
niques to qualify the service life of an equipment, it is appropriate
to use these re:oults to assess qualitatively the vulnerability of
equipment with respect to aging effects. For a more detailed discus-
sion of the theory and piactical application of ALT techniques, the
reader should refer to References 8, 9, and 10.
5.2.4 Environmental Stress Screenig
Analysis of dormant systems has consistently come to the conclusion
5-lb
that most failures that occur during dormancy are the result of latent
defects that were not removed from the product during post assembly
testing. Environmental Stress Screening (ESS) is a technique which
has in recent years come to the forefront as a method to detect the
presence of latent defects at any level of assembly from the component
to the system level before the item is sent to the next level of as-
sembly or to the ultimate user. ESS is finding grjwinn use by defense
electronic contractors as a means to precipitate failures due to la-
tent defects as early as possible to reduce costs and to provide
higher field reliability. ESS is particularly of interest for dormant
systems because it is primarily designed to screen out the latent de-
fects that have historically accounted for the majority of failures
during dormant periods.
The process of ESS detects latent defects by subjecting test items to

specific conditions of environmental stress, so that such defects can
be degraded to a detectable level. Temperature cycling and random
vibration appear to be the most effective tests, however, due to the
varied nature of military electronics equipment and their associated
design, development, and production program elements, a standard ESS
approach is not feasible and a tailoring of the screening process to
the unique elements of a given program is required. (Refs. 11 and 12)
While a standard approach is not feasible, the Institute of Environ-
mental Sciences has made the following conclusions based upon ESS pro-
gram experience to date: (Ref. 12)
" The major objectives of a screening program are to reduce overall

cost, improve productivity, and assure field reliability.
* Each ESS program is by nature hardware unique.
" The tailoring of screens should be considered a normal part of any

ESS program.
" Stress screening parameters should be dynamic to permit their mod-

ification and/or elimination as the product matures.
* As a product matures, the quantity and nature of the defects pre-

cipitated by the screens depends on the effectiveness of the fail-
ure analysis and corrective action program instituted, and on the
adaptation of the screens to current manufacturing processes and
conditions.
* Stress screening levels should not exceed design limits.
" Temperature cycling precipitates 2/3 of all defects and random vi-
bration precipitates the remaining 1/3.
5-17
" The combined environments of temperature cycling and vibration do

not ap2ear to be any more effective than environments applied se-
quentially ana they are considerably moe expensive than sequen-
tially applied environments.
" Random vibration is more effective than either fixed frequency or

swept sine vibration as an environmental screen.
" Module level screening is effective with or without power ap-

plied. The additional defects precipitated by the application of
power do not appear to justify the added costs in most cases.
" Unit and system level screening is more effective if power is ap-
plied to the hardware.
It is beyond the scope of this report to fully describe all of the

steps necessary to develop an effective ESS program. Over the past
several years, various guidance documents have been published towards
this purpose. An overview and comparison of the major documents is
presented in tables 5.2.4-1 and 5.2.4-2. They should be consulted for
further details. In addition, Reference 12 describes a set of three
computer programs which can be used in selecting screens, setting
screening parameters, and adjusting screens on the basis of observed
results in order to find the optimum set of screens to achieve desired
piuyram objectives.
5.3 FIELD EVALUATION METHODS
This section covers two topics regarding the evaluation of fielded

dormant systems. A model useful for assessing the readiness and main-
tainability of dormant systems using field maintenance data is pre-
sented in section 5.3.1. A discussion of the "certified round" con-
cept for conducting periodic test and recertification of dormant sys-
tems is covered in section 5.3.2.
5.3.1 Readiness/Maintainability Assessments Using Field Data
Reference 14 describes a general model for assessing the readiness and

maintainability of stockpiled dormant systems. The model utilizes
field maintenance data available from most field data reporting sys-
tems and permits assessments over any time period for which data is
available. By segregating the field maintenance data into active,
logistic, and administrative time categories, the model can be used to
demonstrate the effects that each time category has on stockpiled dor-
mant system readiness and maintainability. While the model was spe-
cifically developed for stockpiled dormant systems, it is also appli-
cable to actively used equipment.
5-18
ctL~w 0 0 0
V) 0U z _' J_
CC L 0- L W
LJ
< w w C
0 z z )-~L z~
-0 0 C1 L
C[ > J0>W jL
)cr
0 0r
w crww IMCj
LL~~ a 0
< (3 <W 2
4)0 .,0. IO WLI W Cc
W
Cr-W J ~ '- <Ww~
0
1
-jV)
uCl Q)Q
CIZ cCC
0 o C LL >o
CL C z O zc 0
NI -j01 a )LL ~ 0 Nr w C' tr

E~ O Z /
N- LL..C- n_ 1C i-c
*LnD
0n C C,03-~LO (
E~ 0o W ,l < W Q-C> 0 W
0
(- 0
jdJJ
I- <-u 0XCc >W 3
a: 0 Cl) : 0 ua
_E < L U z 0 D s z 3:
'E L3 T-Zi z z Z LU - W
E WCo
C w ~ CDC>5 W 0
i> (1) w C:
0LF-Ln<L <
0E0
Cc
( ~ ~ u< ~ -~f
2 Cc t
u ~ ~ I ~ uU 3
Eu
a: W 31Z~
.0 In -0
z I
0 zCl- Cc
c, - >cj COo > L
Z~~-
-J 'QU 0 oW - (rW
Cn Z LU D > (n
C. ccWcui -n U 0L 0
V)-
Z a. w CJC
C 0
z z z
CL 04 n C N LU
CCa.0 0 LLWZ
Cc "L c
Cl -1 0 0w
0 ,L L 0; Lr
o 0 CD 0a: 0cr 0uD
oD z- < Go0>z a
-o 01~ coC:7 -
0i
- zw ON, x _
H
< 'a4
Za:r-jz
ctLU ?OZL wcO

2_ _ __ _ _ A 0
_ z < LU<
- , -) -> w
-a- i
oo
-' -J I cc c
<0 C
C/ LzW 0 4 L
3-~u U a. L! ZT ) W
L LU -
0 a:
C
z w. < ) < < 0I.LJ Zc . -
- LU L LI) W oou/ i-~

(1
~ 0 Z C C
C> 7 ,'2 LUC/5 Z L
Z L W
4LU wV -
H> 0
<L C LU o L
ccHLL Z c 0 <LU
IOLaQ
o LU cc~ <
c> LLU0-cc --
Ij= nZ)wL L 0 w ww~
u2
i OZ DOOwZ
~ C0 U
:O 4w~ >-~ . O Dr j U, zu'i > w L/V)
ci . .r '*1*0
LJz
0 a -- LLO LU
I - : -U -
w C - 2 )CcZLU 0 4D.<
H~~~ UJ'L 0
- ZLU c
LUL~a
c- >Lc LLCL
60 LU0a C U_(
-j 4 '-
)c:zc H L
<A V) LCm'L < _
2 . Z c -~
LLJ
Z ,uclW uC ci LI :
- LLJLjjF C-2
f fF
z L12L,
O < 0 LU
> cr U(
WOj
QLU
U. J) a-UJ c
S z ~D 0. < Mt QlZU
a~ (n
>0 -a
< 0L
c I L 0 L
0
- LUj 'i 0zL i
I- >- z: L n a,.r,
Coz a
- < -
EEIL DCL I U l
o~ ~ V C): Z <U
,cj
< 2 0 0- u~~I
oU LA0 cL a.
LU
CL0I
- a:crr)Uc 0V
OULu 2 wICJuCw L J LU, _L Q
0 I'CI ~ ~ ~ C a(a:~t
0
3:0-
ZW
0-.JL
r o~~
c' 0<
0 U~(mw~
0W- nJ
Cs (- .zLi 0
U
c U rW( r--r, u C
Z U-C- <c~u 0 <
E - 0 -~~J -'I -J - ( . -I)
w ~ L
011W .0 L C I D Lj L 'r Z
Cc: LIU
2CL wC - jQ c L
Ir) u
I-_jU 00L.C 0 UC
CL x w u C
-L LD-'~W: 0 L)(n1
c5 2
U-1
11w Iaua.
0 6z
x cc w. c
w 0 >
k2 <J C:a: a- 1
Z ~
6r I LLc
0< z
U.a: LIrO >< ir : I - X-
W cc C:c a: < ~ a: r, L
0__u0 5 : j
oL z < .[ C0 C L L ( L .
C. CLI - O: r 0P t'
C ; <~C 7a:301- ZF - ,
k-) LL- :(] () uL -) ZO L )- > a
cI .,wZW '') (D -Z %n z
I. 1I 0,, w ) .1j0 C-J
-W
w
c cn cr
o~0 L 0>U
cm: LL a. -
0J
0) ZDI
Z 4) 0 LL C
wZ
1w < o
0 0 <
z ZJ z
cr, 0
=) Cc0 .- o 0
E - -') 02 0-
The following assumptions provide the basis for the model:
1. An item is assumed ready if it is not undergoing a maintenance

action. Am item is assumed not ready while it is undergoing a
maintenance action.
2. Items undergoing maintenance actions that do not preclude imme-

diate use are considered ready while such maintenance is being
performed (e.g. simple spot painting of a container).
3. There is no distinction between an item in storage (non-operat-

ing) and an item in use when assessing an operating/nonoperat-
ing mixture of items.
4. An item is considered ready if maintenance is completed within

an allowed turnaround time T.
5. 'eadiness is based upon total calendar time.
6. No new maintenance actions are started during the allowed turn-

around time. Once an item becomes ready during this time, it is
considered ready for the duration.
7. In situations where a failure has occurred, an item is assiumed

ready up to the detection of the failed condition.
Readiness and maintainability as used in the context of the model are

defined as follows:
Readiness (U(T))
U(T) = the percentage of items that can he placed in use within

an allowable turnaround time (T) after use is called for
at a random (unknown) point in time.
Maintainability (M(T))
M(T) = the percentage of items not immediately ready for use

that can be made ready for use within an allowable turn-
around time (T) after use is called for at a random (un-
known) point in time.
The basic relationship between readiness and maintainability is then

defined by the following equation:
U(T) = U(O) 4 M(T)[l - U(0)]
where,
5-24
U(T) = stockpile readiness for an allowable turnaround

time T.
U(O) = immediate stockpile readiness (T=O).
(I - U(O)] = immediate stockpile unreadiness (T=O).
M(T) = stockpile maintainability for allowable turnaround

time T.
This equation illustrates that readiness for an allowable turnaround

time is composed of two parts: (1) the fraction rr the stockpile that
is immediately ready (U(O)), and (2) the fraction of the stockpile
that is restored within T (M(T)(1 - U(O)]).
The derivation of the constituent parts then proceeds as follows:
(i - U(O)] = ( tM)/(N Z tCAL )
where,
tM = the total maintenance time for all maintenance

actions occurring during the period for which the
assessment is being performed.
N = the number of items in the stockpile during the

period for which the assessment is being performed.
tCAL a the calendar time of the period for which the

assessment is being performed.
and,
U(O) = 1 - (1 - U(O)]
Maintainability is described by the following equation:
M(T) = [E tM(tM 5T) * N(tM)T) x T]/(E tM)
where,
EtM(tMIT) = the sum of the maintenance time for all maintenance

actions during the period that could be completed
within T.
11(tM>T) = the total number of maintenance actions during the

period that took greater than T time to complete.
T = the allowable turnaround time.
5-25
ZtM = the total maintenance time for all maintenance

actions occurring during the period.
These two equations ais subject to two restrictions. First, the num-
bet of items in the stockpile (N) must be constant over the assessment
period and second, maintenance actions must not overlap the boundaries
of an assessment period. (i.e. All maintenance actions that occur
during the period are assumed to have started and ended during the
period.) To remove these restrictions, the following equations should
be applied:
[1 - U(0)] = (EtM - Zt M - Zt )/(NAVG x tCAL )
= (ZtMC)/(NAVG x tCA L )
and,
M(T) = [EtMC(tM<T) + E(T - tM)(for tMT~tM &
(T - tM) <_tMC) + ItMC(for tM>TtM &
(T - tM)>tMC)]/(EtMC)
where,
tM = the total amount of time expended after the end of

the assessment period for all maintenance actions
occurring during the assessment period.
EtM = the total amount of time expended before the start

of the assessment period for all maintenance
actions occurring during the assessment period.
ZtMC = the total amount of time expended during the

assessment period for all maintenance actions
occurring during the assessment period.
= rtm - tt - t+
NAVG = the weighted-av,-age number of items in the

stockpile over the assessment period.
Reference 14 also contains derivations of these equations that can be

used during development when field data is not available. The conver-
sion terms are as follows:
ZtM = MDT x (NAVG x tCAL)/(MTBMA + MDT)
5-26
ZtM(tM T) = L(NAVG X tCAL)/(MTBMA + MDT)] x 0 fT(m(t) x t)dt
N(tM>T) =[(NAVG X tCAL)/(MTBMA + MDT)] x (1 - 0 fTm(t)dt
where,
MTBMA mean time between maintenance actions
MDT = mean down time.
m(t) = the maintenance probability density function (i.e. the

probability that a maintenance action takes t time to
complete).
After substituting these conversion terms, the equations for readiness

and maintainability become:
M(T) = [ 0 5T(m(t) x t)dt + (1 - 0 fTm(t)dt) x T]/MDT
(1 - U(0)] = MDT/(MTBMA + MDT)
U(O) = MTBMA/(MTBMA + MDT)
5.3.2 Periodic Test and Recertification
As discussed in section 4.3, the objective of periodic testing and

recertification of dormant equipment is to ensure the reliability of
the equipment through the detection and repair of non-operating fail-
ures before the equipment is called upon to perform a mission. It is
not applicable to all dormant equipment and should only be employed
where it is necessary to ensure that equipment reliability does not
fall below the minimum acceptable reliability over the equipment ser-
vice life. (See section 4.3 for further details concerning test/no-
test tradeoffs.) When it is necessary, however, it is very important
that the test and recertification operations be carefully controlled
in order to avoid the introduction of contaminants that can cause
long-term non-operating reliability problems, and to avoid failures
induced by the monitoring and repair operations, some of which may go
undetected.
To minimize the undesirable effects of test and recertification oper-

ations, an approach called the "certified round concept" has been in-
stituted on dormant systems (e.g. IHAWK) with very good results. (Ref.
15) An essential element of this approach is the employment of de-
ployable test and repair stations housed in a controlled environment
which duplicates the conditions under which the equipment was origi-
nally built and tested. On the IHAWK program they are called Theater
Readiness Monitoring Facilities (TRMF). They permit the testing and
5--2 7
repair of fielded equipment in a known, controlled environment so that

maintenance induced failures and the iLtroduction of contaminants are
minimized. In additi- *, since the test conditions duplicate those
during which the equ .:ment was initially tested at the manufacturing
facility, the test results from the TRMF car be directly compared with
baseline tests that were conducted during manufacturing.
The "certified round concept" requires the initial certification of

each lot of equipment produced. The initial certification is per-
formed using equipment of the same type that is employed in the TRMF.
All data from this testing is retained in a central data bank for
future reference.
After initial certification and lot acceptance, the equipment is de-

ployed. At prescribed intervals after deployment (see section 4.3 for
methods of determining optimum test intervals), a random sample of
each lot of equipment is selected for recertification testing at the
TRMF. During recertification, the selected equipment are subjected to
the identical tests that were employed during initial certification
testing. Should an equipment fail, it is repaired using repair test
sets included in the TRMF. All data gathered during recertification
is then stored in a data bank and later compared to the initial certi-
fication data to determine trends in equipment reliability degradation
during its fielded life due to operation, transportation, handling,
and storage.
5.4 SPECIAL EVALUATION METHODS
In addition to the evaluation methods described in the previous sec-

tions, systems that spend most of their service life cycles in non-
operating states are often subjected to special evaluations to deter-
mine the impact of long-term non-operating periods. This section
describes three special evaluation methods. Section 5.4.1 covers
stockpile laboratory testing programs. Section 5.4.2 covers stockpile
flight test programs. Finally, section 5.4.3 covers special real-time
component aging programs.
5.4.1 Stockpile Laboratorv Testing Programs
Stockpile laboratory test programs are detailed test, analysis, and

reporting programs that are designed to compare a dormant system to
established standards for projecting shelf and service life throughout
its life cycle. Under these programs, samples of fielded dormant sys-
tems are periodically removed from storage and examined for degreda-
tion from original specifications or the results of prior tests. The
examinations typically include disassembly, inspection, ATP testing of
all critical components, the actuation of one-shot devices, and root-
cause analysis of all failures.
5-28
The primary objective of these tests and analyses is to discover evi-

dence of deterioration during storage and to assess the readiness and
reliability of the stockpile. If evidence of deterioration is de-
tected, the test results are used to develop corrective action
measures.
Because of the limited number of samples that are evaluated during

stockpile laboratory test programs, the assessment of reliability by
conventional attribute failure classification methods is generally not
feasible. However, in Reference 17, H. Hecht describes a reliability
estimation method which permits the attainment of a given confidence
level with a smaller sample size or the attainment of a higher confi-
dence level with a given sample size utilizing the numerical measure-
ments obtained from laboratory test programs. Called the Design
Margin Approach, it is best employed for equipment where the principal
failure modes are well identified and where these can be attributed to
quantitative levels of a few specific measurements.
For the case of a minimum acceptable parameter value, the Design

Margin Approach computes reliability as follows:
x i = the measurement taken for the iih system in the test

group for the test parameter to be evaluated.
L = the lower acceptance limit specified for the test param-
eter in question.
D = the design margin.
m = the mean of the x i values for the given test parameter

of the given group of systems.
s = the standard deviation of the x i values for the given
test parameter for the given group of systems.
d = the non-dimensional design margin.
n = the sample size - i.e. number of systems under test.
The design margin is defined to be:
D = m - L
The non-dimensional design margin is defined to be:
d = D/s
5-29
If it is assumed that the x i values are normally distributed, then

the probability that x i will fall more than D below the mean is
equivalent to the probability that the standardized normal variate
will fall above the number representing the non-dimensional design
margin. The probability (P(d)) that the standardized normal variate
will fall above d can be found in any table of the cumulative
distribution. Once the P(d) ,:ilue is known, the value of the
relibility (R) of the system can be calculated from:
R = 1 - P(c)
The following sample calculation is included to clarify this calcula-

tion. Suppose a test parameter of a group of systems has a D value of
3.6 units and a d value of 1.68. Then the P(d) value can be found in
any table of the cumulative normal probability distribution as P(d =
1.68) = 0.046. Then the reliability of the system is calculated to be
0.954.
For the case of a maximum acceptable parameter value, the design mar-
gin would be calculated using the upper limit value (U) in place of
the lower limit value (L). The remaining calculations would then pro-
ceed in the same manner. For the case of a two sided limit on a pa-
rameter value, both the upper and lower design margins would be calcu-
lated as described. The reliability would then be calculated as
follows:
R = 1 - P(dL) - P(d U )
where P(dL) and P(dU ) are the standardized normal variate

probabilities for the lower and upper non-dimensional design margin
values.
The design margin approach produces different R values than the con-
ventional approach to calculating R. In particular, in the design
margin approach, the computed reliability does not necessarily change
if an additional system or measurement is added to the data. Addi-
tional information concerning the estimation of reliability confidence
intervals and the estimation of differences using the Design Margin
Approach is presented in Reference 17.
5.4.2 Stockpile Flight Test Programs
Stockpile flight test programs are primarily applied to one-shot sys-

tems such as missiles or munitions. Under these programs, samples of
fielded systems are functioned (e.g. fired, launched, detonated, etc.)
to determine the degree to which they actually perform in accordance
with specified requirements for reliability, accuracy, range, etc.
This information is then used to assess the degree to which the re-
5-30
maining items in the stockpile meet the specified performance require-

ments and to determine if any orrective action Is necessary to im-
prove performance to desired levels.
For the most part, the complexity of the flight test program will be a
function of the complexity of the item being tested. In the case of a
very simple item, such as a projectile fuze, the test program may
simply consist of setting the fuze, firing it, and then observing if
the test projectile detonated as required. For a very complex item,
such as a ballistic missile, the test program may entail the monitor-
ing of a vast array of telemetry data during the flight test supplied
by special telemetry equipment installed prior to launch. Such infor-
mation is vital to determine if the missile performed properly
throughout the test and to isolate the cause in the event of a failure.
The cost of the test item will also impact the number of items that
can be tested. For relatively inexpensive items, the sample test
plans specified in MIL-STD-105 (Ref. 18) can be used to determine the
sample size and accept/reject criteria for a particular level of con-
fidence. For expensive items, Reference 19 provides a Bayesian based
sampling method for reducing the number of sample& necessary to
achieve a desired confidence level.
5.4.3 Real-Time Component Aging Programs
Stockpile laboratory and flight test programs can be expensive to per-

form. In addition to the direct costs associated with performing the
tests, the logistic costs associated with the movement of entire sys-
tems or parts of systems can be substantial. If the testing results
in the destruction of the item being evaluated, its cost must also be
considered.
Real-time component aging has been employed as a means of reducing the

costs associated with the evaluation of the long-term dormant storage
deterioration of fielded systems. The objective of such programs is
to perform continual automated testing of critical devices of a sys-
tem, rather than the systems themselves, at a central location. In
this way the test procedures can be greatly simplified and the test
results can be potentially applied to a variety of different systems.
(Ref. 16)
The test devices are chosen at random from production lots and are
mounted in special test fixtures. The test fixtures are then mounted
in an automated test system consisting of computer controlled measur-
ing instruments. The computer is programmed to periodically test each
component at prescribed intervals and store all measured values for
later analysis for trends.
5-31
When conducting real-time aging programs, it must be realized that the

deterioration rates being measured may be quite small while the random
variations can be quite large. Many random perturbations of the data
can exist. Besides the effects of temperature and humidity, there are
the effects of ac line voltage variations, amplifier drift, and envi-
ronmental electrical equipment transients. One method of coping with
this problem is to examine the mean behavior of all devices within a
device type, rather than particular devices individually. Such aver-
aging will tend to filter random perturbations. (Ref. 16)
The massive amounts of data that must be processed must also be con-
sidered. The data analysis program must be able to manipulate the
data file created by the data measuring system and must be able to
handle the amount of data collected over the projected program time
span.
In order to determine if the devices exhibit a wearout condition due

to the repetitive testing, it is recommended that the test devices be
divided into three groups, each having a different testing rate (such
as differing by a factor of 2). To determine if a correlation exists
betueen testing rate and device deterioration, the percentage of meas-
urements of a group beyond the mean as a function of time should be
examined. For example, if t.est results indicate that a particular
test parameter has increased with time, then, if this is a consequence
of testing and not storage, the group that is tested most often would
increase most rapidly. Since the mean computations make no group dis-
tinctions, the percentage of measurements above the mean for the most
frequently tested group should increase with time. Conversely, the
percentage of measurements above the mean for the least frequently
tested group should decrease with time. The remaining group should
lie somewhere between these two groups. The opposite si uation would
occur if the parameter decreased with time and it exhibited a wearout
condition due to the repetitive testing.
The primary disadvantage of this method is that the environment exper-

ienced by the test components is, in most cases, not the same as the
environment experienced by the fielded systems which are being evalu-
ated. Even if environmental parameters such as temperature and humid-
ity closely approximate the conditions in the field, significant dif-
ferences in other environmental factors may yield results with low
correlation to actual field experience. For example, a corrosive
environment may result in the field due to the outgassing of certain
components in the fielded system. The effects of this corrosive en-
vironment would not be reflected in the component aging program unless
it was known beforehand and was duplicated in the aging program.
Since this is unlikely, it is recommended that real-time component
aging not be relied upon as the sole method for evaluating the effects
of long-term storage on fielded systems.
5-32
5.5 REfERENCE
1. From material provided by R. E. Dehm, RCA/ASD, PO Box 588,

Burlington, Mass., 01803, 1986.
2. "Impact of Nonoperating Periods on Equipment Reliability", David

W. Coit and Mary G. Priore, ITT Research Institute,
RADC-TR-85-91, May 1985.
3. MIL-STD-785, Reliability Program for Systems and Equipment Devel-

opment and Production, Rev. B, Notice 1, 3 July 1986.
4. MIL-HDBK-217, Reliability Prediction of Electronic Equipment,

Rev. E, 27 October 1986.
5. "Storage Reliability Of Missile Materiel Program - Missile

Materiel Reliability Prediction Handbook Parts Count Prediction,
Dennis F. Malik and Joe C. Mitchell, U.S. Army Missile Research
and Development Command, Product Assurance Directorate, Redstone
Arsenal, Alabama, Report No. LC-78-1, February 1978.
6. "An Approach for Assessing Missile System Dormant Reliability",

R. D. Trapp, et al, BDM Corporation, Alburquerque, N.M., NTIS
Report No. AD-A107519, 30 January 1981.
7. "Observed MTBF as a Function of the Usage Profile", Edward Demko,

Singer Kearfott Division, Little Falls, N.J., 1983 Proceedings
Annual Reliability and Maintainability Symposium, pp. 262-267.
8. "A Review of Equipment Aging Theory and Technology", S. P.

Carfagno and R. J. Gibson, Electric Power Research Institute,
Palo Alto, California, EPRI NP-1558, Project 890-1, Final Report,
September 1980.
9. "Proposed Research on Class I Components to Test a General Ap-

proach to Accelerated Aging Under Combined Stress Environments",
K. T. Gillen, et al, Sandia Laboratories, Alburquerque, N.M.,
Sandia Report No. SAND 76-0715, April 1977.
10. "The Analysis of Data from Accelerated Stress Tests", D. S. Peck,

Bell Telephone Laboratories, Allentown, Pennsylvania, Proceedings
9th Annual Reliability Physics Symposium, 1971, pp. 68-83.
11. "Environmental Stress Screening Guidelines for Assemblies", The

Institute of Environmental Sciences, 1984.
5-33
12. "Stress Screening of Electronic Hardware", A. E. Saari, R. E.

Schafer, and S. J. VanDenBerg, Hughes Aircraft Company, RADC-TR-
82-87, May 19S2.
13. "A Consumer Guide to ESS: Making Sense Out of Environmental

Stress Screening Standards and Requirements", Henry Caruso, Tech-
nical Editor, The Journal of Environmental Sciences, Jan./Feb.
1987, pp. 13-17.
14. "A General Model for Assessing Stockpiled Weapon System Readiness
and Maintainability", Robert M. Seman, Lockheed Electronics Com-
pany, Inc., Fifth Annual Lockheed Logistics R&D Symposium
Proceedings, May 1985.
15. "Improved HAWK Certified Round Mid-Life Status Report", Homer L.

Hastings and Billy R. Sparks, U.S. Army Missile Command, Redstone
Arsenal, Alabama, DTIC Report No. AD-A134145, 16 June 1980.
16. "Long Term Dormant Storage Testing, Initial Results", Thaddeus J.

Kobylarz and Albert J. Graf, Proceedings 1976 Annual Reliability
and Maintainability Symposium, pp. 176-181.
17. "Interpretation of Test Data by Design Margin Techniques", H.

Hecht, Aerospace Corp., Annals of Reliability and Maintain-
ability, Vol. 9, June 1970, pp. 107-111.
18. MIL-STD-105, Sampling Procedures and Tables for Inspection by At-

tributes, Rev. D, Notice 2, 20 March 1964.
19. "Planning Guide for Demonstration and Assessment of Reliability

and Durability", U.S. Army Materiel Command, AMCP 702-5, DTIC Re-
port No. AD 710215, February 1969.
5-34
6.0 PRODUCT PERFORMANCE AGREEMENTS
6.1 PRODUCT PERFORMANCE AGREEMENT OBJECTIVES
Product performance agreements (PPAs) are any form of warranty guaran-

tee, or incentive used as a motivational strategy to improve product
performance and support (Reference 1). They are required to be consi-
dered for weapon system production contracts by title 10 section 2403
of the United States Code on all weapon systems purchased after 1
January 1985 costing more than $100K per deliverable end item, or more
than $10M for the total procurement and must guarantee:
- conformance to essential performance requirements,
- conformance to design and manufacturing requirements,
- freedom from defects in materials and workmanship.
This requirement applies to weapon systems in mature full scale pro-

duction. The essential performance requirements coverage is excluded
from the first 10% of the total production quantity, or the first
year's production quantity, whichever is less. In the event that it
is determined that the PPA requirement is not in the interests of na-
tional defense or is not cost effective, the Secretary of Defense may
waive the PPA requirement with proper notification to the House and
Senate committees on Armed Services and Appropriations.
The objective of a PPA is to enforce the performance and (when re-

quired) the reliability and maintainability (R&M) specifications in
order to deliver an effective and available weapon system to the Com-
bat Commander. It seeks to accomplish this by providing the system
manufacturer with a suitable incentive, positive or negative, to as-
sume increased responsibility for post delivery performance.
An optional objective of some PPAs is tu create a consistent pattern

of R&M growth during development and deployment. The target is im-
proved availability during system/item deployment at considerable sav-
ings in Life Cycle Cost (LCC). The trend of Operation & Maintenance
(O&M) costs is an ever increasing share of total system cost. Per
unit cost of a weapon system is not the sole criteria of affordabil-
ity. Total LCC, of wnich O&M is over 50%, determine affordability.
A low reliability and/or difficult to maintain, but moderately priced,
item may not be affordable when the O&M costs are considered over the
expected system life. (Reference 2.)
6-1
System R&M is a key concern because of its impact on mission capabil-

ity. The price of modern weapon systems precludes the excessive O&M
budget required to insure desired mission capability with less than
desired system/item R&M. The maintenance overtime, system cross can-
nibalization, increased purchases of the materiel to effect repairs,
and the increased transportation and administrative costs are re-
flected in the O&M budge. (Reference 2.)
Lacking the protection of a PPA, the Government was forced to absorb

reliability shortfalls which reduced mission capability, and/or
increased support costs. A PPA places greater responsibility for
system field performance on the manufacturer. (Reference 2.)
It is difficult to conclusively prove that reliability measures ob-

served on a program errloying a PPA are the result of the PPA or
whether the benefits obtained by the Government through the PPA exceed
the cost of the PPA. One study (Reference 3) concluded that the
search of warranty experience failed to reveal any overwhelming evi-
dence to support the "success of warranties". It further concluded
that while some benefits have occured in the contracts containing war-
ranties, the separation of the benefits attributable to warranties
vis-a-vis benefits attributable to other factors is unclear. Another
study (Reference 4) investigating the effectiveness of Reliability
Improvement Warranties (RIWs) concluded that the RIW was modcrately
effective for five of eight programs examined. The benefits realized
included the temporary transfer of the risk associated with equipment
field reliability to the production contractor(s), deferred Government
investment of resources to maintain the fielded equipment, and im-
proved ieliability and maintainability over that which the Government
could reasonably have expected to achieve without a warranty.
The effectiveness of PPAs applied to dormant equipment is even less

clear. Reference 3, investigating the use of warranties in the acqui-
sition of conventional munitions, concluded that, because of the
unique features of munitions (one-shot, no maintenance, long dormant
periods, no download provision, all up round concept, etc.), warran-
ties do not provide incentives to contractors for improved product
performance or reduced costs. It found that 3 of the 10 major war-
ranty application criteria and 7 of the 14 secondary warranty appli-
cation criteria could not be met. While reference 3 does not advocate
the use of warranties as a general practice, it does conclude vhAt the
original objective of the product performance agreement to nblt-air. what
was contracted for at the least life cycle cost is still vaijd.
6-2
Product performance agreements are a relatively new phenomenon and

innovative work is being put into these agreements. To be effective,
they should be simple, focused on the primary objectives to be ob-
tained, enforceable, affordable in relation to potential benefits, and
should not cause disruption to existing military systems and proce-
dures. (Reference 1) A correctly structured PPA must be tailored to
the type of procurement, the characteristics of the equipment and mis-
sion, and the anticipated support environment. In many cases, a PPA
may not be applicable or may even be a risky alternative. (Reference
3.)
The following sections present an approach for selecting, analyzing,

structuring, and managing effective proC'mt ierformance agreements in
general with specific guidance where appropriate for applying them to
dormant equipment.
6.2 PRODUCT PERFORMANCE AGREEMENT SELECTION, ANALYSIS, AND STRUCTURE
Creating a Product Performance Agreement (PPA) that is suited to

program requirements and objectives can be divided into five steps.
1. Establishment of a PPA working group.
2. Collection of program characteristics, objective, and equipment

information.
3. Identification of alternative PPAs which best suit the program

parameters.
4. Performance of in-depth cost, benefit, and risk analyses of the

alternatives.
Development of program unique contract provisions, evaluation

methods, and management techniques for the selected
alternative(s).
These steps are described in the following paragraphs. Additional

information concerning these steps can also be found in Reference 2,
the Air Force Product Performance Agreement Guide, and Reference 5,
the Air Force Product Performance Agreement Decision Support Handbook.
6.2.1 Establishment of PPA Working Group
The first step in developing a PPA approach to the acquisition of a

system is the establishmeat of a working group to examine approaches
and develop a PPA implementation strategy. This working group should
be established as early as possible in the -cquisition life cycle.
b-3
PPA concepts must be developed early in the program, along with the
program planning, if they are to be consistent with the overall
program acquisition strategy.
Selecting and tailoring an effective PPA requires a multidisciplinary

approach with participation of the using commands, the acquisition and
supporting commands, and sbbsequently the contractor. The complexity
and potential impact of PPAs on logistics, cost, design trade-offs,
etc., requires a team approach from the outset. Government partici-
pants are generally the same as those needed in developing the acqui-
sition strategy and should include representatives from the following
functional elements: (References 1 & 8.)
- Program Management
- Engineering (including quality assurance)
- Logistics
- Budget and program contrnl
- Cost, pricing, and systems analysis
- Procurement
- Counsel
- Users
The using and supporting commands can offer valuable advice about the
FFA approach to be considered. They are perhaps the most important
members since they are the ones who will have to live with the sys-
tem. (Reference 6.)
The contractor should be included in the working group as early s

possible in the acquisition life cycle. This is very important as it
was concluded in a study of the impact of warranties and guarantees on
electronic system design and development progr.xms (Reference 7) that
delaying contractor invovlement in the development of the detailed
terms and conditions of the PPA until negotiation of the production
contract was found to inhibit important design trade-offs during the
engineering development program. The contractor working group should
include representativeL from the same functional elements as the
Government working group with the exception of the ultimate Users.
One method of accomplishing close contractor involvement is to place

sample PPA provisions in a demonstration and validation RFP to put the
contractor on notice that a PPA will probably be required on the pro-
duction contract (Reference 1). The results from the demonstration
and validation should then be used to tailor PPA requirements. Then
when contractors are required to propose upon government developed
clauses at the time of full-scale development (FSD), or later, they
should also be allowed to propose alternatives to be evaluated by the
government.
6-4
6.2.2 Data Collection
The next step in the development of a PPA implementation strategy is

the collection of data concerning program characteristics, program
objectives, and equipment description and characteristics. This data
is necessary to select alternative PPAs, identify and evaluate poten-
tial PPA implementation problems and concerns, and analyze relevant
PPAs.
Table 6.2.2-1, developed from Reference 5, can be used as an aid in

collecting program characteristic data. It identifies the pertinent
program characteristics and the data required to support the selection
of PPAs.
In addition to the program characteristics data, information must be

collected on the major objectives of the progrem, the details of the
objectives in terms of the program and the specific equipment, and
what their priorities are: (Reference 5.)
" Reducing or imposing a ceiling on costs:

- operations and support costs
- life-cycle cost
- particular maintenance cost
- parts cost, sparing cost, etc.
" Giving incentive to contractors to improve specific aspects of

existing systems:
- operating characteristics
- maintainability
- reliability
- performance in operational environment
- performance in storage (non-operating) environment
" Giv.ng incentive to contractors to meet or exceed system perfor-

mance spocifications:
- reliability (operating, non-operating)
- reliability growth
- mission completion
- mission deppndability
- captive carry
- specific performance task (e.g., satellite location change)
- lifetime of a consumable component (e.g., brakes)
" Giving incentive to contractors to meet or exceed system mainte-

nance specifications
" Basic materials, workmanship, correction of deficiencies coverage
6-5
Table 6.2.2-1. Program Characteristic Data for PPA Selection
Program Characteristic Data Required
1. Multiservice Program * Is the program multiservice?
* Are there differences in objectives,

operational environments, and/or
physical characteristics of the sys-
tem between the services (may neces-
sitate different PPAs for each
service).
2. Program Status * Present status of the program

(source selection, production
FSED, etc.)
& The phase at which the PPA will be

applied.
3. Sole Source Acquisition * Is the acquisition sole source?
* Contractual arrangement with mul-

tiple suppliers (i.e., leader/ fol-
lower, prime with subcontractors,
etc.).
4. Budgetary Issues * Budget Restrictions
* Funding Sources
5. Currently Used PPAs * Are there existing PPAs on the

program?
e Whdt do they cover and for what

period of time?
6. Previously Used PPAs * Types of PPAs used on the program in

the past.
e Experience and results of those PPAs.
7. Program Schedule * The program schedule including all

phases of development, testing, pro-
duction, and operation.
Table 6.2.2-1. Program Characteristic Data for PPA

Selection (Continued)
8. Interim Contractor Support e Length of time and level of support.
9. Program Test Plans * What do the test plans include

(e.g., reliability, maintainability,
performance, etc.)?
e When will the tests be dcne?
* The number of units tested and for

how lo;ag.
* The test environment.
* Is there time and opportunity avail-

able after the tests to modify the
equipment design?
10 Number of Years for * The number of years of program

for the PPA coverage by a PPA.
6-7
" Reducing risk to the government of acquiring substandard items
" Ensuring that the maintenance concept is feasible and realistic:
- BIT requirements
- MTTR parameters
- Support equipment requirements
- Turnaround time requirements
" Giving incentive to achieve or exceed availability/readiness

requirements.
It must also be determined whether the objectives are relevant to:
" The system of unit as a whole
" Specific Line Replaceable Units (LRUs) only
" Specific Ship Replaceable Units (SRUs) only.
Finally, information concerning the equipment functional, operational,

and physical characteristics must be collected. The type of informa-
tion that must be obtained is identified in Table 6.2.2-2. This table
was developed from information provided in Reference 5.
6.2.3 Alternative PPA Selection
The data collected in the previous step is used to determine preli-

minary PPA options for the relevant program using a PPA Selection
Matrix (PSM) such as 3hown in figure 6.2.3-1. This matrix is used to
match program objectives and characteristics with suitable PPAs, and
to generate a preliminary list of suitable PPA options. It is divided
into two parts with program agreement objectives as the rows of Part
I, item characteristics as the rows of Part II, and PPA types as the
column headings. An expanded PSM containing other PPAs is maintained
in the PPA Decision Support System (DSS) at the Air Force Product Per-
formance Agreement Center, Wright-Patterson AFB, Ohio. (Reference 5.)
The four symbols used in the matrix and their meanings are:
" Include this PPA in the list of potential PPAs.
"4" Exclude this PPA from the list of potential PPAs.
"4" Indicates PPA is potentially relevant.
"-" Indicates PPA has possible negative impact to stated program

objective or characteristic.
6-8
Table 6.2.2-2. Equipment Description and Characteristic Data

for PPA Selection
1. Equipment Name * Relevant system, subsystem, or component

name.
2. Parts Breakdown * Equipment parts breakdown into LRU's and

SRU's.
3. Equipment Function 9 The intended function of the equipment.
4. Maturity * If it is existing equipment, how has it

previously performed, what were perform-
ance problems. (This will help determine
what should be incentivized.)
* If it is modified equipment, what are the

modifications, how much do they contri-
bute to the performance of the equipment,
and what are the performance problems of
the original equipment and are they
addressed by the modifications. (This
will help determine if the modifications
affect performance, whether they alone
should be warranted, and whether the
modified system is expected to perform as
the original system did.)
e If it is new egipment, does it include

advanced technology or mature
technology. (The former is high risk in
terms of expected performance, the latter
is medium to low risk depending on the
extent of the maturity and experience
with the technology).
. Government Furnished * Identification of GFE, non-GFE and modi-

Equipment field GFE components. (The contractor
cannot be responsible for performance/
design limitations of GFE.)
6. Production Quantity * The number of units to be produced.
7. Life o The lifetime of a unit.
6-9
Table 6.2.2-2. Equipment Description and Characteristic Data

for PPA Selection (Continued)
Program Characteristic Data Reguired
8. Average Operating * The average operating hours of a unit (per

Hours week, per month, etc.)
9. Life Cycle Environment * Operation, storage, captive carry, etc.
10. Time Usage e Dormant system, one-shot (e.g., missile),

continuous use, fixed or variable use.
11. Maintenance Concept * Wooden round, number of levels, etc.
12. High Risk/Failure * Do they limit performance?

Rate Items
* Are they LRUs or SRUs?
* Do they account for the majority of the

performance risk of the equipment?
13. Performance . Reliability

Specifications
* Maintainability
o Functional
o Design
14. Historical Data o Test results
e Operational data
* Historical data from similar equipment
15. Built-In-Test e Description
* Requirements
16. Software * Description
* Characteristics
17. Relation to Other * Is the equipment tightly integrated with

Equipment other mission critical/significant equip-
ment or does it operate in a self-con-
tained manner?
6-10
w +
-z
1)1 L&J
+ U
2f 0.L
I- ~+ N
4-1
+ +
-LJ
4 4 I U
LLL
+ + .4 +
co
4m LO '-A
uJ U L orm
06 0 0 j
06 0 UcI-
L. V) . LII ix 4
- : +0 CI IA .J C
oL UI.. 4.o
C, 10 A(L kb- C' V) L) 04) L
ow 0). m wn 4 4x 0 M0)
M L. 0. > In x VI-i 0)4 0
-
0-C. 0 .
(il~~
0 In
>..
n4 ~Q.4.
4). .0
.4. CD (D
- ..... .' (A .1 n 6-101 0 ~

CL
zz
C- Ly
< oz
U<
-
CD
M n
-jV -
(A
+ + + + +4.
(AA
r- a)0
0 In :
04 + 0 . L. 06
0 C) Q 0
o= 0v.- do
(A ac L-0 G L .
6-1-
PPAs with no symbol (blanks) beside the objective or characteristic

should be overlooked.
To use the matrix, it is necessary to first identify the agreement

objectives and equipment characteristics relevant to the program using
the information collected in the previous step. Then, beginning with
the rows of Part I, those agreement objectives which fit the program
under examination are selected. The row of each selected objective is
examined to find which PPAs are suitable to the program (denoted as
"= in the matrix) or which may be potentially relevant (" ). For
example, if a program has the objective of achieving a dormant system
availability goal, then reading across that row shows that Storage
Verification Guarantee, Availability Guarantee, and Incentive Award
are options for the program.
After exercising Part I of the PSM to generate the list of potential

PPAs, it is necessary to validate this list by executing Part II of
the PSM. The second part of the PSM contains matrix diagnostics which
are intended to eliminate those preliminary PPA choices that are not
appropriate for the equipment and/or program characteristics given and
to provide insight into the possible positive or negative effects
these characteristics may have on the use of the PPA for the program.
To exercise Part II, the row corresponding to each equipment charac-

teristic of the program is examined. When reading across the row,
only the matrix elements in the columns of the PPAs selected in Part I
of the matrix are relevant. For the above example, only the matrix
elements in the Storage Verification Guarantee, Availability Guaran-
tee, and Incentive Award columns are relevant. Each matrix diagnostic
in the selected rows and columns is then checked to see whether the
PPA should be excluded ("i") or whether the equipment characteristic
will have a positive ("+") or negative ("-") effect on the use of the
PPA.
Continuing the above example, if the equipment under consideration had

a CPFF contract type, reading across the CPFF Contract Type row shows
a "A" in the Storage Verification and Availability guarantee columns
which indicates that these PPAs should be excluded whereas the blank
in the Incentive Award column indicates ambivalence towards this PPA
option. Specific interpretations for each diagnostic element in the
matrix can be found in Reference 5 and can also be obtained from the
Air Force Product Performance Agreement Center, Wright-Patterson AFB,
Ohio.
To complete the preliminary PPA selection process it is also necessary

to consider other PPA types that are not included in the PSM main-
tained in the PPA DSS. Figure 6.2.3-2 contains an expanded list of
Correction of Deficiencies (COD) and Inspection Group
* Inspection Clause
* Inspection of Supplies and Correction of Deficiencies Clause
* Correction of Deficiencies Clause
* Total System Performance Responsibility Guarantee
* Warranty of Supplies
* Warranty of Technical Data
* Rewarranty of Repaired/Overhauled Equipment
* Material and Workmanship Warranty
* Flying Hour Design Warranty
Reliability Group
* Reliability Improvement Warranty

* RIW with Mean Time Between Failure (MTBF) Guarantee
* MTBF Verification Test
* Chronic Line Replaceable Unit Guarantee
* Component Reliability Guarantee
0 Reliability Guarantee
* Logistics Support Cost Guarantee
* Utility Function Warranty
Cost Ceiling Group
0 Commercial Service Life Guarantee

* Logistics Support Cost Guarantee
* Maximum Parts Cost Guarantee
a Ultimate Life Warranty
Basic Maintenance Group
* Commercial Service Life Guarantee

* Repair/Exchange Agreement
Figure 6.2.3-2. PPA Types Grouped by Major Objective.
6-14
DorLmantSstems Performance Group
* * Availability Guarantee
a * Captive Carry Guarantee
a * Mission Dependability Guarantee
* * Storage Verification Guarantee
Availability/Readiness Group
* * Availability Guarantee
Parameter/Performance Goal Incentives Group
* * Incentive Award Payments
• Asterisk Designates PPA Applicable to Dormant Systems.
Figure 6.2.3-2. PPA Types Grouped by Major Objective. (Continued)
6-15
PPA types characterized by major objective of the agreement with those

applicable to dormant systems indicated with an asterisk (*).
To evaluate the applicability of the other PPA types, Reference 2

should be consulted. Appendix E to Reference 2 presents a brief over-
view of each PPA type including the following:
- PPA Objective
- Characteristics
- Applicability
- Description
- Measurement
- Result
- Advantages
- Disadvantages
By comparing the overviews with the information collected in the pre-

vious step, the applicability of a PPA type can be determined. Addi-
tional information from other sources such as the Air Force Product
Performance Agreement Center, Wright-Patterson AFB, Ohio, should be
consulted to aid in the selection of other PPA types.
After the initial set of PPA options for the program has been deter-
vnined, the content of the selected PPAs must be checked for compati-
bility with the program before beginning the quantitative analysis.
Primary aspects of the PPAs which should be closely reviewed include:
(Reference 5.)
Maintenance Concept - Determine whether the PPA type is compatible

with the program maintenance concept or does it require a different
strdtegy. Determine whether it is possible to change the program to
accommodate this different strategy or if it is possible to revise the
PPA to accommodate the program maintenance strategy.
Data Collection Requirements - Identify the general data which is re-

quired to be measured and collected during the warranty/guarantee
period and determine whether it is feasible to collect this data in
terms of the program's logistic support strategy. Is a secific type
of data collection system required for the PPA and if so does this
exist in the DoD or will it have to be developed? (The data collec-
tion is one of the most important aspects of PPA implementation and
must be thoroughly examined for feasibility within the program. Al-
ternatives for implementing data collection systems, whether they are
compatible with the time schedule of the program, and what the cost of
the system is, must be examined.)
6-16
Testing Requiremt_ i - The PPA may include verification testing as

part of the provisions. The testing requirements mu.,t be reviewed to
ensuro that they are feasible with respect to the program or if
changes to program or PPA scheduling are necessary.
For each of the above aspects, the requirements of the PPA must be
reviewed with respect to the program to:
* En-ure that the PPA is consistent with program activities and

requiremeuts.
" Identify requirements for implementing the PPA which must be pro-
vided by the program (either government or contractor).
" Identify alternatives for meeting these requirements (or determine

whether it is impossible to meet them thereby rendering the PPA
type infeasible for the program).
The list of PPX options resulting for this effort should then oe
quantitatively analyzed as described in the next section.
6.2.4 Analysis of Alternatives
The major portion of the quantitative analysis of a PPA centers on

cost because the cost impact of a PPA is of major importance to both
the Government and the contractor and because major benefits and risks
can be stated in terms of cost. To perform the analysis it is useful
to use a Cost Breakdown Structure (see Figure 6.2.4.1-1) and cost
model as an overall framework. Such a framework will assist in iden-
tifying the relevant cost elements of a PPA and will show how to cal-
culate these cost to perform the required analysis among the
alternative PPA options. (Reference 5.)
For specific PPA types, certain other analytic technique are required
to calculate its costs, benefits, and risks. For example, quantita-
tive analysis may involve non-cost measures such as availability for
an Availability Guarantee or mission completion for a Mission Depend-
ability Guarantee.
Since it is always the case that there are at lea3t two alternatives
to be examined (e.g., PPA vs no-PPA), it is important to structure the
analysis to make the selection among competing alternatives an inte-
gral part of it. The following procedure can be used to accomplish
this:
* For each PPA option, identify the relevant cost elements which are
affected by use of the PPA using the Cost Breakdown Structure for
PPA analysis shown in Figure 6.2.4.1-1.
6-17
# COST ELEMENT AG CCG SVG MDG IA
RESEARCH AND DEVELOPMENT
1. Validation
1.1 Government
1.2 Contractor
2. Full Scale Development
2.1 Government
2.2 Contractor
INVESTMENT
4. Government Program Management X X X X X

5. Prime Equipment Acquisition
5.1 Production Hardware
5.2 Production Spt. and Svc.
5.3 Production Test and Eval.
5.4 Initial Transport
6. System Installation
6.1 System Provisioning/Modif.
6.2 Installation Kit Spares
7. Support Equipment
7.1 Hardware Acquisition
8. Initial Support Acquisition
8.1 Initial Spares (Hardware) X
8.2 Software Investment
8.3 New Item Entry
8.4 Facilities
8.5 Documentation
8.6 Initial Training
9. Product Performance Agreement X X X X X
OPERATING AND SUPPORT
10. Corrective Maintenance X X X X X

10.1 Below-Depot Level X X X X X
10.3 Depot Level X X X X X
11. Scheduled Maintenance X
11.1 Labor X
11.2 Material X
12. Packaging and Shipping
Figure 6.2.4.1-1. Cost Breakdown Structure for PPA Analysis
6-18
# COST ELEMENT AG CCG SVG MDG IA
OPERAING AND SUPPORT (Continued)
13. Software Maintenance X X X X X

14. Inventory Storage
15. SE Maintenance
16. Docwentation Maintenance
17. Sustaining Supply Support X X
17.1 Replenishment Spares X X
17.2 Supply System Management X X
18. Recuriing Training
Leaiend:
AG: Availability Guarantee MDG: Mission Dependability

CCG: Captive Carry Guarantee Guarantee
SVG: Storage Verification Guarantee IA: Incentive Award
Figure 6.2.4.1-1. Cost Breakdown Structure for PPA Analysis.

(Continued)
6-19
" Identify the analysis methodology and data requirements for the
PPA types under consideration.
" Perform the requi:ed analyses to obtain the life-cycle costs,

benefits, and risks of the PPA option including the no-PPA
alternative.
" Compare the LCC estimates, savings/benefits estimates, and risk

estimates of each of the PPA options.
" Assess the results and select PPA options for implementation.
An overview of these steps is presented in the following sections.

Further details can be found in References 2 and 5.
6.2.4.1 Identification of Relevant Cost Elements
In comparing two alternatives, it is not enough to look at acquisition

costs or Operating and Support (O&S) Costs for the first year or first
few years. It is necessary to consider the costs of each alternative
over the life of the system. Life cycle costing (LCC) is the tool
most often used for comparisons over a systems life. It is used as a
means to structure the analysis of the various PPAs by identifying tho
relevant cost elements of the PPA and by serving as a valid basis of
comparison.
Not all cost elements need to be considered in the analysis. Only

those costs that are directly impacted by the application of a PPA
should be included. Those costs that would be the same regardless of
which PPA option were implemented are not of importance to the PPA
analysis, and should not be considered.
To identify the relevant cost elements for each PPA type, a common
Cost Breakdown Structure (CBS) must be used. Figure 6.2.4.1-1 gives a
Cost Breakdown Structure that includes all of the cost elements used
in most LCC analyses and indicates the subset of cost elements typi-
cally relevant to the PPA types applicable to dormant systems identi-
field in Figure 6.2.3-2. The cost elements (row names) indicated with
3n "X" in the column headed by the PPA name are those that are rele-
vant to the analysis of the cost of the PPA. Those elements without
an "X" are considered to be non-relevant (i.e., not impacted by the
use of the PPA).
6.2.4.2 Identification of Analysis MethodolQcgy-and Data Re uirements
Figures 6.2.4.2-1 through 6.2.4.2-5 summarize the analysis methodolo-

gies for each of the PPA types identified in Figure 6.2.3-2 as applic-
able to dormant systems. The figures are organized as follows:
6-10
AVAILABILITY GUARANTEE
Overview
The Availability Guarantee (AG) is applicable to dormant systems and

is used to motivate the contractor to design and produce items with
high operational readiness up&. random demand. lt normally applies to
equipment where high reliability is critical and corrective main-
tenance is impractical or extremely costly. Availability is based
upon random sampling checks from dormant storage over an extended
period of time. Corrective action is taken as necessary to raise the
level of availability when deficiencies are discovered. This PPA is
not effective unless the equipment can provide positive indication of
operability through random "go/no-go" checks.
Analysis Procedure
The analysis steps for the analysis required to evaluate the AG are:
* Identify the cost elements associated with the AG

* Calculate the LCC
* Compare this LCC to the other PPA options under consideration.
Data Requirements
The data requirements for the LCC analysis are those elements identi-
fied in the Cost Breakdown Structure in Figure 6.2.4.1-1 with an "X".
This includes the PPA price which includes:
* "Go/No-Go" Testing Costs

* Data Collection/Evaluation Costs
* Administrative/Management Costs
* Corrective Action Costs
* Contractor Fee.
Figure 6.2.4.2-1. Availability Guarantee Analysis Procedure

and Data Requirements Summary
6-21
CAPTIVE CARRY GUARANTEE
Overview
The Captive Carry Guarantee (CCG) is applicable to dormant systems

periodically deployed in a captive carry mode and is used to motivate
the contractor to design and produce items meeting specified captive
carry specification requiements. Correctie action is taken as ne-
cessary to meet the specified performance levels when deficiencies are
discovered. This PPA is not effective unless combined with a pro-
cedure for verification of the specification requirements that have to
be met.
Analysis Procedure
Same as Figure 6.2.4.2-1.
Data Requirements
e Specification verification costs

o Data collection/evaluation costs
o Administrative/management costs
o Corrective action costs
e Contractor fee.
Figure 6.2.4.2-2. Captive Carry Guarantee Analysis Procedure

and Data Requirements Summary
6-22
STORAGE VERIFICATION GUARANTEE
Overview
The Storage Verification Guarantee (SVG) is applicable to dormant sys-

tems and is used to motivate the contractor to design an prod..ce items
with high storage relianility. The guarantee involves testing speci-
fied components of the item after a given storage period. Data from
the test is used to compute a storage reliability parameter and com-
pared to the guaranteed parameter level. Corrective action is taken
as necessary to meet the guaranteed parameter level when deticiencies
are discovered. The SVG PPA is not effective unless it is used in
conjunction with a verification test.
Analysis Procedure
Data Requirements
* Component testing costs

* Data collection/evaluation costs
* Administrative/Management Costs
" Corrective action costs
" Contractor fee.
Figure 6.2.4.2-3. Storage Verification Guarantee Analysis

Procedure and Data Requirements Simnmary.
6-23
MISSION DEPENDABILITY GUARANTEE
Overview
The Mission Dependability Guarantee (MDG) is applicable to dormant

systems and is used to motivate the contractor to design and produce
items which attain specified mission goals (e.g., number of on-target
hits, mission completion success probability etc.) where it is
impractical or impossible to access the system during the mission.
Corrective action is taken as necessary to meet the specified mission
goal when deficiencies are discovered. This PPA is not effective
unless combined with a procedure for verification of the specified
mission goals that have to be met.
Analysis Procedure
Data Requirements
* Mission goal verification costs

* Data collection/evaluation costs
* Administrative/management costs
o Corrective action costs
* Contractor fee.
Figure 6.2.4.2-4. Mission Dependability Guarantee Analysis

Procedure and Data Requirements Summary.
6-24
INCENTIVE AWARD
Overview
The Incentive Award (IA) PPA is applicable to dormant systems and is

used to motivate the contractor to design and produce items which meet
Government objectives related to cost, performance, or schedule such
as: unit cost goals, reliability goals, maintainability goals, readi-
ness goals, performance goals, etc. The agreement may provide for an
inventive payment to the contractor if goals are exceeded and/or a
negative incentive if goals are not met. If multiple incentives are
used it is necessary to ensure that they are compatible. The achieve-
ment by the contractor of one incentive goal should not be at the ex-
pense of another. To encourage beneficial tradeoffs, separate incen-
tive awards should be used for each incentive goal of a multiple in-
centive PPA. A single incentive award should not be coupled to the
achievement of multiple incentive goals.
Analysis Procedure
Once an incentive parameter is selected, it is necessary to perform

statistical analysis to determine the set of values for which the con-
tractor is neither penalized nor rewarded. This zone represents the
"business as usual" range and is based upon technical risk, previously
demonstrated capability, and desired capability. Among the most popu-
lar statistical techniques for accomplishing this are:
" Confidence Interval Analysis

" Risk Analysis
The dollar amount of the incentive payment awarded (1) depends on the
available budget for the PPA and the worth of achieving the improved
parameter levels, and (2) must be large enough for it to serve as a
true incentive to the contractor.
Data Requirements
" Confidence Interval Analysis - The minimum data required for this
analysis are high, low, and "most likely" values for the incentive
parameters, from which data ranges for the desired level of confi-
dence can be determined.
" Risk Analysis - The data required include (1) Government assigned
risk factors for the parameter to be incentivized, and (2) the
specified guaranteed parameter level.
Figure 6.2.4.2-5. Incentive Award Analysis Procedure and

Data Requirements.
6-25
" PPA Overview
" Analysis Procedure and Associated Tools
* Data Requirements
A more detailed discussion of these and other PPA types are available
in References 2 and 5 and from the Product Performance Agreement Cen-
ter, Wright-Patterson AFB, Ohio.
6.2.4.3 Performing the Reauired Analyses
After having identified the analysis methodology and data requirements

for the PPA types under consideration, the next step is to perform the
required analyses to obtain the life-cycle costs, benefits, and risks
of the PPA options including the no-PPA alternative. This section
presents an overview of PPA analysis tools useful for accomplishing
this task. The discussion of these tools is divided into the
following categories:
" Life-Cycle Costing Analysis
" Reliability Analysis
" Cost/Benefit Analysis
" Risk Analysis
A more detailed discussion vf these tools can be found in References 2

and 5.
6.2.4.3.1. Life-Cycle Cost Analysis
Life-cycle costing (LCC) is a method of calculating the total cost of

ownership of equipment considering both the acquisition costs (Re-
search and Development and Investment Phase Costs) and operating and
support (O&S) costs. Acquisition costs include costs such as initial
spares, initial training, installation, and the costs to purchase the
items. O&S costs refer to the costs of operating and supporting the
equipment such as maintenance costs, sustaining spares costs, support
facilities costs, and recurring training costs.
Life cycle costing should be employed whenever a program's life ex-

ceeds more than one year. It explicitly accounts for the total cost
of ownership of equipment through the use of discounting. Discounting
is a technique of converting individual cash flows/costs spread over a
period of time into present values which related the various flows to
6-26
one another just as though they had occurred at the same period in
time. (See Figure 6.2.4.3.1-1.) This technique yields a single LCC
estimate for each competing option on the program providing a basis
for comparing the options.
The major use of LCC in the analysis of PPA's is to assess the total
impact of implementing a PPA upon a program's cost for the life of the
equipment, and to evaluate the cost-effectiveness of the program's
logistic support system given the existence of a PPA. While the life
of the program will, in most cases, substantially exceed the time span
of the PPA, the impact of the PPA on logistics support will be felt
throughout the life of the program. LCC estimates must therefore in-
clude all years in a program's life for which relevant costs are
incurred.
In conducting an LCC analysis or establishing an LCC baseline estimate

for the various PPA options being evaluated, consideration should be
given to LCC estimates already established for the program. The PPA
LCC analysis should be consistent with established LCC estimates in
the following areas as a minimum (Reference 5):
" Cost estimating methodology
" Parameters (system life, quantities, MTBF, etc.)
" Cost breakdown structure
e Data
This will ensure a consistent set of LCC estimates.
If an LCC model is currently being used on the program, it should be

used for the analysis of PPA alternatives providing it meets the fol-
lowing criteria:
• Comprehensiveness - The model must contain a level of detail ap-

propriate for the required analyses.
" Flexibility - The model must permit tailoring to allow inclusion

of the necessary cost elements or levels of detail key to the PPA
analysis.
• Controllability - The model must permit annual parameter changes

to capture the cost impact of changes in parameters that occur
during the program life cycle.
6-27
Case Relationship
Time is not C =P *Q
being considered
Price and C = Pi * Qi" i = 1, .... N

quantity vary
with time
Inflating for C = Pi * (l+P)i * Qi, i = 1, .... N

future values
Discounting for C = Pi *Qi/(l+d)i i = 1, ... , N

present values
Discounting C = Pi * (l+p)i * Qt/(l+d)i, i = 1, .... N

inflating
dollars for
present values.
where,
C = Total Cost, Q = Quantity, p = Inflation Rate

P = Price, N = Years, d = Discount Rate
Figure 6.2.4.3.1-1. Cost Accounting Relationships in

Life-Cycle Costing. (Reference 5.)
6-28
If the current LCC model does not meet these criteria, a new model
must be chosen that does. Reference 9 describes the use of the RCA
price parametric life cycle cost model for the analysis at "PA's.
Additional LCC models that have been utilized for the analysis of PPA
alternatives are described in References 5 and 10.
6.2.4.3.2 Reliability Analvis
Reliability (R(t)) is defined as the probability that a system will

perform in an acceptable fashion, for a specified period of time (t),
when used according to specified conditions. There are several ways
to measure reliability depending upon the characteristics of the
equipment and its intended use. The most frequently used measures in
the analysis of PPA's applied to dormant systems are:
" Mean Time Between Failure (MTBF) -the number of hours between
failures
" Failure rate - the inverse of MTBF
" Success Probability (R(t)) - usually expressed as R(t) =

e - t/M T BF where e is the natural logarithm, and t is a specified
period of time.
MTBF is very important since the ability to predict MTBF accurately is

a key issue in PPA risk. (References 7 and 11.) The Council of
Defense and Space Industries Association (CODSIA), in a letter to the
Air Force in 1976, stated (Reference 11):
"Until the hardware is tested in an operational environment, with

typical operational users, and in all planned vehicles/platforms,
etc., reliability is not 'reasonably predictable.'"
Calculations derived from textbook data sources such as MIL-HBDK-217

for operating periods and RADC-TR-85-91 for non-operating periods are
useful for relative comparisons of alternative designs but they are
not accurate estimates of field reliability under actual use condi-
tions needed for PPA analysis. This is because the environmental fac-
tors used in these data sources represent an ideal state and do not
reflect all of the randomness required for the real use environment
(Reference 12).
Table 6.2.4.3.2-1 lists the data and information sources for failure
data that are typically available during full scale development, test
and evaluation, and operational program phases. The actual sources of
data will vary according to the particular requirements of the pro-
gram. For those systems in development where limited failure data
6-29
Program Phase Data Source Origin
Full Scale * Reliability Demonstration Contractor

Development Test
* Reliability Qualification
Test
* Reliability Prediction
Test and 0 Operational Test Data Government

Evaluation
* Combined Environments Contractor field
Reliability Test (CERT) reports
Operational * Maintenance Data Collecion Government

System (e.g., D056)
0 Base Level Records (e.g.,

AFTO Form 95s and 349s)
Table 6.2.4.3.2-1 Reliability Data Sources (Reference 5)
6-30
exists, the best source is to use experience on analogous fielded sys-

tems. (Reference 5.) This is especially true for systems that will
experience long periods of dormancy since life testing for dormant
periods is impractical and the correlation of accelerated test results
to actual field results is, in many cases, unclear.
An assumption of a constant failure rate, which is often made to sim-

plify reliability calculations, may not be appropriate for PPA analy-
ses. As shown in figure 6.2.4.3.2-1, this assumption is only valid in
the mid-use life of the equipment. It may not be appropriate for PPA
analysis purposes because the PPA generally covers the initial life of
the equipment which may include some quality failures, and the period
of decreasing failure rate during iufant mortality (Reference 9.).
This is especially true for dormant periods since the majority of non-
operating failures are due to latent quality and workmanship defects.
Experience on analogous fielded systems is again the best source of
information in determining the appropriate failure rate to use.
Another method using the Duane reliability growth model to predict the
change in failure rate over the life of the PPA is described in
Reference 5.
6.2.4.3.3 Cost/Benefit Analysis
Cost/Benefit analysis is a technique that employs net present value

analysis to distinguish between two competing alternatives. Net
Present Value (NPV) refers to the difference between the present value
(i.e., discounted) costs of the two alternatives (see Figure
6.2.4.3.1-1). For PPA comparative analysis, the difference between
the costs for two or more alternatives is a measure at the benefit (or
cost avoidance) derived from selecting one alternative over another.
To perform costs/benefit analysis the following procedure should be

used:
a Quantify benefits/costs in terms of dollars for each year affected

by the PPA.
* Calculate present values for the annual dollar benefits/costs for

each alternative (see Figure 6.2.4.3.1-1).
e Subtract the present value costs from the present value benefits
for each alternative. This is the NrV.
* Compare the sum of the NPV's for each alternative. The alterna-
tive with the largest positive NPV is the preferred alternative.
6-31
RATLE IN FAN T CONSTANT FAILURE WERIOO

MORTAITY RATE PERIOD PEIO
PERIOD
I
I If
0
I IM E
Figure 6.2.4.3.2-1. Example Reliability Curve
6-1~2
An example of this procedure from Reference 5 is shown in Figure

6.2.4.3.3-1.
The ratio of total present value costs to total present value benefits
for a particular alternative, called the cost/benefit ratio, is some-
times used to select among alternatives. The project with the lowest
ratio is considered the preferred alternative. The cost/benefit ratio
is a measure of how effectively each dollar of cost is utilized to
create dollar benefits and, as a result, will not necessarily lead to
the same alternative preferred by merely comparing NPV's.
Prior to making a final decision based on NPV's and cost/benefit

analysis, it is advisable to perform cost effectiveness and breakeven
analysis to evaluate those areas where the decision may be switched
from one alternative to the other. In the evaluation of the cost
effectiveness of alternatives, the balance between system effective-
ness (the ability of a system to meet both technical and logistics
requirements) and total LCC is evaluated. The desired balance is
dependent upon the weight that the program places on increments of
system effectiveness vs. increments of LCC. For example, a 10%
increase in availability that results in a 5% increase in LCC may be
acceptable for one program but not another. Cost effectiveness is
often measured as the ratio of the system effectiveness measure (e.g.,
availabi__cy) to the LCC. The cost effectiveness ratios for different
alternatives are then compared to determine the Freferred alternative.
The objective of breakeven analysis is to determine the point in time

when an alternative moves from being not preferred to being preferred
or vice versa. Such a point is called the breakeven point. Breakeven
analysis highlights the differences between alternative choices by
utilizing critical program characteristics (budget and schedule) as
decision factors. The procedure for performing breakeven analysis is
as follows: (Reference 5)
" Plot the cumulative annual present value LCC amounts for two PPA
alternatives on the same axes with cost as the Y axis and time as
the X axis. (See Figure 6.2.4.3.3-2.)
" Determine the crossover point for the two curves. This is the
breakeven point.
" Identify the cost avoidance benefits after the breakeven point
that accrue when the PPA with the least LCC is chosen. This is
the area between the two curves to the right of the breakeven
point.
6-33
Equipment A
Cash Flow Present Value

Year Discount
n Benefits Costs Factor (10%) Benefits Costs
0 $15,000 0.0000 $15,000.00

1 6,000 0.9091 5,454.60
2 $ 5,000 3,000 0.8264 $ 4,132.00 2,479.20
3 12,000 0.7513 9,015.60
4 16,500 0.6830 11,269.50
5 25,800 0.6209 16,019.22
6 23,000 0.5645 12,983.50
Total $82,300 $24,000 $53,419.82 $22,933.80
NPV = $30,486.02
Equipment B
Cash Flow Present Value

Year Discount
n Benefits Costs Factor (10%) Benefits Costs
0 $20,000 0.0000 $20,000.00

1 12,000 0.9091 10,909.20
2 $ 4,000 6,000 0.8264 $ 3,305.60 4,958.40
3 13,000 5,000 0.7513 9,766.90 3,756.50
4 17,000 3,000 0.6830 11,611.00 2,049.00
5 22,000 0.6209 13,569.90
6 20,000 0.5645 11,290.00
Total $76,000 $46,000 $49,633.30 $41,673.10
NPV = $ 7,960.20
Figure 6.2.4.3.3-1. Cost/Benefit Analysis Example. (Reference 5.)
6-34
* Judge the appropriateness of the PPA choice based upon considera-

tions such as: the point in time at which breakeven occurs, the
magnitude of the cost avoidance benefit, and the uncertainty asso-
ciated with the PPA derived benefits for the years preceding and
following the breakeven point.
6.2.4.3.4 Risk Analysis
In general, a risk analysis is required whenever there is uncertainty

associated with the parameter estimates used to represent probable
future values. Reference 5 identifies the following sources of risk
(uncertainty) associated with the analysis of PPA alternatives:
# Estimates of MTBF, repair times, unit cost, repair cost, etc., made
early in the acquisition cycle.
* Changes to the scope of the program (performance, quantity of

items, schedule delivery dates, site activation dates, operating
vs. non-operating time, etc.).
* Technical and technological problems (material or production prob-

lems, resource limitations, etc.).
a Uncertainty inherent in estimating parameters (omission of relevant

parameters, problems in fitting models to data, the random nature
of processes, etc.).
These risk sources must be analyzed and the results must be incor-
porated into the analysis of PPA alternatives in order to minimize the
possibility that an alternative chosen as appropriate for the life of a
system will in the future prove to be a poor choice.
In order to properly account for risk in the analysis of PPA alterna-

tives, it is desirable to bound the risk associated with estimates of
future parameter values. To do this, assumptions must be made about
the magnitude of the error. The following sections describe three
methods for bounding risk that are applicable to the analysis of
PPA's. They are:
v Sensitivity Analysis
* Confidence Interval Analysis
* Beta Distribution Analysis
6.2.4.3.4.1 sensitivity Analysis
Sensitivity Analysis is one technique for assessing the stability of a

choice of alternatives where there is uncertainty associated with the
data and assumptions. It is useful for determining the sensitivity of
6-35
analysis results to variations in the input parameters and for deter-

mining the degree to which input parameters must change in order to
effect a change in the preferred alternative.
In general, as the amount of data available about a program and system

being evaluated increases, the uncertainty concerning future cost
streams for the program decreases. The absense of data for specific
program parameters however usually results in the use of high and low
estimates for these parameters.
In performing sensitivity analysis, it is useful to have a baseline

case for comparison purposes. For PPA analysis, reference 5 recom-
mends using the no-PPA alternative as the baseline. The procedure for
performing the sensitivity analysis is then as follows:
" Calculate the baseline LCC.
" Determine the range of the parameters to be changed. Do this for

either all uncertain parameters or only identified cost drivers
(i.e., parameters having a large influence on total cost).
" Calculate LCC estimates for best and worst case combinations of
the parameters being changed for each alternative.
" Perform cost/benefit analyses (see section 6.2.4.3) and determine

where changes in the preferred alternative occur.
" Select a preferred alternative based upon an assessment of the

likelihood of the parameters changing as defined.
While this procedure provides a convenient method for evaluating the

effects of parameter changes upon LCC and alternative choice, it does
not, by itself, provide a quantified estimate of the uncertainty as-
sociated with a particular estimate or choice. The following two
methods provide a means of quantifying this uncertainty.
6.2.4.3.4.2 Confidence Interval Analysis
Confidence interval analysis is a method of determining an interval

which will include a parameter being estimated, such as MTBF, with a
specified degree of certainty. The interval containing the parameter
is referred to as the confidence interval and the specified degree of
certainty that the interval will include the parameter is referred to
as the confidence level. Confidence intervals are usually defined by
two endpoints which are referred to as the lower limit (LL) and the
upper limit (UL). They can, however, be defined by only one end point.
6-36
By giving a range (in probability) of where the true value of a para-

meter, such as MTBF, lies, confidence intervals permit the expression
of uncertainty about the true mean value of the parameter population
in terms of the bounds on the value. They also allow the testing of
the validity of assumed values of the population parameters. Such
testing is referred to as hypothesis testing.
Confidence interval analysis and hypothesis testing is applicable when

sample data is available from which a parameter of interest is to be
calculated. For example, failure data may be available from a sample
test of a quantity of end items. Rather than merely calculate the
average failure rate for use in the analysis and pricing of a PPA, an
interval can be calculated from the same data. The advantage of using
an interval instead of the single point average failure rate is that
the interval provides information concerning the likelyhood of the
actual failure rate falling within the interval while the single point
value provides no information concerning how much error it is likely
to contain.
A detailed discussion of confidence interval analysis and hypothesis

testing is beyond the scope of this book. For a detailed discussion
of the application of confidence interval analysis to the analysis of
PPA alternatives, refer to Reference 5.
6.2.4.3.4.3 Beta Distribution Analysis
Beta distribution analysis is a method borrowed from PERT/CPM (Program

Evaluation Review Technique/Critical Path Method) used to account for
uncertainty in estimates. In order to account for uncertainty, the
method requires three estimates (optimistic, most probable, and pessi-
mistic) for each parameter. They are defined as:
" Optimistic E5Limate (a) - the value of the parameter if all as-
sumptions and conditions occur in an ideal manner.
" Most Probable Estimate (mn - the value of the parameter most
likely under normal conditions.
" Pessimistic Estimate (b) - the value of the parameter if worst

case conditions occur.
The method gets its name from the assumption that the uncertainty des-
cribed by the estimates is best defined by a beta probability distri-
bution (Reference 13).
Based upon the three estimates, the mean (expected value) and variance
of a parameter (x) being estimated are defined as:
6-37
Mean (x) = 6
6 2
Variance (x) =
The expression for the mean is a result of assuming that the three
estimates describe a beta probability distribution. The expression
for the variance is based on the idea that a standard deviation is
approximately 1/6th of the difference between the extreme values of
the distribution and that the variance is merely the square at the
standard deviation.
After calculating the mean and variance for each parameter, the fol-
lowing relationships from reference 14 are used to calculate the mean
and variance for the sum and product combinations of the parameters
(e.g., X and Y) necessary to complete the PPA analysis. These rela-
tionships assume that the parameters are not correlated.
Sim
Mean [X + Y] = Mean [X) + Mean [Y]
Variance [X + Y) = Variance [X] + Variance [Y]
Product
Mean [XY] = Mean [X] x Mean [Y]

Variance [XY] = (Variance [X] x Variance [Y])
" (Mean (X]2 x Variance [Y])2
" (Variance [X] x Mean [y] )
Based upon the central limit theorem, the resulting mean and variance
values can be assumed to follow a normal distribution and the proba-
bility that a value lies within a certain cange can be calculated us-
ing the standard normal deviate and a table of the normal probability
distribution (Reference 13.) An example of this method is shown in
figure 6.2.4.3.4.3-1.
6.2.4.4 Finpl PPA Selection
After examining the identified PPA options and analyzing their impact
on the program, a final determination of which PPA option(s) to apply
must be made. This involves comparisons across the PPA options in the
following areas (Reference 5 ):
e PPA Costs
6-38
N NI
w I0
-4
1-4 -
0 0
0; -4 0n
0 '4 UM
ot
* i-I 0
to m4 N
-I (44
o 0
.m 0L LA -
E' .0 0 V
4Q- -4
Lni
oX o
LA 0N
12
0 -
>4
).
4d
44
r4
o oz to0
mA 0i + 4 0
04 X- i 1~ > >
to o
0 x - +. ILN --
.,Ii- N >4 4 1
u N>4S. -'4>
m0 0 >
Cz~~~~ ) ) -i~ uIU
:3~!t1 04 N
A.4t 4 4 0 .)Li
>4 H
CLi 0
E) IDN -- E U--m-
0
04 >4 >4- '->
-.
$4~
>4
:3W-
W
) )
0m 0
-Xj >- >
40 4)u 0-u 0- 0 o d)ze' IV
w 04 (d)- U'' &4I li >1 >2 + +
6-39
40)
00
U N,
>>
fo im .) f4
0 0 0
4.) u
014
CD 0 N
-4 0 -A
> -4 0 %0
4A N1.
00
0 o4- N+
4
4) N-
to
0I
Ln r-4
4.)
u ~ 4
0. *
0- 4.) 0 N >
x xA In 04 4toh4
(14C 0-4' 0 ' ~ 4 '
, 0 N 4C two - t
to. LA -q 0 0 .
0.. -4. n4
-4 0
NZ. toI I 01
4
-6 0
" Life-Cycle Costs
" Benefits
" Risks
" Effectiveness in reducing risk, achieving program goals (e.g.,

motivating the contractor to achieve higher levels of performance
for his equipment).
" Cost effectiveness for the program
In addition to these areas, careful consideration must be given to any

possible adverse effects a particular PPA may have on other program
areas. For example, the dedication of incentive money on a captive
carry guarantee may take resources from and possibly weaken other
areas. A PPA may also give a contractor an unintended incentive to
cut corners in other program areas, such as reducing the effectiveness
of BITE to reduce the probability of BITE detected failures under the
PPA.
As stated in section 6.1, effective PPAs must be simple, focused on

the primary objectives to be obtained, enforceable, affordable in
relation to potential benefits, and shouldn't cause disruption to
existing military systems and procedures. Of all the areas to be con-
sidered in the final selection of which PPA options to apply, the most
important one is the cost effectiveness for the program. If a cost
effective PPA cannot be structured for a program then a PPA should not
be applied to the program. Title 10, Section 2403 of the United
States Code specifically provides for the waiver of the PPA
requiremont in the event that it is determined that it is not in the
interests of national defense or is not cost effective.
After the final decision has been made, the contractual provisions for
any selected PPA options must be formulated. This is covered in the
following section.
6.2.5 PA Structuring
There is no "cookbook" approach to structuring a PPA. Each program

has unique needs, characteristics, and objectives that must be care-
fully considered in order to structure an effective agreement. How-
ever, an effective agreement will, in general, contain certain attri-
butes. They are (Reference 5):
" Protect the interests of both the government and the contractor
" Contain tightly defined terms and conditions
6-41
" Minimize government costs
" Quantify risks to government contractors
" Contain specific performance parameters related to program ob-

jeccives and mission
" Specify data requirements, test procedures, analysis, and evalua-

tion requirements.
The actual writing of the sections of a PPA can be a very demanding

task. PPAs, however, generally include certain sections to cover is-
sues common to all agreements. These sections are described in figure
6.2.5-1.
In structuring a PPA, it is very helpful to search out experience on

similar programs. PPA clauses from other programs should be examined
for their applicability and to identify what should be included and to
obtain examples of contractual wording. Table 6.2.5-1 presents typi-
cal performance measures for each of the PPA types identified as ap-
plicable to dormant systems. Other program offices with similar pro-
grams and/or PPAs should be contracted for input on provisions used
and lessons learned to see where provisions should be added or revised
to avoid repeating past mistakes. Other sources for lessons learned
information are the Lessons Learned Programs of the Joint AFLC/AFSC
and PPAC. In addition, the PPAC library at Wright-Patterson AFB,
Ohio, contains PPA Application Criteria including PPA Structuring
Guidelines and Management and Implementation Guidelines.
It is beyond the scope of this book to provide specific contractual

provisions for each of the PPA types identified in section 6.2.3 as
applicable to dormant systems. As previously stated, each program has
unique needs, characteristics, and objectives that must form an inte-
gral part of any resulting agreement. The general guidelines pre-
sented here provide a framework for structuring an effective agree-
ment. The details must be worked out on a program by program basis.
6.3 PRODUCT PERFORMANCE AGREEMENT MANAGEMENT
The management of PPAs applied to equipment procurements .s a key

factor in determining the ultimate cost-effectiveness of these agree-
ments. In order to be cost-effective, the government must be able to
manage the PPA reasonably; that is, have the ability to collect the
data on the items to determine if they are still within the PPA period
before the PPA can be enforced. The cost and effort to accomplish
this must not outweigh the benefits of the PPA. The effects of impro-
per management of a PPA can be severe with the government possibly
loosing many (or even all) of the rights gained under the PPA.
6-44-
PPA Secti on 2eQkrjjiio
1. Introduction Status the Govern-rent's obJw.tives for the program

iiicluding such topics as the syst-,/(juipment type, PPA
ohjactivej, and thr- PPA period.
2. Definitions To Lvo-d anbiguity, all torms should b defined anr Uscd

consistently throughout the :ntract, however, existing
official terms do not need to be redefi,,t.d unless they
are being tailored to provide consiste,(,.
3. Contractor Warranty Generally lengthy but very important i.n ,ininizin q

government and contractor 7i;<. The followirg areas
should be covered:
" Identification of equipment novered.
" The specific perfonance parameters covered with the

necessary equations and models. In general, separate
sections are usually written providing guarantees that
the item provided will" (1) conform to the design and
manufacturing requiremerts in the contrat. (2) be
free from defects in material &nd workmanship at the
time of delivery, and (3) conform to the essetntiai
performance requirements identified.
* The period of the PPA.
a Intended environment.
* Maintenance philosophy for organic and cont.actor

support.
4. Contractor Obligations This section should include terms and conditions

concerning the following:
" Identification plates designating serial number, part

number, installation dates.
" Decals or seals with dates for warranty tracking.
* Bonded storage areas.
Figure 6.2.5-1. Sections of an Effective PPA.
6-43
PPA Section Description
4. Contractor Obligations * Engineering Change Proposals (ECPs) either at no cost

(Cont.) to the Government or shared cost between Government and
contractor.
* Spares levels, requirements, compensation.
a Data tracking.
* Transition period from PPA period to post-PPA period,

including peculiar support equipment, publications,
manuals, and training.
5. Government Obligations This section should describe Government obligations in

sharing the responsibility and risk for a PPA including:
* Shipping costs and facilities for failed units and

spares.
e Provision of specified test equipment, Government

Furnished Equipment.
* Organic maintenance and sunpo-t
* Participation in failure verification proced'ure.
* Ad.nistrati.n and mcitoring of 'FA contrartor

obligations.
# evaluation of PPA-re.ated ECPs or design cnan.es.
6. Notification of Breach Notification must be timely, cLmplete. and accurate.

Notification can be given to the contractor by a
Government. repreentative advising a contractor fie>d
representative, a formal contracting offcer letter (ir
in writing through an appropr-ate service report. The
acceptable method should be specified. The tiire for
notiiication of breach should be based upon the tim
necessary for the Government to: (1) discover pr:lcrjIns,
(2) take administrative steps to report problens, ano
(3) discover and report defective replacements.
Figure 6.2.5-1. Sections of an Effective PPA (Continued)
b-44
PPA section _0jjcription
4. Contractor Obligations * Engineering Change P-opo~ali (FCPs) Pither at no cost

(Cont.) to the Goverinen' or sharad cost between Governmant ana
contractor.
# Spares levels, re,uiremants, compensation.
e Data tracking.
# Tranrition period from PPA period tu post-PPA period,

including peculiar support eq~uipmient, pu'.lications.
manuals, arid training.
5. Government Otltiqationi This sEction should debcribe Government obligations in

sharing the responsibility and ri-. for- a PPA including:
* Shipping costs and facilities for failed units and

spares.
# Provision cf specified test equipment, Government

Furnished Equipment.
* Organic maintL:iance and sopport
6 Participation in failure verification Procedure.
* Administratio~n in.d monitoring of PPA contractor

obligations.
e Evaluation of PPA-related ECPs or design rharges.
5. Notifiration of Breach Notification must be timely, complete, and accurate.

Notification can be given to the contractor by a
Government representativi? advz!ing a contractor field
representativ?, a formal cortracting officer letter, or
in writing through nmapp~ropriate service retort. The
acceptd~ble method shoila be specified. The time for
notification uf brearch shoulW be based jonn, the timc-
necessary for tte Gcve-nment to: 1 d~scovir p.-oblemTs,
(2) take admirnist,-ati-/e steps to report problims, and
(3) discovor ar-I report defactive replacemenTs.
iv re 6. 2.5- 1. Scction; of an Effaciive FPA (Contiru.,d)

PPA Section Cescriotion
7. Remedies This section should clearly define the actions to be

undertake by the contractor in the event of a breach and
should include:
e The scope of the corrective actions for which the

contractor will be responsible.
e The required turnaround time for a warranted item

including steps to be taken if turnaround times are
not met.
e A "disputes" clause identifying what is to be done in

cases of disputes regarding the applicability of the
PPA.
8. Exceptions and Conditions This section should contain exceptions and conditions
voiding the applicability of the PPA. They should be
kept to a minimum and should only include those areas
outside the contractor's control such as: (1) improper
operation or maintenance by Government personnel, (2)
acts of God, and (3) combat damage.
9. Data Requirements The enforcement of a PPA is highly dependent upon

accurate and complete data reporting. Tiis section
should clearly define:
* Types of data to be collected.
* Data reporting format, frequency, forms, etc.
* Data storage and availability to the Government.
* Analysis of data collected, reports to government,

evaluation by government.
Figure 6.2.5-1. Sections of an Effective PPA (Continued)
o-46
PPA Section Description
10. Test Plan Dormant systems typically use a test plan over a
specified period to check the reliability and
availability of the system over time. The test plin
section should include: (1) test articles and
production lots to sample from, (2) test sample size.
(3) test interval including dates and duration, (4),
test location, (5) personnel responsible for testing,
data collection, monitoring, and test result analysis,
(6) retest provisions, (7) failure definition, and (8)
performance measurement definition.
11. Contractor Access To This section should define the contractor's rignt to
Government Maintendnce reasonable access to the facilities and records that
and Operational apply to covered items. However, the contactor should
Facilities not have the right to interfere with the facilities,
operations or unduly burden the government with data
requests.
12. Liability Cap Limitations on liability have generally been used as a

means of restricting the contractor's exposure to a
level that will keep the price of the PPA reasonable.
If used it should be high enough to retain the
motivation for the contractor to want to eliminate
problems and should reflect the degree of risk actually
assumed by the contractor.
Figure 6.2.5-1. Secticns of an Effective PPA (Continued)
6-47
IIr ~~CL
M A1 0 0W
si L£ud
a)
4- &A 'A.
0- V4 04 W 4, 4)
1..- 2 -
ca 041- 41- .- 4
,A 4 W 1L1 0CCC A C A
>.41 4.~ 0% ... ' U 4- A .-
o~~~ 'A 00014 ~~ L ' ~ L ~
4~~~4 0t
- ~ 04.I
0 4 '
CL.-.
0 m. 4 4
C1
.
x1
£0-
L
41. ~ ' C4' 'a MA .04 M
L~~~~ 0 4V 0 4. ~ 0 1 .- 0- L . C
aA 4A U, C4 'A
r_ 1- U C. 0 'a .
4 0 >5 . C - m1 T1C
4.- 0'A 0
&A &4
- , 10-'C
0A IA~
'A 01 40 'A
CA.. f ._ J ld 'A
W .-
Ou* (A Z 4
40 :3A. 4) >4l a 4
0.-
I 4 0. 0 C 'C. 4
>0% 1 414 4) 0C MA
M 0.' Co M 4-'- L,
4V- 4.^4 0
W- Q. M1 0.-4 L
#j~
c (l.-- Io--0 c mv w'' M
4-> ( V .0 4
10 41 L Ln L. WM 0
x >1 0- 4- c4 V C ; L .%
LL
441
L,-
0
4. 41
41 41 C 41
E 0)> 41 Im
1: L-
c M4
0-48
a) D a).M
0. 1. -*a
a) 0 4) 0-
IM4. L. 'AL.
4-> 0
W CL l m w a
n~ -0 t 0 - ) 0 0 na
C aam)UAU# 4.
'4-
in0 4.)a
0. 4 w6 -- 0 x G L
a) >in-CL - 0
CL 4-M& c w L 1 U "
a_, M o c :3c , L -,
- C- ' C m 4
E IA 0a. 'a.
0 4. I- C0 -1
La)
M L.
~~~
-C Ma' )CL'a <a
0...
' to > V.
0 aC ) M4 L- I" C C
W ma in '
S0 .iL f .0- -1
Lr
MA 0 ) -:a), C- C 4. . M)~ M) L C in w
'a *. C0 - (U W C .0 30 m u. 0) m
im 1.. '1 M 4. ' 4.. * W4 .' a)0 . 4- >
t L . -C 10 CD'a L U W) 0 C0
=- A '0 0 > 00.
0) a 5 C a- C %A) M (AC = E ,- C. w -~ L. >
'a-V 0 CLA j a0.. c 'a (V Ma. r--0 0
aJ o )t CL t 0 C n LI)- D 4.) L .) Go 3 c
zl C 4j. L. a)'. I* C a) C >'a 'a
a)) EL A) C
Q A) Z 0 U 03n 0 .
wO)
*n L44-Ca 0 C" 0C C a)
o ~ LA.) ~ 4 L4-0..4. 04 C'- Ci S - -4 C 4 'a 00
L4
a) 0 0 #a CLc0 L)
L. 0 c C m Q ) D IS
n 0- , 1
LI -~ A-- "-jI a4 o -- c 4
0l 'ji Q. CL- X - c .
I- do
(NJ aa
at. *
i) U)-1c
00
a'a
C41
4-, '*- ' 0) X
C
v-
C *- C 'a*
) CL 'a Q6U i .'
(n
a
>
mW
ox
41
) ~ *1
.-
X 0L 0
0)D :
6-49
Perhaps the most important step in making sure that a PPA is admini-
stered properly concerns training. If a PPA isn't administered pro-
perly, potential benefits are lost. Much waste can occur if mainten-
ance, supply, or contract administration personnel; management; and
others are nct given complete information about what is covered by the
PPA and the Government's and Contractor's obligations under the PPA.
All affected personnel should be trained concerning their respective
duties regarding the PPA. Since field personnel are the first to en-
counter potential PPA claims, they should, as a minimum, be able to:
" Identify the covered item.
" Report the breach of a PPA to the proper activities.
" Start the prescribed action leading to the remedy.
To make the identification of covered items easier they should be

suitably marked with some form of identification indicating that they
are covered by a PPA. Such marking could be the word "Warranted" on
the item of just "W" before the part number (Reference 1.). The mark-
ing instructions should be clearly defined in the contract.
It is important to define the tracking system to be used and how data

are to be interpreted. This should be addressed in the development of
the PPA and its management plan. Imposing a PPA without consideration
of the tracking can lead to future problems and discrepancies between
the way a PPA is actually managed and the basis on which the PPA
analysis was made. The ideal situation would be for items to be
tracked within a data system which is programmed to permit use of a
PPA symbol or message while the item is covered under the PPA and
automatically delete the symbol or message when the PPA coverage
period expires. (Reference 1.)
Technical data and manuals should identify items covered by PPAs and
the duration of the PPAs. Equipment log books and record cards should
also state tnat an items is covered by a PPA and should include the
effective period of the PPA.
In monitorilg and assessing PPAs, costs for effort expended on PPA

breach corrections should be segregated by both the contractor and
government. All PPA actions taken by the contractor should be re-
ported according to contract instructions. Each PPA action should be
documented from the time a breach is discovered to the final disposi-
tion of the corrective actions. These data are necessary to evaluate
future PPA proposals and for making a final assessment of the effec-
tiveness and efficiency of the PPA.
6-50
6.4 REFERENCES
1. AFLCP/AFSCP 800-47, Acquisition Management -- Joint Engine War-

ranty Development Guide (for Military Aircraft Turbine Engines),
Departments of the Air Force, the Army, and the Navy, 26 October
1984 (also identified as DARCOM-P 5-1 and NAVMAT P-13700).
2. Product Performance Agreement Guide, the Air Force Product Per-

formance Agreement Center, ASD/ALW-AFALC/XRCP, Wright-Patterson
AFB, Ohio, 1 November 1985.
3. "The Use of Warranties in the Acquisition of Conventional Muni-

tions," USAF Major Bruce D. Mills, Air Command and Staff
College, Student Research Report, DTIC Report No. ADB057697, May
1981.
4. "Assessment of the Reliability Improvement Warranty (RIW),"

Lt. Col. Michael A. Guenther, Air Force Product Performance
Agreement Center, Wright-Patterson AFB, Ohio, 12 July 1985.
5. Product Performance Agreement Decision Support Handbook, Air

Force Product Performance Agreement Center, Wright-Patterson
AFB, Ohio, TR-4632-2R, 1 February 1985.
6. "The United States Air Force Product Performance Agreement

Center," Norman G. Fleig, 1986 Proceedings Annual Reliability
and Maintainability Symposium, pp 449-453.
7. "The Impact of Reliability Guarantees and Warranties on Elec-

tronics Subsystem Design and Development Programs," C. David
Weimer and Paul E. Palatt, Institute for Defense Analyses, Study
S-483, DTIC Report No. ADA035354, October 1976.
8. "An Approach to Selecting Warranties/Incentives," Paul B. Gug-

lielmoni, Eglin AFB, 1986 Proceedings Annual Reliability and
Maintainability Symposium, pp. 166-170.
9. "Warranty Pricing With a Life Cycle Cost Model," Alfred 0. Rotz,

FMC, San Jose, California, 1986 Proceedings Annual Reliability
and Maintainability Symposium, pp. 454-456.
10. "The Use of Warranties for Defense Avionics Procurement,"

Harold S. Balaban, et. al., ARINC Research Corporation,
RADC-TR-73-249, DTIC Report No. AD 769399, June 1973.
11. "Contractor Risk Associated with Reliability Improvement War-

ranty," Harold S. Balaban and Martin A. Meth, Proceedings 1978
Annual Reliability and Maintainability Symposium, pp. 123-129.
*Althouh this document referenres 2; limited document, nn lir;ited

information bas been extracted.
6-51
12. "Case Report: R & M Support of CLS Warranties," J.A. Di

Paol.antonio, et. al., 1986 Proceedings Annual Reliability and
Maintainability Symposium, pp. 457-459.
13. An Introduction to Management Science - Quantitative Approache$

to Decision Makina, David R. Anderson, et. al., West Publishing
Co., St. Paul, MN, Second Edition, 1976.
14. "Modeling Spares/Repairs Calculations for Warranty Pricing of

New System Development," Linda M. Sylvester-Seman and Robert M.
Seman, Proceedings of the Third Annual Old Dominion Chapter of
the Society of Logistics Engineers Logistics Research and
Analysis Symposium, Fort Lee, Virginia, 22 October 1985.
15. "Warranty of Weapon Systems - AGM-65D Maverick," Contract

F33657-85-C-0086, Sectior H, November 1984, pp 42-50.
16. "Warranty of Supplies of a Complex Nature: Pershing II,"

Contract DAAKlO-84-C-0069/PO0002, Part II, Section I, 1984, pp
39-42.
17. "Trident II (D5) Quality Incentives - Methods and Requirements

for Determining Performance and Scoring," OD 55578 Rev. A,
Department of the Navy, Strategic Systems Programs Officc,
Washington, D.C., 11 November 1984.
6-52
APPENDIX A
RELIABILITY CHECKLI ST
RELIABILITY CHECKLIST SECTIONS
PAGE
1.0 General Design Criteria A-2
2.0 Reliability Critical Parts A-3
3.0 Electrical Circuits A-5
4.0 Electrical Parts A-6
5.0 Mechanical Design A-7
6.0 Physical Design and Layout A-10
7.0 Environmental Design A-15
8.0 Electrostatic Discharge (ESD) A-22
A-i
1.0 GENERAL DESIGN CRITERIA
ADE- INADE- N/A I

QUATE QUATE
1. Are the requirements for performance,

signals, environment, life, and reli-
ability established for each unit?
2. Has the theoretical reliability or MTBF

of the unit based on the actual applica-I
tion of the parts been determined?
o Comparison made with reliability

goal?
o Provision for necessary design

adjustments?
3. Has the item or system been designed forI

ease of production and assembly, mainte-1
nance, inspection?
4. Have factors of handling, transporta-

tion, packaging, and environments, other!
than the specified operational environ-
ments, been taken into account?
5. If so, can the proposed design be

considered as optimized under existing
conditions and restraints?
6. Have trade-off studies been conducted

and completed in particular with respectl
to the following areas:
o Performance?
o Reliability?
o Safety?
o Circuit Design?
o Power Consumption?
o Packaging and Integration

with other systems?
o Weight?
o Cost?
A-2
2.0 RELIABILITY CRITICAL PARTS
Reliability critical parts are defined in the following

paragraphs. These paragraphs should be used as a check list to
determine if such parts have been identified and treated as
noted.
Critical items are those whose failure would significantly

affect mission success. Critical items include, but are not
limited to, those identified by reliability, maintainability,
safety, and human factors analyses. As soon as a critical item
is identified, appropriate steps will be taken to assess the
risks involved.
The criteria for defining critical items are:
o Failure effect on safety.
o Failure effect on mission accomplishment.
o Need for special handling or storage.
o Long lead-time.
o High Cost.
o Failure effect on maintenance burden.
o Limited life.
o High technical risk, i.e., the use of

unproven technology.
o Limited availability.
Critical items are generally best handled on a case-by-case

basis. For example...
ADE- INADE- N/A

QUATE QUATE I
1. Items which require special handling or

storage will be reviewed to see if
special containers, etc., are necessary.I
2. Items whose failure might impose a high
maintenance burden will be examined fromi
the viewpoint of reducing mean-time-to-
repair, improving mean-time-between-
failure, or generally improving overall
maintainability.
A-3
ADE- INADE-7 N/Al

QUATE QUATE
3. Limited life items will be avoided to

the greatest extent possible. If
unavoidable, means to overcome this
problem would be investigated by oper-
ating the part at a reduced stress levell
if feasible, if this does not impair
performance.
4. High risk technology v'ill be avoided to

the maximum extent possible. If its
use is unavoidable, more conventional
design concepts will be made available
as a backup.
5. Items with limited availability will be

avoided. If this is not possible,
arrangements will be made to assure thati
the item will be produced for as long asl
it will be required.
A-4
3.0 ELECTRICAL CIRCUITS
ADE- INADE- N/A I

QUATE QUATE
1. Have the operating and non-operating

failure rates of the circuit been
compiled, using approved part
failure rates for the expected circuit
application?
2. Has the application of redundant cir-

cuits been considered as a means of
increasing reliability?
3. Have adjustments been minimized?
4. Are elapsed time indicator:s installed inI

major components?
5. Have self-resting or nronitoring features!

been isolated wherever possible to pre-
vent their causing failure of basic
circuit functions?
6. Is the circuit stable over the entire

operating and tion-operating range?
7. Have design features been incorporated

to suppress parasitic oscillations?
8. Have the effects of long-term

non-operating periods on each component,!
circuit, and assembly been considered?
9. Have the effects of power on/off cyclingI

on each component, circuit, and
assembly been considered?
A-5
4.0 ELECTRICAL PARTS
ADE- TI NADE- N/A

QUATE QUATE
1. Has shelf lire of parto chosen for firtall

design been determined?
2. Have limited life pprts been identifiedj

and inspection and replacement require-
ments specified?
3. Have critical patts which tequire

special procurement, testing, and han-
dling been identified?
4. Has the sclection of electrcenic compo-

nent parts been limited to parts that
have known or predictable parameter var-I
lations, whether resL'ltlng from purchasel
tolerances, soldering, er.vironmental
stresses, or aging?
5. Has the use of environment sensitive

parts been avoided wherever oossible?
6. Have tolerance effccts, includirng those I

due to environment, been considered on
all critical part para:.eters?
7. Are all parts applied wii.hin their elec-I

trical and environnentali ratings?
A-6
.0 MECHANICAL DESIGN
ADE- INADE- N/A

QUATE QUATE
1. Has adequate tolerance analysis been

applied to the mechanical function of
the design? (This includes not only
checking of basic design tolerances but
calculation of functional variations
with expected environmental extremes
mechanical error analysis).
2. Has adequate lubricant life been

ensured an? nave potential aiverse
effects o* bricants on other system
components .een eliminated?
3. Has the danger of hydrogen embrittlementi

been considered in the use of highly
treated steels and alloys?
4. Have flexible parts, oL flexible hoses

and conducts, been routed in a way such
that specified bend-radii and other
installation limitations have not been
exceeded?
5. Has a worst-case adverse tolerance

stackup been performed for each criticall
mechanizal interface?
6. Will tolerances remain within specified

limits at extreme temperature?
7. Has protection from mechanical abuse by

the use of suitable mechanical protec-
tion techniques (packaging, mounting,
and structural) been incorporated?
8. Have either of the 'ollowing been uti-

lized to enable equi'ment to withstand
the shock or vibration environment:
o Vibration isolators and/or shock

mountings?
o Design to witnstand the mechanical

stress?
9. Have any of the to]]oving corrosion

teiniq-ies been utilized on exprsed
equipment:
o Corrosion resistant matetials?
A-7
IADE- INADE- I N/ATI

I PUATE QUATE
o Plating and protective finishes?
o Dissimilar metal contact avoidance?
o Environmental control (water entrap-

ment prevention, atmospheric moisturel
removal, etc.)? I
I
10. Has electrolytic action under specifica-I
tion extremes of humidity & temperature
been minimized?
11. Have the effects of differential pres-

sure on leakage & on system/equipment
configuration itself been considered?
12. Have adequate performance functional andi I I I

dimensional tolerance analyses been
performed? Did this include:
o Checking of basic design tolerances?
o Calculation of functional variations

for expected environmental extremes
(mechanical error analysis)?
13. Have structural materials and their

conditioning and temper been selected
to avoid possible stress corrosion?
14. Have the combined effect of load and

temperature induced stresses been con-
sidered for materials and material-
conditions susceptible to creep or
plasticity?
15. Have major structural joints and bonds I

been prepared and tested under con- 1.
trolled crnditionz? Has this incivied:
o Preparation ',nver con'rolled and

continuou- -nitored and recorded
conditions . environments?
o Verification of adequate quality by

nordestructive inspection, including
standard methods such as x-ray, dye-
penetrant, magnetic particle, ultra-
sonic, and infta-ied tests?
A.-8
ADE- INADE- N/A

QUATE QUATE
o Verification by destruction testing

of test samples or coupons prepared
and conditioned under identical
conditions with the corresponding
joints?
16. Has the proper bolt-torque been speci-

fied where required? This also applies
to all stud mounted devices?
17. Have provisions been made to lock and

secure all fastener elements against
operating and non-operating conditions
which tend to reduce their functional
holding force?
18. Have proper fasteners been employed to

satisfy EMI and drip-proof requirements?l
19. Have the effects of outgassing products

on componeit parts been considered and
properly controlled?
20. Has the existence of residual stresses

introduced during material manufacture
and cunditioning and during subsequent
storage been considered?
21. Has the possibility of flaking, spallingI

and of fretting corrosion of structural
elements been considered?
A-9
6.0 PHYSICAL DESIGN AND LAYOUT
ADE- INADE- T N/A

QUATE QUATE
GENERAL ------------
1. Have the package interfaces and subblock

and subassembly break points been de-
signed with the maintenance philosophy
in mind?
2. Has adequate consideration been given tol

ease of inspection and testing?
ELECTRICAL AND ELECTRONIC LAYOUT
3. Has the part spacing ensured:
o The spacing necessary to reduce

distributed capacitance and coupling
between parts or part and chassis
surfaces in critical electrical
circuits?
o Satisfactory application of harness-

ing branches, wiring leads, & servicel
loops to part terminals, tie-points
and posts?
o Safe terminal and connection clear-

ances to prevent arcing or leakage
under all environmental conditions?
o Sufficient spacing to permit proper

part-lead-forming, strain relief, andl
compcrnent-to-component spacing?
o Sufficient clearance to permit appli-I I

cation of test leads without short
circuiting of part leads?
4. In general, have parts been positioned

to avoid lead lengths that are either
too long or to short for good layout
practice?
PRINTED CIRCjIT LAYOUT
5. Is the thickness of the conformal coat-

ing controlled to prevent overstress or
breakage of components?
A-10
ADE- INADE- N/A

QUATE QUATE
6. Has a qualified base-laminate material

that will remain dimensionally stable inj I
soldering, that has low moisture absorp-I I
tion, and good electrical and mechanicalj
properties been specified for the cir-
cuit board?
7. Are flush mounted parts (flat packs)

protected to prevent penetration of
conformal coating under parts that can
cause breakage of compontent leads?
ELECTROMECHANICAL, MECHANICAL, AND

MISCELLANEOUS LAYOUTS
8. How might the design be modified to

improve reliability, and would this com-I
promise such factors as performance,
cost, weight, availability, schedules,
maintainability, etc.?
9. Has the item or system been designed forl I

ease of production, assembly, mainte-
nance, and inspection?
10. Have handling, transportation, packagingl

and environments other than the speci- I
fied operational environments been takenj
into account?
WIRING AND CONNECTOR LAYOUT
11. Have all interconnections and cables

been arranged so they do not block
removeable units or fasteners?
12. Is wiring mechanically secure against

breakage, chafing, or other vibration
and shock damage that might occur in thel I
expected environment?
13. Has wiring or cabling that is run

through holes in partitions, shields,
etc. been protected by grommets or
tolled edges?
A-11
I ADE- INADE- N/A

QUATE I QUATE
14. Is wiring that passes between two struc-j

tural members mechanically flexible and
securely clamped to each member with
sufficient slack to prevent mechanical
strain?
15. Have specified minimum or less than min-I

imum wire and cable bend radii been per-I
mitted in the layout, especially for RF
cable?
16. Have connectors in areas of heavy mois-

ture, dust, fuel and other fluid see-
page, etc. been protected?
ELECTRONIC CHASSIS, CABINETS, AND

EQUIPMENT ENCLOSURES
17. Has the degrees of enclosure reflected

the environmental requirement, the
electrical shielding requirement, & the
requirements imposed by human factor? I
18. Have qualified materials been selected
on the basis of strength, weight, elec-
trical factors, producibility, & resis-
tance to the expected environment
(humidity, fungus, etc.)?
19. Have all component boards, terminal

boards, and other part-mounting subas-
semblies using insulators been designed
using suitable materials to facilitate
input and output wiring, maintain elec- I
trical properties and strengths, and
provide stable part mounting under all
expected environments?
20. For airborne and space equipment, has

the use of connectors been limited to
necessary intercomponent applications, I
where length and complexity make direct I
wiring impractical? (Multiple connectorl
contacts contribute to system unrelia- I
bility, especially where shock and
vibration environments exists).
A-12
ADE- INADE- N/A

QUATE QUATE
21. Has the design of covers and closures

included:
o Provision of interlocks and automatici

disconnects for high-voltage areas?
o Identification of the items or com-

partments covered?
o Easy access for tool application?
o Permanent attachment of removable

closures, plugs, and small covers?
22. Has a keying system been employed to

prevent faulty interchanging of circuit
boards?
23. Has proper electrical grounding and/or

bonding been provided across corrosion
protected surfaces, mechanical inter-
faces, etc.?
HARDWARE AND MISCELLANEOUS
24. Have part mountings, fasteners, and

assembly items in general used hardware
suitable for the environmental require-
ments, and have the following design
goals been met:
o Adequate quantities and strengths of

"black box" hold-down fasteners?
o Sufficient clearances around and

above fasteners?
o A minimum of inaccessible or blind

fasterners?
o Adequate locking devices for the

mechanical environment?
o No dissimilar metals in contact, or

suitable protection if this must be
done?
A-13
ADE- INADE- N/A I

QUATE QUATE
o Similar types of fasteners in similari

applications, reducing the number of
tools required?
o Specification of torque requirements?I
" Dimensionally correct hardware for

the application? (Example: lengths
of screws, proper openings, & proper
contours on spring clips, etc.?
25. Are materials and finishes consistent

with intended use?
MISCELLANEOUS
26. Have effective radio-frequency-inter-

ference-suppression measures been
incorporated in the design, including:
o Grounding of the chassis or trame in

accordance with good low-resistance
grounding practices?
o By passing of all shock mounts with

a ground conductor?
o Shielding of openings in the externall

wrapper or enclosure?
o Use of noncommutating (brushless)

motors for blowers and other applica-I
tions where possible?
o Use of metal bonding gaskets, under

positive pressure. on all access
doors, drawers, and removable covers?l
o Integral shielding of all oscillator

compartments?
A-14
7.0 ENVIRONMENTAL DESIGN
ADE- INADE- N/A

QUATE QUATE
GENERAL
1. Has the physical design ensured satis-

factory operation in or protection
against all the environments to be
encountered during the desired equipmenti
life?
2. Have the relative advantages and disad-

vantages of designing for operation in
the environment or protection against
the environment been weighed (package
sealing versus environment-resisting
parts)?
3. Have the effects of the nonoperating

environment on electronic and mechanicall
parts been considered (especially non-
reversible effects)?
4. Have the effects of the expected envi-

ronments on the part failure rates used
in equipment reliability analysis been
reviewed?
5. Have corrosion resistant materials or

protective finishes been provided where
required?
6. Have the effects of environmental fac-

tors on material properties been consi-
dered? Has this included:
o Temperature cycles and fluctuations?
o Corrosive atmospheres and solid part-I

icle contaminants?
o Humidity?
o Radiation?
o Life?
o Fatigue?
o Creep & Stress?
o Embrittlement?
A--15
ADE- INADE- I N/A

QUATE QUATE I
o Sublimation?
o Melting?
o Freezing?
o Congealing?
o Segregation?
o Cold Flow?
o Fracture?
THERMAL ENVIRONMENT
7. Have the expected temperature variationsI

per unit time -- both thermal shock and
ambient temperature versus altitude --
been investigated and considered in the
design?
8. Does such protection provide fGr:
o Incidental leaking of coolant, lubri-I

cant or any other fluid heat-convey-
ing medium?
o Local effect of a heater or heat-

source, such as hot-spots, focal
points, or unintentional reflection?
9. Is the equipment either tree of or pro-

tected against:
o Chemical corrosion (reactive pr~o-

cesses)?
o Thermal corrosion (heat fatigue)?
10. Have provisions been made to ensure

equipment success even if limited corro-I
sion or leakage of materials used in thel
equipment should occur?
11. Have thermal stresses and differential

thermal expansion been considered?
A-16
ADE- INADE- N/A

QUATE QUATE
12. Are polished and unpainted radiation

shields provided to protect heat-sensi-
tive parts from "seeing" hot parts?
PRESSURE ENVIRONMENT
13. Has the pressure environment to be

encountered by the equipment, both oper-l
ating & non-operating, been considered?
I I
14. Have all materials used in the design - I
especially plastic laminates or coatingsl
- been checked for out-gassing and gen- I
eration of arc-sustaining atmosphere? I
I
15. Have the diaphragm strengths of the var-I
ious mechanical assemblies and packages I
been checked for adequacy where pressurel
differentials may exist? I
16. In space equipment, have possible explo-i

sive decompressions been avoided by ade-j
quate outlet areas?
17. Has the sublimation of materials been

considered in the system design and in
application of:
o Package surfaces and containers?
o Insulators and wiring?
o Hardware and fastening materials?
o Surface platings and coatings?
o Lubricants, where premiture evapora-

tion might occur?
HUMIDITY
18. Have the worst expected humidity envi-

ronment and the time in the environment
been investigated (both operating and
non-operating?
A-17
ADE- INADE- f N/A I

QUATE QUATE I
19. Where high-humidity and condensation arel
to be encountered, has the design:
o Protected component boards by appli-

cation of conformal plastic coatings
or equivalent moisture barrier?
o Protected RF components & waveguides

by sealing wherever possible?
o Protected appearances and utility of

cabinet or container surfaces and
fastenings by application of moisturel I
resistant platings or coatings?
o Protected grounding and bonding con-

tinuity by overcoating and sealing ofi
ground lug and fastener assemblies?
o Protected cables and interconnection

wiring by suitable potting or encap-
sulation of connectors and by choice
of moisture resistant connectors?
o Generally provided against loosely

overlapping sheet metal surfaces,
folds, tucks, and other unnecessary
moisture entrapment points?
20. Have the relative moisture absorption

characteristics of insulators or insu-
lating bonds in electronic components
been evaluated?
21. In general, have hermetically sealed or
encapsulated parts been chosen wherever
possible to reduce humidity-induced var-I
iation of part parameters?
MECHANICAL SHOCK AND VIBRATION
22. Have the expected mechanical environ-

mental stresses been specified or esti-
mated in terms of:
o Amplitude and approximate waveforms

of shock inputs (equivalent pulse
rise, dwell, and decay)?
A-18
ADE- Ih\DE-TN/A
QUATE QUATE
o Amplitudes, accelerations, and fre-

quency ranges of vibration inputs
(fixed frequencies or random inputs,
wide or narrow band)?
23. Have the principal dynamic stresses on

the package or functional assembly been
calculated, or has au accurate model
been tested with proper and sufficient
instrumentation?
24. Has at least a static stress analysis,

using acceleration levels appropriate tol
the expected dynamic loading and i;iclud-I
ing spe.ified margins of safety, been I
prepared for the package or assembly?
25. Have 311 fasteners and hold-dovn devicesI

been checked foi adequacy under mdximum
shock loading? I
26. Have chassis and sidewall configurations
been designed to provide minimom ampli- I
ficatior, paths for vibration excitation I
of part mountings and component boards? I
I
27. Have statistically meaningfuJ samples ofi
all electLonic component part.. been
qualified in mechanical environments
similar to those eypected in service
(operating and non-operating)?
28. Have the: fundamental resonant frequency

of all componicnt boards been checked and
maintained as high a. feasible to limit I
vibration deflection and resulting
stress on parts and wiring?
29. Have component board resonant trequency

been raised by increasing stiffness
through size reduction, additional
thickness, addition of supports, addi-
tion of edge-beam stiffeners, and/or
cnoice of high-specific-damping mate-
rial?
30. Have isolators or "soft" mounting been

ijnjestigated for use where part sensi-
trvite, to the mechanical environment
,ay be a . ource of unreliability?
A-19
ADE-[ INADE- N/A

QUAIL IQUATE
31. Have the characteristics of isolator ma-iI

terials been checked or tested ovei the
entire exp)ected vibration range, and i
have the cffects of steady accelerationII
on these materials been checked? II
32. Have thE total time in thc mechanical II
elivi ronrrent and the j.es:-sible equipment I
fatigue r-esults; from repeated cycles I
and/or rcpeated stresses (S-N curve
evaiuation) beeti considered?
33. In general. fL elc-ctronic component-I

pant assemblies, have the designs at-
tempted to isolate or mi'anize ampliti-
cations of the vibration inputs so that I
roaxirnum acceleistions on parts do not i
exceed specification limits?
RAI)IAT1ON ENVIRONMENTS
34. It radiation environments are to be en- I

countered, have the different types off
radiation (electromagnetic corpuscular)
been considered, and has the exposure
time irn each been estimated?
35. For space equipment;
" Have the effects of solar, altra- I

violet, cosmic and Van Alien belt II
radiations been c )nSideLE'd? I
" H-ave skin shields such as aluminum, I
magnesium, plexiglass. fiberglass, II
etc. been applied agaiust ionizing
radiation, and have solar ceils beenI
protected with cover glass cc the
equivalent?I
o Where corpuscular riatincau.sing I

secondary emission viil be encounit-
ered, has additiopal Mhelding (hea-
v ier alumninun or copper ) Ov~en pro-
vided or has skin s~hielding been fol< I
lowed by local shielding at sernsicivei
patts?I
ADE- INADE- N/A

QUATE QUATE
o Is the shielding design capable of

stopping or adequately reducing ex-
pected proton, electron, gamma, and
X-radiation?
o Have radiation resisting component

parts - such as wirewound, carbon-
film, and metal-film resistors, cer-
amic and glass capacitors, and thin
base transistors - been used?
o If radiation-sensitive parts must be

used (for example: diodes and some
transistors), have adequated shield-
ing precautions been taken?
MISCELLANEOUS ENVIRONMENTS
36. If sand and dust are to be encountered,

are equipment enclosures, control panelsi
and connector panels adequately sealed;
have open corners been eliminated; and
have cooling-air inlets been adequately
filtered?
37. If salt water or corrosive atmospheres

are to be encountered, have paint fini-
shes & platings that will withstand the
exposure been used, or have non-corro-
sive metals been applied in the design?
38. Where conditions to be encountered will

support fungus growth, have fungus-
nutrient materials been eliminated from
the design?
A-21
8.0 ELECTROSTATIC DISCHARGE (ESD)
Many parts can be damaged by electrostatic discharge (ESU) and must

be protected during the manufacturing process and even after de-
livery of the complete equipment, particularly spare printed circuit
boards.
[DE-INADE- N/A
QUATE QUATE
1. Do the drawings identify ESD devices so

that all personnel involved (receiving,
manufacturing, assembly, test, etc.)
realize that ESD devices are being used
and must be protected?
2. If DOD-STD-1686 applies, have circuits

been divided into Class 1, 2, and 3:
o Has use of Class 1 parts been mini-

mized?
o Have protective cifcuits been used tol

limit assembly sensitivity to Class 21
minimum?
o If protective circuitry can not be

used, has applovedal of the aquiring
activity been obtained?
A-22
ESD CONTROL PROGRAM - SURVEY CHECKLIST
------------------------------------------------------------------------
I ADEQUATE
FUNCTION RESPONSIBILITY ------------
YES NO
---------------------------------------- --------- ------
Engineering/ I1. Is protective circuitry incorporated
Design Servi-I at the lowest practicable assembly
ces level to protect sensitive ESDS items
at the assembly and equipment levels?
2. Do all drawings of ESD$ ems identify!

the item as ESDS; are th=y marked with! I
sensitive electronic device sysmbol of!
Mil-STD-129 and ESD caution; do they I
reference applicable ESD precautionary!
procedures; and do they identify the I
sensitive pins or terminals of assem- I
blies and equipment? I
I
3. Do the drawings show the requirements I
for proper ESD marking of ESDS assem- I
blies and equipment located so that it)
can be readily seen by maintenance
personnel prior to removing/replacing
the ESDS item?
A-23
II ADEQUATE I
FUNCTION RESPONSIBILITY -------------- I
YES NO
---
-----
---
----- -- -- - -- -- --I
-- ----- -------- ----- ----- ------
Equipment/ 1. Is testing performed in ESD protected
ISystem Test I areas which limit static voltages to
and Integra- I 4000 volts or less?I
tionII
2. Are ESD precautionary procedures pro-
vided for personnel who install ESDS I
equipmen~t?II
3. Are ESDS items enclosed in ESD protec-I
tive covering prior to transfez out of I
the ESD protected area?
- - - - -- - - - - - -- - - - - - -- - - - I- - -- - -I - - - I---
A-24
I ADEQUATE
FUNCTION RESPONSIBILITY ----------------
I YES NO
-- ------ ------------------------------------ --------- ------
Production: 1. Is the receiving area equipped as an
Receiving ESD protective area 4here packages of
Inspection ESDS items are opened?

vided for personnel who handle ESDS
items out of their protective pack-
aging?
3. Are ESDS items identified on shipping

documents and labeled as ESDS by the
supplier?
4. Are ESDS items removed from their ESD

protective packaging for inspection oil
test, repackaged in ESD protective
material after inspection?
Processing, 1. Is all processing machinery properly

Assembly, grounded or equipped with ionizers or
Inspection, other means to limit static to safe
Repair and levels?
Rework
2. Are carts or wagons used to transport
ESDS items throughout the plant
equipped with conductive wheels to
control the buildup of unsafe static
charges?
3. Are all part carriers (tote boxes/

trays, etc.) used to transport ESDS
items made of ESD protective (prefer-
ably conductive) materials?

vided for all personnel who handle
ESDS items during the production
operation?
5. Are ESDS items enclosed in ESD protec-(

tive covering material after comple-
tion of assembly and test?
A-25
ADEQUATE
FUNCTION RESPONSIBILITY-----------------
YES NO
I---------I--------------- --------------------------------------------------------- ----
Packaging andj 1. Is handling and packaging performed in
Shipping ESD protected areas?
2. Are ESD precautionary handling and

packaging procedures provided for
personnel who package the ESDS item(s)I
for delivery?
3. Does packaging provide protection

againt:
a. Triboelectric generation?
b. Direct discharge from a charged

person or object?
c. Electrostatic fields?
4. Are packaging containers properly

marked with the MIL-STD-129 electronicl
device sensitive symbol and ESD cau-
tion?
A-26
-------------------------------------------------------------------------
I ADEQUATE
RESPONSIBILITY .....
FUNCTION YES NO
------- --------- ------ I

------------------------------------------
Quality 1. Are all inspection areas where ESDS
Assurance/ items are handled, inspected or
Reliability tested, ESD protected?

vided fcr all quality functions and
documented in the contractor's qualityl I
assurance manual?
3. Are failure analysis laboratories

equipped with ESD protected areas?
A-27/(A-28 blank)
APPENDIX B
MAINTAINABILITY CHECKLIST
B
MAINTAINABILITY CHECKLIST SECTIONS
PAGE
1.0 Concept Definition B-2
2.0 General Design Criteria B-3
3.0 Fault Indicators B-4
4.0 Adjustments B-5
5.0 Consoles, Cases, and Covers B-6
6.0 Packaging B-7
7.0 External Access B-8
8.0 Lubricants B-9
9.0 Tools B-10
10.0 Labelling B-l1
B-i
1.0 CONCEPT DEFINITION
ADE- IINADE- NIA

QUATE QUATE
1. Simple systems (<2000 parts) are better

not tested. The inhertent reliability
of simple dormant systems is apt to be
more harmed than improved by periodic
testing. If the system design is
simple, would it be best to treat the
system as a "wooden round" and have no
periodic maintenance requirements?
2. Have calculations of the expected

operating and non-operating MTBF
for the given system been made?
3. If the complex system inherent low

reliability will most likely result in al
periodic monitoring/repair main-
tenance concept, has the optimum
periodic monitoring frequency been
determined? (Note - Early dormant
systems were often tested to failure
and overhauled to death. Experience
haR Rhown that reducing the number of
tests and inspections often results in
improved reliability.)
4. Is the selected maintenance concept

compatible with the system's mission
and deployment requirements?
B-2
.0 GENERAL DESIGN CRITERIA
ADE- INADE N/A

QUATE QUATE
1. It a "wooden round" maintenance

concept has been chosen. have the
followin',_ Lactors been considered in
the dc. 'gn: i
o V of 100% burn-in, screening
and preconditioning ot parts or
subsystems, on both electronic
and mechanical items.
o Avoidance of storage-time-critical
components and materials.
o The system should be designed for
simplicity using a minimum of parts.
o Use of an inert atmosphere (such I
as nitrogen).
o Use of transport and storage con-
tainers as tactical packaging to
mitigate the effects of long-term
exposute to life cycle environments.
o interconnection plugs shcild only
be used if no other design is
available.
o Reduction to a minimum of built-in
test points for external testing
as they are unnecessary and may
provide paths for moisture and
contamination intrusion.
o Use of compact structure, combining
PC boards and sealing the system.
o Use of digital electronics with no
field adjustment.
o Manufacturing and quality assurance
methods to increase reliability,
through increased usage of automatic
test equipment during manufacture.
o Complete documentation of manufac-
turing parameters and test results.
o If necessary, incorporation of re-
dundancics.
o Replacement of parts which, due to
their present technology, are time
change items. This should be accom-
plished without any disassembly or
test, and the system should remain
in its container. Replacement of
modules should be limited to the
smallest possible individual element.
[3-3
3.0 FAULT INDICATORS
jADE- {7~ADE- { fl/A

I QUATE QUATE I I
I------I------I----I
1. Have direct fault indications been pro- I I
vided? (This means that either a fault I I
light or an audio signal is provided or I I I
there is a direct indication of a mal- I I I
function through a meter reading or j I I I
similar means.) I I I
I I I I
I I I I
I I I
I I I
I I I
I I I I
I I I
I I I I
I I I I
I I I
I I
I I I
I I I
I I I
I I I
I I I I
I I
I I I I
I I I I
I I I
I I I I
I I I
I I I I
I I I
I I I I
I I I I
I I I I
I I I I
I I I
I I I
I I I I
I I I
I I
I I I
I I I I
I I I
I I I I
I I
I I I
I I I
I I I
I I I
I I
B-4
4.0 ADJUSTMENTS
IADE- I NADE- IN/A

QUATE IQUATE
1. Are positive locking devices provided I I
to assure retention of settings of II I
adjustments or alignment devices that j=
are susceptible to vibration or shock? I I I
III
II I III I I I I I I I I I 1
5.0 CONSOLES, CASES AND COVERS
ADE- INADE- N/A I

QUATE QUATE I
1. Is the base of each console reinforced I
to enable lifting with a forklift I
truck? I
I I
2. Are connectors and other parts which I
normally extend beyond outside covers
or cases recessed or adequately
protected by some other means?
3. Are weight labels provided on each rack

and console?
4. Are cases made enough larger than the

units they cover, so that wires and
other components are not likely to be
damaged when the eases are put on and
taken off?
5. Are guides and tracks provided to help

prevent cases from cocking to one side
and causing damage?
6. When the edges of a case must be slid

over rubber stripping or other scaling
material, is the sealing material
adhered tightly enough so that it does
not buckle or tear, damaging the seal
or jamming the case?
B--6
6.0 PACKAGING
ADE- INADE- N/A

QUATE QUATE
1. Does the overall packaging concept allowi

the removal of a replaceable assembly or
part without removing any other?
2. Have guide pins been provided to facili-I

tate installation of plug-in units?
3. Are plug-in units keyed (by some means

other than the connector) to prevent
accidental insertion in the wrong
location?
4. Does all equipment containing synchros,

electrical resolvers, servo motors, slipi I
rings, and other delicate electrical
rotating equipment, delicate ball
bearings, or high precision gears, cams
and levers have a dust-tight enclosure
or are these items in dust-tight
sections of the equipment?
5. Are units light enough for ease of

removal? Have adequate handling devicesl I
been provided?
6. Is each assembly self-supporting in the

desirable position or positions for
maintenance?
7. Can assemblies be laid on a bench in anyl

position without damaging components?
8. Are guards or other protection provided

for easily damaged conductors such as
waveguides, high frequency cables, or
insulated high voltage cables?
9. Can irregular extensions such as bolts,

cables, wave guides, and hoses be easilyl I
removed before the unit is handle?
Such protrusions are easily damaged and
make handling ditficult.
B-7
7.0 EXTERNAL ACCESS
ADE- INADE- N/A

QUATE QUATE
1. Are access openings without covers used

where this is riot likely to degrade
performance?
2. Is a transparent window or quick-opening I

metal cover used for visual inspection
accesses? Are humidity indicators
clearly 'islble from outside the con-
tainer or package?
3. Are panels, covers, access doors, etc.,

secured to the equipment by means of
fastening devices such as captive springl
bolts or screws, trunk fasteners, dogs, I
levers, and latches, in order to providel
quick and easy access to the interiors
of those portions of the equipment?
4. Have the same size and type of fastenersl

been used for covers and cases on a
given equipment where practicable?
Quick-acting fasteners are recommeded
for dust covers. Fine threaded screws
are recommended for pressurized units.
5. Are environmental factors (cold weather,I

darkness, etc.) considered in design andi I
location of all manipulatible items of
equipment?
B-I
B-8
8.0 LUBRICANTS
ADE- INADE- I N/A

QUATE QUATE
1. Is the type of lubricant to be used and

the frequency of lubrication specified
by a label at/near the lubrication port?!
2. Have the adve- " effects of lubricants,

fuels, and fluids during dormant storagel
been investigated and have suitable
measures been taken to mitigate them?
B-9
9.0 TOOLS
9 ADE- INADE- N/A T
QUATE QUATE
1. Is the equipment accompanied by a

comprehensive list of tools needed for
all maintenance tasks?
2. Have tools been specified to cover all

maintenance procedures?
3. Do fasteners for chassis and panels

require special tools?
4. If special tools are required have they

been provided with the equipment?
5. Are each of the provided tools adequate

for efficient performance of maintenancel
jobs?
6. Are tools which must be used near high

voltages adequately insulated at the
handles and other parts of the tools
which the technician is likely to touch?I
7. Are screwdrivers provided which have

clips to hold free screws which cannot
be easily held with the fingers?
B-10
10.0 LABELLING
I ADE- I INADE- IN/AI

QUATE QUATE
1. Are labels placed for full, unobstructed

view?
2. Does displayed printed matter always

appear upright to the technician from
his normal viewing position?
3. Are display labels imprinted, embossed,

or attached in such a way they will not
be lost, mutilated, or become otherwise
unreadable?
4. When parts MUST be replaced at specifiedl

intervals, are they clearly marked?
5. On equipment utilizing color coding, is

the meaning of colors given in manuals
and on an equipment panel?
M I
B-I1103-12 'blank)
APPENDIX C
TESTABILITY CHECKLIST
C
TESTABILITY CHECKLIST SECTIONS
PAGE
1.0 Concept Definition C-2
2.0 Electrical Design C-3
3.0 Test Points C-5
4.0 Terminals C-6
C-1
1.0 CONCEPT DEFINITION
ADE- INADE- 1 N/A

QUATE I QUATE J
1. If a "vooded round" maintenance concept

:ias been chosen, does the testability
concept support its objectives?
o Simplicity of design
o Minimization of parts to maximize
reliability
o Elimination of paths for moisture
oc contamination intrusion
o No checkout before use
o Maintenance free storage
2. If a Periodic Monitoring/Repair
maintenance concept has been chosen,
does the testability concept support
it objectives?
3. Does the testability concept

appropriately address stockpile testing
requiremFents?

minimize "on" time to avoid degrading
reliability?

minimize the possibility of test
induced failures or equipment
degredation?

provide for the retention of the
maximum amount of test parameter data
for failure prognosis and trend
analysis?
C-2
2.0 ELECTRICAL DESICN FOR TESTABILITY
ADE- INADE- N/A

QUATE QUATE
1. Have the following been considered to

make maintenance easier:
o Units should be able to be checked

independently?
o Group the circuits so that a mini-

mum of crisscrossing of signal units
is required?
o Modules should have as few input and

output signals as possible?
o Avoid "trick" or extremely sensitive

circuitry; use standard circuits?
o Separate operational circuitry from

maintenance circuitry?
o Design for maximum use of maintenance!

aids (troubleshooting procedures,
maintenance diagrams, and circuit
data)?
o Design maintenance procedures con-
current with equipment design?
2. Has consideration been given towards

isolating trouble areas without complete! i
equipment power up?
3. Are testing, alignment, and repair pro-

cedures such that a minimum of knowledgel
is required on the part of maintenance
personnel? Can trouble shooting of an
assembly take place without removing it
from a major unit?
4. Is the design such that no unrealistic

requirements for special facilities for
maintenance, storage. or shipment are
imposed?
5. Is the design such that no unnecessary

requirements for a special maintenance
environment (e.g., ground power carts,
cooling, special primary power, etc,)
arc imposed?
C-3
I ADE- I INADE- j N/A

QUATE I
QUATE I
I-----------I----I
6. Are components that are designed to I I I I
respond to environmental stimulae j I I
easily disconnected so that they can I
be connected to simulators for I I I I
individual testing? I I I I
I I I I
I I I I
I I I I
I I I
I I I
I I I
I I I
I I I
I I I
I I I I
I I I I
I I I I
I I I I
I I I I
I I I I
I I I I
I
I I I
I
I I
I I I I
I I I I
I I I I
I I I I
1 I 1 1
I I I I
I I I
I I I I
I I I I
I I I I
1 I I I
I I I I
I I I I
I I I I
I I I
I I I I
I I I I
I I I
I I I
I I I I
I I I I
I I I I
I I I I
I I I I
I I I
I I I I
C-4
3.0 TEST POINTS
ADE- INADE- 1 N/A

QUATE QUATE
1. Are all test points placed or guarded

so that the technician is physically
prevented from contacting high voltage?
2. Are test points accessible in the par-

ticular installation?
3. Are there test points available for the

direct check of all replaceable parts?
4. Are routine test points provided which

are available to the technician without
removing the chassis from the cabinet?
5. If unstable components can be identi-

fied or cxpected in equipment, are
check and adjustment points available
and convenient for checking them?
b. Are test points contiected so that no
damage to the system can occur by
introduction of an incorrect signal?
7. Are test points compatible with planned

test equipment?
8. Are built-in test features provided

wherever standard portable test equip-
ment cannot be used.
9. Have adequate precautions been taken to

prevent test points from providing a
path for moisture and contamination
intrusion in the equipment?
C-5
4.0 TERMINALS
ADE- INADE- N/A

QUATE QUATE
1. Are terminal strips and standoffs

located and labeled so that they can
serve as convenient check points while
the equipment is in operation?
2. Are terminal strips mounted so as to be

accessible from the front of the equip-
ment when opened?
3. Unless otherwise specified, are all

external cables and interconnecting
harnesses used to connect major units ofl
sets or systems, terminated on barrier
type terminal strips?
4. Are terminal strips located in positions

which will best facilitate testing of
the equipment?
C-6
APPENDIX D
LIST OF ACRONYMS
D
List of Acronyms
TERM DEFINITION
ac Alternating Current
AFSC Air Force System Command
AG Availability Guarantee
ALT Accelerated Life Testing
ATE Automatic Test Equipment
AUR All-Up-Round
BIT Built-In-Test
BTU British Thermal Units
CBS Cost Breakdown Structure
CCG Captive Carry Guarantee
CERT Combined Environments Reliability Test
CMOS Complementary Metal Oxide Semi.conductor
COD Correction Of Deficiencies
CODSIA Council of Defense and Space Industries Association
CPFF Cost Plus Fixed Fee
dB Decibels
dc Direct Current
DH Design Handbook
DoD Department of Defense
DSS Decision Support System
ECAP Electronic Circuit Analysis Program
ECP Engineering Change Proposal
EMP Electromagnetic Pulse
ESS Environmental Stress Screening
FSD Full-Scale Development
GFE Government Furnished Equipment
GN2 Gaseous Nitrogen
IA Incentive Award
ILS Integrated Logistic Support
LCC Life Cycle Cost
LEC Lockheed Electronics Company, Inc.
D1
List of Acronyms
TERM DEFINITION
LL Lower Limit
LRU Line Replaceable Unit
LSI Large Scale Integration
LTPD Lot Tolerance Percent Defective
MDG Mission Dependability Guarantee
MDT Mean Down Time
MIL Military
MIL-HDBK Military Handbook
MIL-STD Military Standaid
MOS Metal Oxide Semiconductor
mph Miles Per Hour
MSI Medium Scale Integration
MTBF Mean Time Between Failure
MTBMA Mean Time Between Maintenance Actions
NPV Net Present Value
O&M Operation and Maintenance
O&S Operation and Support

PAWS Platoon Eatly Warning System
PEDs Plastic Encapsulated Devices
PERT/CPM Program Evaluation Review Technique/Critical Path
Method
PIND Particle Impact Noise Detection
PPA Product Performance Agreenent
psi Pounds Per Square Inch
PSM PPA Selection Matrix
QPL Qualified Parts List
R&M Reliability and Maintainability
RADC Rome Air Developm,:.nt Centet
RIW ReliabilitY Improvement Warranty
D-:1
List of Acronyms
TERM DEFINITION
SCEPTRE Systems for Circuit Evaluation and Prediction of

Transient Radiation Effects
SRU Ship Replaceable Unit
SSI Small Scale Integration
SVG Storage Verification Guarantee
TREE Transient Radiation Effects on Electronics
TRMF Theater Readiness Monitoring Facility
UL Upper Limit
VHSIC Very High Speed Integrated Circuit
VLSI Very Large Scale Integration
4
D_3/)-- blank)

Radc TR 88 110 - 1988

Uploaded by

Copyright:

Available Formats

Radc TR 88 110 - 1988

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Radc TR 88 110 - 1988

Uploaded by

Copyright:

Available Formats

Downloaded from http://www.everyspec.

Lockheed Electronics Company, Inc. DT[7

APPROVED FOR PUBLIC RELEASE; DISTRIBU7ION UNLIMITED.

ROME AIR DEVELOPMENT CENTER

Rome Air Development Center RJXE-2 P30602.-85-C-0040

Security Cawication)620F 1 3802G

IinI FROM ~a L&L.TOEF .. 8 I Max 1988

i.0 INTRODUCTION 1-1

1.1 The Need for Dormancy Design Guidelines 1-1

2.0 HOW TO USE THESE GUIDELINES 2-1

3.0 DORMANCY ENVIRONMENTS 3-1

3.1 System Life Cycle Profile 3-1

3.2.1 Natural Environmental Factors 3-14

3.2.3.1 Temperature 3-19

3.2.4 Induced Environmental Factors 3-37

3.2.5.1 Mechanical Factors 3-41

3.2.5.1.1 Vibration 3-41

TABLE OF CONTENTS (Continued)

3.2.5.2 Nuclear Radiation 3-45

3.3 Dormant Storage 3-51

3.3.1 Storage Types, Capacities, and Limitations 3-51

3.4 References 3-54

4.0 DESIGN GUIDELINES 4.1-1

4.1 Part Selection/Control 4.1-1

4.1.1 Resisters 4.1-7

4.1.1.1 General Characteristics of Fixed Resistors 4.1-13

4.1.1.1.1 Fixed Composition Resistors 4.1-13

4.1.1.2 Variable Resistors 4.1-22

4.1.1.2.1 Low Precision Variable Composition

4.1.1.3 Special Resistors 4.1-31

4.1.1.3.1 Fixed Film Network Resistors 4.1-31

4.1.2 Capacitors 4.1-32

4.1.2.1 General Characteristics of Fixed Capacitors 4.1-35

4.1.2.1.1 Fixed Paper and Plastic Film Dielectric

TABLE OF CONTENTS (Continued)

4.1.2.1.2 Fixed Glass and Mica Dielectric

4.1.i.2 General Characteristics of Variable

4.1.2.2.1 Variable Ceramic Dielectric Capacitors 4.1-53

4.1.3 Microci.cuits 4.1-53

4.1.3.1 Characteristics of Hermetic Microcircuits 4.1-66

4.1.3.1.1 Hermetic Monolithic Microcircuits 4.1-67

4.1.3.2 General Characteristics of Non-Hermetic

4.1.3.2.1 Plastic Encapsulated Devices 4.1-81

4.1.3.3 VLSI/VHSIC Microcircuits 4.1-85

4.1.4 Discrete Semiconductors 4.1-91

4.1.6.1 Solid-State Relays 4.1-96

4.1.6.2.1 Design Guidelines 4.1-97

4.1.7 Switches 4.1-99

4.1.7.1 Guidelines 4.1-99

TABLE OF CONTENTS (Continued)

4.1.7.1.1 Design Guidelines 4.1-100

4.1.7.2 Environmental Effects/Guidelines 4.1-102

4.1.8 Connectors 4.1-105

4.1.8.1 Guidelines 4.1-106

4.1.9 Cables/Wires 4.1-110

4.1.9.1 Environmental Effects/Guidelines 4.1-111

4.1.10 Batteries 4.1-113

4.1.10.1 Design Guidelines 4.1-117

4.1.11 Fiber Optic Components 4.1-124