FOURAH BAY COLLEGE

INSTITUTION

ELECTRICAL AND ELECTRONIC ENGINEERING

DEPARTMENT

ADVANCED DIGITAL SYSTEMS DESIGN

MODULE TITLE

EENG424
MODULE CODE

PROF JONAS A S REDWOOD-SAWYERR

MODULE LECTURER
Mobile : +232 76 670904, +232 88 001019, +232 25 273401
Email : [email protected], [email protected],
CONTACTS
[email protected]

LECTURE HOURS/LAB. 3-0-3

PRACTICALS/CREDIT HOURS
Examination will account for 70% of the overall grade whilst continuous assessment will account
ASSESSMENT for 30%. Question sets at the end of each lecture must be returned for grading one week after
receipt of lectures and will account for 10% of the continuous assessment grade.
 This module draws from a number of concepts met in earlier years of the programme and seeks
to show how these engineering tools can be used in digital systems design. You may wish to
revise Modules in Digital Systems at Year IV so as to better cope with the discussions as these
OBJECTIVES AND OUTCOMES concepts will be assumed during our lectures.

References have been made to a number of text books which are listed in the section for Further
Reading, and modifications made for a coherent and reader-friendly presentation with worked
examples and assignments provided.

Below is a list of the topics to be covered.

 RELIABILITY: Failure Rate Curve – Burn-in; Useful life and Wear-out; Systems Reliability (R(t)
= exp(-λt), where λ is defined as the Failure Rate
 Mean time between failures-MTBF and Mean time to failures-MTTF; Mean Time To Repair-
MTTR; Availability and the factors that affect it; Type of failures; Reliability function (series,
MODULE DESCRIPTION
parallel and series-parallel); Factors that affect reliability; Maintainability.
 FAULT DIAGNOSIS AND TESTING: Fault detection and Location; Gate sensitivity; Fault-
test for a 2-input AND gate; Undetectable faults; Bridging faults; Fault detection table; Fault
Library; Two-level circuit fault detection in AND/OR circuit; Two-level circuit fault detection
in OR/AND circuit; Boolean difference; Testing Techniques; Designing for Testability.

SUGGESTED References
1. http://www.reliabilityeducation.com/ReliabilityPredictionBasics.pdf
READINGREFERENCE
2. http://www.mtl-inst.com/images/uploads/datasheets/App_Notes/AN9030.pdf
TEXTS/MANUALS/WEBSITES 3. http://en.wikipedia.org/wiki/Failure_rate
 4. G C Loveday,1989. Electronic Testing and Fault Diagnosis. Longman Scientific & Technical. ISBN
0-582-03865-0

02 – Systems Reliability analysis

LECTURE NUMBER
LECTURE DURATION (HOURS) 01
SPECIFIC INSTRUCTIONAL At the end of this lecture the students should be able to :
OBJECTIVES AND LEARNING
1. Determine composite(overall) reliability figures for combinational systems
OUTCOMES
2. Calculate the overall reliability of systems using the Un-Reliability method

1.6 Systems’ models for Reliability determination

The Reliability of a system can be derived in terms of the reliabilities or the failure rates of the sub-systems used to build it. Two limiting
cases of system design are available, i.e..

a) Systems in which each sub-system must perform or function properly, if the system as a whole is to function properly.
b) Systems in which the correct operation of just one of the sub-system is sufficient for the system as a whole to function satisfactorily,
i.e. the system consists of redundant sub-systems and will fail only if all sub-systems fail.

Scenario A

This can be modelled as a series system as shown :

1 2 3 N

If the sub-system failures are independent and Ri is the reliability of sub-system i, then the overall system reliability is given by
 𝑁

𝑅𝑜𝑣 = ∏ 𝑅𝑖
𝑖=1
...1.19
In the useful life period or constant failure rate case, we obtain

𝑅𝑖 = exp⁡(−𝜆𝑖 𝑡)
Giving

𝑅𝑜𝑣 = ∏ exp⁡(−𝜆𝑖 𝑡)
𝑖=1
...1.20
i.e
𝑁

𝑅𝑜𝑣 = exp⁡[∑ −𝜆𝑖 𝑡]

𝑖=1
...1.21
If the N sub-systems have identical failure rates, i.e. λi = λ, then Ri = R. We therefore obtain

𝑅𝑜𝑣 = exp(−𝑁𝜆𝑡) = 𝑅 𝑁
...1.22
Giving

1
𝑀𝑇𝐵𝐹 =
𝑁𝜆
...1.23
NB.
a) The overall reliability of a series connected system is decreased by N-fold
b) The MTBF of the system is given by (1/N) times that of the sub-system.
Example 1.3

(i) Let Ri = 99% and N = 10. Find Rov in a series connected system.

Solution

Rov = RN = 0.9910 ≅ 0.9

(ii) Let Ri = 0.5 and N = 10, find Rov

Solution

Rov = 0.510 = 9.77 x 10-4

Hence in a series connected system the reliabilities of the individual sub-systems must be very high if the overall reliability of the system is
to be high.

Scenario II

Parallel connected sub-systems

If the sub-system failures are independent and Ri is the reliability of sub-system i, then the overall system reliability is given by
 𝑁

𝑅𝑜𝑣 = 1 − ∏(1 − 𝑅𝑖 )
𝑖=1
...1.24

= 1 – {(1-R1)(1-R2)(1-R3)…..(1-RN)}

Also if all the sub-systems are identical and each with a constant failure rate of λ, then we obtain

𝑅𝑜𝑣 = 1 − [1 − exp(−𝜆𝑡)]𝑁
...1.25
i.e.

Rov = 1 – (1 – R)N
...1.26
Example 1.4

(i) Let N = 10 and Ri = 0.75, Find Rov for the system.

Solution

For this case the overall Reliability is given by

Rov = 1 – (1 – 0.75)10 = 0.99999

(ii) Let N = 10 and Ri = 0.5, Find Rov.

Substituting in Eqn 1.26 yields Rov = 0.99902

In general the MTBF of a parallel system with N identical sub-systems is

 𝑁
1
𝑀𝑇𝐵𝐹𝑜𝑣𝑒𝑟𝑎𝑙𝑙 = [∑ ⁡] 𝑥⁡𝑀𝑇𝐵𝐹𝑖
𝑖
𝑖=1
..1.27
where MTBFi is the MTBF of each of the identical sub systems.

Example 1.5.

Let N = 2 and Failure rate = λ for each sub system. Find Rov

Now
𝑅𝑜𝑣 = 1 − [1 − exp(−𝜆𝑡)]𝑁

i.e. for N=2 we obtain

𝑅𝑜𝑣 = 1 − [1 − exp(−𝜆𝑡)]2

= 2 exp(−𝜆𝑡) − exp⁡(−2𝜆𝑡)
Now the MTBF of the system is given by

∞ ∞
3
𝑀𝑇𝐵𝐹𝑠𝑦𝑠𝑡𝑒𝑚 = ∫ 𝑅(𝑡)𝑑𝑡 = ∫[2 exp(−𝜆𝑡) − exp⁡(−2𝜆𝑡)]𝑑𝑡 =
2𝜆
0 0
Using Eqn. 1.27 we can write

𝑁
1 3
𝑀𝑇𝐵𝐹𝑜𝑣𝑒𝑟𝑎𝑙𝑙 = [∑ ⁡] 𝑥⁡𝑀𝑇𝐵𝐹𝑖 =
𝑖 2𝜆
𝑖=1

Combination of parallel and series sub systems

In practice a system normally consist of a combination of series and parallel sub systems as shown
Case A

A B

C D

i.e. a parallel to series interconnection

Case B.

A series to parallel interconnection

A B

C D

Let us consider Case A as shown

A B

C D

The Reliability of the system in case A, i.e. the parallel to series interconnection can be obtained as follows through a system reduction
technique:
 A B

C D

RA//C = 1-(1-RA)(1-RC)
A/C B/D
RB//D = 1-(1-RB)(1-RD)

Giving

RPS = RAC * RBD = [1-(1-RA)(1-RC)][1-(1-RB)(1-RD)]

..1.28

For case B i.e. the series to parallel interconnection we obtain

A B

C D
 A-B

C-D
RAB = RA.RB

RCD = RC.RD

Giving

RSP = [1-(1-RARB)(1-RCRD)]
...1.29

If RA=RB=RC=RD=R, then we obtain

RPS = R4-4R3-4R2

and

RSP = 2R2-R4

In general RPS > RSP.

Alternative calculation of overall Reliability of parallel interconnected systems

The Unreliability of a system is given by

Q(t) = 1 – R(t)
Let us consider two systems, X and Y connected in parallel with reliabilities Rx and Ry.

We therefore obtain for subsystem X, the Unreliability is given by

Qx = 1 – Rx

and for subsystem Y we obtain

Qy = 1 - Ry

The probability of failure of the units in parallel is the product of their Un-reliabilities

i.e. Qxy = Qx.Qy ...1.30

The system reliability is given by

Rxy = 1 – Qxy = 1 – QxQy ...1.31

= 1-(1-Rx)(1-Ry)

= 1-(1-Ry-Rx-RxRy)

= Rx + Ry – RxRy ...1.32

This confirms one of the laws of probabilities., which states :

If a and b are two events which can occur together, then the probability of either or both events taking place is given by

pa + pb - papb ...1.33
Example 1.6

Let Rx = 0.85 and Ry = 0.75, Find Rxy.

Solution

Qx = 1 – Rx = 1 – 0.85 = 0.15, i.e. the unit’s unreliability

Qy = 1 – Ry = 1 – 0.75 = 0.25

Qxy = Qx.Qy ; from Eqn 1.30

i.e. Qxy = 0.15 x 0.25 = 0.0375

The system reliability over 1000 hours is therefore given as

Rxy = 1 - Qxy = 1 – 0.0375 = 0.9625 = 96.25%.

As before the system reliability has been greatly enhanced through a parallel interconnection of its sub systems, i.e. redundancy has been
introduced into the system’s operation.

(discuss redundancy in power systems and telecommunications.)

For the same units connected in series we obtain

Rxy = Rx.Ry = 0.85 x 0.75 = 63.75%.

This is an overall reduction in the system’s reliability.

Question Set

1. A system comprising a series interconnection of a power supply, an oscillator and an amplifier is to be analysed. Calculate the individual and
overall system reliabilities for a 1000 hrs operating period if the MTBF for the sub systems are : 20,000 hrs, 100,000 hrs, and 50,000 hrs
respectively. [6 marks]
2. The system in Fig.Q2 has the specifications given in Table 1. Determine the following :

a. The overall system’s Reliability [6 marks]

b. The Mean Time Between Failure (MTBF) of the system for an observation period of 1000 hours [3 marks]
c. The constraint time for the A-B-C-D-E-F sub-system, given that its overall failure rate is 60x10-6 per hour. [4 marks]

A B G
Fig. Q2
C D H

E F I

Table Q1
System A B C D E F G H I
R(t) 0.75 0.75 0.84 0.88 0.90 0.95 0.93 0.85 0.65

3i. What is the key defining characteristic of a Maintainable system? Derive the components that constitute a system’s repair time. [3 marks]
ii. A maintenance engineer observes the following periods (in hours) between failures of four subsystems in a plant given in Table Q1.
 Table Q1.
Subsystem T1(x103) T2(x103) T3(x103) T4(x103)
1 40 40 55 60
2 40 30 45 50
3 25 25 45 40
4 30 35 40 45

i. Find the Reliability of the complete system for a constraint time of 1000 hours when
a) All subsystems are connected in series. [5 marks]
b) All subsystems are connected in parallel [7 marks]
ii. For a system Availability of 85%, and a constraint time of 1000 hours, find the Repair Time of the interconnected systems in Part a) above. [5 marks]

MODE AND DEADLINE FOR SUBMISSION OF

ASSIGNMENT Send by Whatsapp images or email. Due date: TBD.