Laboratory Soils Testing

This document downloaded from All of the information, data and computer software
vulcanhammer.net
("information") presented on this web site is for
general information only. While every effort will
be made to insure its accuracy, this information
should not be used or relied on for any specific
application without independent, competent
professional examination and verification of its
accuracy, suitability and applicability by a licensed
professional. Anyone making use of this
since 1997, information does so at his or her own risk and
assumes any and all liability resulting from such
your source for engineering information use. The entire risk as to quality or usability of the
information contained within is with the reader. In
for the deep foundation and marine no event will this web page or webmaster be held
liable, nor does this web page or its webmaster
construction industries, and the historical provide insurance against liability, for any
damages including lost profits, lost savings or any
site for Vulcan Iron Works Inc. other incidental or consequential damages arising
from the use or inability to use the information
contained within.
Use subject to the fine print to the This site is not an official site of Prentice-Hall, the
University of Tennessee at Chattanooga, Vulcan
right. Foundation Equipment or Vulcan Iron Works Inc.
(Tennessee Corporation). All references to
sources of equipment, parts, service or repairs do
not constitute an endorsement.
Dont forget to visit our companion site http://www.vulcanhammer.org

REPRINT WITH CORRECTED COPIES INCORPORATED,
ENGINEER MANUAL E M 1110-2-1906 _

30 November 1970
Change 2 dated 20 August 1986

has been inserted.
ENGINEERING AND DESIGN
LABORATORY SOILS TESTING
HEADGUARTERS, DEPARTMENT OF THE ARMY

OFFICE OF THE CHIEF OF ENGINEERS
-Foreword
Thin manual wao prepared at the U. S. Army Engineer

Waterways Experiment Station (WES) under dipection from the
Office, Chief of Engineer8 (OCE). General supervision and tech-
nical guidance were provided by R. A. Barron, Chief, Soil Me-
chanics Branch, Engineering Division, OCE, and M. M. Johnston
and R. R. W. Beene of that Branch.
The manual was prepared by the staff of the Embankment
an+ Foundation Branch, Soil6 Division, WES; the principal authors
were B. N. MacIver and G. P. Hale.
-
DEPARTMENT OF TEE ARMY EM 1110-2-1906
Office of the Chief of Engineers Change 1
DAEN-CWE-SS Washington, D.C. 20314
Engineer Manual
No. 1110-2-1906 3 May 1980
Engineering and Design

1. This change to EM 1110-2-1906, 30 November 1970, provides the

updating of the TABLE OF CONTENTS and the addition of the following
four (4) APPENDICES:
a. Appendix VIII A; Swell and Swell Pressure Tests
b. Appendix XA: Cyclic Triaxial Tests
c. Appendix XB: Determination of Critical Void Ratio
d. Appendix XIII: Pinhole Erosion Test for Identification of

Dispersive Clays
2. Substitute the attached pages as shown below:
Remove pages i, ii, iii, Insert new pages i, ii, iii, iv,
iv, v, vi, and vii v, vi, vii, viii, ix, x, xi, xii,
and xiii
3. Add Appendices VIII A, XA, XB, and XIII.
4. File this change sheet in front of the publication for future

reference purposes.
FOR THE GRIEF OF ENGINEERS:
FORREST T. GAY, II:

Colonel, Corps of Engineers
Executive Director, Engineer Staff
THIS CONSTITUTES CmNcx 1 TO EM 11 lO-2--1906, 30 NOv 70
DEPARTMENT OF THE ARMY E M iiiO-2-i906

U. S. Army Corps of Engineers Change 1
DAEN-ECE-G Washington, D. C. 20314-1000 1 I&y 80
Engineer Manual 30 November 1970

No. 1110-2-1906
ENGINEERING AND DESIGN
TABLE OF CONTENTS
Paragraph Page
1. PURPOSE .................................................
2. APPLICABILITY ...........................................
3. NOTATIONS ...............................................
4. REFERENCES ..............................................
5. TESTING PROCEDURES ......................................
6. RELIABILITY OF TESTING APPARATUS AND
RESPONSIBILITY OF PERSONNEL ............................
7. LABORATORY FACILITIES ...................................
8. SAMPLE HANDLING AND STORAGE .............................
9. SELECTION AND PREPARATION OF
TEST SPECIMENS ......................................... 5
10. DATA SHEETS AND REPORT FORMS ............................ 6
APPENDIX I:
WATER CONTENT - GENERAL
* 1. DEFINITION .............................................. l-l

2. APPARATUS ............................................... I-l
3. SPECIMEN ................................................ I-l
4. PROCEDURE ............................................... I-2 'X
This regulation supersedes EM 1110-2-1906, 10 May 65.
i
EM 1110-2- 1906
Change 2
20 Aug 86
Paragraph Page
5. COMPUTATIONS . . . . . . . . . . . . . . . . . . . . . . . . I- 3
6. POSSIBLE ERRORS. . . . . . . . . . . . . . . . . . . . . . . I- 3
APPENDIX II:
UNIT WEIGHTS, VOID RATIO, POROSITY, AND
DEGREE OF SATURATION
1. RELATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . II- 1
2. DEFINITIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . II- 2
3. VOLUMETRIC METHOD . . . . . . . . . . . . . . . . . . . II- 2
4. DISPLACEMENT METHOD. . . . . . . . . . . . . . . . . . II- 8
5. POSSIBLE ERRORS. . . . . . . . . . . . . . . . . . . . . . . II- 11
APPENDIX III:
LIQUID AND PLASTIC LIMITS
* 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . III- 1
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . . . III- 2
3. CALIBRATION OF APPARATUS . . . . . . . . . . . . . . III- 7
4. PREPARATION OF MATERIAL. . . . . . . . . . . . . . . III- 10
5. LIQUID LIMIT . . . . . . . . . . . . . . . . . . . . . . . . . . III- 13
6. PLASTIC LIMIT . . . . . . . . . . . . . . . . . . . . . . . . . III- 17
7. PLASTICITY INDEX . . . . . . . . . . . . . . . . . . . . . . III-20
8. REPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III- 2 0
9. PREPARING CLAY SHALE MATERIAL

FOR TESTING.. . . . . . . . . . . . . . . . . . . . . . . . III- 2 1 >k
APPENDIX IIIA:
ONE-POINT LIQUID LIMIT TEST
:: 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . IIIA- 1
2. APPARATUS AND PREPARATION OF SAMPLE. . . . m.A- 1
3. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . . . IIIA- 1
4. CALCULATIONS . . . . . . . . . . . . . . . . . . . . . . . . . IIJA-2
5. POSSIBLE ERRORS. . . . . . . . . . . . . . . . . . . . . . . IIIA-3 *
ii
EM 1110-2-1906
Change 2
20 Aug 86
Paragraph Page
APPENDIX IIIB:
SHRINKAGE LIMIT TEST
1. DEFINITION . . . . . . . . . . . . . . . . . . . . . . IIIB-1
APPARATUS IIIB-1
:: PREPARATION.0; MA;ERIAL* : : : : : : : : : : : : : : : IIIB-1
4. PROCEDURE 1118-3
5. COMPUTATIONS : : : : : : : : : : : : : : : : : : : : : 1118-4
POSSI8LE ERRORS IIIB-6
;: HANDLING OF MERCURY. : : : : : : : : : : : : : : : : : IIIB-6
APPENDIX IV:
SPECIFIC GRAVITY
TYPES OF*SPECIFIC GRAVITY .............. IV-1

;: SPECIFIC GRAVITY OF SOLIDS .............. IV-1
3. APPARENT AND BULK SPECIFIC GRAVITY . . . . . . . . . . IV-7
4. POSSIBLE ERRORS . . . . . . . . . . . . . . . . . . . IV-10
APPENDIX V:
GRAIN-SIZE ANALYSIS
1. DEFINITION . . . . . . . . . . . . . . . . . . . . . . V-l
SIEVE ANALYSIS . . . . . . . . . . . . . . . . . . . . V-l
:: HYDROMETER ANALYSIS ................. V-8
4. COMBINED ANALYSIS v-21
5. PROCEDURES FOR PREPARiNG CLAY' l l l l
SHALE MATERIAL ................... V-24

6. POSSIBLE ERRORS ................... V-24
APPENDIX VI:
COMPACTION TESTS
1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . VI-1
* STANDARD COMPACTION TEST . . . . . . . . . . . . . . . VI-1
:: MODIFIED COMPACTION TEST . . . . . . . . . . . . . . . VI-15
4. 15-BLOW COMPACTION TEST . . . . . . . . . . . . . . . VI-16
5. POSSI8LE ERRORS . . . . . . . . . . . . . . . . . . . VI-16 *
iii
EM 1110-2-1906
Change 1
1 May 80
Paragraph Page
APPENDIX VIA:
COMPACTION TEST FOR EARTH-ROCK MIXTURES
5; 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . VIA- 1
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . VIA- 1
3. QUANTITY OF SAMPLE ................. VIA-4
4. PROCESSING OF SAMPLE ................ VIA-4
5. SPECIAL CONSIDERATIONS .............. VIA- 6
6. PREPARATION OF TEST SPECIMEN ........ VIA- 6
7. COMPACTIONPROCEDURE . . . . . . . . . . . . . . VIA- 6
a. COMPUTATIONS ...................... WA-0
9. PRESENTATION OF RESULTS ............. VIA- 8
10. POSSIBLE ERRORS ..................... V I A - % 9r
APPENDIX VII:
PERMEABILITY TESTS
1. DARCYS LAW FOR FLOW OF WATER THROUGH

SOIL.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII- 1
2. TYPES OF TESTS AND EQUIPMENT ......... VII- 2
3. CONSTANT-HEAD PERMEABILITY TEST
WITH PERMEAMETER CYLINDER ......... VII- 5
4. FALLING-HEAD PERMEABILITY TEST
WITH PERMEAMETER CYLINDER ......... VII- I3
5. PERMEABILITY TESTS WITH SAMPLING
TUBES . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII-i6
6. PERMEABILITY TEST WITH PRESSURE
CHAMBER.. . . . . . . . . . . . . . . . . . . . . . . . . VII-17
7. PERMEABILITY TESTS WITH BACK
PRESSURE . . . . . . . . . . . . . . . . . . . . . . . . . VII- I7
8. PERMEABILITY TESTS WITH
CONSOLlDOMETER . . . . . . . . . . . . . . . . . . . . VII- 22
9. POSSIBLE ERRORS . . . . . . . . . . . . . . . . . . . . . VII- 24
iv
E M 1430-2-4906
Change 1
1 May 80
Paragraph Page
APPENDIX VIII:
CONSOLIDATION TEST
* 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . VIII- 1
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . VIII- 1
3. CALIBRATION OF EQUIPMENT ............ VIII- 5
4. PREPARATION OF SPECIMENS ............ VIII- 6
5. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . VIII - 7
6. COMPUTATIONS ...................... VIII- 11
7. PRESENTATION OF RESULTS ............. VIII-43
8. POSSIBLE ERRORS ..................... V I I I - 1 5 +t
* APPENDIX VIIIA:
SWELL AND SWELL PRESSURE TESTS
1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . VIIIA- 1
2. APPARATUS, CALIBRATION OF EQUIPMENT,

A N D P R E P A R A T I O N O F S P E C I M E N S ........ VIIIA- 1
3. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . VIIIA- 2
4. COMPUTATIONS ...................... VIIIA- 4
5. PRESENTATION OF RESULTS ............. VIIIA- 5
6. POSSIBLE ERRORS ..................... V I I I A - 8 st
APPENDIX IX:
DRAINED (S) DIRECT SHEAR TEST
* 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . IX-4
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . IX-1
3. PREPARATION OF SPECIMEN ............. IX-5
4. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . IX-~
5. COMPUTATIONS ...................... IX-12
6. PRESENTATION OF RESULTS ............. IX-i4
7. SIMPLIFIED PROCEDURE FOR DRAINED (S)

DIRECT SHEAR . . . . . . . . . . . . . . . . . . . . . . IX-15 *
V
EM iflO-2-i906
Change 1
PMay 80
Paragraph Page
* 8. POSSIBLE ERRORS ..................... IX-16
APPENDIX IXA:
DRAINED (S) REPEATED DIRECT SHEAR TEST
* 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . IXA- 1
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . IXA- 1
3. SPECIMEN PREPARATION ................ IXA-i
4. PROCEDURE . . . . . . . . . . . ..< . . . . . . . . . . . IXA- 2
5. C O M P U T A T I O N S ....................... IXA- 5
6. PRESENTATION OF RESULTS .............. IXA- 5
7. POSSIBLE ERRORS ..................... I X A - 5 +c
APPENDIX X:
TRIAXIAL COMPRESSION TESTS
1. PRINCIPLES OF THE TRIAXIAL COMPRESSION

TEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X-i
2. TYPES OF TESTS ...................... x-2
3. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . x-3
4. P R E P A R A T I O N O F S P E C I M E N S ............. X-i1
5. Q TEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X-i8
6. Q TEST WITH BACK-PRESSURE SATURATION. .. X-27
7. RTEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X-29
8. S TEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X-38
9. POSSIBLE ERRORS ..................... x-44
* APPENDIX XA:
CYCLIC TRIAXIAL TESTS
4. PRINCIPLES OF CYCLIC TRIAXIAL TESTING l . l

XA- i
2. SPECIMENS .......................... XA-3
3. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . XA-4
4. T E S T I N G P R O C E D U R E S .................. XA-15 *
vi
E M 1110-2-1906
Change 1
1May 80
Paragraph Page
* 5. COMPUTATIONS ...................... XA-24

6. PRESENTATION OF RESULTS ............. XA-25
7. CALIBRATION OF EQUIPMENT AND TESTING
PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . XA-27
8. POSSIBLE ERRORS ..................... XA-29 *
* APPENDIX XB:
DETERMINATION OF CRITICAL VOID RATIO
i. DEFINITION OF CRITICAL VOID RATIO ....... XB-1

2. PRINCIPLES OF MONOTONIC TRIAXIAL
TESTING .......................... XB-1
3. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . XB-3
4. T E S T I N G P R O C E D U R E S .................. XB-6
5. COMPUTATIONS ...................... XB- 9
6. PRESENTATION OF RESULTS ............. XB-9
7. POSSIBLE ERRORS ..................... XB-10 sc
APPENDIX XI:
UNCONFINED COMPRESSION TEST
* 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . XI- 1
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . XI- 1
3. PREPARATION OF SPECIMENS ............ XI- 3
4. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . XI- 6
5. C O M P U T A T I O N S ....................... XI- 6
6. PRESENTATION OF RESULTS ............. XI- 7
7. POSSIBLE ERRORS ..................... XI- 8
8. USE OF OTHER TYPES OF EQUIPMENT FOR
UNDRAINED SHEAR STRENGTH
D E T E R M I N A T I O N S .................... X I - 8 .tt
vii
E M itlO-2-1906
Change 1
1 May 80
Paragraph Page
APPENDIX XII:
RELATIVE DENSITY
* 4. GENERAL . . . . . . . . . . . . . . . . . . . . . . . . . . . XII- 1
2. APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . XII- 2
3. CALIBRATION OF EQUIPMENT ............ XII- 6
4. PREPARATION OF SAMPLE .............. XII- 6
5. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . XII- 7
6. COMPUTATIONS .............. 1 ....... XII-ii
7. POSSIBLE ERRORS . . . . . . . . . . . . . . . . . . . . . XII-i2 *
APPENDIX XIIA:
MODIFIED PROVIDENCE VIBRATED DENSITY TEST
* 1. INTRODUCTION . . . . . . . . . . . . . . . . .e. . . . . XIIA- 1

2. APPARATUS . . . . . . . . . . . . . . . . . . . . ..d.. XIIA- 1
3. CALIBRATION OF EQUIPMENT . . . . . . . . . . . . XIIA- 4
4. PREPARATION OF SAMPLE . . . . . . . . . . . . . . XIIA- 5
5. PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . XIIA- 5
6. COMPUTATIONS . . . . . . . . . . . . . . . . . . . . . . XIIA- 6
7. POSSIBLE ERRORS . . . . . . . . . . . . . . . . . . . . . XIIA-6 *
* APPENDIX XIII:
PINHOLE EROSION TEST FOR IDENTIFICATION OF
DISPERSIVE CLAYS
i. CHARACTERISTICS OF DISPERSIVE CLAYS. ... XIII-:

2. IDENTIFICATION OF DISPERSIVE CLAYS ..... XHI- ii
3. WES PINHOLE EROSION APPARATUS ........ XIII-S
4. CALIBRATION OF PINHOLE EROSION
APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . X :II- .1
5. PREPARATION OF COMPACTED SPECIMENS
FOR PINHOLE EROSION TESTING . . . . . . . . . xII-11 *
...
Vlll
EMifiO-2-4906
Change 1
1 May 88
Paragraph Page
* 6. PREPARATION OF UNDISTURBED SPECIMENS

FOR PINHOLE EROSION TESTING . . . . . . . . . XIII- 14
7. PINHOLE EROSION TEST PROCEDURE . . . . . . . XIII- 15
8. PRESENTATION AND ANALYSIS OF PINHOLE
EROSION TEST RESULTS . . . . . . . . . . . . . . . XIII-i8
9. POSSIBLE ERRORS WITH PINHOLE EROSION
TEST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIII-18 *
ix
E M 1410-2-1906
Change 1
1 May 80
AH Change in height during consolidation or before shear

0
h Hydraulic head
i Hydraulic gradient
k Coefficient of permeability
L Length
Linear shrinkage
Ls
LL Liquid limit
N Number of blows of liquid limit cup
* NDj,ND2 Nondispersive soil classification--pinhole erosion
test
* ND3,ND4 Intermediate soil classification--pinhole erosion
test
n Porosity
P Applied axial load
PI Plasticity index
PL Plastic limit
P Pressure
Preconsolidation pressure
pc
Overburden pressure
PO
Q Quantity of flow
q Rate of discharge
Unconfined compressive strength
%l
R Shrinkage ratio
S Degree of saturation
* SAR Sodium absorption ratio
SL Shrinkage limit
S Shear strength
S Undrained shear strength
U
t Time
Time to 50 percent primary consolidation
t50
xi
EM IiiO-2-1906
Chang9 1
1 May 80
Time to failure
tf
U Degree of consolidation
U Atmospheric pressure (absolute)
a
U Required back pressure (gage)
v Volume
Volume of air
va
vs
V
Volume of soil solids
Volume of voids
VV Volume of water
LX? Change in volume
7L.V Change in volume of water
W
W Weight of wet soil
W Weight of dry soil
WS Weight of water
W
W Water content
Water content at N blows
wN
F Slope angle of flow line
Dry unit weight (or dry density) of soil
yd
Wet unit weight (or wet density) of soil
yrn
Unit weight of water
YW
& Axial strain
Double amplitude axial strain
da
n Coefficient of viscosity
u Normal stress
Major principal stress
5
Minor principal stress
u3
Deviator stress
Y - u3
Cyclic deviator stress
udc
-dc Cyclic stress ratio
2 u3c
xii
E M i410-2-1906
Change 1
1May 80
* Effective confining pressure at failure
c3f
* Consolidation stress on the failure plane
cfc
7 Shear stress
* T Cyclic shear stress

cvclic
4J Angle of internal friction (or slope angle of strength
envelope) based on total stresses
Angle of internal friction (or slope angle of strength
envelope) based on effective stresses
Angle of friction determined from the drained
repeated direct shear test
xiii
DEPARTMENT OF THE ARMY E M 1110-2-1906
Office of the Chief of Engineers
ENGCW-ES Washington, D. C. 263i4
Manual 30 November i970

No. 4110-2-19C6
E N G I N E E R I N G A N D DESIGN
1. PURPOSE. This manual presents recommended testing procedures for

making determinations of the soil properties to be used in the design of
civil works projects. It is not intended to be a t ext book on soils testing
or to supplant the judgment of design engineers i n specifying procedures
to satisfy the requirements of a particular proJect.
2. APPLICABILITY. The provisions of this manual are applicable to all
divisions and districts having Civil Works functions.
3. NOTATIONS, A partial list of the symbols used in this manual ap-
pears following the Table of Contents. Not included in the list are sym-
bols used arbitrarily in formulas to facilate computations and those
having a special meaning within a particular appendix. All symbols are
defined where they first appear in an appendix and wherever restatement
may be needed for the sake of clarity. Whenever possible, these symbols
correspond to those recommended by the Committee on Glossary of
Terms and Definitions in Soil Mechanics of the Soil Mechanics and Foun-
dations Division of the American Society of Civil Engineers.t
4. REFERENCES. The material presented in this manual has been
drawn from many sources, persons, and organizations; wherever possi-
ble, specific references are given by footnotes. In general, the proce-
dures and practices herein have been taken, under the guidance of the
7 Glossary of terms and definitions in soil mechanics, Proceedings&

ASCE[ vol. 84, No, SM4 (October i958).
E M iiiO-2-1906
30 Nov 70
Office, Chief of Engineers, from the experience of the U. S. Army Engi-

neer Waterways Experiment Station and the U. S. Army Engineer Divi-
sions. Further contributions have come from Harvard University, the
Massachusetts Institute of Technology, the U. S. Bureau of Reclamation,
and the American Society for Testing and Materials.
5. TESTING PROGEDURES. Since soils exist in an enormous variety,
and since the problems of applied soil mechanics also exist in a very
great variety, testing procedures for determining the engineering proper-
ties of soils (such as strength-deformation relationships) must not, in
fact, cannot, be standardized. Before any soils testing is requested of a
laboratory, the design engineer responsible for formulating the testing
program must clearly define the purpose of each test to himself and to
the person who will supervise the testing.
It is generally necessary to adapt the testing procedures to the specific
requirements of an investigation. For example, the consolidation test can
be performed in various ways. What is often called the standard con-
solidation &St is performed by always doubling the previous load on the
specimen. This procedure will produce tin-&-consolidation curves that
usually permit the most precise evaluation of the coefficients of perme-
ability and consolidation. However, these load increments are not always
satisfactory for defining the preconsolidation pressure from the shape of
the void ratio-pre,ssure curve; for this purpose a much smaller factor
than 2.0 should be used during incremental loading. Also, the maximum
load to which a consolidation test should be continued will depend on the
consistency and stress history of the soil and the requirements of the
project. For example, if a clay which had been normally consolidated under
an effective overburden pressure of 0.5 ton per sq ft is to be loaded by an
embankment which will exert an additional pressure of i.0 ton per sq ft,
the consolidation test need not be continued beyond a load of 4.0 tons per
sq ft to fulfill the purpose of the test. On the other hand, a highly over-
consolidated clay which will be loaded by an embankment of substantial
height may require that the consolidation test be continued to a loading
2
E M i1iO-2-4906
30 Nov 70
An even greater variety in testing procedures exists for measuring

the strength of soils, and the purpose of the tests must be constantly re-
viewed to insure that the results have meaning with respect to design.
Tests which do not measure clearly defined engineering properties
(such as Atterberg limits, specific gravity, grain-size analyses, and
compaction), however, do require adherence to standardized procedures.
Even here the dangers of injudicious testing must be recognized As an
example, compaction test results-must be carefully evaluated if the mate-
rial coarser than 3/4 in. (or some other size) ila~ been removed according
to the standard method.
Procedures for soils tests necessary for the design of Civil Works
projects appear as appendixes to this manual. The procedures are con-
sidered to represent the b.est current guidance for obtaining acceptable
de sign data, Deviations from these procedures may be necessary on
occasion, according to the judgment of testing or design engineers,
-
their experience with local soils, or peculiarities of a project. How-
ever, to insure that the test methods remain compatible with the pur-
pose of the tests and that the resuIts will be acceptable, every such
deviation should be discussed in advance with the design office request-
ing the tests or, if in the judgment of the design engineer such devia-
tion represents a major departure from the conventional procedure,
should be first approved by the Office, Chief of Engineers. Also, a
description of any non-conventional procedure must accompany the test
data.
6. RELIABILITY OF TESTING APPARATUS AND RESPONSIBILITY OF
PERSONNEL, All who are engaged in soils testing must constantly be
aware of the importance of accuracy in measurements. Inaccurate meas-
urements will produce test results which are not only valueless but are
misleading. Each appendix to this manual contains a list of the more com-
mon possible errors associated with the procedures described in that ap-
pendix. Serious errors can be caused by poorly constructed apparatus (for
E M 1110-2-1906
30 Nov 70
e x a m p l e p if3tcJn f r i c t i o n i n t r i a x i a l c o m p r e s s i o n c h a m b e r s o r rough-
f i n i s h e d c o n s o l i d o m e t e r ringa), by maladjusted apparatus (liquid limit
d e v i c e s , p r o v i n g r i n g s , or mechanical compactors), and by worn parts
(liquid limit cup or grooving tool or knife edges of lever systems).
Regular calibration and inspection must be a standard practice in all
laboratories.
The personnel performing the tests must be thoroughly familiar with
the apparatus, the testing procedures, and good laboratory technique in
general. They must be conscientious in the handling of soils and must
appreciate the purpose of each test they perform. Neat, thoughtfui w o r k ,
with the recording of.all test data and a continuous watchfulness for
irregularities c a n p r e v e n t m o s t e r r o r s . The philosophy should be that
one good test is not only far better than many poor tests, but is also less
expensive and less likely to permit a misjudgment in design.
7. LABORATORY FACILITIES. A laboratory preferably should be on a
ground floor or basement with a solid floor and should be free of traffic
.-
and machinery vibrations. Separate areas should be designated for dust-
producing activities such as sieve analyses and sample processing.
Temperature control of the entire laboratory is to be preferred. If the
temperature-controlled space is limited, this space should be used for
t r i a x i a l c o m p r e s s i o n , consolidation, and permeability testing. A humid
room large enough to permit the storage of samples and the preparation
of test specimens should be available.
8. SAMPLE HANDLING AND STORAGE. The identification markings of
all samples should be verified immediately upon their receipt at the
laboratory, and an inve n t o r y, of the samples received should be mainA
tained. Samples should be examined and tested as soon as possible after
r e c e i p t ; h o w e v e r , it is often necessary to store samples for several days
or even weeks to complete a large testing program. Every care must be
taken to protect undisturbed samples against damage or changes in water
content Such samples should be stored in a humid room and may require
rewaxing and relabeling before storage. Except for special purposes,
4
EM 1110-2-1906
30 Nov 70
such as for viewing by designers or contractors or for research, soil samples

should not be retained for long periods; even the most careful sealing and
storing of undisturbed samples cannot prevent the physical and chemical
changes which, in time, would invalidate any subsequent determinations of
their engineering properties.
9. SELECTION AND PREPARATION OF TEST SPECIMENS. Under the most favorable
circumstances, a laboratory determination of the engineering properties of
a small specimen of undisturbed soil gives but an approximate guide to the
behavior of an extensive nonhomogeneous geological formation under the
complex system of stresses induced by the construction of an embankment or
other structure; under the worst circumstances such a determination may
have no meaning. Also, the strength, compressibility, and permeability of
a soil in place may vary severalfold within a few inches. No other aspect
of laboratory soils testing is as important as the selection of test
specimens to best represent those features of a foundation soil which
influence the design of a project. The selection cannot be based on boring
logs alone, but requires personal inspection of the samples and the closest
teamwork of the laboratory personnel and the design engineer. This cooper-
ation must be continued throughout the testing program since, as quanti-
tative data become available, changes in the initial allocation of samples
or the securing of additional samples may be necessary.
Second in importance only to the selection of the most representative
undisturbed material is the preparation and handling of the test specimens
to preserve in every way possible the natural structure and water content
of the material. Indifferent handling of undisturbed soils can result in
test data that are erroneous by several times any errors caused by faulty
testing apparatus. With but few exceptions, test specimens should always
be prepared in a humid room. Trimming instruments should be sharp
and clean and the specimens should be adequately supported at all times;
details of the preparation equipment and procedures are presented in the
appendixes to this manual. What cannot be gained from any manual,
EM 1110-2-1906
Change 2
20 Aug 86
however, is the judgment and awareness necessary to adjust the tech-

niques for each type of material in order to secure the most satisfactory
specimens.
During the preparation of specimens, the laboratory personnel have
the best opportunity to record a complete description of the material and
to judge whether the material is truly undisturbed. The description
should include an identification of the material, its color and
consistency, the brittleness of the material and the loss of strength upon
remolding, and any heterogeneity or unusual characteristics which might
prove valuable in analyzing the test results. Also, any indication of
disturbance of boring samples (strata deformed at periphery or distortions
concentric with axis of sample) must be noted. Often these distortions
cannot be seen except by slowly drying a slice of the material to a water
content at which the differences between strata show clearly.+ Photo-
graphs of such partially dried slices may be helpful when evaluating the
test data and can contribute to improvements in sampling equipment and
techniques. Disturbed samples should never be used for any tests other
than classification, specific gravity or water content.
10. DATA SHEETS AND REPORT FORMS. Examples of suggested form sheets for
recording and computing test data are presented in the appendixes hereto,
and some appendixes show the forms to be used for reporting test results.
The data sheets shown may be satisfactory in many instances, though each
laboratory should adopt whatever data sheets are most suitable for their
* practices and apparatus. ENG Form 2086 should be used for presentation of
a summary of soil test data in design documents. Well-planned data sheets *
can improve the efficiency of testing and, by encouraging the recording of
data which otherwise might be lost, can lead to better testing.
The report forms shown in the appendixes have been developed to
facilitate the review of design memoranda of Civil Works projects.
+ M. J. Hvorslev, Subsurface Exploration and Sampling of Soils for

Civil Engineering Purposes, U. S. Army Engineer Waterways Experiment
Station, CE (Vicksburg, Miss., November 1949).
6
EM iitO-2-1906
30 Nov 70
Because they are intended for review purposes, these forms often do not
display the test results in sufficient detail for interpretation by the design
engineer. Therefore, each laboratory should include with the standard
report forms whatever tabulated or plotted data are necessary to satisfy
the purpose of a testing program. Graphs should show all the plotted
points, not just smooth curves, and be given scales in easily read units,
such as i, 2, or 5 divisions per unit. The report form should contain a
complete description of the material, not just the classification, and
sketches to illustrate the mode of failure of strength test specimens.
FOR THE CHIEF OF ENGINEERS:
RICHARD F. McADOO
Colonel, Corps of Engineers
Executive
15 Appendixes
APP I - Water Content - General
APP II - Unit Weights, Void
Ratio, Porosity, and
- Degree of Saturation
APP III - Liquid and Plastic Limits
A P P IIIA - One-Point Liquid Limit
Test
A P P IIIB - Shrinkage Limit Test
APP IV - Specific Gravity
APP V - Grain-Size Analysis
APP VI - Compaction Tests
A P P V I A - Compaction Test, Earth-
Rock Mixture s
APP VII - Permeability Tests
APP VIII - Consolidation T e s t
APP IX - Drained (S) Direct Shear
Test
A P P I X A - Drained (S) Repeated
Direct Shear Test
APP X - T riaxial Compr e s sion
Tests
APP XI - Unconfined Compr e s sion
Test
APP XII - Relative Density
APP XIIA - Modified Providence
Vibrated Density Test
7
EM 1i10-2-1906
30 Nov 70
APPENDIX I:
WATER CONTENT - GENERAL
i. D EFINITION. Water content, w, is defined as the ratio, expressed

as a lbercentage, of the weight of water in a given soil mass to the weight
of solid.particles.
2. APPARATUS The apparatus should consist of the following:
-a. Oven,- preferably of the forced-draft t y p e automatically con-

trolled to maintain a uniform temperature of 110 f 5 C throughout the
oven.
b. Balances, sensitive to 0.01 g for samples weighing less than
-
50 g; 0.1 g for samples weighing 5G to 500 g; i.0 g for samples weighing
over 500 g.
-
C. Specimen Containers. Seamless metal containers with lids
are recommended. The containers should be of a metal resistant to
corrosion (aluminum is satisfactory). They should be as small and light
in weight as practicable in relation to the amount of material to be used
in the determination. For routine water content determinations in which
specimens weighing between iO0 and 200 g are used, a 2-in.-high by
3-i/2-in. diameter container is adequate.
3. SPECIMEN. The amount of material used in the water content deter-
mination will generally depend on the maximum size of particles, the
amount of material available, and the requirement that the specimen be
representative of the material for which the determination is made. When
the water is not uniformly distributed throughout the sample, larger speci-
mens will be needed than would otherwise be required. For routine water
content determinations on material passing a No. 4 sieve, specimens
weighing between iO0 and 200 g are adequate, A minimum specimen
weight of 500 g is recommended for material having a maximum particle
size in the range of the No. 4 to 3/4-in, sieves, and a minimum specimen
I-l
E M iliO-2-i906
Appendix I
30 Nov 70
weight of 1000 g is recommended for material having a maximum particle

size in the range of the 3/4-in. to i-1/2-in. sieves. Specific amounts of
material are required for water content determinations for other labora-
tory tests; the test procedures should be consulted to determine the
proper amounts.
4. PROCEDURE. The procedure shall consist of the following steps:
a. Record all identifying information for the specimen, such as
project, boring number, sample number, or other pertinent data, on a data
sheet (Plate I-i is a suggested form).
b. Record the number and tare weight of the specimen container.
C. Place the specimen in the container, set the lid securely in
p o s i t i o n and immediately determine the weight of the container and wet
soil by weighing on an appropriate balance.
d. Before the specimen is placed in the oven, remove the lid; the
lid is usually placed under the container in the oven. Then place the
specimen and container in the oven heated to 4iO * 5 C.7 Leave the speci-
men in the oven until it has dried to a constant weight. The time required
for drying will vary depending on the type of soil, size of specimen, oven
type and capacity, and other factors. The influence of these factors
generally can be established by good judgment, and experience with the
soils being tested and the equipment available in the laboratory. When in
doubt, reweigh the oven-dried specimens at periodic intervals to establish
the minimum drying time required to attain a constant weight. For routine
water content determinations, specimens consisting of clean sands and
7 Laboratory oven drying at 110 C does not result in reliable water con-
tent values for soils containing gypsum or significant amounts of organic
material. Reliable water content values for these soils can be obtained
by drying in oven at 60 C*, or by vacuum desiccation. See: U. S. Army
Engineer Waterways Experiment Station, CE, A Study of Moisture- -
Content
- ~ Determinations on Selected Soils, Miscellaneous Paper No.
4-73 (Vicksburg, Miss., September 1954).
I-2
EM iiiO-2-1906
Appendix I
30 Nov 70
gravels should be oven-dried for a minimum of 4 hr. For most other

soils a minimum drying time of 16 hr is adequate, Dry soil may absorb
moisture from wet specimens; therefore, any dried specimens must be
removed before wet specimens are placed in the oven.
e. After the specimen has dried to constant weight, remove the
container from the oven and replace the lid. Allow the specimen to cool
until the container can be handled comfortably with bare hands. If the
specimen cannot be weighed immediately after cooling it should be placed
in a desiccator; if a sample is left in the open air for a considerable
length of time it will absorb moisture.
f. After the specimen has cooled, determine its dry weight and
record it on the data sheet.
5. COMPUTATIONS. The following quantities are obtained by direct
weighing:
a. Weight of tare plus wet soil, g
b. Weight of tare plus dry soil, g
The water content in percent of oven-dry weight of the soil is equal to:
(weight of tare plus wet soil) - (weight of tare plus dry soil) x 1oo
(weight of tare plus dry soil) - (tare)
W
or w = zx ioo
wS
where W = water content, percent

WW = weight of water, g
WS
= weight of dry soil, g
6. POSSIBLE ERRORS.. Following are possible errors that would cause

inaccurate determinations of water content:
I-3
E M 1110-2-1906
Appendix I
30 Nov 70
a. Specimen not representative. The specimen must be rep-

-
resentative of the sample as required for the purpose of the determi-
nation. For example, a stratified soil may have a great variation in,
water content between adjacent strata; were it intended to evaluate the
strength of the soil on the basis of water content, a large specimen
that included material from several strata would not be representative
of the weakest stratum. As another example, to determine the average
water content of a gravelly clay, the specimen must be large enough to
contain representative amounts of both coarse and fine fractions.
b.
- Specimen too small. As a rule, the larger the specimen,
the more accurate the determination because of the larger weights
involved.
C. Loss of moisture before weighing wet specimen. Even in
a covered container a specimen can lose a significant amount of water
unless weighed within a short period.
d.
- Incorrect temperature of oven. The oven-dry weight of
many soils is dependent on the temperature of the oven, so variations in
temperature throughout the interior of an oven can cause large variations
in the computed water c0ntent.t
-e. Specimen removed from oven before obtaining a constant oven-

dry weight.
f.
-- Cain of moisture before weighing oven-dry specimen.
g* Weighing oven-dry specimen while still hot. The accuracy

of a sensitive balance may be affected by a hot specimen container.
h. Incorrect tare weight. The weights of specimen containers
should be checked periodically and should be scratched on the containers
to avoid possible errors in reading such weights from lists.
t T. W, Lambe, Soil Testing for Engineers, John Wiley & Sons, Inc.
(Net:? York, i95i).
I-4
EM lliO-2-i906
Appendix I
30 Nov 70
m m!rE
PRoJEcr
BORING m*
1
se.r@e or specllmx r?o.
Tare No.
Tam plue wet a011
8 Tare plue dry soil

w
Saqple or Spedmen No.
Tam tie.
Tare plun wet 8011
Tare plue dry 0011
Water "%I
Tamz
Dry soil Wm
Water content w $ $ $ $ $
Water content " % $ $ 6 $ $

:'
W
tam plue wt'moil) - (tax% phle dry eo11 xloo=~xloo
4J (tam plue &y eoll] - (time)
"e
Remarka
Tecbniclan Computed by Checkedby

mG FORM
1 J U N 65 3835 I-5 PLATE I-1
EM ii40-2-iqO6
30 Nov 70
APPENDIX II:
UNIT WEIGHTS, VOID RATIO, POROSITY, AND

1. RELATIONS. A soil mass is considered to consist of solid particles

enclosing voids of varying sizes. The voids may be filled with air, water,
or both. The fundamental relations of the weights a.uo volumes of the
various components of a soil
mass can be derived using the
simplified sketches shown in
.Figure i. Some of the more
important relations used in
rIW
%I#
soils engineering calculations

are unit weights, void ratio,
porosity, and degree of satura-
tion. The quantities which must
MOIST SOIL
iL W
SATURATED SUL
be known to compute these rela-

Ww
WATERCONTENT
tions are the weight and volume
*z-
ws
of the wet specimen, the weight IORY OENSllYb
w.
-ORY UNIT WEIGMT
- Yd=T
of the same specimen after

oven-drying, and the specific WET UNIT WEIGMT IWET OENSlTYb Ym=;
gravity of the solids (see

.=-
VOID RATIO
Appendix IV, SPECIFIC .
GRAVITY). The weights of the

Z_ZL
POROSITY
V 1to
specimens usually can be ob- W

tained without difficulty. The s=r

volume of the wet specimen is SPECIFIC GRAVITY OF SOLIOS

W.
G,=-
. * yw
determined by linear measure-
NOTE: N,T WE,GT O F W A T E R . y_ = 62.4 Lrn,C F T .
ment (volumetric method), or
by measurements of the volume Figure i. Fundamental relations
II-i
o r w e i g h t of w a t e r displac e d b y t h e specimen (dis~J~a~_erlJcIlt method).
Definitions of the relations to be determined and detailed procedures for
determining these values, using the volumetric and displacement methods,
are given in the following paragraphs.
2. DEFINITIONS. The unit weights, void ratio, porosity, and degree of
saturation are defined below.
-a. Dry unit weight, Yd, or dry density, is the weight of oven-
dried soil solids per unit of total volume of soil mass, and is usually ex-
pressed in pounds per cubic foot.
b.
- Wet unit weight, fm, or wet density, is the weight (solids
plus water) per unit of total volume of soil mass, irrespective of the de-
gre 2 of saturation (see -
e below). The wet unit weight is usually expressed
in pounds per cubic foot.
C.
- V o i d r a t i o , e, is the ratio of the? volume of voids to the volume
of solid particles in a given soil mass.
d.
- Porosity, n, is the ratio (usually expressed as a percentage)
of the volume of voids of a giv-en soil mass to the total volume of the soil
mass.
-e. Degree of saturation, S, is the ratio (expressed as a percent-

age) of the volume of water in a given soil mass to the total volume of
voids.
3. VOLIJMETRIC M E T H O D . a-. De s c ription. The volumetric method
consists of computing the total volume of soil from linear measurements
of a regularly shaped mass. In general, the method is applied to soils
which can be cut or formed into a cylinder or parallelepiped. Specimens
of this type are used i._ _,:er laboratory tests, and methods for preparing
them are described under the individual test procedures. The procedure
presented below is based on obtaining a cylindrical specimen by progres-
sive trimming in front of a calibrated ring-shaped specimen cutter. How-
ever\ other methods for obtaining a regularly shaped mass, such as
cutting and trimming or punching, can often be used successfully. The
II-2 .-
E M 1110-2-i906
Appendix II .I
30 Nov 70
volumetric method should not be used for soils containing gravel, shells,
or foreign materials which would interfere with advance trimming. The
calibrated specimen cutter method is particularly suitable for obtaining
volumes of silty and sandy soils having little cohesion.
b.
- Apparatus. The apparatus should consist of the following:
(i) Calibrated ring-shaped specimen cutter, hereinafter re-
ferred to as a volumetric cylinder. Types and sizes of volumetric cylin-
ders may vary widely; two types are shown in Figure 2. General require-
ments are that a volumetric cylinder be made of materials not susceptible
of rapid corrosion and that it be as large as possible in relation to the
samples being tested. The inside of the cylinder should be polished to a
smooth finish, and sharp cutting edges should be provided on the base. It
is very important that no voids form between the sample and cylinder; to
facilitate detection of such voids, a volumetric cylinder of transparent
Lucite with detachable steel cutting edges may be used.
(2) Guide cylinder for guiding cutter into soil (not absolutely
necessary).
( 3 ) T r i m m i n g t o o l s , such as wire saw, straightedge, or knife.
(4) Oven (see Appendix I, WATER CONTENT - G E N E R A L ) .
(5) Specimen container. The container should be of metal
that is resistant to corrosion. Seamless aluminum pans with lids are
satisfactory.
( 6 ) B a l a n c e , sensitive to 0.1 g.
(7) Glass plate, large enough to cover top of specimen.
C.
- Procedure. The procedure shall consist of the following steps:
(i) Record on a data sheet (Plate II-i is a suggested form)
all identifying information for the sample, such as project, boring number,
and other pertinent data.
(2) Measure and record the height, H, and inside diameter,
D, of the volumetric cylinder. In general, linear measurements shall be
II-3
E M iiIO-2-1906
Appendix II
3C-I Nov -/cJ
NOTE, DIMENSIONS AND MATERIALS OF

CONSTRUCTION MAY VARY
DEPENDING ON SIZE OF SAMPLES
AVAILABLE AND TYPES OF SOILS
TO BC TESTED
LUMINUU WIDE CVLINDER
C VL INQER Wl Tff
CUTTING .%WE
(a) VOLUMETRIC CYLINDER
CUTTER Cx-4 SAM P L E JUST &FORE CUTTING CUTTER PARTIALLY FORCED DOWN INTO SAMPLE
(b) VOLUMETRIC CYLINDER OF SPLIT-RING TYPE, SHOWING METHOD OF USE
Figure 2. Examples of volumetric cylinders
II-4
E M 1110-2-j906
Appendix II
30 Nov 70
tnade with an accuracy which will result in a volumetric error, dV/V,

of less than 1 percent. The volumetric error is represented by the
expression:
dV
FP ercent = 2 0 0
where dH = accuracy of height measurement

dD = accuracy of diameter measure ..t
(3) Center the volumetric cylinder on top of the sample. The

sample may be roughly trimmed to a size somewhat larger than the cylin-
der (see Fig. 2b) or the entire available sample may be used (see Fig. 3).
Push the cylinder vertically into the sam-
ple not more than I/4 in. and carefully
trim the soil from the edge of the cylinder
(see Fig. 3). Repeat the operation until
the specimen protrudes above the top, of
the calibrated cylinder. Care should be
taken that no voids are formed between
the cylinder and specimen. Using a wire
saw for soft specimens and a knife,
straightedge, or other convenient tool for
harder specimens, trim the top of the
specimen flush with the top of the cylin-
der. Invert the specimen, place it on a
glass plate, and trim the bottom of the
specimen.
Figure 3. Determining the
(4) Remove the specimen
unit weight of a soil speci-
from the volumetric cylinder using a men with the volumetric
guide cylinder, if available, and place it cylinder (split-ring type)
in a container. Weigh the specimen and container and record this weight
II-5
,
EM iiiO-2-i906
Appendix II
30 Nov 70
on the data sheet as the weight of tare plus wet soil. Alternatively, the
wet weight of the specimen may be determined by weighing the volumet-
ric cylinder with the specimen therein and then removing the material and
placing it in a container for a water content deter.mination.
(5) Place the soil and container in an oven and oven-dry the
specimen at 110 C f 5, allow it to cool, and then weigh. Record this
weight as weight of tare plus dry soil.
d. Computations. (1) Q uantities obtained in test. The following
-
quantities are obtained in the test.
(a) Weight of tare (specimen container or cylinder) plus
wet s o i l The tare weight is subtracted from this value to obtain the
weight of wet soil, W.
(b) Weight of tare (specimen container plus dry soil. The
tare weight is subtracted from this value to obtain the weight of dry
soil, Ws , or if the alternate procedure is used, dry weight of specimen --
is computed by the following formula:
wet weight of specimen

Dry weight of specimen = water content of specimen
it 100
ws= w
1 -t o.oiw
(c) The inside volume of the volumetric cylinder. Volume,

V , ,:yf the wet soil specimen is equal to this volume. The volume, V , may
also be computed from linear measurements of a specimen in the form of
a cylinder or parallelepiped.
(2) Unit weigilcs. The wet unit weight, ym, and the dry unit
weight, vd, expressed in terms of pounds per cubic foot, are computed
by the for lowing formulas:
Wet unit weight = weight in g of wet specimen x 62.4

volume in cc of wet specimen
= $ x 62.4
ym
II-6
EM illO-2-i906.
Appendix II
30 Nov 70
weight in g of dry specimen

Dry unit weight =
volume in cc of wet specimen x 62.4
W
= +X 62.4
yd
(3) Void ratio. The void ratio, e , is computed by the following

following formula:
volume in cc of wet specimen - volunle

. ... in cc of solids
Void ratio =
volume in cc of ::.:,Lds
v - vs
e=
vS
wS
where vs = volume of solids = -
GS
G = specific gravity of solids (see Appendix IV, SPECIFIC

S
GRAVITY)
(4) Porosity. The porosity, n, is computed by the following
formula:
volume in cc of wet specimen - volume in
cc of solids
Porosity, percent = x ioo
volume in cc of wet specimen
v - vs
n, percent = x too
V
(5) Degree of saturation. The degree of saturation, S, is

computed by the following formula:
volume in cc of water
Degree of saturation, percent = volume in cc of wet specimen x ioo
- volume in cc of solids
vW
S, percent = nx iO0
S
where V = ww = difference between the wet weight of the .soil speci-

W
men and the oven-dried weight
II-7
E M 11$0-2-1906
Appendix II
30 Nov 70
In the metric system, the volume of water, VW, is approximately ecIua1

numerically to the weight of water, Ww.
4. D I S P L A C E M E N T M E T H O D . fi. Description. The displacement
method consists of determining the total volume of a soil by measuring
the volume or weight of water displaced by the soil mass. The method is
particularly adaptable to irregularly shaped specimens and soils con-

taining gravel, shells, etc.
b. Apparatus. The apparatus should consist of the following:
( 1 ) B a l a n c e , sensitive to 0.i g.
(2) Wire basket of sufficient size to contain the soil
specimen.
(3) Can, or container, of sufficient size to submerge the wire
basket and specimen.
(4) Oven (see Appendix I, WATER CONTENT - G E N E R A L ) .
(5) Specimen container. The container should be of metal
that is resistant to corrosion, Seamless aluminum pans with lids are
satisfactory.
(6) P a i n t b r u s h .
(7) Microcrystalline wax or paraffin.t
(8) Container for melting wax, preferably with a self-
contained thermostat.
(9) Thermometer, range 0 to 50 C, graduated in 0.1 deg.
C.
- Procedure. The procedure shall consist of the following steps:
(1) Record all identifying information for the sample, such as
t Among the many microcrystalline wa.xes found satisfactory are Product

2300 of the Mobil Oil Company, Microwax 75 of the Gulf Oil Corpora-
tion, and Wax 1290 of the Sun Oil Company. Paraffin alone is not as
suitable for sealing soil specimens because its brittleness and shrink-
age upon cooling will cause cracking, especially in thin sections and at
corners; a mixture of 50 percent paraffin and 50 percent petrolatum
has properties that approach those of a microcrystailine wax.
II-8
EM 1i10-2-1906
Appendix II
30 Nov 70
project, boring number, or other pertinent data, on the data sheet

(Plate II-2).
(2) Determine, if not previously established, the specific
gravity of the wax to be used. (About 0.9 g per cc, but should be deter-
mined for each batch of wax.)
(3) Cut a specimen from the sample to be tested. (Th e size
of the specimen is not very important provided the capacity of the balance
is not exceeded. In general, more accurate results will be obtained with
larger specimens.) Trim the specimen to a fairly regular shape, Re-
entrant angles should be avoided, and any cavities formed by large parti-
cles being pulled out should be patched carefully with material from the
trimmings.
(4) Determine and record the wet weight of the soil specimen.
(5) Cover the specimen with a thin coat of melted wax, either
with a paintbrush or by dipping the specimen in a container of melted wax.
Apply a second coat of wax after the first coat has hardened. The wax
should be sufficiently warm to flow when brushed on the soil specimen,
yet it should not be so hot that it penetrates the pores of the soil. If hot
wax comes in contact with the soil specimen it may cause the moisture to
vaporize and form air bubbles under the wax.
(6) Determine and record the weight of the wax-coated
specimen in air.
(7) Determine and record the submerged weight of the wax-
coated specimen. This is done by placing the specimen in a wire basket
hooked onto a balance and immersing the basket and specimen in a can of
water as shown in Figure 4. In order to directly measure the submerged
weight of the wet soil and wax, the balance must have been previously bal-
anced with the wire basket completely submerged in the can of water. Ensure
that the specimen is fully submerged, and that the basket is not touching the
sides or bottom of the container. Measure the temperature of the water.
(8) Remove the wax from the specimen. It can be peeled off
II-9
EM 1110-2-19()6
Appendix 11
30 Nov 70
after a break is made in the wax sur-

face. Use the entire sample, or as
much as is free of wax inclusions, for
a water content determination (see
Appendix I, WATER CONTENT -
GENERAL).
d. Computations. The follow-
ing quantities are obtained directly in
the test:
(1) Weight of uncoated
specimen, W.
(2; Weight of soil plus
wax. The weight of uncoated speci-
men, W, is subtracted from this
value to obtain the weight of wax.
Figure 4. Determining the weight (3) Weight of soil plus wax

of a wax-coated specimen sub- in water.
merged in water
The following computations
shall be made:
(I) Divide the weight of the wax by its specific gravity. This
gives the volume of the wax.
(2) Subtract the weight of the wax-coated specimen in water
from its weight in air. The difference divided by the density of water at the
test temperature (see Table IV- 1, Appendix IV, SPECIFIC GRAVITY) is
numerically equal to the volume of the coated specimen in cubic centimeters.
(3) Subtract the volume of wax from the volume of the coated
specimen to obtain the total volume of the soil specimen, V.
(4) Compute the water content of the specimen (see Appendix I,
WATER CONTENT - GENERA-L). If the entire specimen is used for the
w a t e r content determination, obtain the dry weight of specimen, Ws, di-
rectly. If only a portion of the initial specimen is used for the water
II-10
EM iiiO-2-i906
Appendix II
30 Nov 70
content determination, compute the dry weight of specimen according to

the following formula:
wet weight of uncoated soil

Dry weight of specimen =
water content of wet soil
1 +
C ioo 1
ws= w
@k5
Based on the above information, compute the unit weights, void

ratio, porosity, and degree of saturation as specified hereinbefore.
5. POSSIBLE ERRORS. Following are possible errors that would
cause inaccurate determinations af the total volume:
a. Volumetric Method.. (i) Imprecise measurement of volu-
metric cylinder (or of cylindrical specimen trimmed by other methods).
Three height measurements and nine diameter measurements should be
made to determine the average height and diameter of the cylinder.
Precise calipers should be used for these measurements rather than flat
scales.
(2) Voids formed on side of specimen by trimming beyond
cutting edge.
(3) Material lost while removing specimen from cylinder.
be Displacement Method. Voids on surface of specimen not

filled by wax or air bubbles formed beneath wax.
II-ii
EM iiiO-2-i906
Appendix II
30 Nov 70
I I I
PLATE II-4
EM 4ilO-2-1906
Ap.pendix II
30 Nov 70
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
APPENDIX III:
LIQUID AND PLASTIC LIMITS
* 10 INTRODUCTION
a. The Atterberg Limits. The Atterberg limits are water

contents which define the limits of various stages of consistency
for fine-grained soils. The liquid limit (LL) and the plastic
limit (PL) define the upper and lower limits, respectively, of
the plastic range of a soil; the numerical difference between
these two limits expresses the plasticity of a soil and is termed
the plasticity index (PI). Detailed procedures for determining
the liquid and plastic limits for use in classifying soils and
developing correlations with engineering properties of soils are
given below, and a simplified method for determining the liquid
limit is described in Appendix IIIA, ONE-POINT LIQUID LIMIT TEST.
A detailed procedure for determining the shrinkage limit is given
in Appendix IIIB, SHRINKAGE LIMIT TEST.
b l Definitions.
(1) Liquid Limit. The liquid limit of a soil is the

water content, expressed as a percentage of the weight of oven-
dried soil at which two halves of a soil pat separated by a
groove of standard dimensions will close at the bottom of the
groove along a distance of 1/2 in. under the impact of 25 blows
in a standard liquid limit device.
(2) Plastic Limit. The plastic limit of a soil is the

water content, expressed as a percentage of the weight of oven
dried soil at which the soil just begins to crumble into short
pieces when rolled into a thread 1/8 in.~ in diameter.
III-1
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
20 APPARATUS
a. Liquid Limit Device. A mechanical device consisting of

a brass cup suspended from a carriage designed to control its
drop onto a hard rubber base. A drawing showing the essential
features of the device and the critical dimensions is given in
Figure 1. The design of the device may vary provided that the
essential functions are preserved. The device may be operated
either by a hand crank or by an electric motor.
(1) Base. The base shall be hard rubbert having a D

Durometer hardness of 80 to 90, and a resilience such that an
80mm (5/16=in.) diameter polished steel ball, when dropped from a
height of 25 cm (9.84 in.) will have an average rebound of at
least 80% but no more than 90%. The tests shall be conducted on
---
the finished base with feet attached.
(2) Feet. The base shall be supported by rubber feet

designed to provide isolation of the base from the work surface
and having an A Durometer hardness no greater than 60 as measured
on the finished feet attached to the base.
(3) Cupa The cup shall be brass and have a weight,

including cup hanger, of 185 to 215 g.
(4) Cam. The cam shall raise the cup smoothly and
continuously to its maximum height, over a distance of at least
180' of cam rotation. The preferred cam motion is a uniformly *
t Micarta No. 22lA has been used in the past. It is satisfac-

tory as long as it meets the resilience requirement set forth
for hard rubber.
III-2
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* accelerated lift curve.* The design of the cam and follower com-
bination shall be such that there is no upward or downward veloc-
ity of the cup when the cam follower leaves the cam.
(5) Carriage. The cup carriage shall be constructed

in a way that allows convenient but secure adjustment of the
height of drop of . the cup to 10 mm (0.394 in.). The cup hanger
shall be attached to the carriage by means of a pin which allows
removal of the cup and cup hanger for cleaning and inspection.
(6) Optional Motor Drive. As an alternative to the

hand crank shown in Figure 1, the device may be equipped with a
motor to turn the cam. Such a motor must turn the cam at
2 k 0.1 revolutions per second, and must be isolated from the
rest of the device by rubber mounts or in some other way that
prevents vibration from the motor being transmitted to the rest
of the apparatus. It must be equipped with an ON-OFF switch and
means of conveniently positioning the cam for height of drop
adjustments. The results obtained using a motor-driven device
must not differ from those obtained using a manually operated
device.
b Grooving Tool. A grooving tool having dimensions as

l
shown in Figure 2. The tool shall be made of plastic or *
* The cam and follower design in Figure 1 is for uniformly

accelerated (parabolic) motion. after contact and assures that
the cup has no velocity at drop off. Other cam designs also
provide this feature and may be used. However, if the cam-
follower lift pattern is not known, zero velocity ab drop off
can be assured by carefully filing or machining the cam and
follower so that the cup height remains constant over the last
20 to 45' of cam rotation.
III-4
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
DIMENSIONS
t 1
L E T T E R A* BA CA DA EA FA *
MM 2 I I 4 0 8 50 2
. * 0.1 & 0.2 5 0.5 k 0.1 & 0.5 2 0.1
LETTER G H J KA LA N
MM IO I3 60 IO 6 0 DEG 20
.
I IMINIMUM k 0.05 * I DEG
J
* ESSENTIAL DIMENSIONS
BACK AT LEAST I5 MM FROM TIP
NOTE: DIMENSION A SHOULD BE 1.9-2.0 AND DIMENSION D

SHOULD BE 8.0-8. I WHEN NEW TO ALLOW FOR
A D E Q U A T E SE&/ICE LIFE
4Dk-
I
t .
F K,
\
L
SECTION
K
I \\ \
I-
Figure 2. Grooving tool (optional height-of-drop gage attached)
noncorroding metal.* The design of the tool may vary as long as

the essential dimensions are maintained. The tool may, but need
not, incorporate the gage for adjusting the height of drop of the
liquid limit device. *
* Polycarbonate plastic grooving tools meeting the dimensional

requirements given above are available to US Government
agencies through the US Army Engineer Division Laboratory,
Southwestern, 4815 Cass Street, Dallas, TX 75235.
III-5
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* c. Gage* A metal gage block for adjusting the height of

drop of the cup, having dimensions as shown in Figure 3. The
design of the tool may vary provided the gage will rest securely
on the base without being susceptible to rocking, and the edge
which contacts the cup during adjustment is straight, at least
10 mm (3/8 in.) wide, and without bevel or radius.
DIMENSIONS IN MILLIMETERS
Figure 3. Height of drop gage
dl Containers. Small corrosion-resistant containers with

snug-fitting lids for water content specimens. Aluminum or
stainless steel cans 2.5 cm (1 in.) high by 5 cm (2 in.) in
diameter are appropriate.
e. Balance. A balance readable to at least 0.01 g and

having an accuracy of 0.03 g within three standard deviations
within the range of use. Within any 159g range, a difference
between readings shall be accurate within 0.01 g.
f Storage Container. A container in which to store the

l
prepared soil specimen that will not contaminate the specimen in

any way@ and which prevents moisture loss. A porcelain, glass,
or plastic dish about 11.4 cm (401/2 in.) in diameter and a
plastic bag large enough to enclose the dish and be folded over
is adequate. *
III-6
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* g Ground Glass Plate. A ground glass plate at least
l
30 cm (12 in.) square by 1 cm (3/8 in.) thick for mixing soil and
rolling plastic limit threads.
h l*atula* A spatula or pill knife having a blade about

2 cm (3/4 in.) wide by about 10 cm (4 in.) long. In addition, a
spatula having a blade about 2.5 cm (1 in.) wide and 15 cm
(6 in.) long has been found useful for initial mixing ofsamples.
i. Sieve* A 20.3 cm (8 in.) diameter, 4250pm (No. 40)

sieve conforming to the requirements of ASTM Specification El1
and having a rim at least 5 cm (2 in.) above the mesh. A 2-mm
(No. 10) sieve meeting the same requirements may also be needed.
.
I Wash Bottle. A wash bottle or similar container for
0
adding controlled amounts of water to soil and washing fines from

coarse particles.
k 0 Drying Oven. A thermostatically controlled oven,

preferably of the forced-draft type, capable of continuously
maintaining a temperature of 110 & 5O C throughout the drying
chamber. The oven shall be equipped with a thermometer of
suitable range and accuracy for monitoring oven temperature.
1 0 Washing Pan. A round, flat-bottomed pan at least

7.6 cm (3 in.) deep, slightly larger at the bottom than a 20.3-cm
(8-in.) diameter sieve.
m. Rod (Optional). A metal or plastic rod or tube 3.2 mm

(1/8 in.) Diameter and about 10 cm (4 in.) long for judging
the size of plastic limit threads. *
III-7
EM 1110-2~1906 -
Appendix III
Change 2
20. Aug 86
* n. Mixing Water. A supply of distilled or demineralized

water.
Blender (Optional). A single speed blender with

0.
1,000 ml container for preparing clay shale materials.
3 l CALIBRATION OF APPARATUS
a. Inspection for Wear
(1) Liquid Limit Device. Determine that the liquid

limit device is clean and in good working order. The following
specific points should be checked:
(2) Wear of Base. The spot on the base where the cup
makes contact should be worn no greater than 10 mm (3/8 in.) in 1-
diameter. If the wear spot is greater than this, the base can be
machined to remove the wear spot provided the resurfacing does
not decrease base thickness to less than that specified in 2(a)
and the other dimensional relationships are maintained.
(3) Wear of Cup. The cup must be replaced when the

grooving tool has work a depression in the cup 0.1 mm (0.004 in.)
deep or when the edge of the cup has been reduced to half its
original thickness. Verify that the cup is firmly attached to
the cup hanger.
(4) Wear of Cup Hanger. Verify that the cup hanger

pivot does not bind and is notworn to an extent that allows more
than 30mm (1/8 in.) side-to-side movement of the lowest point on
the rim. *
III-8
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* (5) Wear of Cam. The cam shall not be worn to an

extent that the cup drops before the cup hanger (cam follower)
loses contact with cam.
(6)Grooving Tools. Inspect grooving tools for wear

on a frequent and regular basis. The rapidity of wear depends on
the material from which the tool is made and the types of soils
being tested. Sandy soils cause rapid wear of grooving tools;
therefore, when testing these materials, tools should be in-
spected more frequently than for other soils. Any tool with a
tip width greater than 2.1 mm must not be used. The depth of the
tip of the grooving tool must be 7.9 to 8.1 mm. The width of the
tip of grooving tools is conveniently checked using a pocket-
sized measuring magnifier equipped with a millimeter scale.
Magnifiers of this type are available for most laboratory supply
companies. The depth of the tip of grooving tools can be checked
using the depth measuring feature of vernier calipers.
(7) Blender Blades. Blender blades should be replaced

when their overall length becomes 3 mm (1/8 in.) less than their
original length.
bl Adjustment of Height of Drop. Adjust the height of

drop of the cup so that the point of the cup that comes in con-
tact with the base rises to a height of 10 ? 0.2 mm. A conve-
nient procedure for adjusting the height of drop is as follows:
place a piece of masking tape across the outside bottom of the
cup parallel with the axis of the cup hanger pivot. The edge of
the tape away from the cup hanger should bisect the spot on the
cup that contacts the base. For new cups, placing a piece of
carbon paper on the base and allowing the cup to drop several
times will mark the contact spot. Attach the cup to the device
and turn the crank until the cup is raised to its maximum height.
III-9
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* Slide the height gage under the cup from the front, and observe
whether the gage contacts the cup or the tape (see Figure 4). If
the tape and cup are both contacted, the height of drop is ap-
proximately correct. If not, adjust the cup until simultaneous
contact is made. Check adjustment by turning the crank at
2 revolutions per second while holding the gage in position
against the tape and cup. If a ringing or clicking sound is
heard without the cup rising from the gage, the adjustment is
correct. If no ringing is heard or if the cup rises from the
gage1 readjust the height of drop. If the cup rocks on the gage
during this checking operation, the cam follower pivot is exces-
sively worn and the worn parts should be replaced. Always remove
tape after completion of adjustment operation.
POINT WHERE
HEIGHT GAUGE
i-MASKING TAPE APPLIED AS AID

IN ADJUSTMENT OPERATION
Figure 4. Calibration for height of drop *
III-10
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* 4 0 PREPARATION OF MATERIAL
a. Selection of Material. It is essential that the same

carefully prepared soil mixture be used for determining both the
liquid and plastic limits. Layers of soil of different plastic-
ity should not be mixed. Furthermore, if the natural water con-
tent is to be determined, the specimen must be taken from an
identical mixture to permit valid correlations. If other test
results are to be correlated with the liquid and plastic limits,
the material used for the determinations must be the same as that
tested. Clay shale materials require special preparation as dis-
cussed in Paragraph 9.
b 0 Effects of Dryins. Whenever possible, soils should be

at the natural water content when preparation for testing is
begun. If drying has occurred before testing; the limit values
may change. The plasticity of soils containing organic colloids
and certain types of inorganic colloids derived from volcanic
rocks is highly sensitive to drying. The effects of drying can
be determined by comparing the liquid limit values of specimens
in undried," "airdried," and "ovendried" states.
c. General Preparation of Material
(1) Samples Passing the 4250pm (No. 40) Sieve. When

by visual and manual procedures, it is determined that the sample
has little or no material retained on the 425-urn (No. 40) sieve,
prepare a specimen of 150 to 200 g by mixing thoroughly with
distilled or demineralized water on the glass plate using the
spatula. If desired, soak soil in a storage dish with small
amount of water to soften the soil before the start of mixing.
Adjust the water content of the soil to bring it to a consistency
that would require 15 to 25 blows of the liquid limit device to *
III-11
EM 1110-2~1906 .
Appendix III
Change 2
20 AUCJ 86
* closethe groove. The time taken to adequately mix a soil will
vary greatly depending on the plasticity and initial water
content. Initial mixing times of more than 30 min may be needed
for stiff, fat clays. If, during mixing, a small percentage of
material is encountered that would be retained on a 4250pm
(No 40) sieve, remove these particles by hand, if possible. If
l
it is impractical to remove the coarser material by hand, remove

small percentages (less than about 15%) of coarser material by
working the specimen through a 425-pm (No. 40) sieve using a
piece of rubber sheeting, a rubber stopper,or other convenient
device provided the operation does not distort the sieve or
degrade material that would be retained if the washing method
described in the next paragraph were used. If larger percentages
of coarse material are encountered during mixing, or it is con-
sidered impractical to remove the coarser material by the methods
just described, wash the sample as described in the next para-
graph. When the coarse particles found during mixing are concre-
tions, shells, or other fragile particles, do not crush these
particles to make them pass a 425-pm (No. 40) sieve, but remove
by hand or by washing. Place the mixed soil in the storage dish,
cover to prevent loss of moisture, and allow to stand for at
least 16 hr (overnight). After the standing period and immedi-
ately before starting the test, thoroughly remix the soil.
w Samples Containing Material Retained on a 425-urn

(No a 40) Sieve
(a) Select a sufficient quantity of soil at

natural water content to provide 150 to 200 g of material passing
the 4250urn (No. 40) sieve. Place in a pan or dish and add
sufficient distilled or demineralized water to cover the soil.
Allow to soak until all lumps have softened and the fines no
longer adhere to the surfaces of the coarse particles. *
III-12
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
.* (b) When the sample contains a large percentage
of material retained on the 4250urn (No. 40) sieve, perform the
following washing operations in increments, washing no more than
0.5 kg (1 lb) of material at one time. Place . the 4250pm (No. 40)
sieve in the bottom of the clean pan. Pour the soil water mix-
ture onto the sieve. If gravel or coarse sand particles are
present, rinse as many of these as possibleLwith small quantities
of water from a wash bottle and discard. Alternatively, pour the
soil water mixture over a 2-mm (No. 10) sieve nested atop the
425-pm (No. 40) sieve, rinse , the fine material through and remove
the 2-mm (No. 10) sieve. After washing and removing as much of
the coarser material as possible, add sufficient water to the pan
to bring the level to about 13 mm (1/2 in.) above the surface of
the 425-urn (No. 40) sieve. Agitate the slurry by stirring with
the fingers while raising and lowering the sieve in the pan . and
swirling the suspension so that fine material is washed from the
i
coarser particles. Disaggregate fine soil lumps that have not
slaked by gently rubbing them over the sieve with the fingertips.
Complete the washing operation by raising the sieve above the
watersurface and rinsing the material retained with a small
amount of clean water. Discard material retained on the 425-pm
(No 0 .a:40) sieve.
(c) Reduce the water content of the material

passing the 450pm (No. 40) sieve until it approaches the liquid
limit. Reduction of water content may be accomplished by one or
a combination of the following methods: (a) exposing the air
currents at ordinary room temperature, (b) exposing to warm air
currents from a source such as an electric hair dryer,(c) fil-
tering in a Buckner funnel or using filter candles, (d) decanting
clear water from surface of suspension, or (e) draining in a
colander or plaster of paris dish lined with high retentivity,
III-13
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* high wet-strength filter paper.* If a plaster of paris dish is

used1 take care that the dish never becomes sufficiently satu-
rated that it fails to actively absorb water into its surface.
Thoroughly dry dishes between uses. During evaporation and cool-
ing, stir the sample often enough to prevent overdrying of the
fringes and soil pennacles on the surface of the mixture. For
soil samples containing soluble salts, use a method of water
reduction such as (a) or (b) that will not eliminate the soluble
salts from the test specimen.
h W Thoroughly mix the material passing the

425-um ( . 0. 40) sieve on the glass plate using the spatula.
Adjust t e water content of the mixture, if necessary, by adding
ii
small increments of distilled or demineralized water or by allow-

<;
ing the mixture to dry at room temperature while mixing on the
glass plate. The soil should be at a water content that will --.-
i
result in closure of the groove in 15 to 25 blows. Return the
mixed soil to the mixing dish, cover to prevent loss of moisture,
and allow to stand for at least 16 hr. After the standing
period, and immediately before starting the test, remix the soil
thoroughly.
5. LIQUID LIMIT
a. Procedure
(1) Place a portion of the prepared soil in the cup of

the liquid limit device at the point where the cup rests on the
baset squeeze it down# and spread it into the cup to a depth of
about 10 mm at its deepest point, tapering to form an *
* S&S 595 filter paper in 32 cm circles has been found

satisfactory.
III-14
EM 1110-24906
Appendix III
Change 2
23 Aug 86
* approximately horizontal surface. Take care to eliminate air

bubbles from the soil pat, but form the pat with as few strokes
as possible. Heap the unused soil on the glass plate and cover
with the inverted storage dish or a wet towel.
(2) Form a groove in the soil pat by drawing the tool,

beveled edge, forward through the soil on a line joining the
highest point to the lowest point on the rim of the cup. When
c u t t i n g the groove, hold the grooving tool against the
, . surface of
the cup and draw in an arc maintaining the tool perpendicular to
the surface of the cup throughout its movement (see Figure i).
p:
Figure 5. Grooved soil pat in liquid limit device
In soils wnere a groove cannot be made in one stroke without

tearing the soil cut the groove with several strokes of the
grooving tool. Alternately, cut the groove to slightly less than
required dimensions with a spatula and use the grooving tool to
bring the groove to final dimensions. Exercise extreme care to
prevent sliding the soil pat relative to the surface or the cup. *
III-15
E M 1110-2-1906 ,
Appendix III
Change 2
20 Aug 86
* ,(3) Verify that no crumbs of soil are present on the
base or the underside of the cup. Lift and drop the cup by turn-
ing the crank at a rate of 1.9 to 2.1 drops per second until the
two halves of the soil pat come in contact at the bottom of the
groove along a distance of 13 mm (1/2 in.) (see Figure 6). Use
the end of the grooving tool (Figure 2) or a scale to verify that
the groove has closed 13 mm (1/2 in.)@
Figure 6. Soil pat after groove has closed
(4) Verify that an air bubble has not caused premature

closing of the groove by observing that both sides of the groove
have flowed together with approximately the same shape. If a
bubble has caused premature closing of the groove, reform the
soil in the cup by adding a small, amount of soil to make up for
that lost in the grooving operation and repeat 5a(l) through
5a(3). If the soil slides on the surface of the cup,+repeat
5a(l) through 5a(3) at a higher water content. If, after several
trials at successively higher water contents, the soil pat *
III-16
EM 1110-2-1906
Appendix III
Change 2
2 0 Aug 86
* continues to slide in the cup or if the number of blows required
to close the groove is always less than 25, record that the
liquid limit could not be determined, and report the soil as non-
plastic without performing the plastic limit test.
w Record the number of drops (N) required to close

the groove. Remove a slice of soil approximately the width of
the spatula extending from edge to edge of the soil cake at right
angles to the groove and including that portion of the groove in
which the soil flowed together, place in a weighed container, and
cover.
(6) Return the soil remaining in the cup to the glass

plate. Wash and dry the cup and grooving tool and reattach the
cup to the carriage in preparation for the next trial.
(7) Remix the entire soil specimen on the glass plate

to reduce the water content of the soil and increase the number
0. of blows required to close the groove. Repeat 5 through
5a(6) for at least three additional trials producing successively
1 Fester numbers of blows to close the groove. Preferably, two
trials should produce closure in 25 blows or less, and two trials
should produce closure in 25 blows or more.
(8) Determine the water content (WN) of the soil

specimen from each trial in accordance with the procedure in
Appendix I, WATER CONTENT - GENEML. Make all weighings on the
same balance. Initial weighings should be performed immediately
after completion of the test. If the test is to be interrupted
for more than about 15 min, the specimens already obtained should
be weighed at the time of the interruption.
III-17
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
b0 Calculations.
(1) Plot the relationship between the water content,

'N and the corresponding number of drops, N I of the cup on a
semilogarithmic graph with the water content as ordinates on the
arithmetical scale, and the number of drops as abscissas on the
logarithmic scale. See Plate III-1 for an example data form.
Draw the best straight line through the four or more plotted
points.
(2) Take the water content corresponding to the

intersection of the line with the 250drop abscissa as the liquid
limit of the soil. Computational methods may be substituted for
the graphical method for fitting a straight line to the data and
determining the liquid limit.
60 PLASTIC LIMIT
a. Preparation of Test Specimen. Select a 20-g portion of

soil from the material prepared for the liquid limit test, either
after the second mixing before the test, or from the soil remain-
ing after completion of the test. Reduce the water content of
the soil to a consistency at which it can be rolled without
sticking to the hands by spreading and mixing continuously on the
glass plate. The drying process may be accelerated by exposing
the soil to the air current from an electric fan or by blotting
with paper that does not add any fiber to the soil such as hard
surface paper toweling or high wet strength filter paper.
b0 Procedure
(1) From the 20-g mass, select a portion of 1.5 to

2.0 g. Form the test specimen into an ellipsoidal mass. Roll * -..-
III-18
EM 1110~24906
Appendix III
ChqTge ?
20 Aug 86
* this mass between the palm or fingers and the ground-glass plate
with just sufficient pressure to roll the mass into a thread of
uniform diameter throughout its length. A normal rate of rolling
for most soils should be 80 to 90 strokes per minute counting a
stroke ,4as one complete motion of the hand forward and back to the
starting position. This rate of rolling may have to be decreased
for very fragile soils. The thread shall be further deformed on
each stroke so that its diameter is continuously reduced; and its
length extended until the diameter reaches 3.2 2 0.5 mm
!
(0.125 i 0.020 in.), taking no more than 2 minutes to complete
the rolling operation. A 3.2~mm (1/84n.) diameter rod or tube
1
is useful for frequent comparison with the soil threadto aster-
tain when the thread has reached the proper diameter especially
for inexperienced operators. The amount of hand or finger pres-
sure required will vary greatly according to the soil. Fragile
soils of low plasticity are best rolled under the outer edge of
the palm or at the base of the thumb. When the diameter of the
thread becomes 3.2 mm, break the thread into several pieces.
Squeeze the pieces together, knead between the thumb and first
finger of each hand, reform into an elliposidal mass, and reroll.
Continue this alternate rolling to a thread 3.2 mm in diameter,
gathering together, kneading and rerolling, until the thread
crumbles under the pressure required for rolling, and the soil
can no longer be rolled into a 3.20mm diameter thread (see
-Figure 7)
l It has no significance if the thread breaks into
threads of shorter length. Roll each of these shorter threads to
3.2 mm in diameter. The only requirement for continuing the test
is that they are able to be reformed into an ellipsoidal mass and
rolled out again. The operator shall at no time attempt to pro-
duce failure at exactly 3.2 mm diameter by allowing the thread to
reach 3.2 mm, then reducing the rate of rolling or the hand pres-
sure! or both, while continuing the rolling without further
deformation until the thread falls apart. It is permissible, *
III-19
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
*
Figure 1. Lean clay soil at the plastic limit
however, to reduce the total amount of deformation for feebly

plastic soils by making the initial diameter of the elliposidal
mass nearer to the required 3.2 mm final diameter. If crumbling
occurs when the thread has a diameter greater than 3.2 mm, this
shall be considered a satisfactory end point provided the soil
has been previously rolled into a thread 3.2 mm in diameter.
Crumbling of the thread will manifest itself differently with the
various types of soil. Some soils fall apart in numerous small
aggregations of particles, others may form an outside tubular
layer that starts splitting at both ends. The splitting pro-
gresses toward the middle, and finally,the thread falls apart in
many small platy particles. Fat clay soils require much pressure
to deform the thread, particularly as they approach the plastic *
III-20
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* limit. With these soils, the thread breaks into a series of
barrel-shaped segments about 3.2 to 9.5 mm (1/8 to 3/8 in.) in
length*?
(2) Gather the portions of the crumbled thread

together and place in a weighed container. Immediately cover the
container.
(3) Select another 1.5 to 2.0 g portion of soil from

the original 209g specimen and repeat the operations described in
17.1 and 17.2 until the container has at least 9 g of soil.
w Repeat 17.1 through 17.3 to make another container

holding at least 9 g of soil. Determine the water content, in
percent, of the soil contained in the containers in accordance
with the procedure given in Appendix I, WATER CONTENT - GENERAL.
Make all weighings on the same balance.
c. Calculations. Compute the average of the two water

contents. If the difference between the two water contents is
greater than two percentage points, repeat the test. The plastic
limit is the average of the two water contents.
70 PLASTICITY INDEX
a. Calculations. Calculate the plasticity index as

follows:
PI = LL - PL *
t A. Casagrande, R. C. Hirschfield, and S. J. POU~OS, Third

Progress Report on Investigation of Stress-Deformation and
Strength Characteristics of Compacted Clays, Soil Mechanics
Series No. 70, Harvard University (Cambridge, Mass., November
1963).
III-21
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* where
LL = the liquid limit
PL = the plastic limit
Both LL and PL are whole numbers. If either the liquid

limit or plastic limit could not be determined, or if the plastic
limit is equal to or greater than the liquid limit, report the
soil as nonplastic, NP.
b Plasticity chart. Errors in computing the liquid or

l
plastic limits sometimes can be detected by plotting the values

of liquid limit versus plasticity index on the plasticity chart?
as shown in Figure 8. The upper limit line starts from a liquid
limit of 8 at a plasticity index of 0 and rises toward the right
with a slope of 9 vertically on 10 horizontally; the equation of
the upper limit line, therefore, is PI = 0.9 (LL - 8). A plot of
liquid limit versus plasticity index for natural soils has never
been known to fall above the upper limit line. Plate III-2 is a
suggested form for the graphical correlation of the various
Atterberg limits data within a project or testing assignment.
8 l REPORT
a. Report the following information:
(1) Sample identifying information, *
f US Army Engineer Waterways Experiment Station, The Unified

Soil Classification System, Technical Memorandum NOe 3-357,
VOW 1 (Vicksburg, Misse, March 1953, revised April 1960). An
abridged version of the material in this report is presented in -
Military Standard MIL-STD-619A, 20 March 1962e
III-22
EM 1110-2-1906
Appendix III
Change 2
Zu Aug 86
*
60
lllllllllllllJ
0 10 20 30 40 50 60 70 80 90 100
LIQUID LIMIT
Figure 8. Plasticity chart showing classification

group symbols
(2) Any special specimen selection process used such

as removal of sand lenses from undisturbed sample,
(3)
Liquid limit, plastic limit, and plasticity index
to the nearest whole number and omitting the percent designation.
If the liquid limit or plastic limit is equal to or greater than
the liquid limit, report the soil as nonplastic, NP,
(4) An estimate of the percentage of sample retained

on the 425-pm (No. 40) sieve, and
wProcedure by which liquid limit was performed, if

it differs from the multipoint method.
III-23
EM 1110-2-1906
Appendix III
Change 2
20 Auq 86
9 0 PREPARING CLAY SHALE MATERIAL FOR TESTING
a. General. Investigations have shown that classification

indexes of clay shale materials are affected by air-drying and
slaking, by oven-drying and slaking, by the type and duration of
mechanical dispersing, and by other variations in procedure.
While the methods for preparing clay shale material for testing
should cover a sufficient range of disaggregation efforts to
assess the strength of interparticle bonds, the number of
variables allowed to influence the indexes must be minimized by
standardized procedures to prevent the classification of each
material becoming a minor research project in itself. Therefore,
three standard methods of processing clay shale material will be
used* These will be referred to as the blenderized, undried, and
air-dried methods.
The primary method is to test material that has been

essentially completely disaggregated by high-speed blenderizing;
this method will provide a reference value and it should be used
for all clay shale samples on which Atterberg limits are to be
determined. To provide additional indexes as desired, material
that has not been subjected to any drying and material that has
been subjected to a single cycle of air-drying and soaking may be
tested. These two methods should be used on sufficient represen-
tative samples to cover the range of samples identified by the
primary method.
b
Standard Methods. When material is to be prepared by
l
all three processing procedures, exercise care that the parent

material for the batches is similar. Divide the piece of sample
selected by a vertical cut into two parts with one piece about
twice as large as the other. Shave the smaller piece into dis-
tilled water to produce the undried batch, and use the larger * --
III-24
EM 1110-2-1906
Appendix III
Change 2
20 Aug 8 6
* piece to produce the other two batches. Figure 9 shows a flow

diagram of the three preparation methods and indicates when
separation of batches is required. Material may be taken from
each of the three batches and used for Atterberg limits deter-
minations without further processing. Details of each procedure
are as follows.
(1) Blenderized (primary method). Shave or shred

material at essentially natural water content and dry to a
constant weight in an atmosphere with a temperature less than
50' C and a relative humidity less than 30 percent. After a
constant weight is attained (and after a drying period of at
least 48 hr), soak the material in distilled water for at least
48 hr.
(a) Place about 500 ml of the slurry in the

l,OOO-ml glass container (available from any laboratory supply
company) of a Waring single-speed blender. Make the initial
water content of the slurry above 300 percent or more than twice
the estimated liquid limit (blenderized)! whichever is greater.
Typically, the weight of dry soil in the blender at any one time
should not exceed 150 g.
(b) Blenderize the slurry without interruption

for 10 min and then wash through a 4250pm (No. 40) sieve. Remove
excess water using a plaster of Paris dish lined with filter
paper. Work material at a water content above the liquid limit
in a thin layer on a glass plate with a steel spatula until no
further reduction in the size of lumps can be achieved.
(2) Undried Shave or shred material at essentially

natural water content, 1,mme diately place in distilled water, and
soak for at least 48 hr. A.fter removing excess water by
III-25
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
SELECT REPRESENTATIVE SAMPLE AT ESSENTIALLY NATURAL WATER CONTENT

.
v 1/3 SAMPLE 2/3 SAMPLE $
,
SHAVE OR SHRED 1JSlNG A KNIFE OR SALAD GRATER
IMMEDIATELY
3
I I .
SOAK IN DISTILLED WATER FOR 48 HR
.
r 1/2 P A R T - 1 /2 P A R T
1
I
BLENDERIZED M E T H O D
,
I .
. I I
REMOVE EXCESS WATER BY DECANTING
GRIND, WHILE WET, IN A MORTAR WITH

A RUBBER-TIPPED PESTLE TO PASS A
NO. 40 SIEVE
. .
WASH THROUGH A NO. 40 SIEVE
I I I
DRY TO WATER CONTENT WELL ABOVE LIQUID LIMIT USING PLASTER DISHES
b
WORK IN A THIN LAYER ON A GLASS PLATE WITH A STEEL SPATULA
Figure 9. Flow diagram showing the three standard

methods of preparing clay shale materials
III-26
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
-,k decanting, grind the wet material in a mortar with a rubber-

tipped pestle and wash through the 4250pm (No. 40) sieve. Remove
excess water using a plaster-of-Paris dish lined with filter
paper. Work material at a water content above the liquid limit
in a thin layer on a glass plate with a steel spatula until no
further reduction in the size of lumps can be achieved.
(3) Air-dried. Shave or shred material at essentially

natural water content and dry to a constant weight in an atmo-
sphere with a temperature less than 50C (120F) and a relative
humidity less than 30 percent. After a constant weight is
attained (and after a drying period at least 48 h& soak the
material in distilled water for at least 48 hr. After removing
excess water by decanting, grind the wet material in a mortar
with a rubber-tipped pestle and wash through the 4250urn (No. 40)
sieve. Remove excess water using a plaster-of-Paris dish lined
with filter paper. Work material at a water content above the
liquid limit in a thin layer on a glass plate with a steel
spatula until no further reduction in the size of lumps can be
achieved.
10 POSSIBLE ERRORS. Following are possible errors that would

0
cause inaccurate determinations of liquid and plastic limits:
a. General.
(1) Specimen not representative. As described in

paragraph 4a.t the liquid and plastic limits must be determined
using the same mixture of soil as that used for determinations of
natural water content or for other tests. Care should be taken
when using the trimmings from preparation of other test specimens
that material is as close as possible. *
III-27
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* (2)
Specimen improperly prepared. The specimens must
be thoroughly mixed and be permitted to cure for a sufficient
period before testing. Erroneous results may be caused by the
loss of colloidal material when removing particles coarser than
the No. 40 sieve or by testing air-dried or oven-dried soils.
(3 Inaccurate determination of water contents. The

possible errors described in paragraph 6 of Appendix I, WATER
CONTENT - GENERAL, would greatly affect the computed liquid and
plastic limits because of the small quantities of material
available for the water content determinations.
(4) Computational mistakes.
bI Liquid Limit Test
w Improperly constructed or adjusted liquid limit

device.
(2)
Worn parts of liquid limit device, especially at
point of contact between the cup and the base or worn tip of
grooving tool.
w Soil at point of contact between the cup and the

base. Removal of the cup for shaping and grooving the soil pat
will also ensure that the bottom of the cup and the top of the
base are clean. Any soil that has dropped onto the base can be
removed with one stroke of the back of the hand just before
replacing the cup.
(4) Loss of moisture during test. Erratic and errone-

ous results may be causing by drying of some soil mixtures unless
the test is performed in a humid room. 7? %-
III-28
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
* c. Plastic Limit Test
(11 Incorrect final thread diameter. A length of

1/8=in. -diameter metal rod close at hand will help in estimating
this diameter accurately.
(2) Stopping the rolling process too soon. If there

is any doubt as to whether the thread has crumbled sufficiently,
it is better to roll the thread once more than to stop the
process too soon.
III-29
EM 1110-2-1906
Appendix III
Change 2
20 Aug 86
*
LIQUID AND PLASTIC LIMIT TESTS
For use of this form, see EM 11 lC&2-1906.
DATE
PROJECT
BORING NO. SAMPLE NO.
LIQUID LIMIT
RUN NO. 1 2 3 4 5 6
TARE NO.
TARE PLUS WET SOIL
TARE
DRY SOIL
WATER CONTENT, % W
NUMBER OF BLOWS
LL
--.
PL
PI
Symbol from
plasticity chart
NUMBER OF BLOWS
PLASTIC LIMIT NATURAL

WATER
RUN NO. 1 2 3 4 5 CONTENT
TARE NO.
T A R E P L U S WETSOIL
WATER CONTENT, % W
PLASTIC LIMIT
1 A
REMARKS
TECHNICIAN COMPUTED BY CHECKED BY

\
GPO : 1983 0 - 407-134 PLATE I II-1

ENG , ;;;;5 3838
I
III-30
EM 1110-2-1906
Appendix III
Change 2
20 Auq. 86
I d XXINI ALl3l.LSVld
III-31
EM 1110-2-1906
Appendix IIIA
Change 2
20 /Jug 86
APPENDIX IIIA:
ONE-POINT LIQUID LIMIT TEST
-+ 1. INTRODUCTION: The one-point liquid limit test is based on

the experience that the slope of the liquid limit flow line for
soils within a given geologic environment is essentially a
constant on a logarithmic plot. Thus, the liquid limit can be
determined from one test point provided the constant defining the
slope has been established from correlations on the soil in
question. The one-point liquid limit test shall be used only in
those areas where the soils are geologically similar and adequate
correlations defining the slope of the liquid limit flow line
have been made.
This test is best performed by technicians who have

experience performing the four-point liquid limit described in
Appendix III and who can judge closely the consistency required
to cause closure at 20-30 blows. It is generally simpler and
faster for inexperienced technicians to perform the four-point
limit until experience has been gained.
2. APPARATUS AND PREPARATION OF SAMPLE. The apparatus required

is the same as that listed in paragraph 2 of Appendix III, LIQUID
AND PLASTIC LIMITS. The sample is prepared in the same manner as
that described in paragraph 4 for the standard liquid limit test
except that the soil to be tested is prepared initially to a con-
sistency that will require between 20 and 30 blows to cause
closure.
3. PROCEDURE
a. Proceed as described in Appendix III, LIQUID AND

PLASTIC LIMITS, paragraph 5a(l) through 5a(5) except that the *
IIIA-1
EM 1110-2-1906
Appendix IIIA
Change 2
20 Aug 86
* number of blows required to close the groove shall be 20 to 30.

If less than 20 or more than 30 blows are required, adjust the
water content of the soil and repeat the procedure.*
b. Immediately after removing a water content specimen as

described in paragraph 5a(5) of Appendix III, reform the soil in
the cup adding a small amount of soil to make up for that lost in
the grooving and water . content sampling
s-v. operations. Repeat
through 5a(5) of the above appendix, and, if the second closing
of the groove requires the same number of drops or no more than
two drops difference, secure another water content specimen.
Otherwise, remix the entire specimen and repeat.
C.Determine water contents of specimens as described in

paragraph 5a(8) of the above appendix.
4. CALCUGATIONS
a. (Determine the liquid limit for each water content

specimen using the following equation:
LL = WN
where
wN = water content
N = the number of blows causing closure of the groove

at water content
tan 6 = slope of the flow line
* Excess drying or inadequate mixing will cause the number of -

blows to vary.
IIIA-2
EM 1110-2-1906
Appendix IIIA
Change 2
23 Aug 86
* For those soils having an average slope of the liquid limit
flow line of tan fi = 0.121,t the equation LL = K(WN) may be used
where K = a factor given in Table 1. The liquid limit is the
average of the two trial liquid limit values.
b. If the difference between the two trial liquid limit

values is greater than one percentage point, repeat the test.
5. POSSIBLE ERRORS. Since this test is identical to the liquid

limit test except for the method of calculating the liquid limit,
it would be subject to the same possible errors listed in
paragraph 10 of Appendix III, LIQUID AND PLASTIC LIMITS.
TABLE 1. Factors for Obtaining Liquid Limit

-- from Water Content and Number of Drops
Causing Closure of Groove
N K
Number of Drops Factor for Liquid Limit
20 0.974
21 0.979
22 0.985
23 0.990
24 0.995
25 1.000
26 1.005
27 1.009
28 1.014
29 1.018
30 1.022
t U.S. Army Engineer Waterways Experiment Station, CE,

Comparison of Results of Liquid Limit Tests by Standard and
One-Point Methods, Miscellaneous Paper No. 3-488 (Vicksburg,
Miss., April 1962).
IIIA-3
E M 1440-2-1906
30 Nov 70
A P P E N D I X IIIB:
SHRINKAGE LIMIT TEST
1. DEFINITION. The shrinkage limit of a soil is the water content, ex-

pressed as a percentage of the weight of the oven-dried soil, at which
further loss in moisture will not cause a decrease in its volume. As part
of the shrinkage limit test, the shrinkage ratio, R, and linear shrinkage,
I s, are also usually determined._ The shrinkage ratio is defined as the
ratio between a given volume change and the corresponding change in
water content above the shrinkage limit. The linear shrinkage is defined
as the decrease in one dimension of a soil mass, expressed as a percent-
age of the original dimension, when the water content is reduced from a
given value to the shrinkage limit.
2. APPARATUS. The apparatus should consist of the following:
a.
- Evaporating dish; a porcelain evaporating dish approximately
4-I/2 in. in diameter is recommended,
b.
- Shrinkage dish; a circular porcelain or monel metal dish
I-3/4 in. in diameter and I/2 in. in height is recommended.
c. Glass cup, about 2 in. in diameter and about 1 in. in height
with the top rim ground smooth and flat.
d.
- Glass plate, 3 in. by 3 in. by i/l6 in. fitted with three metal
prongs for immersing the soil pat in mercury as shown in Figure i.
e.
- Mercury, sufficient to fill the glass cup to overflowing.
f
-* Spatula, having a blade about 4 in. long and about 3/4 in. wide.
g* Steel straightedge.
h. Balances, (1) sensitive to O.Oi g, and (2) sensitive to 0.1 g.
i. Oven (see Appendix I, WATER CONTENT - G E N E R A L ) .
3. PREPARATXON OF MATERIAL. Approximately 30 g of soil shall be
obtained from the thoroughly mixed portion of the material passing the
No. 40 sieve. The material to be used in the test should be prepared in
IIIB - 1
E M 1110-2-1906
Appendix IIIB -.
30 Nov 70
I 1/16 ti 3 X 3 G L A S S
R A D I U S = 0.54
BRASS PINS SECURED

WITH BALSAM
l/B DlAM\_
SHRINKAGE
DISH
GLASS PLATE EVAPORATING DISH
M E R C U R Y DISPL A C E D
TOP OF GLASS DISH
BY SOIL PAT
GROUND SURFACE
DRY SOIL PAT
Figure i. Apparatus for determining the volume of dry soil pat

of shrinkage limit test
IIIB - 2
EM ii10-2-1906
Appendix IIIB
30 Nov 70
the same manner as that described for the liquid limit test.
a. Record all identifying information for the specimen on a data
sheet; see Plate IIIB-i for suggested form.
b. Place the specimen in the evaporating dish and mix it thor-
oughly with distilled water. The amount of water added shall be sufficient
to make the soil wet enough to be readily worked into the shrinkage dish
without inclusion of air bubbles. The amount of water required to give
friable soils the desired consistency is equal to or slightly greater than
the liquid limit; the amount necessary to g i v e plastic soils the desired
consistency may exceed the liquid limit as much as i0 percent.
c. Coat the inside surface of the shrinkage dish with a thin layer
of petroleum jelly or similar compound to prevent the soil from adhering
to the dish. Place an amount of the wetted soil equal to about one-third the
volume of the dish in the center of the dish and tap the dish on a firm sur-
face, causing the soil to flow to the outer edges, Continue tapping the dish
until all air bubbles are eliminated from the soil. Repeat this step for
two more layers. The final layer shall fill the dish completely, with some
excess soil allowed to stand above the rim of the dish. Strike off the ex-
cess soil with a straightedge and remove all soil adhering to the outside
of the dish.
cl. Weigh the full dish of soil immediately and record the weight
on the data sheet as the weight of dish and wet soil. Allow the soil pat to
air-dry until a definite color change takes place and then oven-dry it to a
constant weight, Record the oven-dried weight as the weight of the dish
and dry soil. Determine and record the weight of the empty dish.
e. Determine the volume of the shrinkage dish by filling the dish
to overflowing with mercury,t removing the excess by pressing a glass
t Caution should be exercised in handling mercury. Mercury may have

toxic effects, particularly if spilled on the floor in areas without good
ventilation. See paragraph 7, HANDLING OF MERCURY.
UIB-3
E M j110-2-1906
Appendix IIIB
30 Nov 70
plate firmly over the top of the dish, and weighing the amount of mercury
required to completely fill the dish. The weight of the mercury divided
by its density (i3.53 g per cc) equals the volume of the inside of the
shrinkage dish. Record the volume of the shrinkage dish, which is equal
to the volume of the wet soil pat.
f. Place the glass cup in the evaporating dish and fill it with
mercury to overflowing. Remove the excess mercury by placing the glass
plate with the three metal prongs firmly over the cup; take care not to
trap air under the plate. Empty the excess mercury from the evaporating
dish and remove all mercury adhering to both the glass cup and the
evaporating dish with a brush.
g* Determine the volume of the soil pat by immersing the pat in

the mercury contained in the cup, using the glass plate with the three
metal prongs as shown in Figure i. Take care not to trap air under the
soil pat or glass plate. D e t e r m i n e the weight of the displaced mercury -
and compute its volume, as indicated in step e above, and record it as the
volume of the dry soil pat.
h. Record all information pertaining to the soil specimen such as
w e i g h t s , v o l u m e s , e t c . , on the data sheet, Plate IIIB-1.
5. COMPUTATIONS. a. Water Content. The water content, w, of
the soil at the time it was placed in the shrinkage dish is determined as
follows:
W
w =-Jx ioo
W
S
where w = weight of water in g, obtained by subtracting the weight

W
of the shrinkage dish plus dry soil from the weight of the
dish plus wet soil
ws = weight of oven-dried soil in g, obtained by subtracting

the weight of the shrinkage dish from the weight of the
dish plus dry soil
EM iiiO-2.-4906
Appendix IIIB
30 Nov 70
b. Shrinkage Limit. The shrinkage limit, SL, is calculated as

follows:
v - vs
SL = W-
C wS
x 400
)
where SL = shrinkage limit
w = water content of wet soil pat when placed in shrinkage
dish, expressed as a percentage of the weight of oven-
dried soil (see paragraph 2)
v = volume of wet soil pat, cc
weight of displaced mer-
cury in evaporating dish
vs = specific gravity of mer-
= volume of oven-dried soil pat, cc
cury (13.53 g per cc)
ws = weight of oven-dried soil pat, g
c. Shrinkage Ratio, The shrinkage ratio, R, shall be determined

by the following formula:
where and Vs are the same as given in paragraph b_.

wS
d. Linear Shrinkage. The linear shrinkage, Ls, shall be deter-

mined by the following formula:
Ls l ioo (i - Q&)
where c = volumetric change from a given water content, w (usually

the liquid limit)
c = (w - SL) R
IIIB-5
E M 1jjO-2-4906
Appendix IIIB
30 .Nov 70
6. POSSIBLE ERRORS. Besides errors in the preparation of soil mix-

tures given in paragraph 45 of Appendix III, LIQUID AND PLASTIC
LIMITS, following are possible errors that would cause inaccurate deter-
minations of shrinkage limit:
a. Inside of shrinkage dish not lubricated. If the soil adheres
to the shrinkage dish, the soil pat may crack during drying.
b. Air bubbles included in soil pat.

C. Soil pat dried too rapidly. To prevent the soil pat from
cracking, it should be dried slowly, first in the humid room and then in
the air of the laboratory, until a definite change in color is noted. Only
then should it be placed in the oven.
55 Air bubbles trapped beneath soil pat or glass plate when

immersing pat in mercury.
7. HANDLING OF MERCURY.* 2. Properties. Mercury is an odor-
less silver-white liquid at normal temperature and pressure. Mercury
has the property of forming amalgams with most metals, with the excep-
tion of platinum and iron. It is highly volatile, vaporizing at room tem-
perature to form vapors that are highly toxic. Mercury has a specific
gravity of i3.6, a boiling point of 356.9 C, and a freezing point of -38.9 C.
&!a H e a l t h H a z a r d s . The air concentration of mercury vapor

corresponding to the equilibrium vapor pressure at room temperature
(20 C) is approximately 20 mg per cu m, or 200 times the safe concentra-
tion of 0.i mg per cu m for continuous exposure (8 hr per day, 5 days per
week); safe concentrations for shorter periods have not been proposed.
Inhalation of mercury vapor of concentrations greater than 0.4 mg per
cu m over a long period of time can cause chronic poisoning. The initial
symptoms of poisoning may include gingivitis, digestive disturbance, fine
tremor of the extremities, irritability, excessive emotional response, and
* M e r c u r y Data Sheet 2Q3! National Safety Council, 425 N. Michigan

Ave., Chicago? Ill. 6flf$i.
IIIB-6
EM lliO-2-i906
Appendix IIIB
30 Nov 70
exaggerated salivation. These symptoms may increase in severity and

may result in permanent disability. Exposure to concentrations far
greater than the maximum allowable concentration for short periods of
time can cause acute illness. The exact nature, concentration, and dura-
tion of exposure determine the type and severity of symptoms, Although
inhalation of mercury vapor is by far the greatest avenue of entry to the
body, ingestion and absorption through the skin are also possible
C. Use. Whenever possible, use of mercury for teat purposes
should be avoided. Personnel working in a n area where mercury is used
should be made aware of its hazards.
ZL Detection. .Direct reading units for determining the concen-
tration of mercury vapor are available from the following sources:
Mine Safety Appliances Company
2ii N. Braddock Ave.
Pittsburgh, Pennsylvania 15208
Union Industrial Equipment Corporation (UNICO)
i50 Cove St.
Fall River, Massachusetts 02720
Acton Associates
Ii80 Raymond Blvd.
Newark, New Jersey 07102
Beckman Instruments, Inc.
2500 Harbor Blvd.
Fullerton, California 92634
20 Handling. The precautions listed below should be o b s e r v e d

when handling mercury.
(i) Mercury shouldnot be heated without elaborate control be-
cause of the rapid increase of vapor pressure with increase in temperature.
(2) Eating and smoking should not be permitted in areas
where mercury is handled. Hands should be thoroughly washed after
handling mercury.
(3) If clothing becomes contaminated, a change of clothes
should be available.
IIIB-7
EM 1110-2-1906
Appendix IIIB
30 Nov 70 -.
(4) Respiratory protection should be available where there is

a possibility of contamination.
(5) All laboratories handling mercury should have a precise
plan to be followed in decontamination after a mercury spill. Some
general proposals are:
(a) Maximum general exhaust ventilation and local ex-
haust should be effected. Windows should be opened.
(b) A vacuum flask or a vacuum cleaner designed for
removal of mercury should be put into service immediately to recover
the mercury.
(c) The area should be decontaminated by treating with
flowers of sulfur or other decontaminant.
(d) Effectiveness of decontamination should be verified
with a mercury vapor detector.
L* Facilities. In facilities where mercury is handled the meas-
ures listed below should be taken.
(1) Floors of areas should be free of cracks and the inter -
section of the wall and floor should be fitted with a cove,
(2) Recirculation of air in room should be discouraged be-
cause of the possibility of buildup of mercury vapor.
(3) Mercurymanometers shouldbe provided with traps to en-
sure that there will be no spillage of mercury into a process line or into room.
(4) Precision equipment should be removed from areas
where contamination with mercury is possible.
_g* Transportation and Storage. Whentransporting and storing
mercury the following precautions should be taken.
(i) Mercury containers should be placed in a tray when
transported within the laboratory; metal or continuous type (nonwelded)
plastic containers are preferable.to glass ones.
.(2) Mercury containers should be stored in pans that will
contain any spillage.
IIIB - 8
EM H&2-i906
30 Nov 70
APPENDIX IV:
SPECIFIC GRAVITY
i. TYPES OF SPECIFIC GRAVlTY, The specific gravity of a soil mass

for use in soils engineering calculations is usually expressed in three dif-
ferent forms: (a) the specific gravity of solids, Gs, applied to soils
finer than the No. 4 sieve; (b) the apparent specific gravity, Ga; and (c)
the bulk specific gravity, Gm, both (b) and (c) being applied to soils
coarser than the No. 4 sieve. The specific gravity of solids is not applied
to coarse particles because they normally contain voids from which air
cannot be displaced unless the particles are ground into finer particles to
eliminate the voids. Thus, when dealing with coarser particles it is more
convenient to work with the apparent specific gravity of the particle mass.
T h e v a l u e Gs or Ga is used in all calculations involving fundamental
properties of a soil mass. The bulk specific gravity is used in special
calculations, such as corrections of density and water content for soils
containing gravel sizes. Definitions and detailed procedures for deter-
mining the values of specific gravity of solids, apparent specific gravity,
and bulk specific gravity are given below.
2. S P E C I F I C G R A V I T Y OF S O L I D S , 2. Definition. The specific
gravity of solids, G of a soil is the ratio of the weight in air of a given
9
volume of soil particles at a stated temperature to the weight in air of an
equal volume of distilled water at a stated temperature.
b* Apparatus. The apparatus should consist of the following:

(1) Volumetric flask, 500-cc capacity
(2) Vacuum pump, with piping and tubing for connections to
each flask (as shown in Figure 2, page 4). The connection to each flask
should be provided with a tras to c.atch any water drawn from,the flask.
(3) Oven (see Appendix I, WATER CONTENT - GENERAL)
(4) Balance, sensitive to 0.01 g
EM ii40-2-i906
Appendix IV
30 Nov 70
(5) Thermometer, range 0 to 50 C? graduated in 0.1 \degree

(6) Evaporating dish
(7) Water bath
(8) Sieve, U. S. Standard No. 4 conforming to ASTM Designa-
tion: E 44, Standard Specifications for Sieves for Testing Purposes
C. Calibration of Volumetric Flask. The volumetric flask shall
be calibrated for the weight of the flask and water at various temperatures.
The flask and water are calibrated by direct weighing at the range of b
temperatures likely to be encountered in the laboratory. The calibration
procedure is as follows:
(4) Fill the flask with deaired-distilled (or deaired-
demineralized) water to slightly below the calibration mark and place
in a water bath which is at a temperature between 30 and 35 C. Allow
the flask to remain in the bath until the water in the flask reaches the
temperature of the water bath. This may take several hours. Remove
the flask from the water bath, and adjust the water level in the flask so 1
that the bottom of the meniscus is even with the calibration mark on
the neck of the fJask. Thoroughly dry the outside of the flask
I
and re-
move any water adhering to the inside of the neck above the graduation;
then weigh the flask and water to the nearest 0.01 g. Immediately after
weighing, shake the flask gently and determine the temperature of the
water to the nearest 0.i C by immersing a thermometer to the middepth
of the flask.
(2) Repeat the procedure outlined in step (i) at approxi-
mately the same temperature. Then make two more determinations, one
at room temperature and the other at approximately 5 degrees less than
room temperature.
(3) Draw a calibration curve showing the relation between
temperature and corresponding weights of the flask plus water. Prepare
a calibration curve for each flask used for specific gravity determinations
and maintain the curves as a permanent record. A typical calibration
lV-2
EM IilO-2-1906
Appendix IV
30 Nov 70
832
15 20 &REES 20
TEMPERATURE IN CENTIGRADE
Figure 1. Typical calibration curve of volumetric flask
curve (omitting the fine grid necessary for accurate determinations) is

shown in Figure I.
LL Preparation of Sample. Particular care should be taken to
obtain representative samples for determination of specific gravity of
solids. The sample of soil may be at its natural water content or oven-
dried; however, some soils, particularly those with a high organic content,
are sometimes difficult to rewet after having been oven-dried. These
soils may be tested without first being oven-dried, in which case the
oven-dry weight of sample is determined at the end of the test. When
the sample contains particles both larger and smaller than the No. 4
sieve, the sample shall be separated on the No. 4 sieve and a determi-
nation made of the specific gravity of the fine fraction and the apparent
specific gravity of the coarse fraction. The specific gravity value for the
sample shall be the composite specific gravity relation on the basis of
the solid volume of the components. When the specific gravity value is
to be used in calculations in connection with the hydrometer analysis
( s e e Appendix V, GRAIN-SIZE ANALYSIS), the specific gravity shall be
IV-3
EM ii10-2-i706
Appendix IV
30 Nov 70
d e t e r m i n e d on that portion o f the soil used for the hydrometer analysis

(usually that which passes the No. 200 sieve). It may be necessary to
use other liquids s u c h as kerosene) in lieu of distilled water for testing
soils containing soluble salts.
-e. Procedure, (1) Soils at natural water content. The proce-
dure for determining the specific gravity of soils at natural water content
shall consist of the following steps:
(a) Record all identifying
information for the sample
such as project, boring num-
ber, sample number, and other
pertinent data, on a data sheet
(see Plate IV-1 for suggested
form).
(b) Place a representative
sample of soil equivalent to
approximately 50 to 86 g oven-
dry weight in a dish and, by
means of a spatula, mix with
sufficient distilled or deminer -
alized water to form a slurry.
Place the slurry in a volumet-
ric flask and fill the flask ap-
proximately half full with ciis-
tilled water.
(c) Connect the flask to the
Figure 2. Evacuating air from samples vacuum line as shown in
in determination of specific gravity. A,
flask; B, splash trap; C, vacuum line Figure 2 and apply a vacuum
of approximately 29.0 in.
mercury. Agitate the flask gently at intervals during the evacuation proc-
ess; commercially available mechanical agitators have been used for
IV-4
E M 1110-2-1906
Appendix IV
30 Nov 70
this purpose. The length of time that vacuum should be applied will de-
pend on the type of .soil being tested. Soils of high plasticity and organic t
soils usually require 6 to 8 hr; some soils may require less time for re-
moval of air but this should be verified by experimentation. To ensure
continuous boiling, the temperature of the flask and contents may be ele-
vated somewhat above room temperature by immersing in a water bath at
approximately 35 C. Alternatively, entrapped air may be removed by
boiling$ the suspension gently for at least 10 min while occasionally roll-
ing the flask to assist in the removal of air The boiling process should
be observed closely as loss of material may occur. Allow flask and con-
tents to cool, preferably overnight, before filling and checking.
(d) Fill the flask with deaired distilled water to about
3/4 in. below the 500-cc graduation and apply a vacuum slightly less than
that which will cause vigorous boiling (as vigorous boiling may result in
a loss of solids). To determine if the suspension is desired, slowly re-
lease the.vacuum and observe the lowering of the water surface in the
neck of the flask. If the water surface is lowered less than I/8 in., the
suspension can be considered sufficiently deaired.
(e) Fill the flask until the bottom of the meniscus is
coincident with the calibration line on the neck of the flask. Thoroughly
dry the outside of the flask and remove the moisture on the inside of the
neck by wiping with a paper towel. Weigh the flask and contents to the
nearest 0.01 g. Immediately after weighing, stir the suspension to assure
uniform temperature, and determine the temperature of the suspension
to the nearest 0.1 C by immersing a thermometer to the middepth of the
flask.
t Air removal from organic soils usually cannot be accomplished by the

application of vacuum. In this case it will be necessary to boil the sus-
pension contained in the flask for about 30 min, adding distilled or de-
mineralized water carefully from time to time to prevent boiling the
sample dry. The flask should at all times be approximately half full.
1 Use of indirect heat such as a sand bath is recommended.
IV-5
EM 1110-2-4906
Appendix IV
30 Nov 70 -
(f) Carefully transfer the contents of the flask to an

evaporating dish. Rinse the flask with distilled water to ensure removal
of all of the sample from the flask. Oven-dry the sample to a constant
weight at a temperature of. i40 f 5 C. Allow the soil to cool to room tempera-
ture in a desiccator and determine the weight of the soil to the nearest 0.01 g.
(g) Record all weights on the data sheet.
(2) Oven-dried soils. The procedure for determining the spe-
cific gravity of solids for oven-dried soils shall consist of the following steps:
(a) Record information identifying the sample on the
data sheet (see Plate IV- 1).
(b) Oven-dry the soil toa constant weight at 110 f 5 C and
cool it to room temperature in a desiccator. Select a representative sample
of between 50 g (for cohesive soils) and 150 g (for cohesionless soils) and
weigh the sample to the nearest 0.04 g. After weighing, transfer the soil to a
volumetric flask, taking care not to lose any material during this operation.
To avoid possible loss of preweighed soil, the sample may be weighed -
after transfer to the flask. Fill the flask approximately half full with de-
aired distilled water and allow the suspension to stand overnight.
(c) Connect the flask to the vacuum line and apply a
vacuum of approximately 29.0 in. mercury for approximately 2 to 4 hr.
Entrapped air may also be removed by boiling as previously discussed;
however, the process should be observed closely to avoid loss of material
during boiling. Allow flask and contents to cool, preferably overnight,
before filling and checking.
(d) Perform the remainder of the test as outlined in
paragraph 2$(l)(d) and 2z($)(e).
(e) Record all weights on the data sheet.
L Computations. The following quantities are obtained by direct
weighing :
(4) Weight of flask plus water plus solids at test temperature
W in grams.
bws
IV-6
EM 1410-2-1906
Appendix IV
30 Nov 70
(2) Weight of tare plus dry soil in grams. The tare weight
_&J subtracted from this value to obtain the weight of dry soil, Ws. The
specific gravity of solids is computed to two decimal places by the
formula:
G = ws K- W
S
ws + wbw bws
where
K = correction factor based on the density of water at 20 C (see
Table IV-i). Unless otherwise required, specific gravity
values reported shall be based on water at 20 c.
= weight of flask plus water at test temperature .in grams (ob-
wbw
tained from calibration curve as shown in Figure 1).
3. APPARENT AND BULK SPECIFIC GRAVITY. 5. Definitions. The

apparent specific gravity, Ga, of a soil is the ratio of the weight in air of
a given volume of the impermeable portion of a permeable material (that
is, the solid matter including its impermeable pores or voids) at a stated
temperature to the weight in air of an equal volume of distilled water at a
stated temperature.
The bulk specific gravity of a soil, G , is the ratio of the
weight in air of a given volume of a permeable mzerial (including both
permeable and impermeable voids normal to the material) at a stated
temperature to the weight in air of an equal volume of distilled water at a
stated temperature.
b.
(f) Balance, having capacity of 5 kg or more and sensitive to
i.0 g
(2) Wire basket of No. 6 mesh, approximately 8 in. in diam-
eter and 8 in. high
(3) Suitable container for immersing the wire basket in
water, and suitable apparatus for suspending the wire basket from
IV-7
E M 1310:2-1906
Appendix IV
30 Nov 70
the center of the balance scale pan

( 4 ) T h e r m o m e t e r , range 0 to 50 C, graduated in 0.1 degree
C. Sample. The material to be tested shall be separated on the
No. 4 sieve and the material retained on the sieve used for the test. A
representative sample of approximately 2 kg is required. Samples may
be air -dried; however, oven-drying the sample before the: test may affect
the results and should be avoided when possible.
d.
- Procedure. The procedure for determinIng the apparent and
bulk specific gravity of soils retained on the No. 4 sieve shall consist of
the following steps:
(1) R e c o r d information identifying the specimen on the data
sheet (see Plate IV-i).
(2) Wash the specimen thoroughly to remove dust or other
coatings from the surfaces of the particles.
(3) Immerse the specimen in water at 15 to 25 C for a period
of 24 hr.
(4) Remove the specimen from the water and roll it in a
large absorbent cloth until all visible films of water are removed, although
the surfaces of the par,ticles may still appear to be damp. Wipe large
particles individually. Take care to avoid excess evaporation during the
operation of surface drying.
(5) Obtain the weight in grams of the saturated surface-dry
specimen. The specimen in this and subsequent weighings should be
weighed to the nearest 1.0 g.
(6) Immediately after weighing, place the specimen in the
wire basket and determine the weight of the specimen in water. Deter -
mine and record the temperature of the water in which the specimen is
immersed.
(7) Oven-dry the specimen to a constant weight at ii0 f 5 C.
After cooling to room temperature, weigh the specimen.
(81 Record all weights on the data sheet.
IV-9
E M iilO-2-1906
Appendix IV
30 Nov 70
e.
- Computations. The following quantities are obtained by
direct weighing:
(i) W e i g h t o f tare plus oven-dried soil in grams. The tare
weight is subtracted from this value to obtain the weight of dry soil, A,
on the data sheet.
( 2 ) W e i g h t o f tare plus saturated surface-dry soil in grams.
The tare weight is subtracted from this value to obtain the weight of satu-
rated surface-dry soil, B.
(3) Weight of wire basket plus saturated soil in water in
grams. The weight of wire basket in water is subtracted from this value
to obtain the weight of saturated soil in water, C.
The apparent specific gravity is computed to two decimal
places by the formula:
G =&
a
where
-
K = correction factor based on the density of water at 20 C (see
Table IV-i).
The bulk specific gravity is computed to two decimal places
by the formula:
G Z&
m B - C
When a soil is composed of particles both larger and smaller

than the No. 4 sieve, the specific gravity of the soil for use in engineering
calculations shall be computed as follows:
ioo
G=
70 passing No. 4 sieve + 7o retained on No. 4 sieve
G G
S a
4. POSSIBLE ERRORS. Follow$ng are possible errors that would cause

inaccurate determinations of specific gravity:
a. Specific Gravity of Solids.
- (i) I m p r e c i s e w e i g h i n g o f
IV-10
E M IiiO-2-i906
Appendix IV
30 Nov 70
flask and contents. Since the computation of the specific gravity of solids
is based on a difference in weights which is small in comparison with the
weights themselves, the same balance should be used for calibrating the
volumetric flask and for determining the specific gravity whenever the
calibration curve is used.
(2) Temperature of flask and contents not uniform. Both in
calibrating the flask and determining the specific gravity, utmost care
should be taken to insure that measured temperatures are representative
of the flask and contents during the times when the weighings are made.
(3) Flask not clean. The calibration curve will not remain
valid if dirt accumulation changes the weight of the flask. Also, if the in-
side of the neck is not clean, an irregular meniscus may form,
(4) Moisture on outside of flask or inside of neck. When
calibrating the flask for a temperature lower than room temperature,
there is a tendency for condensation to form on the flask despite careful
drying and rapid weighing. Whenever possible, weighing should be done at
approximately the same temperature as that of the flask.
(5) Meniscus not coincident with mark on neck of flask. One
drop of water too much makes an error of approximately 0.05 g. This
error can be minimized by taking the average of several readings at the
same temperature. When the suspension is opaque, a strong light behind
the neck is helpful in seeing the bottom of the meniscus.
(6) Use of water containing dissolved solids. It is essential
that distilled or demineralized water be used exclusively to insure the
continued validity of the flask calibration curve.
(7) Incomplete removal of entrapped air from soil suspension.
This is the most serious source of error in the specific gravity deter-
mination and will tend to lower the computed specific gravity. The sus-
pension must be thoroughly evacuated or boiled and the absence of
entrapped air verified as described in paragraph 2z(i)(d). (It should be
noted that air dissolved in the water will not affect the results, so it is not
IV-ii
E M 1110-2-1906
Appendix IV
30 Nov 70
necessary to apply vacuum to the flask when calibdating or after filling the
flask to the calibration mark.)
(8) Gain in moisture of oven-dried s ecimen before weighing.
If the specimen is oven-dried before the specific ravity determination, it
must be protected against a gain in moisture unti I it can be weighed and
placed in the flask.
(9) Loss of material from oven-dri4 d specimen. If the
specimen is oven-dried and weighed before being placed in the flask, any
loss of material will lower the computed specific gravity.
b.
- Apparent and Bulk Specific Gravity. (1) Loss of moisture
from saturated surface-dry particles before weighing. Unless the satu-
rated surface-dry material is weighed promptly, evaporation may cause
an increase in the computed bulk specific gravity,
(2) Failure to correct for the change in density of water with
temperature. This correction is often overlooked when computing either
-
the apparent or bulk specific gravity.
IV-12
EM 1110-2-1906
Appendix IV
30 Nov 7t
StiEClZJIC GRAVITY TESTS
Date
'roJect-- _ _
mlnij NL _ .~
.-.
SPECIFIC GRAWl'Y OF
APPAPmT (C$ AND EWIX (Cm) SPECIFIC GRAVITY
re basket
-. + ~011) I
--_-~~~--
Computed by Checked by
PLATE IV-i
IV-13
EM IiiO-Z-i906
3 0 Nov 70
APPENDIX V:
GRAIN-SIZE ANALYSIS
i. DEFINITION. Grain-size analysis is a process in which the propor-

tion of material of each grain size present in a given soil (grain-size
distribution) is determined. The grain- size distribution of coarse -grained
soils is determined directly by sieve analysis, while that of fine-grained
soils is determined indirectly by hydrometer analysis. The grain-size
distribution of mixed soils is determined by combined sieve and hydrometer
analyses, Detailed procedures for determining the grain-size distribution
of soils by sieve, hydrometer, and combined analyses are given below.
2. SIEVE ANALYSIS. a. Description. A sieve analysis consists of
passing a sample through a set of sieves and weighing the amount of mate-
rial retained on each sieve, Sieves are constructed of wire screens with
square openings of standard sizes. The sieve analysis is performed on
material retained on a U. S. Standard No. 200 sieve. The sieve analysis,
in itself, is applicable to soils containing smali amounts of material
passing the No. 200 sieve provided the grain-size distribution of that por-
tion of the sample passing the No. 200 sieve is not of interest.
b
-* Apparatus. The apparatus should consist of the following:
(i) A series of U. S. standard sieves with openings ranging
from 3 in. to 0.074 mm (No. 200), including a cover plate and bottom pan,
conforming to ASTM Designation: E ii, Standard Specifications for Sieves
for Testing Purposes.* The number and sizes of sieves used for testing a
given soil will depend on the range of soil sizes in the material, and the
intended use of the gradation curve.
(2) Sieve shaker, a mechanical unit which can produce on
duplicate samples the same consistent results as those obtained by the
circular and tapping motion used in hand sieving. Typical commercially
available mechanical shakers are the Tvler Ro-Tan and the Combs and
* See page V-26 for U. S. Standard Sieve Sizes or numbers and sieve
openings in inches and millimeters.
v-i
EM 1110-2-1906 -
Appendix V
Change 2
20 Aug 86
Syntron machines; there appears to be no significant differences in the
resultsobtained among these machines.t
(3) Balances, sensitive to 0.i g for samples weighing less
than 500 g, and to i.0 g for samples weighing over 500 g.
(4) Paintbrush, 1 in., or soft wire brush, for cleaning sieves.
(5) Sample splitter or riffle for dividing samples.
(6) Mortar and rubber-covered pestle, for breaking up
aggregations of soil particles.
(7) Oven, similar to that described in Appendix 1, WATER
CONTENT - GENERAL.
C. Preparation of Sample.* The material to be treated is first air-
dried, after which the aggregations present in the sample are thoroughly
broken up with the fingers or with the mortar and pestle. A representative
sample is then obtained by dividing, using the sample splitter or riffle.
The size of the sample to be used will depend on the maximum particle -
size in the sample and the requirement that the sample be representative
of the material to be tested. The sample should be limited .in weight so
that no sieve in the series will be overloaded. Overloading of a sieve will
result in incomplete separation with errors in the test. The following
tabulation will be used as a guide in obtaining a minimum-weight sample:
Maximum Particle Size Minimum Weight of Sample, g
3 in. 64,000 g
2 in. 19,000 g
l-l/2 in. 8,000 g
1 in. 2,400 g
3/4 in. 1,000 g
112 in. 300 g
3/8 in. 150 g
No. 4 50 g *
7 U.S. Army Engineer Waterways Experiment Station, CE, Sieve Analy-

ses of Granular Soils by Division Laboratories, Engineering Study 516
(Vicksburg, Miss., October 1963).
$ Clay shale materials require special preparation. See paragraph 5.
v-2
EM iiiO-2-1906
Appendix V
30 Nov 70
If the sample contains more than about 40% of sizes larger

than the vo. 4 sieve, it is generally advisable to separate the material on
the No. 4 sieve, retaining both fractions for independent sieve analysis as
subsequently described. If the sample contains plastic fines which tend to
form hard lumps or to coat the coarser particles during air-drying, the
entire sample should be placed in a pan filled with water and allowed to
soak until all the soil lumps or the coatings have disintegrated, before it
is separated on the No. 4 sieve. The coarser fraction and the fraction
passing the No. 4 sieve including the fines and water should be retained
for independent sieve analysis ai subsequently described.
d
-0 Procedure. (i) Material predominantly finer than the No. 4
sieve. The procedure for samples predominantly finer than the No. 4
sieve consists of the following steps:
(a) Record all identifying information for the sample,
such as project, boring number, or other pertinent data, *on a data sheet
(see Plate V-i for suggested form).
(b) Oven-dry the sample at ii0 f 5 C, allow to cool, and
weigh. If the sample weighs less than 500 g, weigh it to the nearest 0.1 g;
if the sample weighs over 500 g, weigh to the nearest 1 g. Record the dry
weight of the sample on the data sheet.
If the sample consists of clean sands or gravels,
proceed with step (f).t If the sample contains plastic fines which tend to
form hard lumps or to coat the coarser particles during oven-drying,
place the oven-dry sample in a pan filled with enough water to cover all
the material and allow it to soak until all the soil lumps or coatings have
disintegrated. The length of time required for soaking will vary from
about 2 to 24 hr, depending in general on the amount and plasticity of the
fines.
(d) Transfer the sample and water from the pan to a
No. 200 sieve, or if the sample contains an appreciable amount of coarse
t If there is any doubt concerning the cleanness of a sand or gravel, i.e.

whether or not the particles may be coated with fines, or if the test is
performed to determine whether or not a material complies .with speci-
fications, then the sample should be treated as subsequently described
in steps (c) through (e).
v-3
EM IiIO-2-i906
Appendix V
30 Nov 70
particles, to a combined set of No. 4 and No. 200 sieves. Care should be
taken not to overload the No. 200 sieve; if necessary, transfer the sample
in increments. Wash the sample thoroughly through the sieves, discarding
the material passing the No. 200 sieve. Larger particles in the sample
may be individually washed and removed from the sieves.
(e) Oven-dry the combined material retained on the No. 4
and the No. 200 sieves and, after the sample has cooled, weigh. Record on
the data sheet in the Weight Retained in grams column the difference
between the original oven-dry weight aad the oven-dry weight after wash-
ing. Use the washed sample for the remainder of the analysis.
( f) Select a nest of sieves suitable to the soil being
tested. The choice of sieves usually depends on experience and judgment,
and the use for which the grain-size curve is intended. Select as the top
sieve. one with openings
slightly larger than the
diameter of the largest
particle in the sample.
Arrange the nest of
sieves according to size
as shown in Figure i,
with decreasing open-
Figure 1. Arrangement of sieves for grain-size
ings from top to bottom.
analysis
Attach the bottom pan
to the bottom of the smallest sieve used. Place the sample on the top
sieve of the nest as shown in Figure 2 and put the cover plate over the top
sieve.
(g) Place the nest of sieves in the shaking machine as
shown in Figure 3 and shake them for i0 min, more or less, or until addi-
tional shaking does not produce appreciable changes in the amounts of
material on each sieve. If a shaking machine is not available, the nest of
sieves may be shaken by hand. In the hand operation, shake the nest of
V-4
E M 11~0-2-1906
Appendix V
30 Nov 70
sieves with a lateral and

vertical motion, accotn-
panied by jarring, to
keep the material rnov-
ing continuously over the
surfaces of the sieves.
Jarring is accomplished
by occasionally dropping
the nest lightly on sev-
eral thicknesses of
..
magazines. The nest i:
should not be broken to

Figure 2. Placing soil on sieves
rearrange particles or to
manipulate them through a sieve by hand. Hand-shaking should be con-
tinued for at least 15 min.
(h) Remove the nest of sieves from the mechanical shak-
er, if used. Beginning
with the top sieve, trans-
fer the contents of the
sieve to a piece of heavy
paper approximately 1 ft
square. ~-Carefully invert
the sieve on the paper
and gently brush the bot-
tom of the sieve, as
shown in Figure 4, to re-
move all the sample.
Transfer the sample
from the paper to the bal-
ance and weigh in accord-
Figure 3. Nest of sieves placed in typical
machLIe for shaking ance with requirements in
v-5
EM 1110-2-1906
Appendix V
30 Nov 70
step (b). Care should be

exercised that no loss of
material occurs during
th& transfer. Coarser
fractions may be trans-
ferred more readily from
the sieves directly onto
the balance pan. Record
the weight of material re-
Figure 4. Removing soil from sieves tained on each sieve on
the data sheet.
(i) Repeat step (h) for each sieve. The sum of the
weights retained on each sieve and pa? should equal the initial total weight
of the sample within 1 percent. If the difference is greater than 1 percent,
the sieving should be repeated.
(2) Material split on No. 4 sieve. The procedure for samples
which have been split on the No. 4 sieve consists of the following steps:
(a) Record pertinent information fbr the sample on a
data sheet (see Plate V-i for suggested form).
(b) Oven-dry the sample, allow it to cool, and weigh the
fraction retained,on the No. 4 sieve. Record the oven-dry weight on the
data sheet. Alternatively, the air-dry weights of the total sample and the
fraction retained on the No. 4 sieve may be utilized and the air-dry mate-
rial retained on the No. 4 sieve used in the sieve analysis as in step (c)
below. In the latter procedure, the relative percentages of materials
greater than the No. 4 sieve are determined on an air-dry basis. This
method is satisfactory provided the air-dry water contents of the plus and
minus No. 4 portions of the sample are approximately equal.
(c) Proceed as in paragraphs 2g(i)(f) through 2i(i)(i).
In general, it is advisable to use large sieves and a Ty-Lab or Gilson
shaker for the coarse fraction.
V-6
EM iIiO-2-i906
Appendix V
30 Nov 70
(d) If the sample has not been washed during the pre-
liminary treatment, process the material passing the No. 4 sieve accord-
ing to paragraphs 2d( i)(b) through 2d(
- I)( i). If the material has been
washed as part of the preliminary treatment, proceed with paragraphs
2cJ I)(d) through 2$ i)( i), except that the material passing the No. 200
sieve in paragraph 2d(f)(d) should be oven-dried and weighed. This
weight is added to the oven-dry w,eight of the plus No. 200 material to ob-
tain the total weight of sample.
e. Computations. The percentage of material by weight retained
on the various sieves is computed as follows:
weight in g retained on a sieve

Percent retained =
total weight in g of oven-dry sample joo
If the sample has been split on the No. 4 sieve during preliminary treat-
ment and the air-dried coarser fraction sieved independently, the percent
retained for the coarser fraction is computed as follows:
air-dry weight in g retained on a sieve

Percent retained = x 100
air-dry weight in g of total sample
Similarly, for the finer fraction when oven-dry weights are used:
weight in g retained on a sieve

Percent retained = x percent passing No. 4
oven-dry weight in g of sample
passing No. 4 sieve
where the percentage passing No. 4 sieve is computed on an air-dry basis.

The values of percent retained based on the above formulas refer to the
total weight of sample. Computation of a partial percent retained as indi-
cated in Plate V-i is necessary only when the sample is initially separated
on the No. 200 sieve for purposes of a combined analysis, as subsequently
described. The cumulative percent finer by weight than an individual
sieve size (percent finer) is calculated by subtracting the percent retained
v-7
EM IijO-2-*906
Appendix V
30 Nov 70
on the individual sieve from the cumulative percent finer than the next
larger sieve.
-f. Presentation of Results. The results of the sieve analysis

are presented in the form of a grain-size distribution curve on a semi-
logarithmic chart as shown in Plate V-2, The grain-size distribution
curve is obtained by plotting particle diameter (sieve opening) on the
abscissa (logarithmic scale) and the percent finer by weight on the ordi-
nate (arithmetic scale).
3. HYDROMETER ANALYSIS. a. Description. The hydrometer
method of analysis is based on Stokes law, which relates the terminal
velocity of a sphere falling freely through a fluid to the diameter. The
relation is expressed according to the equation:
where v = terminal velocity of sphere, cm per set

Ys = density of sphere, g per cm3
Yf = density of fluid, g per cm3
2
J-l = viscosity of fluid, g-set per cm
D = diameter of sphere, mm
It is assumed that Stokes law can be applied to a mass of dispersed soil

particles of various shapes and sizes. The hydrometer is used to deter-
mine the p.ercentage of dispersed soil particles remaining in suspension
at a given time. The maximum grain size equivalent to a spherical parti-
cle is computed for each hydrometer reading using Stokes law. The
hydrometer analysis is applicable to soils passing the No. IO sieve for
routine classification purposes; E,hen greater accuracy is required (such
as in the study of frost-susceptible soils), the hydrometer analysis should
be performed on only the fraction passing the No. 200 sieve (see para-
graph COMBINED ANALYSIS).
v-l3
EM .lIiO-2-4906
Appendix V
30 Nov 70
b.
(i) Hydrometer, calibrated at 20/20 C (68/68 F), graduated
in specific gravity or grams per liter with a range of 0.995 to i.040 and
0 to 50, respectively. The accuracy of the specific gravity hydrometer
shall be l O.OOI and of the
gram-per-liter hydrome-
ter, *I.
(2) Dispersion
apparatus, either of two
types may be used:
(a) A me-
chanically operated stirring
device in which a suitably DETAILS Of BAFFLE
STIRRING PADDLf LOCATION PLAN
mounted electric motor turns
a vertical shaft at a speed
of not less than 10,000 rpm
without load. The shaft shall
be equipped with a replace-
able stirring paddle of metal,
plastic, or hard rubber. De-
PERMANENT
-l- ,\ ROM
tails of a typical paddle are

shown in Figure 5. A special
dispersion cup conforming
to either of the designs
shown in Figure 5 shall be
provided to hold the sam-
ple while it is being
dispersed. !+----25*Lwf--4
DlSPfRSlON CUPS
(b) An air
dispersion device such as the
Figure 5. Detail of stirring piaddle
air -jet dispersion tube device and dispersion cups
v-9
EM 1140-2-i906
Appendix V
30 Nov 70
developed at Iowa State Col1ege.t

(3) Sedimentation cylinder, of glass, essentially 18 in, high
and 2-i/2 in. in diameter and marked for a volume of iOO0 ml.
(4) Centigrade thermometer, range 0 to 50 C, accurate to
0.5 c.
(5) Timing device, a watch or clock with a second hand.
(6) Balance, sensitive to 0.i g.
(7) Oven (see Appendix I, WATER CONTENT - GENERAL).
-C. Hydrometer Calibration. ,The hydrometer shall be calibratedt

to determine its true depth in terms of the hydrometer reading (see Fig. 6)
in the following steps:.
(I) Determine the volume of the hydrometer bulb, VR. This
may be determined in either of two ways:
(a) By measuring the volume of water displaced. Fill a
IOOO-cc graduate with water to approximately 700 cc. The water should
-
be at about 20 C, Observe and record the reading of the water level. ,.
Insert the hydrometer and again observe and record the reading. The dif-
ference in these two readings equals the volume of the bulb plus the part
of the stem that is submerged. The error due to inclusion of this latter
quantity is so small that it may be neglected for practical purposes.
(b) By determining the volume from the weight of\\the
hydrometer. Weigh the hydrometer to 0.01 g on the laboratory balance.
Since the specific gravity of a hydrometer is, about unity, the weight in
grams may be recorded as the volume in cubic centimeters. This volume
includes the volume of the bulb plus the volume of the stem. The error
7 T. Y. Chu and D. T. Davidson, Simplified air-jet dispersion apparatus.

for mechanical analysis of soils, Proceedings, Highway Research
Board, vol. 32 (1953), pp. 541-547.
$ ASTM hydrometers 151 H or 152 H (ASTM Designation: E 100) have a
uniform size; therefore, only a single calibration is required, which
can be applied to all ASTM hydrometers of this type.
V-IO
EM iiiO-2-1906
Appendix V
30 Nov 70
due to inclusion of the stem

volume is negligible.
(2) Determine the
area, A, of the graduate in
which the hydrometer is to be
used by measuring the dis-
tance between two gradua-
tions. The area, A, is equal
to the volume included between
~SPENS,O,, A0OE CCNTER
the graduations divided by the

measured distance.
(3) Measure and
record the distances from the
lowest calibration mark on the
stem of the hydrometer to
each of the other major cali-
bration marks, R.
(4) Measure and
record the distance from the
neck of the bulb to the lowest
calibration mark. The dis-
tance, HI, car re sponding to
Figure 6. Hydrometer calibration
a reading, R, equals the sum
of the two distances measured in steps (3) and (4).
(5) Measure the distance from the neck to the tip of the bulb.
Record this as h, the height of the bulb. The distance, h/2, locates the
center of volume of a symmetrical bulb. If a nonsymmetrical bulb is
used, the center of volume can be determined with sufficient accuracy by
projecting the shape of the bulb on a sheet of paper and locating the center
of gravity of this projected area.
(6) Compute the true distances, HR, corresponding to each
EM 1110-2-3906
Appendix V
30 Nov 70
of the major calibration marks, R, from the formula:
(7) Plot the curve expressing the relation between HR and

R as shown in Figure 6. The relation is essentially a straight line for
hydrometers having a streamlined shape.
d. Meniscus Correction. Hydrometers are calibrated to read
correctly at the surface of the liquid. Soil suspensions are not transparent
and a reading at the surface is not possible; therefore, the hydrometer
reading must be made at the upper rim of the meniscus. The meniscus
car re ction, Cm, which is a constant for a given hydrometer, is deter-
mined by immersing the hydrometer in distilled or demineralized water
and observing the height to which the meniscus rises on the stem above the
water surface, For most hydrometers it will be found that Cm is equal
to approximately 0.5, and this value can be assumed for routine testing.
e. Preparation of Sample. The approximate size of sample to be
used.for the hydrometer analysis varies according to the type of soil being
tested, as shown in the tabulation below:
Soil Type Dry Weight, g

Fat clays 30
Le?n clays and silty soils 50
Sandy soils 75t
7 Up to i50 g of sandy soil can be used for the hy-
drometer ,analysis provided no more than 50 g of
the sample is finer than-the No. 200 sieve.
The exact dry weight of the sample in suspension may be determined

either before or after the test. However, oven-drying some clays
before the test may c+use permanent changes in the apparent grain sizes.
Samples of such soils should, if possible, be preserved at the natural
v-42
EM lllO-2- 1906
Appendix V
Change 2
ZU Aug 86
water content and tested without first being oven-dried, the dry weight either
being obtained after the hydrometer analysis or computed according to
the formula:
weight of wet soil
Dry weight of specimen =
water content
it
ioo
ws= w
i t o.oi w
w having been determined on an untested portion of the sample.
Furthermore, if samples are dried and weighed before the test, any loss
of material during the test will affect the results.
f. Dispersing Agent. Very fine soil grains in a suspension nor-
mally will tend to flocculate, i.e. to adhere with sufficient force that they
settle together. Consequently, a dispersing agent to pr ev ent flocculation
of the soil grains during the test should be added to all samples. The fol-
lowing dispersing agents, listed in approximate order of effectiveness,
have been found to be satisfactory for most types of soi1s.f
Stock Solution
Concen- g Per
Dispersing Agents tration liter Manufacturer
1 Sodium tripolyphosphate 0.4 N 29 Blockson Chem. Co.,
Joliet, Ill.
2 Sodium polyphosphate 0.4 N 36 Blockson Chem. Co.,
Joliet, Ill.
3 Sodium tetraphosphate 0.4 N 3i Rumford Chem. Works,
(trade name Quadrafos) Rumford, R. I.
* 4 Sodium Hexametaphosphate 0.4 N 41 Most laboratory
(sometimes called sodium chemical supply c OS .
metaphosphate) adjusted to
pH8-9 with Na2C03
The chemical product Calgon available in grocery stores shall not be used
as a dispersing agent as it no longer contains sodium hexametaphosphate.
Sodium silicate shall not be used as a dispersing agent since it gives un-
satisfactory dispersion while at the same time permitting flocculation to a
t A. M. Wintermyer and E. B. Kinter, A study of dispersing agents for
particle-size analysis of soils, Public Roads, vol. 28, No. 3 (August
1954), pp 55-62.
v-i3
EM i1i0-2-i906
Appendix V
30 Nov 70 -.
point where it is not apparent to visual examination. Phosphate solutions are

somewhat unstable and therefore shouldnot be stored for extended periods of
time. In most instances, i5 ml of a dispersing agent solution is adequate.
However, should flocculation tend to continue, a second or third addition
of i5 ml of solution may be added.
The addition of a dispersing agent to the soil suspension re-
sults in an increase in density of the liquid and necessitates a correction
to the observed hydrometer reading. The correction factor, Cd, is deter-
mined by adding to a IOOO-ml graduate partially filled with distilled or de-
mineralized water the amount of dispersing agent to be used for the par-
ticular test, adding additional distilled water to the lOOO-ml mark, then
inserting a hydrometer and observing the reading. The correction factor,
is equal to the difference between this reading and the hydrometer
=d8
reading in pure di6tilled or demineralized water.
The addition of a dispersing agent also increases the weight o
solids in the suspension. If the oven-dry weight of soil used for the hy- -
drometer analysis is obtained at the end of the test, this weight must be
corrected by subtracting the dry weight of the dispersing agent used.
&a Procedure. The procedure shall consist of the following steps:
(1) Record all identifying information for the sample, such
as project, boring number, or other pertinent data, on a data sheet (see
Plate V-3 for suggested form).
(2) Determine the dispersing agent correction, Cd, and the
meniscus correction, C unless they have been previously established.
m
Record this information on the data sheet.
(3) D e t e r m i n e or estimate the specific gravity of solids and
record on the data sheet.
(4) If the oven-dry weight is to be obtained at the start of the
test, oven-dry the sample, allow to cool, and weigh to nearest 0.i g. Re-
cord the dry weight on the data sheet. Place the sample in a numbered dish
and add distilled or demineralized water until the sample is submerged. Add
v-i4
EM IilO-2-1906
Appendix V
30 Nov 70
the dispersing agent at this time. Allow the sample to soak overnight or
until all soil lumps have disintegrated. Highly organic soils require special
treatment, and it may be necessary to oxidize the organic matter in order to
perform a hydrometer analysis on these soils. Oxidation is accomplished
by mixing the sample with a solution of 30 percent hydrogen peroxide; this
solution will oxidize all the organic matter. If only small amounts of or-
ganic matter are present, treatment with hydrogen peroxide may be omitted.
(5) Transfer the soil-water slurry from the dish to a dispersion
cup (Fig. 5), washingt any residue from the dish with distilled or deminer-
alized water. Add distilled water to the dispersion cup, if necessary, until
the water surface is 2 or 3 in. below the top of the cup; if the cup contains
too much water, it will splash out while mixing. Place the cup in the dispers-
ing machine and disperse the suspension for i to i0 min. The lower the plas-
ticity of the soil the shorter the time required to disperse it in the cup.*
(6) Transfer the suspension into a iOOO-ml sedimentation
cylinder and add distilled or d.emineralized water until the volume of the
uspension equals iOO0 ml. The suspension should be brought to the tem-
perature expected to prevail during the test.
(7) One minute before starting the test, take the graduate in
one hand and, using the palm of the other hand or a suitable rubber cap as
a stopper, shake the suspension vigorously for a few seconds in order to
transfer the sediment on the bottom of the graduate into a uniform suspen-
sion. Continue the agitation for the remainder of the minute by turning the
cylinder upside down and back. Sometimes it is necessary to loosen the
t A large syringe or wash-water bottle is a convenient device for. handling

the water in the washing operation.
3 Air dispersion may be used in place of mechanical dispersion. A dis-
persion time of i0 min is recommended, using an air pressure of 25 psi
for clays and silts and i0 psi for sands. Several comparative tests in-
dicate that the air dispersion apparatus gives a higher degree of dis-
persion of clayey soils while causing less degradation of sands than the
mechanical stirring apparatus. See: Chu and Davidson, op. cit., and
U. S. Bureau of Reclamation, Comparison of Dispersion Methods for
Soil Gradation Analysis< Earth Laboratory Report No. EM-6i8 (Denver,
Cola., May 1961).
v-is
EM 1110-2-i906
Appendix V
30 Nov 70
sediment at the bottom of the cylinder by means of a glass rod before

shaking. Alternatively, the suspension may be agitated by means of a hand
agitator for one minute prior to testing. A schematic drawing of a hand
agitator is shown in Figure 7. A uniform distribution of the soil particles
in the suspension is accomplished by moving the hand agitator up and down
through the suspension for one minute. This process also prevents the
accumulation of sediment on the base and sides of the graduate.
#-+ DIA BRASS R O D
LOOP POR l INGCR
Figure 7. Hand agitator for hydrometer cylinder -
(8) At the end of i min, set the cylinder on a

table. If foam is present, remove it from the top of
the suspension by lightly touching it with a piece of
soap. Slowly immerse the hydrometer in the liquid 20
to 25 set before each reading, as shown in Figure 8.
Care should be exercised when inserting and remov-
ing the hydrometer to prevent disturbance of the
suspension.
(9) Observe and record the hydrometer read-
ings on the data sheet after 1 and 2 min have elapsed
from the time the cylinder is placed on the table. As.
soon as the 2-min reading has been taken, carefully
Figure 8. Immersing remove the hydrometer from the suspension and place
hydrometer in sus-
it in a graduate of clean water. (If a hydrometer is left in
pension prior to mak-
ing observation a soil suspension for any length of time, material will
v-i6
E M IIIO-2-i906
Appendix V
30 Nov 70
settle on or adhere to the hydrometer bulb and this will cause a significant
error in the reading,) Again insert the hydrometer in the suspension and
record readings after elapsed times of 4, 45, 30, 60, i20,t 240, and i440
min, removing the hydrometer from the suspension after each reading and
placing it in a graduate of clean water. Make all hydrometer readings at
the top of the meniscus. For hydrometers graduated to read in specific

gravity of the suspension, read only the last two figures and estimate the
. .
third. Record the indicated specific gravity, minus i, multiplied by $000
(example: the reading i.0225 should be recorded as 22.5). For hydrometers
graduated to read grams per liter of suspension, record the actual reading.
(i0) At the end of 2 min and after each subsequent hydrometer
reading, place a thermometer in the suspension and record the tempera-
ture reading on the data sheet. The temperature shall be recorded to *0.5
C. Temperature changes of the soil suspension during the test will affect
the test results. Variations in temperature should be minimized by keep-
ing the suspension away from heat sources such as radiators, sunlight, or
open windows. A constant-temperature bath provides a convenient means
of controlling temperature effects.
(ii) If the dry weight of the sample is to be obtained at the end
of the test, carefully wash all the suspension into an evaporating dish.
Oven-dry the material, allow to cool, and determine the sample weight.
Subtract the dry weight of dispersing agent used.from this weight to obtain
the oven-dry weight of soil.
h.
- Computations. (i) C o r r e c t e d h y d r o m e t e r r e a d i n g . Compute
the corrected hydrometer readings, R, for use in computing particle
diameter by adding the meniscus correction, C
to the actual hydrom-
m
e t e r r e a d i n g s , R. Record the corrected reading, R, on the data sheet.
(2) Computation of particle diameter. Calculate the particle
t A final reading after i20 min is sufficient for most soils when hydrom-
eter analysis is used for classification purposes.
EM iiiO-2-1906
Appendi* V
30 Nov 70
diameter corresponding to a given hydrometer reading on the basis of

Stokes equation, using the nomograph shown in Figure 9. The R-scale cor-
responding to the distances HR is prepared using the hydrometer calibra-
tion curves as shown in Figure 6. The R-scale shall be designed for the
particular hydrometer used in the test. A key showing the steps to follow
in computing D for various values of R is shown on the chart. Record
the particle diameters, D, on the data sheet.
(3) Percent finer. To compute the percent of particle diameters
finer than that corresponding to a given hydrometer reading, subtract the dis-
persing agent correction, Cd, from the corrected hydrometer reading, R. A
temperature correction factor, m, must also be added algebraically to each
of the readings. This factor can be either positive or negative depending on
,
the temperature of the suspension at the time of each reading. Obtain the
temperature correction factors from Table V-i and record them on the data
sheet. Record the values of R - Cd t m on the data sheet. The R - Cd t n -
values are used to compute percent finer according to the following formulas:
Hydrometer calibrated in specific gravity:
Percent finer by weight = G

Gs iXg(R-Cdtm)
s - S
Hydrometer caltbrated in grams per liter:
Percent finer by weight = F(R - Cd + m)

s
where Gs = specific gravity of solids

ws = w. = oven-dry weight in g of soil used for hydrometer analysis
R- Cdtm= corrected hydrometer reading minus dispersing agent
correction plus, algebraically, temperature correction
Calculations for routine work can b??greatly facilitated by using charts,

tables, and other simplifying aids based on a given oven-dry weight of the
sample and average specific gravity values for the major soil groups.
EM lilO-2-tt906
Appendix V
30 Nov 70
E M ii$O-2-i906
Appencji? N -
3d IQOV to
Table V-l
Temperattire Correction Factor, m j for
Use i.Q Computing Percent Finer
Degrees Degrees Degrees Degrees

C F Correction C F Correction
14.0 57.2 -0.9 24.0 75.2 +0.8
14.5 58.1 -0.8 24.5 76.1 +o.g
15.0 59.0 -0.8 25.0 77.0 +1.0
15.5 59.9 -0.7 25.5 77.9 +1.1
16.0 60.8 -0.6 26.0 78.8 +1.3
16.5 61.7 -0.6 26.5 79.7 +1.4
17.0 62.6 -0.5 27.0 8o.G +1.5
17.5 63.5 -0.4 27.5 81.5 +1.6
16.0 64.4 -0.4 28.0 82.4 +1.8
18.5 65.3 YO.3 28.5 83.3 +1.9
19.0 66.2 -0.2 29.0 84-2 t2.1
19.5 67.1 -0.1 29.5 85.1 +2.2
20.0 68.0 0.0 30.0 86.0 +2.3
20.5 68.9 +0.1 30.5 86.9 +2.5

21.0 69.8 +0.2 31.0 87.8 +2.6
21.5 70.7 +0.3 31.5 88.7 t2.8

22.0 71.6 +0.4 32.0 89.6 +2.g
22.5 72.5 +0.5 32.5 90.5 +3.0

23.0 73.4 +0.6 33.0 91.4 +3.2
23.5 7+.3 +0.7 33.5 92.3 +3*3

34.0 93.2 +3.5
v-20
EM iiiO-2-i906
Appendix V
30 Nov 70
-
1. Presentation of Results. The data obtained from the hydrom-
eter analysis are presented in the form of a grain-size distribution curve
on a semilogarithmic chart, as shown in Plate V-2.
4. COMBINED ANALYSIS. a. Description. A combined analysis is
necessary for soils containing material finer than the U. S. Standard No. 200
sieve when the grain-size distribution of the material passing the No. 200
sieve is of interest. A sieve analysis is performed on the material re-
tained on the No. 200 sieve, and a hydrometer test is performed on the
material passing the No. 200 sieve.
b. Apparatus. The apparatus for the combined analysis is the
same as that used for both the hydrometer and sieve analyses.
C. Preparation of Sample. A representative sample for the
combined analysis is selected and prepared in the manner described in
pa rag raph 25 The total amount of sample should be sufficient to yield
required amounts of material for both the sieve and hydrometer analyees.
A visual inspection of the sample will usually suffice to indicate the need
for intermediate steps such as large screen processing for the plus No. 4
fraction, washing, etc. Samples of soils having fines with little or no
plasticity are oven-dried, weighed, and then separated on the No. 200
sieve. The plus and minus No. 200 sieve fractions are preserved for the
sieve and hydrometer analyses, respectively.
Soils containing plastic fines may also be oven-dried initially.
However, if the sample contains plastic fines which tend to form hard
lumps or to coat the coarser particles during oven-drying, the sample is
placed in a pan filled with enough water to cover all the material and
allowed to soak until all the lumps or coatings have been reduced to indi-
vidual particles. The length of time required for soaking will vary from
2 to 24 hr, depending in general on the amount and plasticity of the fines,
The water and soil mixture is then washed over a No. 200 sieve (and No. 4
sieve, if necessary), The coarser fractions are preserved for a sieve
analysis, and the soil and water passing the No. 200 sieve are preserved
V-21
EM 1110-2-1906
Appendix V
Change 2
20 Aug 86
for a hydrometer analysis. Excess water with the lines is removed by

evaporation, filtration, or wicking. If the grain size of the plastic
fines would be altered by oven-drying. The oven-dry weight of the fines is
determined after the hydrometer test.
In routine testing when all soil particles are finer than the
No. 10 sieve size, the hydrometer test may be performed on a total sample
of known dry weight; the sample is then washed through the No. 200 sieve,
and finally the sieve analysis is performed on the oven-dried fraction
retained on the No. 200 sieve.
d
-0 Procedure. The procedure shall consist of the following steps:
* (1) Record identifying information for the sample on both the
*
sieve and hydrometer analysis data sheets (see Plate V-l).
(2) Perform a sieve analysis on a representative portion of the
sample retained on the No. 200 sieve, using the procedures described in
paragraphs 2d_(l) and 2g(.2). -
(3) Perform a hydrometer analysis on a portion (see para-
graph 3~ for approximate weight) of the sample passing the No. 200 sieve,
using the procedure described in paragraph 32.
c. Computations. The computations consist of the following steps:
(1) Compute the percentage retained on the No. 200 sieve for
the total sample used in the combined analysis as follows:
Percent retained on No. 200 sieve = wl_ x 100
where
wl = dry weight of sample retained on No. 200 sieve
= total dry weight of sample used for combined analysis
wS
(2) Compute the data from the sieve analysis in the same
manner as outlined in paragraphs 2i(l) and 2i(2), except that the percent
retained for each sieve shall be based only on that portion of the total
material used for the sieve analysis. As the amount of material used in
the sieve analysis may be less than WI, it will be necessary to compute
v-22
E M 11iO-2-1906
Appendix V
30 Nov 70
in terms of particle diameter and total percent finer by weight are pre-
sented in the form of grain-size distribution curves on a semilogarithmic
chart as shown in Plate V-2. The curves obtained from the sieve and hy-
drometer analyses are joined by constructing a smooth curve between them.
5. PR%EDURES F O R P R E P A R I N G C L A Y S H A L E M A T E R I A L . T h e p r o -
cedures for preparing clay shale material shall be the same as those de-
scribed in paragraph 4, page III- 14, Appendix III, LIQUID AND PLASTIC
LIMITS. Material for a particle-size distribution test should be removed
from a processed batch and the test performed in accordance with the
procedures described in this appendix. However, the material should not
be oven-dried before testing, and the hydrometer analysis should be of
duration sufficient to determine the percent finer than 2-p size.
4. POSSIBLE ERRORS. Following are possible errors that would cause
inaccurate determinations of grain-size distribution:
a. Sieve Analysis. (i) Aggregations of particles not thoroughly
-
broken. If the material contains plastic fines, the sample should be slaked
-before sieving.
(2) Overloading sieves. This is the most common and most
serious error associated with the sieve analysis and will tend to indicate
that a material is coarser than it actually is. Large samples may have to
be sieved in several portions, and the portions retained on each sieve re-
combined afterwards for weighing.
(3) Sieves shaken for too short a period or with inadequate
horizontal or jarring motions. The sieves must be shaken so that each
particle is exposed to the sieve openings with various orientations and has
every opportunity to fall through.
(4) Broken or deformed sieve screens. Sieves must be fre-
quently inspected to ensure they contain no openings larger than the
standard.
(5) Loss of material when removing soil from each sieve.
b.
- Hydrometer Analysis. (1) Soil oven-dried before test. Except
V-24
EM i410-2-i906
Appendix V
30 Nov 70
for inorganic soils of low dry strength, oven-drying may cause permanent
changes in the particle sizes.
(2) Unsatisfactory type or quantity of dispersing agent.
Whenever new or unusual soils are tested, trials may be necessary to de-
termine the type and quantity of chemical which gives the most effective
dispersion and deflocculation.
(3) Incomplete dispersion of. soil into suspension.
(4) Insufficient shaking or agitating of suspension in cylinder
at start of test.
(5) Too much soil in suspension. The results of the hy-
drometer analysis will be affected if the size of the sample exceeds the
recommendations given in paragraph 3e.
-
(6) Disturbance of suspension while inserting or removing
hydrometer. Such disturbance is most likely to result when the hy-
drometer is withdrawn too rapidly after a reading.
(7) Stem of hydrometer not clean. Dirt or grea.se on the
stem may prevent full development of the meniscus.
(8) Nonsymmetrical heating of suspension,
(9) Excessive variation in temperature of suspensionduring test.
(IO) Loss of material after test. If the oven-dry weight of the
soil is obtained after the test, all of the suspension must be washed care-
fully from the cylinder.
-
C. Combined Analysis. (I) I n s u f f i c i e n t w a s h i n g o f m a t e r i a l
over the No. 200 sieve. The dispersing agent should be added to the water
in which the sample is soaked and the soil-water mixture should be fre-
quently manipulated to aid the separation of particles; coarser particles
may be removed from the mixture and washed free of fines by hand to re-
duce the quantity of material to be washed on the sieve. While the addi-
tional water used for washing should be held to a minimum, enough must
be added to insure adequate removal of the fines.
(2) ~0s~ of suspension passing the No. 200 s i e v e .
V-25
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
APPENDIX VI:
COMPACTION TESTS
1. INTRODUCTION. In the laboratory compaction test, a soil at a

known water content is placed in a specified manner in a mold of
given dimensions and subjected to a compactive effort of con-
trolled magnitude after which the resulting unit weight of the
soil is determined. The procedure is repeated at various water
contents until a relation between water content and unit weight
of the soil is established.
The laboratory compaction procedure is intended to simulate

the compactive effort anticipated in the field. As a general
rule the standard compaction test shall be used to simulate field
compaction for routine foundation and embankment design. In
special cases, to suit anticipated construction procedures, it
may be necessary to use higher or lower compactive efforts on the
soil. For a higher compactive effort the modified compaction
test, and for a lower compactive effort the 15-blow compaction
test shall be used. Details of the standard, modified, and
15-blow compaction tests are given below.
2. STANDARD COMPACTION TEST.
* a. Apparatus. The apparatus consists of the following:
(1) Molds, cylindrical, metal. Molds shall have a

detachable base and a collar assembly extending approximately
2-1/2 in. above the top of the mold to retail soil during pre-
paration of compacted specimens of the desired height and volume.
Molds having a slight taper to facilitate removal of the specimen
after the compaction test are satisfactory provided the taper *
VI-1
EM 1110-2-1906
Appendix VI
Change 2
20 AUCI 86
* is no greater than 0.200 in. in diameter per foot of mold height.
Capacities and dimensions of the molds shall be as follows:
(a) Mold with an average inside diameter of

4.0 & 0.016 in. and a capacity of 1/30 * 0.0004 cu ft. Details
of a typical mold are shown in Figure 1.
(b) Mold with an average inside diameter of

6.0 * 0.016 in. and a capacity of 3/40 ? 0.0009 cu ft. The
6.0-in. mold may be similar in construction to that shown in
Figure 1, and shall be used for compacting samples containing
material that would be retained on the No. 4 sieve but passing
the 3/4-in. sieve.
(cl The exact volume of molds should be

determined before use and periodically thereafter, and this -
measured volume is used in calculations.
(2) Rammer, manually or mechanically operated. The

rammer shall consist of a drop weight which can be released to
fall freely and strike the soil surface. The height of drop
shall be controlled so that the weight falls from a height of
12 2 l/l6 above the surface of the soil. The mass of the free
falling part of the rammer shall be 5.5 & 0.02 lb and the
striking face of the rammer shall be flat. Rammers must also
meet the following requirements:
(a) Manual rammer. The striking face shall be

circular with a diameter of 2.0 ? 0.005 in. The rammer shall be
equipped with a guide sleeve having sufficient clearance so,that
the free fall of the rammer shaft and head will not b% restricted.
The guidesleeve shall have at least four vent holes at each end
(eight holes total) located with centers 3/4 ? l/l6 in. from each * _
VI-2
*
EM 1110-2-1906
Appendix VI
Change 2
20 PUg 86
PLAN
j--------
7------
I
ELEVATION
Figure 1. 4.0-in. diameter compaction mold
VI-3
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* end and space 90 deg apart. The minimum diameter of the vent
holes shall be 3/8 in. Additional vent holes or slots may be
incorporated in the guidesleeve if desired. Figure 2 illustrates
a typical manual rammer.
(b) Mechanical rammer. A mechanical rammer must

operate in such a manner as to provide uniform and complete
coverage of the specimen surface. The clearance between the ram-
mer and the inside surface of the mold at its smallest diameter
shall be 0.10 k 0.03 in. When used with the 4-in. mold, the
specimen contact face shall be circular with a diameter of
2.000 ? 0.005 in. When used with the 6.0-in. mold, the specimen
contact face shall be either circular or sector shaped?; if sec-
tor shaped, it shall have a radius of 2.90 k 0.02 in. The sector
face rammer shall operate in such a manner that the vertex of the
sector is positioned at the center of the specimen. -
(c) Calibration of mechanical rammer compactors.

The mechanical rammer compactor must be calibrated periodically
against the results obtained with the manual rammer. The compac-
tor must be calibrated for the circular foot and, if used, the
sector foot. The mechanical compactor shall be calibrate-d before
initial use, near the end of each period during which the mold
was filled 500 times before use after anything including repairs
that may affect test results whenever test results are question-
able, and before use after any 6-month period during which the
rammer was not calibrated. Procedures for calibrating mechanical
compactors are given in Engineer Manual EM 1110-2-1909, Calibra-
tion of Laboratory Soils Testinq Equipment. *
t The mechanical rammer equipped with a sector shaped foot

should not be used for compacting specimens for the California
Bearing Ratio (CBR) test described in MIL-STD-62lA as CBR
values may differ substantially from those obtained on
specimens compacted with a rammer having a circular foot.
VI-4
EM 1110-2-1906
Appendix VI
Change 2
ZU Aug 86
* (31 Balance having a readability
of 1 gt an accuracy of 2 g, and having a
capacity sufficient for weighing compacted
samples.
(4) Oven (see Appendix I, WATER

CONTENT - GENERAL).
(5) Sieves, US Standard 3/4-in.

and No. 4 (0.187 in.) conforming to ASTM
Designation: E 11, Standard Specification
for Wire-Cloth Sieves for Testing Purposes.
Large sieves are generally more suitable
for this purpose.
(6) Straightedge, steel, at

least 1/8 x l-3/8 x 10 in. and having a
beveled edge.
(7) Mixing tools, such as mixing

pan, spoon, trowel, spatula, etc. A suit-
able mechanical device may be used for mix-
ing fine-grained soils with water.
(8) Specimen containers. Seam- Figure 2. Manual

rammer for stan-
less metal containers with lids are recom- dard compaction
mended. The containers should be of a test
metal resistant to corrosion such as alumi-
num or stainless steel. Containers 2 in.
high by 3-1/2 in. in diameter are adequate.
VI-5
EM 1110-2-1906
Appendix VI
Change 2
20 hug 86
* ('9) Sample splitter or riffle for dividing the
samples.
(10) Glass jars, metal cans, or plastic buckets with

airtight lids in which to store and cure soil prepared for
compaction.
(11) Equipment for determining water contents (see

Appendix I, WATER CONTENT - GENERAL).
b. Preparation of Sample. The amount of soil required for

the standard compaction test varies with the kind and gradation
of the soil to be tested. For soils passing the No. 4 sieve that
are to be tested in the 4.0-in. mold, 20 lb of soil is normally
sufficient for the test. For samples containing gravel that are
to be tested in the 6.0-in. mold, approximately 75 lb of pro- -
cessed material is required. Ordinarily, the soil to be tested

shall be air-dried, or dried by means of drying apparatus pro-
vided the apparatus will not raise the temperature' of the sample
above 60 C (140' F). The requirement for fully air-drying soils
in preparation for compaction is intended to facilitate soil pro-
cessing and reduce variability in testing procedures. However in
some construction control operations, it may not be practical to
completely air dry, rewet, and cure the soil in preparation for
compaction. Inthese instances, the soil is air-dried to some
water content near the driest point on the compaction curve and
water for preparation of individual test specimens added as
needed to obtain the desired range of water contents. Partial
air drying of some soils during preparation may lead to compac-
tion results different from those which would be obtained if the
soil had been completely air-dried during preparation*. If a pro-
cedure other than the standard (fully air-dry, rewet, and cure)
procedure is used, comparison tests must be performed for each of * _-
VI-6
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* the soil types encountered at a given project to verify that
there is no differences in results. If differences in results do
appear, a procedure that reflects the actual field conditions
must be adopted for both design and construction control testing.
Aggregations present in the sample shall be thoroughly

broken, but care should be taken that the natural size of the
individual particles is not reduced. The material shall then be
screened through a 3/4-in. and a No. 4 sieve. For some soils, it
may be desirable to reduce aggregations before the sample is
dried. If all the material passes the No. 4 sieve, the sample
shall be mixed thoroughly and a representative sample taken to
determine the initial water content (see Appendix I, WATER
CONTENT - GENERAL). The sample shall then be stored in an air-
tight container until ready for processing at different water
contents for compaction in the 4.0-in. mold.
If all the sample passes the 3/4-in. sieve and contains

,5 percent or less material larger than the No. 4 sieve, the plus
No. 4 fraction shall be discarded and the test performed using
the 4.0-in. compaction mold. If all the sample passes the
3/4-in. sieve but contains more than 5 percent material retained
on the No. 4 sieve, it shall be tested in the 6-in. mold. The
sample shall be mixed thoroughly after which its initial water
content shall be determined. The sample shall then be stored in
an airtight container until ready for processing at different
water contents for compaction.
If the sample contains some material retained on the

3/4-in. sieve, but the amount is 5 percent or less, the plus
3/4-in. fraction shall be removed and discarded and the sample
tested in the 6-in. mold. The initial water content of the
sample shall be determined and the sample stored in an airtight *
VI-7
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* container until ready fo'rprocessing at different water contents
for compaction.
If the sample contains more than 5 percent material re-

tained on the 3/4-in. sieve, the test should be performed using
the 12-in. compaction mold, the procedures for which are given in
Appendix VIA: COMPACTION TEST FOR EARTH-ROCK MIXTURES.
C. Procedure.
(1) Material finer than No. 4 sieve. The procedure

for soils finer than the No. 4 sieve shall consist of the
following:
(a) Record all identifying information for the

sample such as project name or number, boring number, and other -
pertinent data on a data sheet (see Plate VI-l for suggested

form). Record the compactive effort to be used, size of mold,
and initial water content of processed sample.
(b) From the previously prepared sample, weight

a quantity of air-dry soil equivalent to 2,500.g oven-dry weight
(see paragraph 2d(l)). Thoroughly mix the material with a mea-
sured quantity of water sufficient to produce a water content 4
to 6 percentage points below estimated optimum water content. At
this water, nonplastic soils tightly squeezed in the palm of the
hand will form a cast which will withstand only slight pressure
applied by the thumb and fingertips without crumbling; plastic
soils will ball noticeably. Store the soil in an airtight con-
tainer for a sufficient length of time to permit it to absorb the
moisture. The time required for complete absorption will vary
depending on the type of soil. For nonplastic soils in which
VI-8
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* moisture is readily absorbed, storage is not necessary. For most
other soils a minimum curing time of 16 hr is usually adequate.
(c) Repeat step (b) for at least eour additional

specimens. Increase the water content for each specimen by
approximately 2 percentage points over that of the previous
specimen.
(d) Weigh the 4.0-in. compaction mold to the

nearest gram, ,and record the weight on the data sheet.
lel Attach the mold, with collar, to the base

plate and place the mold on a uniform, rigid foundation, such as
a block or cylinder of concrete weighing not less than 200 lb.
(f) Place an amount of the pretiiously prepared

sample in the 4.0-in. mold such that when three such layers have
been compacted in the mold, the total compacted'height is between
4-518 in. and 5 in.? Compact each layer by 25 uniformly distrib-
uted blows from the rammer, with the drop weight falling freely
from a height of 12.0 in. In operating the manual rammer, take
care to hold the rammer vertical and avoid rebounding the rammer
drop weight from the top of the guidesleeve. Apply the blows at
a uniform rate not exceeding 1.4 set per blow. The compaction
procedure is illustrated in Figure 3.
w Remove the extension collar from the mold.

Remove the exposed compacted soil with a knife and carefully trim *
t It is important-that the compacted soil just fill the mold

with little excess to be struck off. As the amount of
material to be struck off varies, the mass of soil to which a
constant amount of energy is supplied varies. When the amount
of material to be struck off is more than about 1/4 in., the
test results become less accurate.
VI-9
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
*
Figure 3. Compacting soil specimen.
the surface even with the top of the mold by means of a straight-
edge. Any cavities formed by large particles being pulled out
should be carefully patched with material from the trimmings.
(h) Remove the mold with the compacted specimen

therein from the base plate, weigh the mold plus wet soil to the
nearest gram, and record the weight on the data sheet. When
cohesionless soils are being tested there is a possibility of
losing the sample if the base plate is removed. For i?hese soils,
weigh the entire unit. *
VI-10
EM 1110-2-1906
Appendix VI
Change 2
20 Auq 86
* til Remove the compacted specimen from the mold,
and slice it vertically through the center. Take a representa-
tive specimen of the material from each of the two parts and
determine the water content of each (see Appendix I, WATER
CONTENT - GENERAL). The water content specimens shall weigh not
less than 100 g. Alternatively, the entire compaction specimen
may be used for the water content determination. In this case,
the wet weight of specimen for use in computing water content
should be redetermined after the specimen is extruded from the
compaction mold as some loss of material may occur during trans-
fer of the specimen._
tjl Repeat steps (d) through (i) for remaining

specimens. Compact a sufficient number of test specimens over a
range of water contents to establish definitely the optimum water
content and maximum density. Generally, five compacted specimens
prepared according to the above-described procedure should com-
pletely define a compaction curve. However, sometimes more
specimens are necessary. To determine if the optimum water con-
tent has been reached, compare the wet weights of the various
compacted specimens. The optimum water content and maximum
density have been reached if the wettest specimens compacted
indicate a decrease in weight in relation to drier specimens.
(2) Materials larger than 3/4 in. sieve. The proce-

dure for determining the density and optimum water content of
soils containing material retained on the 3/4 in. sieve is the
same as that for the finer than 3/4 in. sieve material, except
that the test is performed in the 6.0-in.-diam mold and the
number of blows of the compaction rammer is 56 per soil layer
instead of 25. This results in equal compactive efforts for the
two molds. It is advisable to use the entire compacted specimen
for the water content determination. The quantity of soil
VI-11
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* required for each compacted sample will be equivalent to about
5,500 g of oven-dry material.
d. Computations.
(1) Preparation of specimen. The required weight of

soil, W' in grams necessary to produce 2,500 g of oven-dry
0'
soil is computed as follows:
where
W
0
= initial water content of material (after air-drying)
-
w; = desired weight of oven-dry soil = 2,500 g
The amount of water, Ww , in cc, to be added to the weight of

soil, W'0) to produce specimens at the desired test water
contents is computed as follows:
w; lw - woI
ww = 100
where
W = desired test water content
(2) Quantities obtained in compaction test. The

following quantities are obtained for each specimen in the
compaction test:
VI-12
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
(a) Weight of compaction mold plus wet soil. The
weight of the compaction mold is subtracted from this value to
obtain the weight of the soil, W .
(b) The inside volume of the compaction mold.

This volume is equal to the volume, V , of the wet soil
specimen.
(c) Weight of water content specimen plus tare

before and after oven-drying. The tare weight is subtracted from
these values to obtain the weight of wet and dry soils for
computing water content.
(3) Water content and density. The water content,

w, of each compacted specimen shall be computed in accordance
with Appendix I, WATER CONTENT - GENERAL. The weight of oven-dry
soil, W
s' of each compacted specimen shall be computed accord-
ing to the formula:
Dry weight of specimen = weight of wet soil

1+ water content
100
ws = W
1+&
The dry weight of the specimen is obtained directly if the entire

compacted specimen is used for the water content determination
and no loss of material occurs during removal of the specimen
from the mold.
The wet unit weight, ym , (optional) and the dry unit

weight, yd , expressed in pounds per cubic foot, shall be
computed by the following formulas: *
VI-13
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* dry density, and this value shall be recorded to the nearest
0.1 lb per cu ft.
(2) Air voids curves. The zero air voids curve (see
example in Figure 4) represents the dry density and water content
of a soil completely saturated
with water. The zero air voids
and 90 percent saturation curves
shall be shown with the compac-
tion curve in Plate VI-2. Data
for plotting these curves for :.
<
soils with different specific :
2
gravities are given in 9
Table VI-l. The specific gravity
of the soil used in the compac-
tion test shall be determined in
Appendix IV, SPECIFIC GRAVITY.
Figure 4. Determination of
3. MODIFIED COMPACTION TEST. maximum density and opti-
mum water content
The modified compaction test dif-
fers from the standard test in that
a greater compactive effort is used which results in higher
maximum densities and lower optimum water contents. The appara-
tus, preparation of sample, and procedure are the same as those
used in the standard compaction test, with the following
modifications:
a. sparatus. The rammer shall consist of a lO.OO-lb

weight with an 18.0-in. free drop. If a mechanical rammer is
used in performing these tests, the rammer must be calibrated
separately for this test in accordance with procedures given in
Engineer Manual EM 1110-2-1909, Calibration of Laboratory Soils
Testing Equipment. *
VI-15
EM 1110-2-1906
Appendix VI
Change 2
20 Au9 86
* b. Procedure. The soil shall be compacted in five layers
of equal thickness. The number of blows per layer shall be the
same as for the standard compaction test: 25 blows per layer in
the 4.0-in.-diameter mold, and 56 blows per layer in the 6.0-in.-
diameter mold. The computations and presentation of results
shall be the same as those used in the standard compaction test.
4. 15-BLOW COMPACTION TEST. The 15-blow compaction test differs

from the standard compaction test in that a lesser compactive
effort is used resulting in lower maximum densities and higher
optimum water contents. The apparatus, preparation of samples,
and procedures shall be the same as those used in the standard
compaction test (5.50-lb weight with a 12.0-in. free drop) with
the following modifications:
a. The 6-in. mold shall not be used.
b. The number of blows per layer shall be 15.
The computations and presentation of results shall be the same as

those used in the standard compaction test.

cause inaccurate determinations of compaction curves for any
compactive effort:
a. Aggregations of dried soil not completely broken.
b. Water not thoroughly absorbed into dried soil. Consis-

tent results cannot be obtained unless the soil and water are
complete mixed and sufficient time allowed for the soil to absorb
the water uniformly. *
VI-16
EM 1110-2-1906
Appendix VI
Change 2
20 Aug 86
* C. Soil reused. Since some soils are affected by recom-
paction, fresh material must be used for each specimen. Recom-
paction tends to increase the maximum dry unit weight of some
clays and, therefore, decrease the apparent optimum water
content.
d. Insufficient number of range of water contents to

define compaction curve accurately. See paragraph 2c(l) (j).
e. Improper foundation for compaction mold.
f. Incorrect volume of compaction mold used. The exact

inside volume of each mold must be determined before being used.
go Mechanical compactor not properly calibrated.
h. Human factors in the operation of hand rammer. Varia-

tions in results can be caused by not bringing the drop weight to
a complete stop before releasing it to fall and compact the soil.
If raising and releasing the rammer's drop weight is done too
quickly, the drop weight will not be brought to rest before
release. If the rammer is not held vertical during operation,
the compactive effort will be reduced. The tendency to press the
sleeve of the manual rammer into the soil specimen, the way the
blows are distributed over the surface of the specimen, and other
individual operator characteristics all tend to affect compaction
results. By proper instruction and supervision, uniform tech-
nique can be maintained within a laboratory; however, it is
preferable that all specimens of a given test be compacted by the
same person with the same rammer in one sitting.
i. Excessive variation in total depth of compacted speci-

men. The extension of the specimen into the collar of the mold *
VI-17
EM 1110-2-1906
Appendix VI
Change 2
20 iiug 86
* should not exceed about 1/4 in., and care should be taken that
each layer is nearly equal in weight.
j- Water content determination not representative of

specimen. This error can be avoided by using the entire specimen
for the water content determination.
VI-18
EM iiiO-2-i906
Change 2
20 kl0 Wj
APPENDIX VIA:
COMPACTION TEST FOR EARTH-ROCK MIXTURES
i. INTRODUCTION. In Appendix VI, procedures are given for the

standard effort compaction test using samples having particles finer than
the 3/4-in. sieve sizes. The procedures outlined in this appendix are for
the standard effort compaction test using material with particles larger
* than 3/4-in. and finer than 2-in. sieve sizes. This method should be
used for testing material containing particles larger than 3/4-in. *
sieve sizes if these particles exceed i0 percent by weight of the total
sample. If less than 5 percent by weight of the total sample is finer than
the No. 200 sieve, maximum density should be determined by vibratory
methods.
The test method outlined is comparable to the standard test in that
(a) the compactive effort applied is i2,300 ft-lb/cu ft and (b) the equipment
has been devised to maintain rati.os between mold diameter, rammer
diameter, and maximum particle size of the test specimen similar to
those ratios used in the standard test in Appendix VI.
2. APPARATUS. The apparatus shall consist of the following:
a. Cylindrical mold, with an ID of i2.0 in. f 0.1 in., a height of
12.0 in. f 0.1 in., and a detachable collar approximately 2-i/2 in. high.
The mold and collar assembly should be constructed to fasten to a de-
tat hable baseplate . Details of a typical assembly are shown in Figure 1.
k* Hand rammer, metal, of the sliding-weight, fixed-head type

pith a 4-in.-diameter face and a free-falling weight of ii.50 lb f 0.05 lb.
The rammer should be equipped with a guide such that the height of fall
of the sliding weight is 24.00 in. * 0.05 in. Details of a typical rammer
are shown in Figure 2.
C. Balances sensitive to 0.i lb with a capacity of 250 lb.
& Oven, forced-draft type, iO- to iZ-cu-ft capacity: automatically
VIA-1
EM $liO-2-i906
Appendix VIA
30 Nov 70
EM iiiO-2-1906
Appendix VIA
30 Nov 70
VIA- 3
EMiilO-2-1906
Appendix VIA
30 Nov 70
controlled to maintain a uniform temperature of 110 f 5 C.

-e. Pans, drying, of aluminum or other corrosion-resistant metal,
with a capacity of at least 0.5 cu ft. Roasting pans i8 in. by 24 in. by
4 in. are satisfactory.
-f. Sieves, U. S. Standard, large diameter type, ranging from 4-in.
openings to the No. 4 size, and a mechanical sieve shaker. Sieves with
3/8-, 1/2-, 3/4-, I-, I-1/2-, 2-, 3-, and 4-in. openings are normally
required.
g* Containers, corrosion resistant, with a capacity of at least
1 cu ft and having airtight lids.
!? Shovel, hand, square-edged, and a mortar box having a capa-
city of at least 4 cu ft.
-1. Straightedge, steel, at least 16 in. long, 3/8 in. thick, and
1 in. wide with a beveled edge.
is Graduates, hand scoop, trimming knife, wire brush, and
rubber-head hammer.
3. QUANTITY OF SAMPLE. At least 700 lb.of sample is required
having particles finer than the 2-in. sieve sizes, If the field sample
contains quantities of particles larger than the 2-in. sizes, the total
sample weight required must be increased to permit removal of over-
size particles.
4. PROCESSING OF SAMPLE. a_. Record on a work sheet
(Plate VIA-l) identifying information for the sample, including visual
classification.
-b. Spread the material in flat pans and air-dry the entire sample.
Other means, such as ovens and heat lamps, may be used to accelerate
drying if the maximum drying temperature is 60 C.
C. Reduce all aggregates, or lumps formed during drying, of
fine-grained material to particles finer than the No. 4 sieve. With a
wire brush or other means, remove all fine-grained material that may
be clinging to rock sizes, taking care not to lose the fine-grainedmaterial.
VIA-4
EMl110-2-1906
Appendix VIA
30 Nov 70
6 Separate all the material using a set of sieves ranging from

the largest particle size in the sample to the No. 4 sieve. The total sample
must be processed to determine the as-received gradation.
e.
- Place the material retained on each sieve and that passing the
No. 4 sieve in separate containers, weigh the contents of each, and com-
pute the percent of the total sample retained on each sieve as follows:
dry weight of material retained on sieve x iooy

% Retained = 0
dry weight of total sample
f. If i0 percent or less of a field sample is retained on the 2-in.

sieve, the particles larger than this size should be discarded and re-
placement is not necessary.
iZ* If more than i0 percent of a field sample is retained on the
2-in. sieve, it will be necessary to remove the plus 2-in. sizes and re-
place them with an equal weight of material between the 2-in. and No. 4
sieve sizes. The gradation of the replacement m.aterial must be the same
relative gradation as that of the total sample between the 2-in, and the
No. 4 sieve sizes. The percent passing the No. 4 sieve remains constant
and is equal to the percent passing the No. 4 sieve for the total as-
received sample. For each sievebetween the 2-in. and the No. 4 sizes,
the percent required to replace the plus 2-in, sizes is computed as
follows:
$ retained on
one sieve?
Replacement %I = total % of t2-in. sizes X
total % between 2-in. and
No. 4 sieve
For each sieve, add the Replacement $ to the $ Retained on that

sieve initially. This gives the percent by weight of a test specimen re-
quired for each sieve size in order to reconstitute a specimen with the
t Any sieve between 2-in. and No. 4 sieve sizes.
VIA- 5
EM ii10-2-1906
Appendix VIA
30 Nov 70
t2-in. sizes replaced with sizes ranging from the 2-in. to the No. 4 sizes.
Typical results are tabulated in Plate VIA- 1. A typical as-received
gradation and test gradation is shown in Plate VIA-2.
5. SPECIAL CONSIDERATIONS. In materials of a heterogeneous
nature, such as mixtures of sandstones, siltstones, and shale, the large
particles may be siltstone or sandstone, while the smaller size particles
may be shale. For materials of this type, when particles larger than the
2-in. sieve sizes are removed for preparation of the test specimen, re-
placement must be made using the same types of materials scalped
off, or removed. For example, oversize sandstone particles must be
removed and replaced, where applicable, with smaller particles of
sandstone.
6. PREPARATION OF TEST SPECIMEN. 2. Prepare 130 lb of proc-
essed air-dried material for the test specimen by combining the weight
of material required from each sieve size (refer to typical work sheet,
-
Plate VIA- 1).
-b. Thoroughly mix the material for the test specimen with a
measured quantity of water sufficient to produce a water content 4 or 5
percentage points below the estimated optimum water content of the en-
tire sample. This can be determined only by judgment and experience.
c. Store the moistened sample in an airtight container for a
minimum of 16 hr.
G* Prepare material for at least four additional test specimens
by repeating steps in paragraphs 65 through 6c_. Increase the water con-
tent of each specimen by approximately 2 percentage points over that of
the previous specimen.
7. C O M P A C T I O N PRO.CEDURE. -a. Weigh the compaction mold to
the nearest 0.1 lb, determine its inside volume to the nearest 0.001 cu ft,
and record the data.
-b. Attach the collar to the compactor mold, clamp the mold se-
curely to the baseplate and place the assembly on a level, rigid foundation
VIA- 6
E M iiIO-2-f906
Appendix VIA
30 Nov 70
made of a concrete cube having a total weight of at least 200 lb. This
concrete foundation should not be covered with a rnetal plate.
-
C. Mix the cured material thoroughly to attain a uniform water
distribution.
d.
- Place a sufficient quantity of the test specimen in the mold to
give a compacted layer between 4.0 and 4.5 in. thick. Compact each layer
by applying 440 uniformly distributed blows of the rammer, with the
11.5-lb weight falling freely from a height of 24.0 in compact three
equal layers in this manner, taking care to seat the rammer face flush
with the soil surface before each blow and to keep the rammer assembly
vertical during testing. Use just enough material to finish with less than
1 -in. of sample protruding above the top of mold.
e.
- Detach the extension collar, taking ca:re not to disturb the soil
mass extending above the top of the mold. Trim the surface exactly even
with the top of the mold. Fill any cavities formed by removal of particles
during trimming with material from the trimrnings and press this filling
material firmly into place. Clean excess material from the lip of the mold.
f. Weigh the mold and compacted specimen to the nearest 0.1 lb
and record the data.
g* Remove the entire test specimen from the mold, spread it in

flat drying pans, and determine its water content. For most specimens,
this requires at least 16 hr oven-drying time. Shorter drying times may
be used if a constant weight is attained.
-h. Repeat the steps in paragraphs 75 through 7g for a sufficient

number of specimens over a range of water contents to establish the op-
timum water content and dry density. Five specimens will usually define
the compaction curve accurately. Fresh material, not previously com-
pacted, should be used for all tests.
-1 . For tests in which degradation of particles due to compaction

is significant, determine the after-compaction gradation of at least two
total specimens from each test series.
VIA- 7
EM IiiO-2-i906
Appendix VIA -
30 Nov 70
8. COMPUTATIONS. The computations shall consist of the following:

a.
- Compute the water content of each compacted specimen as
follows :
w
Water content, W, % = + X dO0
S
where
W = wet weight of total specimen minus its oven-dry weight (lb)
WI = oven-dry weight of specimen (lb)
&* Compute the dry unit weight of each compacted specimen as
follows:
W
Dry unit weight, y, pcf = -$
where V = volume of the compaction mold (cu ft).

-
9. PRESENTATION OF RESULTS. Present the results of the test on
ENG FORM 209i.
a. Compaction Curve.
- Plot the dry unit weight, in pounds per
cubic foot, as the ordinate and the corresponding water contents, in per-
centages of dry weight, as the abscissa, on an arithmetic plot. Connect
the plotted points with a smooth curve. The water content at the peak of
the curve is the optimum water content and the corresponding dry unit
weight is the maximum. Record the optimum water content to the nearest
0.i percent and the maximum dry unit weight to the nearest 0.i pcf.
!? Zero Air Voids and 90 Percent Saturation Curves. Using the

weighted average of the specific gravity of the plus No. 4 and the minus
No. 4 material, compute and plot the zero air voids curve and the curve
representing the line of 90 percent saturation.
i0. POSSIBLE ERRORS. The following errors can cause inaccurate
results:
a. Aggregations of air-dried soil not completely reduced to
VIA- 8
EMiiiO-2-i906
Appendix VIA
30 Nov 70
finer particles during processing.

LL* Water not thoroughly absorbed into dried material due to in-
sufficient mixing and curing time.
-
C. Material reused after compaction.
& Insufficient number of tests to define compaction curve

accurately.
e.
- Improper foundation for mold during compaction.
L* Incorrect volume or weight of compac:tion maid.

iS* Incorrect rammer weight and height of fali.
*h. Excessive material extending into the extension collar at
the end of compaction.
-i. Improper or insufficient distribution of blows over the soil
surface.
L Tendency to press the head of the rammer against the
specimen before letting the weight fall.
k. Insufficient drying of sample for water content determination.
-0 eliminate the need for triming away the excess amterial extending into
the collar, an alternative may be used. Determine, using the sand-cone
method, the volume of the empty collar above the top of the mold and the
volume of the collar partially filled with the protruding specimen. The
difference is the volume of the compacted excess material. Weigh the
entire specimen without trimming the excess and determine the moist
unit weight.
VIA-9
EMillO-2-1906
Appendix VIA
30 Nov 70
PROJECT _
LA0. N O . _
DATE
TESTED BY _
COMPACTION TEST
WORK SHEET
GRADATION
TOTAL
8 7. 0 1
R E M A R K S --_h!.& 8LOWS/EACH OF_d__LAERS WITH jA5-Lf3 RAMMER AN0 SlANDARD /
Td-lN. DROP. OIAMETER O F ,,,O,_O =&,,,.
MOOlFlE3 _
PLATE VIA-I
VIA-10
EM il40-2-i906
Appendix VIA
30 Nov 70
.-. --- + -..4 -_- .+._._ _ ..+..
VIA-ii
E M iiiO-2-1906
30 Nov 70
APPENDIX VII:
PERMEABILITY TESTS
i. DARCYS LAW FOR FLOW OF WATER THROUGH SOIL. The flow of

water through a soil medium is assumed to follow Darcys law:
q = kiA
where q= rate of discharge through a soil of cross-sectional area A

k = coefficient of permeability
iz hydraulic gradient: the loss of hydraulic head per unit
distance of flow
The application of Darcys law to a specimen of soil in the laboratory is

illustrated in Figure i. The coefficient of permeability, k (often termed
WATER SUPPLY
/-
J
0 -
OVERFLOW
---
*---
- WHERE q z RATE OF DISCHARGE
= QUANTITY OF FLOW, 0,
PER UNIT OF TIME, t
TO MAINTAIN
CONSTANT HEAD k = COEFFtClENT OF
PERMEAEILITY
i = HYDRAULIC GRADIENT
h =h/L
.-* *:.*.*.
.
.
.
. .
.
.
.
*.
.
A = CROSS-SECTIONAL AREA
OF SPECIMEN
. A.
. . .
+
. . . . .
. . . .
GRADUATE
SCREEN 1
I
Figure i. Flow of water through soil
VII- 1
EM iiiO-2-1906
Appendix VII
30 Nov 70
permeability), is defined as the rate of discharge of water at a tem-

perature of 20 C under conditions of laminar flow through a unit cross-
sectional area of a soil medium under a unit hydraulic gradient. The
coefficient of permeability has the dimensions of a velocity and is usually
expressed in centimeters per second. The permeability of a soil depends
primarily on the size and shape of the soil grains, the void ratio of the
soil, the shape and arrangement of the voids, and the degree of saturation.
Permeability computed on the basis of Darcys law is limited to the
conditions of laminar flow and complete saturation of the voids. In turbu-
lent flow, the flow is no longer proportional to the first power of the
hydraulic gradient. Under conditions of incomplete saturation, the flow is
in a transient state and is time-dependent. The laboratory procedures
presented herein for determining the coefficient of permeability are based
on the Darcy conditions of flow. Unless otherwise required, the coefficient
of permeability shall be determined for a condition of complete saturation
-
of the specimen. Departure from the Darcy flow conditions to simulate
natural conditions is sometimes necessary; however, the effects of turbu-
lent flow and incomplete saturation on the permeability should be recognized
and taken into consideration.
2. TYPES OF TESTS AND EQUIPMENT. -a. Types of Tests. (1) Constant-
head test. The simplest of all methods for determining the coefficient of
permeability is the constant-head type of test illustrated in Figure f. This
test is performed by measuring the quantity of water, Q, flowing through
the soil specimen, the length of the soil specimen, L, the head of water,
h, and the elapsed time, t. The head of water is kept constant throughout
the test. For fine-grained soils, Q is small and may be difficult to mea-
sure accurately. Therefore, the constant-head test is used principally for
coarse-grained soils (clean sands and gravels) with k values greater than
about 10 X iOm4 cm per sec.
(2) Falling-head test. The principle of the falling-head test
is illustrated in Figure 2, This test is conducted in the same manner as
VII-2
EM ii40-2-i906
Appendix VII
30 Nov 70
STANDPIPE
-I
OVERFLOW
USING SETUP SHOWN IN (a), THE COEFFICIENT OF PERMEAEILITY IS

DETERMINE0 AS FOLLOWS:
ho h
k = z in- = *.s~s 5 i09,~*
hf f
USING SETUP SHOWN IN (b), THE COEFFICIENT OF PERMEABILITY ls

DETERMINED AS FOLLOW%
WHERE: hC = HEIGHT OF CAPILLARY RISE

0 = INSIDE AREA OF STANDPIPE
A = CROSS-SECTIONAL AREA OF SPECIMEN
L = LENGTH OF SPECIMEN
= HEIGHT OF WATER IN STANDPIPE ABOVE
ho
DISCHARGE LEVEL MINUS hC AT TIME, to
hf = HEIGHT OF WATER IN STANDPIPE ABOVE

D IS CH A R G E LEVE L MINUS hC AT TIME , tf
t = ELAPSED TIME, tt - to
Figure 2. Principle of falling-head test
VII-3
EM 1110-2-1906
Appendix VII
30 Nov 70
the constant-head test, except that the head of water is not maintained
constant but is permitted to fall within the upper part of the specimen
container or in a standpipe directly connected to the specime-n. The quan-
tity of water flowing through the specimen is determined indirectly by
computation. The falling-head test is generally used for less pervicus
soils (fine sands to fat clays) with k values less than IO X IOw4 cm
per sec.
b. Equipment.
- The apparatus used for permeability testing may
vary considerably in detail depending primarily on the condition and
character of the sample to be tested. Whether the sample is fine-grained
or coarse-grained, undisturbed, remolded, or compacted, saturated or
nonsaturated will influence the type of apparatus to be employed. The
basic types of apparatus, grouped according to the type of specimen con-
tainer (permeameter), are as follows:
(4) Permeameter cylinders
(2) Sampling tubes
(3) Pressure cylinders
(4) Consolidometers
The permeability of remolded cohesionless soils is determined in
permeameter cylinders, while the permeability of undisturbed cohesion-
less soils in a vertical direction can be determined using the sampling
tube as a permeameter. The permeability of remolded cohesionless soils
is generally used to approximate the permeability of undisturbed cohesion-
less soils in a horizontal direction. Pressure cylinders and consolidome-
ters are used for fine-grained soils in the remolded, undisturbed, or
compacted state. Fine-grained soils can be tested with the specimen
oriented to obtain the permeability in either the vertical or horizontal di-
rection. The above-listed devices are described in detail under the
individual test procedures. Permeability tests,utilizing the different types
of apparatus, together with recommendations regarding their use, are
discussed in the following paragraphs.
VII-4
EM ii10-2-i906
Appendix VII
30 Nov 70
3. CONSTANT-HEAD PERMEABILITY TEST WXTH PERMEAMETER

C Y L I N D E R . 2. U s e . The constant-head permeability test with the
permeameter cylinder shall in general be used for determining the per-
meability of remolded samples of coarse-grained soils such as clean
sands and gravels having a permeability greater than about i0 X 1O-4 cm
per sec.
!z* Apparatus. The apparatus and accessory equipment should
consist of the following:
(1) A permeameter cylinder similar to that shown schemati-
cally in Figure 3a. The permeameter cylinder should be constructed of a
transparent plastic material, The inside diameter of the cylinder should
be not less than about i0 times the diameter of the largest soil particles,
except when the specimenis encased ina rubber membrane as in the perm-e-
ability test with pressure chamber, in which case the diameter of the cylin-
der should be at least six times the diameter of the largest soil particles.
- Ckrrhmd Dewrd OtdAd Hbfv L$,+
b
- =
Cal C O N S T A N T - H E A D A P P A R A T U S (bl fALLiNC-HEAD A P P A R A T U S
Figure 3. Schematic diagram of constant-head and falling -head

permeability apparatus
VII-5
EM itlO-2-1906
Appendix VII
30 Nov 70
Piezometer taps along the side of the permeameter within limits to be oc-
cupied by the sample are advantageous in that the head loss within the
sample is always measured across a fixed distance and rapid determina-
tion of hydraulic gradient can be made.
(2) Perforated metal or plastic disks and circular wire
screens, 35 to 400 mesh, cut for a close fit inside the permeameter. .
(3) Glass tubing, rubber or plastic tubing, stoppers, screw
clamps, etc., necessary to make connections as shown in Figure 3a.
(4) Filter materials such as Ottawa sand, coarse sand, and
gravel of various gradations.
(5) A device for maintaining a constant-head water supply.
(6) Deaired distilledt water. Tapwater contains dissolved
air and gases which separate from solution in the initial layers of a test
specimen of soil in the form of small bubbles. These bubbles reduce the
permeability of the soil by decreasing the void space available for the
flow of water. The most common method for removing dissolved air
from water is by boiling the water and then cooling it at reduced pres-
sures. This method is applicable oniy with small quantities of water.
Freshly distilled water also has a very negligible amount of air. Large
quantities of deaired distilled water may be prepared and retained for
subsequent use by spraying distilled water in a fine stream into a con-
tainer from which the air has been evacuated (see Fig. 4). Permeability
tests on saturated specimens should show no significant decrease in
permeability with time if properly desired distilled water is used. How-
ever, if such a decrease in permeability occurs during a test, then a pre-
filter, consisting of a layer of the same material as the test specimen,
should be used between the deaired distilled water reservoir and the test
specimen to remove the air remaining in solution. $
t Demineralized water or tapwater when it is known to be relatively free
of minerals may be used in place of distilled water.
$ G. E. Bertram, An Experimental Investigation of Protective Filters,
Soil Mechanics Series No. 7, Harvard University (Cambridge, Mass.,
January i940, reprinted May i959).
VII- 6
EM liiO-2-i906
Appendix VII
30 Nov 70
FRO D I S T I L L E D WATER
SUPPLY
DEAIREO DtSTlLLEO
W A T E R RESkRV0lf.l
II-
VACUUMWATERTRAP
~ FRDM V A C U U M SUPFLY
T O T E S T S P E C I M E N -
RELEASE
Figure 4. Schematic diagram of apparatus for preparing

deaired distilled water
(7) Manometer board with tubing leading from the piezom-

eter taps. If piezometer taps are not provided, equipment to measure
the distance between the constant-head source and tailwater is required.
(8) Timing device, a watch or clock with second hand.
(9) Graduated cylinder, 100 -ml capacity.
(i0) Centigrade thermometer, range 0 to 50 C, accurate
to 0.i c.
(ii) Balance, sensitive to 0.i g.
VII-7
E M 3110-2-i906
Appendix VII
30 Nov 70
(13) Scale, graduated in centimeters.

c. Placement and Saturation of Specimen. Placement and satura-
tion of the specimen shall be done in the following steps:
(1) Record all identifying information for the specimen* s u c h
as project, boring number, sample number, or other pertinent data, on a
data sheet (Plate VII-1 is a suggested form).
(2) Oven-dry the specimen, Allow it to cool and weigh to the
nearest 0.1 g. Record the oven-dry weight of material on the data sheet
opposite W s. The amount of mate_rial should be sufficient to provide a
specimen in the permeameter having a minimum length of about one to two
times the diameter of the specimen.
(3) Place a wire screen, with openings small enough to retain
t h e s p e c i m e n , over a perforated disk near the bottom of the permeameter
above the inlet. The screen openings should be approximately equal to the
i0 percent size of the specimen.
(4) A l l o w deaired distilled water to enter the water inlet of the
permeameter to a height o f about i/2 in. above the bottom of the screen,
taking care that no air bubbles are trapped under the screen.
(5) Mix the material thoroughly and place in the permeameter
to avoid segregation. The material should be dropped just at the water
surface, keeping the water surface about i/2 in. above the top of the soil
during placement. A funnel or a special spoon as shown in Figure 5 is
convenient for this purpose.
(6) The placement procedure outlined above will Rsult in a
saturated specimen of uniform density although in a relatively loose condi-
tion. To produce a higher density in the specimen, the sides of the perme-
ameter containing the soil sample are tapped uniformly along its circum-
ference and length with a rubber mallet to produce an increase in density;
however, extreme caution should be exercised so that fines are not put into
suspension and segregated within the sample. As an alternative to this pro-
cedure, the specimen may be placed in the in the dry using a funnel or
VII-8
EM ii10-2-1906
Appendix VII
30 Nov 70
spoon which permits the material

to fall a constant height. The de-
sired density may be achieved by
vibrating the specimen to obtain a
specimen of predetermined height,
Compacting the specimen in layers
is not recommended as a film of
dust may be formed at the surface
of the compacted layer which might
affect the permeability results,
After placement, apply a vacuum ,CONNEClWG
to the top of the specimen and per-

WRE
mit water to enter the evacuated

specimen through the base of the
permeameter.
(7) After the speci-
men has been placed, weigh the
excess material, if any, and the
container. The specimen weight is
the difference between the original
weight of sarnple and the weight of
the excess material. Care must
be taken so that no material is lost
da ring placement of the specimen. FRONT VIEW
If there is evidence that material has

Figure 5. Spoon for placing
been lost, oven-dry the specimen cohesionless soils
and weigh after the test as a check.
(8) Level the top of the specimen, cover with a wire screen
similar to that used at the base, and fill the remainder of the permeameter
with a filter material.
(9) Measure the length of the specimen and inside diameter of
the permeameter to the nearest 0.3 cm and record on the data sheet as
initial height and diameter of specimen.
(10) Test the specimen at the estimated natural void ratio or
VU-9
EM iiI0-2-1906
Appendix VII -
30 Nov 70
at a series of different void ratios, produced by increasing the amount of

vibration after each permeability determination. Measure and record the
length (height) of specimen in the permeameter prior to each determina-
tion. Permeability determinations at three different void ratios are
usually sufficient to establish the relation of void ratio to permeability.
d. Procedure. The procedure shall consist of the following steps:
(I) Measure the distance, L i, between the centers of the
piezometer taps to the nearest O.Oi cm and record on the data sheet.
(2) Adjust the height of the constant-head tank to obtain the
desired hydraulic gradient. The hydraulic gradient should be selected so
that the flow through the specimen is laminar. The range of laminar flow
conditions can be determined by plotting discharge versus hydraulic gra-
dient. A straight-line relation indicates laminar flow, while deviations
from the straight-line at high gradients indicate turbulent flow. Laminar
flow for fine sands is limited to hydraulic gradients less than approxi-
mately 0.3. It is usually not practicable to achieve laminar flow for
coarser soils, and the tests generally should be run at the hydraulic
gradient anticipated in the field.
(3) Open valve A (see Fig. 3a) and record the initial piezom-
eter readings after the flow has become stable. Exercise care in building
up heads in the permeameter so that the specimen is not disturbed.
(4) After allowing a few minutes for equilibrium conditions
to be reached, measure by means of a graduate the quantity of discharge
corresponding to a given time interval. Measure the piezometric heads
and the water temperature in the permeameter.
(5) Record the quantity of flow, piezometer readings, water
temperature, and the time interval during which the quantity of flow was
measured on the data sheet, Plate VII-j..
(6) Repeat steps (4) and (5) several times over a period of
about i hr, and compute the coefficient of permeability corresponding to
each set of measured data. If there is no substantial change in the
EM iilO-2-i906
Appendix VII
30 Nov 70
permeability, then the computed permeability is probably reliable. If there

is a slight decrease in the permeability, then the permeability computed
from the initial measurements, rather than the average, should be reported,
so long as a plot of permeability versus time shows that the initial measure-
ments are consistent with the subsequent measurements; a difference in
permeability may result from a change in density caused by inadvertent
jarring of the specimen in the permeameter. If there is any substantial de-
crease of the permeability with time, a prefilter should be used between the
water reservoir and the permeameter (see paragraph 3b(6)). The criterion
for judging whether a change in the computed permeability is substantial
depends on the desired accuracy of the coefficient of permeability.
(7) If desired, reduce the void ratio as previously described
and repeat the constant-head test.
e.
- Computations. The computations consist of the following steps:
(i) Compute the test void ratios in accordance with Appendix
II, UNIT WEIGI-ITS, VOID RATIO, POROSITY, AND DEGREE OF SATURA-
TION, The specific gravity shall be estimated or determined in accord-
ance with Appendix IV, SPECIFIC GRAVITY.
(2) Compute the coefficient of permeability, k, by means of
the following equation: QXLXRT
k
20= hXAXt
where k coefficient of permeability, cm per set at 20 C
20 =
Q= quantity of flow, cc
L = length of specimen over which head loss is measured, cm.
If piezometer taps are used, L = L 1 = distance between
piezometer taps, cm
RT = temperature correction factor for viscosity of water ob-
tained from Table VII-1
h = loss of head in length, L, or difference in piezometer
readings = hi _ h2, c m
A = cross-sectional area of specimen, sq cm
t = elapsed time, set
EM iiiO-2-i906
Appendix VII
30 Nov 70
-f. Presentation of Results. The coefficient of permeability shall

be reported in units with coefficients of i.0, i X iOw4, and 1 X iOm9 cm per
sec. The void ratio of the specimen shall be reported with all values of k.
The coefficient of permeability, k, is logarithmically dependent upon the
void ratio of the soil. Where k is determined at several void.ratios, the
test results shall be presented on a semilogarithmic chart as shown in
Figure 6 in which k is plotted on the abscissa (logarithmic scale) and the
void ratio is plotted on the ordinate (arithmetic scalpj.
i iiiiii A t
10 100
COEFFICIENT OF PERMEABILITY IN CM/SEC * IO-
Figure 6. Relation between permeability and void ratio

for cohesionless soils
4. FALLING-HEAD PERMEABILITY TEST WITH PERMEAMETER

C Y L I N D E R , z. U s e . The falling-head test with t&e permeameter
VII-i3
E M 11iO-2-1906
Appendix VII
30 Nov 70 I
cylinder should in general be used for determining the permeability Of

remolded samples of cohesionless soils having a permeability less than
about IO x iOm4 cm per sec.
b.
- Apparatus. The apparatus and accessory equipment should
consist of the following:
(I) A permeameter cylinder similar to that shown schemat-
ically in Figure 3b, or modified versions thereof. The permeameter
cylinder should be constructed of a transparent plastic material. The in-
side diameter of the cylinder should be not less than about IO times the
diameter of the largest soil particles. The use of two piezometer taps, as
shown by Figure 3b, conn.ected to a standpipe and discharge level tube
eliminates the necessity for taking into account the height of capillary rise
which would be necessary in the case of a single standpipe of small size.
The height of capillary rise for a given tube and condition can be mea-
sured simply by standing the tube upright in a beaker full of water. The
size of standpipe to be used is generally based on experience with the -
equipment used and soils tested. In order to accelerate testing, air pres-
sure may be applied to the standpipe to increase the hydraulic gradient.
(2) Perforated metal or plastic disks and circular wire
s c r e e n s , 35 to 100 mesh, cut for a close fit inside the permeameter.
(3) Glass tubing, rubber or plastic tubing, stoppers, screw
clamps, etc., necessary to make connections as shown in Figure 3b.
(4) Filter materials such as Ottawa sand, coarse sand, and
gravel of various gradations.
(5) Deaired distilled water, prepared according to para-
graph 3k(6).
(6) Manometer board or suitable scales for measuring levels
in piezometers or standpipe.
( 7 ) T i m i n g d e v i c e , a watch or clock with second hand.
( 8 ) C e n t i g r a d e t h e r m o m e t e r , range 0 to 50 C, accurate to 0.1 C.
( 9 ) B a l a n c e , sensitive to 0.1 g.
VII-14
EM iiiO-2-1906
Appendix VII
30 Nov 70

(if) Scale, graduated in centimeters.
C. Placement and Saturation of Specimen. Placement and satu-
ration of the specimen shall be done as described in paragraph 3c.
Identifying information for the sample and test data shall be entered on
a data sheet similar to Plate VII-2.
d.
- Procedure.
- The procedure shall consist of the following steps:
(i) Measure and record the height of the specimen, L, and
the cross-sectional area of the specimen, A.
(2) With valve B open (see Fig. 3b), crack valve A and
slowly .bring the water level up to the discharge level of the permeameter.
(3) Raise the head of water in the standpipe above the dis-
charge level of the permeameter. The difference in head should not result
in an excessively high hydraulic gradient during the test. Close valves
A and B.
(4) Begin the test by opening valve B. Start the timer. As
the water flows through the specimen, measure and record the height of
water in the standpipe above the discharge level, ho, in centi.meters, at
time to, an,d the height of water above the discharge level, hf, in
centimeters, at time tf.
(5) Observe and record the temperature of the water in the
permeameter.
(6) Repeat the determination of permeability, and if the com-
puted values differ by an appreciable amount, repeat the test until con-
sistent values of permeability are obtained.
e. Computations. The computations consist of the following steps:
(i) Compute the test void ratios as outlined in paragraph*3c(i).
(2) Compute the coefficient of permeability, k, by means of
the following equation:
k = 2.303 ;+
VII-f 5
E M i110-2-1906
Appendix VII
30 Nov 70
where a = inside area of standpipe, sq cm

A = cross-sectional area of specimen, sq cm
L = length of specimen, cm
t = elapsed time (tf - to), set
ho = height of water in standpipe above discharge level at time
to, c m
hf = height of water in standpipe above discharge level at time
tf, c m
= temperature correction factor for viscosity of water ob-
RT
t a i n e d f r o m T a b l e VII-I, degrees C
If a single standpipe of small diameter is used as shown in Figure 2, the
height of capillary rise, hc, should be subtracted from the standpipe
readings to obtain ho and hf.
f
- Presentation of Results. The results of the falling-head
permeability test shall be reported as described in paragraph 3 f.
5. PERMEABILITY TESTS WITH SAMPLING TUBES. Permeability
tests may be performed directly on undisturbed samples without removing
them from the sampling tubes. The sampling tube serves as the per-
meamete r cylinder. The method is applicable primarily to cohesionless
soils which cannot be removed from the sampling tube without excessive
disturbance. The permeability obtained is in the direction in which the
sample was taken, i.e. generally vertical. The permeability obtained in a
vertical direction may be substantially less than that obtained in a hori-
zontal direction.
Permeability tests with sampling tubes may be performed under
constant-head or falling-head conditions of flow, depending on the esti-
mated permeability of the sa&mple (see paragraph 22). T h e e q u i p m e n t
should be capable of reproducing the conditions of flow in the constant-
head or falling-head tests. It is important that all disturbed material or
material containing drilling mud he removed from the top and bottom of
the sample. The ends of the sample should be protected by screens held
in place by perforated packers. The test procedure and computations are
VII-f 6
E M iiiO-2-i$O6
Appendix VII
30 Nov 70
the same as those described previously for each test.

6. PERMEABILITY TEST WITH PRESSURE CHAMBER. In the perme-
ability test with a pressure chamber, see Figure 7, a cylindrical specimen
is confined in a rubber membrane and subjected to an external hydrostatic
pressure during the permeability test. The advantages of this type of
test are: (a) leakage along the sides of the specimen, which would occur
if the specimen were tested in a permeameter, is prevented, and (b) the
specimen can be tested under conditions of 1oadin.c expected in the field.
The test is applicable primarily to cohesive soils in the undisturbed,
r e m o l d e d or compacted state. Complete saturation of the specimen, if it
is not fully saturated initially, is practically impossible. Consequently,
this test should be used only for soils that are fully saturated, unless
values of permeability are purposely desired for soils in an unsaturated
condition. The permeability test with the pressure chamber is usually
performed as a falling-head test.
The permeability specimens for use in the,pressure chamber generally
should be 2.8 in. in diameter, as rubber membranes and equipment for
cutting and trimming specimens of this size are available for triaxial
testing apparatus (see Appendix X, TRIAXIAL COMPRESSION TESTS). A
specimen length of about 4 in. is adequate. (The dimensions of a test
specimen may be varied if equipment and supplies are available to make
a suitable test setup.) The pressure in the chamber should not be less
than the maximum head on the specimen during the test. The other test
procedure and computations are the same as those described for the
falling-head test. The linear relation between permeability and void ratio
on a semilogarithmic plot as shown in Figure 6 is usually not applicable
to fine-grained soils, particularly when compacted. Other methods of
presenting permeability-void ratio data may be desirable.
7. PERME_4BILITY TESTS WITH BACK PRESSURE.
a. Description. Gas bubbles in the pores of a compacted or un-
disturbed specimen of fine-grained soil will invalidate the results of the
VII-17
EM 1110-2-1906
Appendix VII
30 Nov 70
HIGMER LATERAl. PRfSSURCS

I N PLACE OF i_EVELlNC EI!_8
WING NUT
POROUS PLATE ----#+
-FLASlPZ .UI?t!VE AN0

T U B I N G USI!0 FOR
SOIL SPl!ClU!N SMURATI~ O F
SPECIMN.
TAILWATCR
Y
/ l_EVCL
~------_-
.
,
--1---r--- 1 I. 0.
Figure 7. Pressure chamber for permeability test
VII -18
E M IilO-2-i906
Appendix VII
30 Nov 7 0
permeability tests described in the preceding paragraphs. It is known

that an increase in pressure will cause a reduction in volume of gas bub-
bles and also an increased weight of gas dissolved in water. To each de-
gree of saturation there corresponds a certain additional pressure (back
pressure) which, if applied to the pore fluid of the specimen, will ca.use
complete saturation. The permeability test with back pressure is per-
formed in a pressure chamber such as that shown in,Figure 8, utilizing
equipment that permits increasing the chamber pressure and pore pres-
sure simultaneously, maintaining their difference constant The method
is generally applicable to fine-grained soils that are not fully saturated.
Apparatus and procedures have been described by A. Casagrandet and
L. Bjerrum and J. Huder.$
b.
- Procedure (see Fig. 8). The procedure shall, consist of the
following steps :
(I) After having determined the dimensions and wet weight
of the test specimen, place it in the triaxial apparatus, using the same
procedure as for setting up a specimen for an R triaxial test with pore
pressure measurements except that filter strips should not be used
( s e e para 7, A P P E N D I X X, T R I A X I A L C O M P R E S S I O N T E S T S ) .
(2) Saturate the specimen and verify 400 percent saturation
using the procedure described in paragraph 7b, APPENDIX X, TRIAXIAL
C O M P R E S S I O N T E S T S . B u r e t t e A is utilized during this operation.
(3) With the drainage valves closed, increase the chamber
Casagrande, A., Third Progress Report on Investigation of Stress

Deformation and Strength Characteristics of Compacted Clays, Soil
Mechanics Series No. 70, Nov 1963, Harvard University, C,ambridge,
M a s s . , pp 30 and 3i.
Bierrum, L. and Huder, J., Measurement of the Permeability of

Compactid Clays , Proceedings, Fourth International Conference on
Soil Mechanics and Foundation Engineering, London, Vol 1,
Aug 1957, pp 6-8.
VII- 49
E M iilOF2-1906
Appendix VII
30 Nov 70
pressure to attain the desired effective consolidation pressure (chamber

pressure minus back pressure). At zero elapsed time, open valves E
and F.
(4) Record time, dial indicator reading, and burette reading
at elapsed times of 0, i5, and 30 set, 1, 2, 4, 8, and 15 min, and 1, 2, 4,
and 8 hr, etc. Plot the dial indicator readings and burette readings on an
arithmetic scale versus elapsed time on a log scale. When the consoli-
dation curves indicate that primary consolidation is c o m p l e t e close
valves E and F.
(5) Apply a pressure to burette B greater than that in
burette A. The difference between the pressures in burettes B and A
is equal to the head loss h; h divided by the height of the specimen after
consolidation, L, is the hydraulic gradient. The difference between the
two pressures should be kept as small as practicable, consistent with the
requirement that the rate of flow be large enough to make accurate
measurements of the quantity of flow within a reasonable period of time.
Because the difference in the two pressures may be very small in com-
parison to the pressures at the ends of the specimen, and because the
head loss must be maintained constant throughout the test, the difference
between the pressures within the burettes must be measured accurately;
a differential pressure gage is very useful for this purpose. The dif-
ference between the elevations of the water within the burettes should
also be considered (I in. of water = 0.036 psi of pressure).
(6) Open valves D and F. Record the burette readings at any
zero elapsed time. Make readings of burettes A and B and of temperature
at various elapsed times (the interval between successive readings de-
pends upon the permeability of the soil and the dimensions of the speci-.
men). Plot arithmetically the change in readings of both burettes versus
time. Continue making readings until the two curves become parallel
and straight over a sufficient length of time to accurately determine the
rate of flow (slope of the curves).
VII-21
CM 1110-2-1906
\ppendix VII
30 Nov 70
(7) If it is desired to determine the permeability at several

void ratios, steps 3 through 6 can be repeated, using different consolida-
tion pressures in step 3.
(8) At the end of the permeability determinations, close all
drainage valves and reduce the chamber pressure to zero; disassemble
the apparatus.
(9) Determine the wet and dry weights of the specimen.
C. Computations. The computations consist of the following
steps.
(i) Compute the test void ratios as outlined in para-
graph 3~( 1).
(2) Computations of coefficients of permeability are the
same as those described for the constant-head permeability test.
8. PERMEABILITY TESTS WITH CONSOLIDOMETER. A perme-
ability test in a consolidometer (see Appendix VIII, CONSOLIDATION
-
TEST) is essentially similar to that conducted in a pressure chamber,
except that the specimen is placed within a relatively rigid ring and is
loaded vertically. The test can be used as an alternate to the perme-
ability test in the pressure chamber. The test is applicable primarily
to cohesive soils in a fully saturated condition. Testing is usually per-
formed under falling-head conditions.
A schematic diagram of the consolidation apparatus set up for a
falling-head permeability test is shown in Figure 9. Identifying informa-
tion for the specimen and subsequent test data are entered on a data sheet
(Plate VII-3 is a suggested form). The specimen should be placed in the
specimen ring and the apparatus assembled as outlined under Appen-
dix VIII, CONSOLIDATION TEST. The specimen is consolidated under
the desired load and the falling-head test is performed as previously
described. The
VII-22
E M iiiO-Z-i906
Appendix VII
30 Nov 70
STANDPIPE AREA = a
NOTE: VALVE BETWEEN DEAIRED DISTILLED WATER

JUG AND CONSOLIDATION APPARATUS SHOULD
BE CLOSED DURING THE PERMEABILITY TEST.
DIAL INDICATOR
STILLED WATE
SOIL SAMPLE
Figure 9. Schematic diagram of falling-head device for

permeability test in consolidometer
net head on the specimen may be increased by use of air pressure; how-
ever, the pressure on the pore water should not exceed 25 to 30 percent
of the vertical pressure under which the specimen has consolidated. Dial
indicator readings are observed before and after consolidation to permit
computation of void ratios. The determination of the coefficient of perme-
ability may be made in conjunction with the consolidation test, in which
case the test is performed at the end of the consolidation phase under each
load increment. Computations are similar to those described for the
VII- 23
EM iiiO-2-i906
Appendix VII I
30 Nov 70
falling-head test with the permeameter cylinder.

The permeability may also be determined indirectly from computa-
tions using data obtained during the consolidation test; however the as-
sumptions on which the method is based are seldom satisfied, and conse-
quently, the direct determination of permeability should be employed
where reliable values of permeability are required.
inaccurate determinations of the coefficient of permeability:
a. Stratification or nonuniform compaction of cohesionless soils.
If the specimen is compacted in layers, any accumulation of fines at the
surface of the layers will reduce the measured coefficient of permeability.
b.
- Incomplete initial saturation of specimen.
-C. Excessive hydraulic gradient. Darcys law is applicable only

to conditions of laminar flow.
d. Air dissolved in water. No other source of error is as
-
troublesome as the accumulation of air in the specimen from the flowing
water. As water enters the specimen, small quantities of air dissolved
in the water will tend to collect as fine bubbles at the soil-water interface
and reduce the permeability at this interface with increasing time. The
method for detecting and avoiding this problem is described in paragraph
32(6). (It should be noted that air accumulation will not affect the coef-
ficient of permeability determined by the constant-head test if piezometer
taps along the side o*4 the specimen are used to measure the head loss.)
e.
- Leakage along side of specimen in permeameter. One major
advantage to the use of the triaxial compression chamber for permeability
tests (see paragraphs 6 and 7) is that the specimen is confined by a
flexible membrane which is pressed tightly against the specimen by the
chamber pressure.
VII-24
EM iilO-2-i906
Appendix V
30 Nov 7
~.- -_-~ _--._--- --- --. - -- -_ _______
rONSTANJ'-HEAD 1'XRMlXBIY~ TEST
DATK-_---~----
RCUEC~
3XNG NO.
__.~~..___
._ --.-..--.~.. -----=
or Specimen No. -~--
oid ratio =
--~ .--.---
(1) Correction factor for vitxon1t.y of water at. 20 C obtained from table VII-l.
QxLx%
(2) km = hxAxt *
where L = height of specimen or di6t6nce betveen plezaneter tap6 if used.
Remarks
-
*ChIliClEUl- Ccmputed by Checked by -
iNG FORM
I JUN 65 3w PLATE VII- 1
VII-25
E M iilO-2-1906
Appendix VII
30 Nov 70
__ ~__
lo,laC 70 =el
PLATE VII-2
E M iliO-2-1906
Appendix VII
30 Nov 70
FALLlNG+lEAD PERMEABILITY T&T

WITH CONSOLIDOMETER
DATE
P L A T E VII- 3
VII-27
E M iliO-2-i906
30 Nov 70
APPENDIX VIII:
CONSOLIDATION TEST
1. INTRODUCTION. Consolidation is the process of gradual transfer of

an applied load from the pore water to the soil structure as pore water is
squeezed out of the voids. The amount of water that escapes depends on
the size of the load and compressibility of the soil. The rate at which it
escapes depends on the coefficient of permeability, thickness, and com-
pressibility of the soil. The rate and amount of consolidation with load
are usually determined in the laboratory by the one-dimensional consol-
idation test. In this test, a laterally confined soil is subjected to suc-
cessively increased vertical pressure, allowing free drainage from the
top and bottom surfaces.
a.
- Consolidometer shall consist of a rigid base, a consolidation
ring, porous stones, a rigid loading plate, and a support for a dial indica-
tor (Fig. 1). It may be either the fixed-ring or the floating -ring type both
of which are shown in Figure 2.
II OVERFLOW tl-
Figure 1. Typical consolidometer
VIII - 1
E M 1110-2-1906
Appendix VIII
30 Nov 70 -
r-l r I n
II I LOADING PLATE I II
SOIL SPECIMEN
L BASE
a. FIXED-RING CONSOLIDOMETER
-FL OA T/W
l?/lve
SOIL SPEC.1 MEN
-
BASE
I
b. FLOATING-RING CONSOLIDOMETER
Figure 2. Schematic diagrams of fixed-ring and

floating-ring consolidometers
The various metal parts of the consolidometer shall be of the same non-
corrosive material. All-plastic or combination plastic and metal consoli-
dometers may also be used to reduce electrochemical effects. The con-
solidometer shall conform to the following requirements.
(3) Fixed-ring consolidometer shall have a rigid base with a
recess for supporting the bottom porous stone and for seating and attach-
ing the consolidation ring. The upper surface of therecess shall be
grooved to permit drainage. The base shall also have (a) an inundation
VIII - 2
EM iiiO-2-1906
A.ppendix VIII
30 Nov 70
ring to permit submergence of the specimen in water to prevent evapora-

tion of water from the specimen during the test, and (b) suitable connec-
tions and a standpipe for making permeability tests.
(2) Floating-ring consolidometert shall have a rigid base
for supporting the bottom porous stone. The base shall be large enough
to permit free vertical movement of the consolidation ring and shall have
a chamber surrounding the ring for submergence of the specimen.
b. Consolidation ring shall completely and rigidly confine and
-
support the specimen laterally. The inside diameter of the ring should
not be less than 2-3/4 in. and preferably not less than 4 in.; use of larger
rings for specimens of larger diameter, particularly with the fixed-ring
consolidometer, will reduce the percentage of applied load carried by
side friction and consequently will provide more accurate results.
Normally, the ratio of the height of ring to inside diameter of ring should
be between i/4 and 1/6. The consolidation ring may be lined with a ma-
terial such as Teflon to reduce the friction between the ring and a speci-
men of fine-grained soil. A stainless steel ring is preferable for speci-
mens containing abrasive particles.
-C. Porous stones more pervious than the specimen of soil should
be lised to permit effective drainage. For routine testing, stones of
medium porosity are satisfactory. The diameter of the porous stones
-1 In the floating-ring consolidometer the friction between the inside of the

ring and the specimen is less than that in the fixed-ring type. However,
when very soft soils are tested with the floating-ring consolidometer,
the side friction will not support the weight of the ring, and compression
occurs toward the middle of the specimen from top and bottom. The
floating-ring device is suitable only for comparatively stiff soils, and
has the disadvantage that it cannot be used for permeability tests.
VIII- 3
EM IilO-2-1906
Appendix VIII
30 Nov 70
shall be such as to prevent the squeezing out of soil through the clear-
ance spaces between the ring and stone and to permit free compression
of the specimen without binding; to
minimize the possibility of binding,
the sides of the upper porous stone
A A of the fixed-ring consolidometer

should be slightly tapered away
L - - - - - - ~~ J
from the specimen, while both po-
rous stones of the floating-ring
consolidometer should be tapered.
A clearance of about 0.010 to O.Oi5
PLAN in. around the stone generally will
,u be adequate; however, if very soft
I ++X
&=?I - -
,,,oU WAWlc 0, c0NSo#_lMTloN WG - 0.006* soils are tested, a smaller clear-
ante may be desirable or retainer
-
SECTION A-A
rings may be used as shown in

Figure 3. Typical retainer
ring Figure 1. Details of a typical re-
tainer ring are shown in Figure 3.
The porous stones should be cleaned after every test, preferably in an
ultrasonic cleaner or by boiling and flushing.
z!* Loading devices of various types may be used to apply load to

the specimen. The most commonly used is the beam-and-weight mech-
anism. The loading device should be capable of transmitting axial load to
the specimen quickly and gently. Also, the equipment should be capable of
maintaining the load constant for at least 24 hr. The equipment should be
calibrated to ensure that the loads indicated are those actually applied to
the soil specimen.
VIII - 4
EM i110-2-1906
Appendix VIII
30 Nov 70
-c. Dial Indicator. A dial indicator reading counterclockwise,

with a range of 0.50 in. and graduated to O.OOOi in., is recommended.
-f. Equipment for Preparing Specimens. A trimming turntable

operated as a vertical lathe is commonly used in preparing specimens
(see Fig. 4). Suitable trimming
knives notched to fit the thickness
of the consolidation ring, a wire
saw with 0.01 -in. -diameter wire,
and a metal straightedge or screed
are also required.
g* Other i t e m s needed are:
(I) Balances, sensitive
to 0.1 g and 0.01 g
(2) Timing device, a Figure 4. Cutting. specimen into
specimen ring
watch or clock with second hand
(3) Centigrade thermometer,, range 0 to 50 C, accurate to
0.1 c
(4) Distilled or demineralized water
(5) Filter papers and glass plates
(6) Apparatus necessary to determine water content and spe-
cific gravity (see Appendices I, WATER CONTENT - GENERAL, and IV,
SPECIFIC GRAVITY)
3. CALIBRATION OF EQUIPMENT. In the consolidation test it is de-
sired to measure only the volume change of the specimen; therefore, cor-
rections must be applied for any significant deformation due to the com-
pressibility of the apparatus itself. In sandy and stiff soils, an appreciable
proportion of the total deformation may be caused by this factor. There-
fore, a calibration curve should be prepared for each consolidometer when
testing such soils. This is done by placing the consolidometer with sub-
merged porous stones and filter papers in the loading device, applying the
load increments to be used in the consolidation test, and reading the dial .
VIII- 5
E M 1110-2-1906
Appendix VIII
30 Nov 70 -
indicator for each load. After the maximum load has been applied, the
loads are decreased in the same order as that in which they were applied,
and the dial indicator reading is again recorded. Since the deformations
are almost instantaneous, the effect of time can be ignored. The total
change in dial reading for each load is the correction to be applied to the
dial reading recorded during the consolidation test under that same load.
Generally, a single cycle will be sufficient for the calibration.
4. PREPARATION OF SPECIMENS. Specimens shall be prepared in a
humid room to prevent evaporation of soil moisture. Extreme care shall
be taken in preparing specimens of sensitive soils to prevent disturbance
of their natural structure. Specimens of relatively soft soils may be pre-
pared by progressive trimming in front of a calibrated, ring-shaped speci-
men cutter as shown in Figure 2 of Appendix II, UNIT WEIGHTS, VOID
RATIO, POROSITY, AND DEGREE OF SATURATION. More commonly,
specimens are prepared using the trimming turntable shown in Figure 4
herein; the procedure, based on the use of this equipment, shall be as -
described in the following subparagraphs. Preferably, specimens of com-

pacted soil should be compacted to the desired density and wa.ter content
directly into the consolidation ring, in thin (1/4 to 3/8 in.) layers, using a
pressing or kneading action 0f.a tamper having an area less than one-sixth
the area of the specimen and thoroughly scarifying the surface of each
layer before placing the next. Alternatively, specimens may be trimmed
from samples compacted in a compaction mold by a similar kneading action.
a.
- Using a wire saw, knives, or other tools, trim the specimen into
approximately cylindrical shape with a diameter about I/2 in. greater than
the inside diameter of the specimen ring. Care should be taken to disturb
the specimen as little as possible during trimming. Chamfer the lower
edge of the specimen until the bottom will fit exactly into the specimen ring.
-b. Place the specimen ring on the rotating wheel and the,specimen
on the ring, starting the bottom into the ring as shown in Figure 4. Use a
cutting tool to trim the specimen to .accurate dimensions, place a glass
VIII-6
E M 11iO-2-1906
Appendix VIII
30 Nov 70
plate on top of the specimen, and gently force the specimen down during
the trimming operation. The specimen shou1.d fit snug in the consolidation
ring.
C.
- Cut off the portion of the specimen remaining above the ring
with a wire saw or knife (or other convenient tool for harder specimens).
Extreme care must be taken for many soils, especially fissured clays, in
cutting off this portion. Carefully true the surface flush with the speci-
men ring with a straightedge. If a pebble is encountered in the surface,
remove it and fill the void with soil. Place a glass piate ( p r e v i o u s l y
weighed) over the ring and turn the specimen over.* Cut off the soil ex-
tending beyond the bottom of the ring in the same manner as that described
for the surface portion. Place another glass plate on this surface, and
again invert the specimen to an upright position, removing the metal disk
if one was used.
a.
- Record all identifying information for the specimen, such as
project number, boring number, and other pertinent data, on the data
sheet (Plate VIII-1 is a suggested form); note any difficulties encountered
in preparation of the specimen. Measure and record the height and cross-
sectional area of the specimen. Record weight of specimen ring and glass
plates. After specimen is prepared, record the weight of the specimen
plus tare (ring and glass plates), and from the soil trimmings obtain 200 g
of material for specific gravityt and water content determinations. Record
* It may be found convenient after the top surface has been prepared to
place over the specimen a circular metal plate, approximately 0.05 in.
thick and of the same diameter as the specimen, and force it down until
it is flush with the top of the ring. This provides a recess for the top
porous stone and prevents the specimen from squeezing out of the con-
solidation ring.
t It is recommended that a specific gravity test be made on representa-
tive material from every consolidation test specimen.
VIII- 7
EM iiiO-2-i906
Appendix VIII
30 Nov 70 -
the wet weight of the material used for the water content determination on
the data sheet.
!?* Fill the grooves in the base of the consolidometer with water.
Fit the porous stone (previously saturated with water) into the base of the
consolidometer. Add sufficient water so that the water level is at the top
of the porous stone. Place a moist filter paper (Whatman No. i or equal)
over the porous stone. (Be very careful to avoid entrapping any air during
the assembly operations .) Place the ring with the specimen therein on
top of the porous stone. If the fixed-ring consolidometer is used, secure
the ring to the base by means of clamps and screws.
C. Place a moist filter paper on top of the specimen, and then
place the previously saturated top porous stone and the loading plate in
position.
-d. Place the consolidometer containing the specimen in the load-
ing device.
e. Attach the dial indicator support to the consolidometer, and -
-
adjust it so that the stem of the dial indicator is centered with respect to
the specimen. Adjust the dial indicator to permit the approximate maxi-
mum travel of the gage but still allow measurement of any swelling.
L Adjust the loading device until it just makes contact with the
specimen. The seating load should not exceed about 9.04 ton per sq ft.
g* Read the dial indicator, and record the reading on a data sheet
(Plate VIII-2 is a suggested form). This is the initial reading of the dial
indicator.
& With the specimen assembled in the loading device, apply a
load of 0.25 ton per sq ft to the specimen and immediately inundate the
specimen by filling the volume within the inundation ring or the chamber
surrounding the sp.ecimen with water. If a fixed-ring device is used,
a low head of water should be applied to the base of the specimen and
maintained during the test by means of the standpipe. Place a
thermometer in the water, and record the temperature at 2-hr intervals.
VIII - 8
E M 1iiO-2-1906
Appendix VIII
30 Nov 70
To obtain reliable time-consolidation curves the temperature should not

vary more than h 2 C during the test. For most fine-grained soils a load
of 0.25 ton per sq ft is usually enough to prevent swelling, but if swelling
occurs apply additional load increments until swelling ceases. Were the
specimen permitted to swell, the resulting void ratio-pressure curve
would have a more gradual curvature and the preconsolidation pressure
would not be well defined. Alternatively to applying a large initial load
increment, swelling can be prevented by not inundating the specimen un-
til the load on the specimen has reached such a ievel that consolidation is
obviously occurring along the straight-line portion of the void ratio-
pressure curve, During the stages before water is added, the humidity
around the specimen should be maintained at 100 percent to prevent evap-
oration; a moist paper towel, cotton batting, or other cellular material
wrapped around the specimen is usually adequate. This alternative pro-
cedure permits an initial load increment less than 0.25 ton per sq ft to be
applied to the specimen.
1. Continue consolidation of the specimen by applying the next
load increment. The following loading schedule is considered satisfactory
for routine tests: 0.25, 0.5, 1.0, 2.0, 4.0, 8.0, and 16.0 tons per sq ft, the
total load being doubled by each load increment. The maximum load
should be great enough to establish the straight-line portion of the void
ratio-pressure curve, subsequently described. The loading schedule may
be modified by the designer to simulate the loading sequence anticipated
in the field.
L Observe and record on the data sheet (Plate VIII-2) the defor-
mation as determined from dial indicator readings after various elapsed
times. Readings at 0.1, 0.2, 0.5, 1.0, 2.0, 4.0, 8.0, 15.0, and 30.0 min, and
1, 2, 4, 8, and 24 hr for each load increment are usually satisfactory. A
timing device should be located near the consolidometer to insure accu-
rately timed measurements. Allow each load increment to remain on the
specimen for a minimum of 24 hr until it is determined that the primary
VIII - 9
E M itlO-2-1906
Appendix VIII
30 Nov 70
consolidation is completed. For most plastic, fine-grained soils a time

interval of 24 hr will be sufficient. It is desirable that the duration of all
load increments be the same. During the course of the test, plot the dial
reading versus time data for each load increment on a semilogarithmic
plot as shown in Plate VIII-3. P lot the dial reading on an arithmetic sea le
(ordinate) and the corresponding elapsed time on a logarithmic scale
(abscissa) as shown in Figure 5. For saturated fine -grained soils, the
dial reading versus time curve will generally be similar to the curve
shown in Figure 5 and can be converted into a time-consolidation curve
using the theory of consolidation. The 100 percent consolidation or the com-
pletion of the primary consolidation is arbitrarily defined as the inter-
section of the tangent to the curve at the point of inflection, with the
tangent to the straight-line portion representing the secondary time effect.
The construction necessary for determination of the coordinates
L A B O R A T O R Y TIME-CONSOLlDATiON CRF
FOR A GIVEN LOAD ,NCRE,ENT
SECONDARY CON<OLl,,AT,N
NOTE. U = DEGREE C
Figure 5. Time -consolidation curve
-
VIII- 10
EM iliO-2-i906
Appendix VIII
30 Nov 70
representing iO0 percent consolidation and other degrees of consoli-

dation is shown in Figure 5 (see also EM 1110 -2 -i 904, Settlement
Analysis).
k. Record on a data sheet (Plate VIII-4 is a suggested form)
the dial reading for each load increment corresponding to a selected
time (usually 24 hr) at which primary consolidation has been completed
for all increments.
-1. After the specimen has consolidated under the maximum

load, remove the load in decrements, taking three-quarters of the
load off successively for each of the first two decrements and as
considered desirable thereafter. Take readings of the dial indicator as
each decrement is removed to determine the rebound of the specimen.
Observe, record, and plot the dial readings versus time; loads should not
be removed until the dial readings are relatively constant with time or
until the dial reading versus logarithm of tim.e curve indicates completion
of rebound. The final load at the end of the rebound cycle should be 0.i
ton per sq ft or less, and this load should be maintained for 24 hr in order
to reduce to a tolerable amount the error in the final water content deter-
mination caused by swelling.
m. When the dial readings indicate no further significant rebound,
remove the dial indicator and disassemble the apparatus. Carefully blot
any excess water from the ring and surface of the specimen, eject the
specimen into a dish of known weight, and weigh the dish and wet speci-
men; th(*n oven-dry the wet specimen to constant weight (see Appendix I,
WATER CONTENT - GENERAL).
a.
- From the recorded data compute and record on the data sheet,
Plate VIII-l, the initial and final water contents (see Appendix I, WATER
C O N T E N T - GENERAL). Compute also the height of solids, void ratio
before and after test, initial and final degree of saturation, and dry density
VIII - 11
E M ii10-2-1906
Appendix VIII
30 Nov 70
tests shall be shown on the report forms, Plates VIII-3 and VIII-5. The
data shall be shown graphically in terms of time-consolidation curves

in the form shown as Plate VIII -3 and in terms of void ratio -pressure
curves in the form shown as Plate VIII-5. To obtain the void ratio-
pressure curve, the void ratio, e, is plotted on the arithmetic scale
(ordinate) and the corresponding pressure, p, in tons per square foot
on the logarithmic scale (abscissa) as shown in Figure 6. The over-

burden pressure, p preconsolidation pressure, p and compression
0 C
index, C shall be determined and shown on the report form (Plate

C
VIII-5). The determinations of the preconsolidation and overburden pres -
sures of a soil are normally made by design engineers; procedures are
1.0
0.6
0.1 0.5 1 .o 5.0 10 20
P R E S S U R E , p, TONS/SC? FT
Figure 6. V o i d r a t i o - p r e s s u r e curve
VIII - 14
EM iiiO-2-i906
Appendix VIII
30 Nov 70
discussed in detai 1 in EM 1110 -2 -1904, Settlement Analysis.

The slope of the straight-line portion of the pressure-void ratio
cuz-ve on a semilogarithmic diagram is known as the compression index,
C The compression index is defined by the equation
C.
ei - e2
cc =
%~O P2 - log40 PI
where p and p are selected pressures from the straight-line portion

1 2
of the curve, and e and e a,re the corresponding void ratios. The
i 2
compression index is a measure of the compressibility of a soil. An ex-
ample of the computation of Cc i.s shown in Figure 6. For simplification,
p2 is often chosen to be i0 times pi, in which case the denominator be-
comes unity.
If permeability tests are performed in conjunction with the consolida-
tion test (see Appendix VII, PERMEABILITY TESTS), the coefficient of
permeability for each load increment shall also be plotted in the form
shown as Plate VIII -5.
A brief description of undisturbed specimens should be given on
the report form. The description should include color, approximate con-
sistency, and any unusual features (such as stratification, fissures, shells,
roots, sand pockets, etc.). For compacted specimens, give the method of
compaction used and the relation to maximum density and optimum water
content.
inaccurate determinations of consolidation characteristics:
a. Specimen disturbance during trimming. As in all laboratory
determinations of the engineering properties of undisturbed soils, the
largest errors are caused by changing the natural structure of the soil
while preparing the test specimen; disturbance will affect the time-
deformation relation and will obscure the preconsolidation pressure.
VIII-15
E M 11iO-2-i906
Appendix VIII
30 Nov 70
Trimming must be done in the humid room with every care taken to mini-
mize the disturbance. Since the zone of disturbance caused by trimming
a given soil is essentially constant in depth, the effect of this zone can be
reduced by using a larger specimen.
b.
- Specimen not completely filling ring. The volume of the
specimen must be exactly that of the consolidation ring, 0therwis.e there
will not be complete lateral confinement.
C. Galvanic action in consolidometer. To prevent changes in
the consolidation characteristics of the specimen due to galvanic currents,
all metal parts of the consolidometer should be of the same noncorrosive
material; it is preferable that all such parts be made of plastic.
ZL Permeability of porous stones too low. The measured rate

of consolidation can be markedly affected by the permeability of the porous
stones. The stones should be cleaned after every test to remove embedded
soil particles.
-
e. Friction between specimen and consolidation ring. Tests
have shown that over 20 percent of the load applied to a specimen can be
lost by side friction in a fixed-ring consolidometer and about one-half this
amount in a floating-ring conso1idometer.t The effect of side friction can
be reduced by (i) using a larger diameter specimen, (2) using a thinner
specimen, and (3) lining the consolidation_ ring with Teflon.
L* Inappropriate load increment factor. Depending on the pur -

pose of the test, a load increment factor of 2.0 (that is, of doubling the
total load by each load increment) may not be satisfactory; for example,
to sharply define the preconsolidation pressure, a factor of 1.50 or even
1.25 may be better.
l%.* Unsatisfactory height (or thickness) of specimen. The height

of the specimen will determine how clearly can be detected the break in
t D. W. Taylor, Research on the Consolidation of Clays, Serial 82, De-

partment of Civil and Sanitary Engineering, Massachusetts Institute of
Technology (Cambridge, Mass., i942).
VIII-i6
EMiiiO-2-i906
A@pendix VIII
30 Nov 70
the time-consolidation curve that represents completion of primary con-

solidation. Depending on the character of the soil, if the specimen is too
thin, the time to iO0 percent consolidation may be too rapid, while if too
thick, the break in the curve may be obscured by secondary compression.
Also, when a load increment factor smaller than 2.0 is used, the thickness
of the specimen may have to be increased to cause enough deformation
during primary consolidation to define the break in the curve.
VU-i7
E M 1410-2-1906
Change 1
1 May 80
APPENDIX VIIIA:
SWELL AND SWELL PRESSURE TESTS
4. INTRODUCTION. Swell is the process of imbibing available mois-

ture to cause an increase in soil volume until the pore water pressure
increases to an equilibrium determined by the environment. The
amount of swell to satisfy the new pore pressure equilibrium depends
on the magnitude of the vertical loading and soil properties that in-
clude soil composition, natural water content and density, and soil
structure. The rate of swell depends on the coefficient of perme-
ability (hydraulic conductivity), thickness, and soil properties. Soils
that are more likely to swell appreciably include clays and clay
shales with plasticity indices greater than 25, liquid limits greater
than 40, and natural water contents near the plastic limit or less.
The presence of capillary stress or negative pore water pressure
arising from molecular forces in swelling soils causes available
moisture to be absorbed. The vertical confining pressure required to
prevent volume expansion from absorbed moisture is defined as the
swell pressure.
The swell and swell pressure are generally determined in the
laboratory with the one-dimensional consolidometer (Appendix VIII,
CONSOLIDATION TEST). Swell is determined by subjecting the
laterally confined soil specimen to a constant vertical pressure and
by giving both the top and bottom of the specimen access to free
water (usually distilled) to cause swell. The swell pressure is
determined by subjecting the laterally confined soil specimen to in-
creasing vertical pressures, following inundation, to prevent swell.
2. APPARATUS, CALIBRATION OF EQUIPMENT, AND PREPARA-
TION OF SPECIMENS. The apparatus is essentially the same as that
listed in paragraph 2, Appendix VIII, CONSOLIDATION TEST.
VIIIA- 1
E M lllO-2-4906
Appendix VIIIA
Change 1
1 M&Xy 80
Smoothly ground porous stones should be used in the consolidometer
to minimize seating displacements. Filter paper should not be used
because it is compressible and contributes to displacements. The
equipment is calibrated and the sample prepared in the same manner
as described for the consolidation test.
3. PROCEDURE.
2. S w e l l T e s t . (1) Record all identifying information for the
s p e c i m e n , such as project number, boring number, and other pertinent
data, on the data sheet (Plate VIII-J, p VIII-18, is a suggested form);
note any difficulties encountered in preparation of the specimen.
Measure and record the height and cross-sectional area of the
specimen. Record weight of specimen ring and glass plates. After
the specimen is prepared, record the weight of the specimen plus
tare (ring and glass plates), and from the soil trimmings obtain 200 g
of material for specific gravity and water content determinations.
Record the weight of the material used for the water content deter-
-
mination on the data sheet.
(2) Fit an air-dried, smoothly ground porous stone into
the base of a dry consolidometer. Place the ring with the specimen
therein on top of the porous stone. If the fixed-ring consolidometer
is used, secure the ring to the base by means of clamps and screws.
(3) Place the top air-dried, smoothly ground porous stone
and loading plate in position. The inside of the reservoir should be
moistened to promote a high-humidity environment. The reservoir
and loading plate should subsequently be covered with a sheet of
impervious material such as plastic film or moist paper towel to
inhibit loss of moisture.
(4) Place the consolidometer containing the specimen in
the loading device.
(5) Attach the dial indicator support to the consolidometer,
and adjust it so that the stem of the dial indicator is centered with
VIIIA - 2
E M iiiO-2-1906
Appendix VIIIA
Change
1 MaY 8lO
respect to the specimen. Adjust the dial indicator to allow for both
swell and consolidation measurements.
(6) Adjust the loading device until it just makes contact
with the specimen. The seating load should not exceed about 0.01 ton
per sq ft.
(7) Read the dial indicator and record the reading on a data
sheet (Plate VIII-2, p VIII-i9, is a suggested form). This is the
initial reading of the dial indicator.
(8) Depending on the particular design considerations, a
specific load (e.g. overburden plus design load) is applied to the speci-
men. After a period of at least 5 min but less than 30 min (to avoid
shrinkage from drying), record the dial reading on the data sheet
(Plate VIII-2) and inundate the specimen.
(9) Inundate the specimen by filling the reservoir, within the
inundation ring that surrounds the specimen, with water (distilled,
tap, or field pore water, actual or reconstituted). Cover with the
plastic films and moist paper towel or equivalent. If a fixed-ring
device is used, a low head of water should be applied to the base
of the specimen and maintained during the test by means of the
s andpipe.
(40) Observe and record on the data sheet (Plate VIII-2) the
deformation as determined from dial indicator readings after various
elapsed times. Readings at 0.1, 0.2, 0.5, 1.0, 2.0, 4.0, 8.0, 15.0, and
30.0 min, and 1, 2, 4, 8, 24, 48, and 72 hr are usually satisfactory.
A timing device should be located near the consolidometer to ensure
accurately timed measurements. Allow the load increment to remain
on the specimen until it is determined that the primary swell is com-
pleted. Time to complete the primary swell of heavy clay soils and
clay shales often requires three or more days. Plot the dial reading
versus time data on a semilogarithmic plot as shown in Plate VIIIA-1
to ascertain the completion of primary swell. The completion of
VIIIA- 3
E M iiiO-2-1906
Appendix VIIIA
Change 1
1 May 80
primary swell is arbitrarily defined as the intersection of the tangent
to the curve at the point of inflection, with the tangent to the straight-
line portion representing a secondary time effect as shown in Figure i.
This is similar to the procedure in paragraph 5j, Appendix VIII,
CONSOLIDATION TEST.
(ii) Although a falling-head permeability test may be per-
formed at this point of the swell test (see paragraph 8, Appendix VII,
PERMEABILITY TESTS), the specimen may not be fully saturated and
the permeability results consequently affected. A.fter primary swell
is complete,t the load should be removed in decrements according to
the procedure in paragraphs 51- and 53 Appendix VIII, CONSOLIDA-
TION TEST. The final load should be the seating load.
-b. Swell Pressure Test. The procedure of this test is identical

with the preceding swell test through (9). Following (9), increments
of load are applied as needed to prevent swell. Variations from the
dial reading at the time the specimen is inundated with water should -
be kept preferably within 0.0002 in. and not more than 0.0005 in.
Following 24 hr after the specimen exhibits no further tendency to
swell, a falling-head permeability test may be performed (see para-
graph 8, Appendix VII, PERMEABILITY TESTS) and the final load
(which is also the swell pressure) should be removed in decrements
according to the procedure in paragraphs 52 and 5m,
- Appendix VIII,
CONSOLIDATION TEST. The final load should be the seating load.
4. COMPUTATIONS. The computations for the swell tests are similar
to those presented in paragraph 6, Appendix VIII, CONSOLIDATION TEST.
After primary swell is complete, the loading pressure may be in-

creased to consolidate the specimen until the void ratio is less than
the initial void ratio under the overburden pressure; thereafter,
loads may subsequently be removed to determine rebound charac-
t e r i s t i c s . This procedure permits an alternative measure of the
swell pressure, defined as the total pressure required to reduce
the void ratio to the initial void ratio.
VIIIA- 4
E M i110-2-1906
Appendix VIIIA
Change 1
1 May 80
and optimum water content.
6. POSSIBLE ERRORS. In addition to the possible errors discussed
in paragraph 8, Appendix VIII, CONSOLIDATION TEST, the following
may also cause inaccurate determination of swelling characteristics:
a. Displacements caused by seating of the specimen against
-
the surface of the porous stones may be significant, especially if
swell displacements and loading pressures are small. Thus, smoothly
ground porous stones are recommended.
k. Filter paper is highly compressible and contributes to the
observed displacements and hysteresis in displacements on loading
and .rebound. Filter paper is consequently not recommended.
c. The compressibility characteristics of the consolidometer

-
and the test procedures influence the swell pressure results. Because
very small expansions in volume greatly relieve swell pressures, the
stiffness of the consolidometer should be as large as possible, and
-
variations in displacements that occur during determinaiion of the
swell pressure should be as small as possible.
d. Swelling char.acteristics determined by consolidometer swell
-
tests for the purpose of predicting heave of foundation and compacted
soils are not representations of many field conditions because:
(1) Lateral swell and lateral confining pressures are not
simulated.
(2) The actual availability of water to the foundation soils
may be cyclic or intermittent. Field swell usually occurs under
constant overburden pressure depending on the availability of water.
The swell index, in contrast, is evaluated by observing swell due to
decreases in overburden pressure while the soil specimen is inun-
dated with water.
(3) Rates of swell indicated by swell tests are not reliable
indicators of field rates of swell due to fissures in the mass soil and
inadequate simulation of the actual availability of water to the soil.
VIIIA- 8
EM iiiO-2-i906
Appendix VIIIA
Change 1
L,l May 80
(4) Secondary or long-term swell, which is not evaluated by
these test procedures, may be significant for some clays and clay
shales . These soils may not be fully saturated at the conclusion of
the swell test.
(5) Chemical content of the inundating water affects results;
e.g., when testing shales, distilled water may give radically different
results than natural or reconstituted pore water.
VIIIA- 9
EMiilO-2-1906
30 Nov 70
APPENDIX IX:
DRAINED (S) DIRECT SHEAR TEST
t. INTRODUCTION. The direct shear test is used to measure the shear

strength of a soil under drained conditions. In this test, a relatively thin,
square specimen of soil is placed in a rigid box that is divided horizon-
tally into two frames, the specimen is confined under a vertical or normal
stress, and a horizontal force is applied so as to fail the specimen along a
horizontal plane at its midheight. Generally, a m i n i m u m of three speci-
mens, each under a different normal stress, are tested to establish the
.relation between shear strength and normal stress. The magnitude of the
normal stresses used depends on the range of stresses anticipated for de-
sign. Because of the difficulties involved in controlling drainage of the
soil specimen during the direct shear test, only the drained (S) test
method, in which complete consolidation is permitted under each incre-
ment of normal and shear stress, shall be used.
a.
- Shear box[ of bronze or stainless steel, open at the top and
divided horizontally into two frames that can be fitted together accu-
rately with alignment pins and elevating screws. A schematic diagram
of a direct shear box is shown in Figure 1. The lower frame of the
NORMAL fORCC
Figure i. Schematic diagram of direct shear box
IX-i
EMillO-2-1906
Appendix IX
30 Nov 70
shear box shall contain a reservoir for water, with the bottom grooved or
provided with a grooved area to permit drainage. The upper frame of the
shear box should contain an accurately machined piston, the bottom of
which is also grooved to permit drainage. The upper frame shall be pro-
vided with horizontal locking screws to lock it to the piston within the
upper frame of the box. The various metal parts of the shear box shall
be of a like, noncorrosive material. A typical shear box, assembled and
unassembled, is shown in Figure 2; a narrow-edged shear box as shown
is preferable to a wide-edged one. Generally, shear boxes for direct
shear tests shall have minimum inside dimensions of 3 by 3 in. The
maximum thickness of a 3- by 3-in. specimen shall be I/2 in. after con-
solidation. If the soil to be tested contains particles larger than the
No. 4 sieve, the test should be performed in a larger shear box or else
the shear strength should be determined by means of the triaxial test.
b.
-- Porous stones shall be smooth, coarse grade Alundum or
-
CarborundumB. finish-ground except for the surface in contact with the -
specimen, which shall be rough-finished by sandblasting or by using hand-
tools. Porous metal plates of similar porosity and texture may also be
used. It is very important that the permeability of porous stones not be
reduced by the collection of soil particles in the pores of the stones;
hence, frequent checking and cleaning (by flushing and boiling, or by
ultrasonic cleaner) are required to ensure the necessary permeability
(see page VIII-3).
C . Loading devices for applying the normal load and horizontal
shear fzrce% the s p e c i m e n . Any type of loading device may be used that
meets the following requirements:
(1) For applying normal load. The equipment for applying
the normal load shaS1 be capable of transmitting the load to the specimen
quickly, without impact, and maintaining the load constant for the duration
of the test. The equipment should be calibrated to ensure that the loads
indicated are those actually applied to the soil specimen.
(2) For applying shear force. The horizontal shear force
may be applied by either controlled-stress or controlled-strain
methods,
IX-2
EM fIiO-2-1906
Appendix IX
30 Nov 70
E M lllO-Z-l906
Appendix IX
30 Nov 70 -
though the controlled-strain method is preferred in that the ultimate, as

we11 as the maximum, stress can be determined. Controlled-stress equip-
ment should be capable of applying the horizontal force in increments to the
specimen in the same manner as that described above for the normal load.
Controlled-strain equipment should be capable of shearing the specimen at
a unifo1.m rate of strain and should permit adjustment of the rate of strain
over a relatively wide r a n g e . The controlled straining of the specimen is
us.ually done with a motor and gear box arrangement, and the shear force
is deternljned by a load-indicating device such as a proving ring or frame.
d.
- -D i a l i n d i c a t o r s f o r m e a s u r i n g (i) vertical deformation of the
spccirnen, having a range of 0.25 in. and an accuracy of 0.0001 in.; and
(2) horizontal displacement of the specimen, having a range of 0.5 in. and
an accliracy of 0.001 in.
e.
- -Ezquipment f o r p r e p a r i n g specimen including a specimen cut-
ter with sharp cutting edges. The cutter shall have inside dimensions the
same as those of the inside of the shear box if the specimen is to be trans- -
ferred from the cutter to the shear box. A metal or rigid plastic plate
having the same dimensions as the spec i men and having a short handle
attached at the center of one face is ,required for transferring the speci-
men from the cutter to the shear box. Knives, wire saws with O.OIO-in.-
diarneler wire, and other cutting equipment are also required.
-f. -Other items needed are:

(3) B a l a n c e s , s e n s i t i v e t o 0.1 and 0.01 g.
(,2) T i m i n g d e v i c e , a watch or clock with second hand.
(3) Centigradethermometer, range 0 to 50 C, a c c u r a t e to
0.d C.
(4) D i s t i l l e d o r d e m i n e r a l i z e d w a t e r .
(5) Glass plates.
(6) Apparatus necessary to determine water content.and spe-
cific gravity (see Appendixes I, W A T E R C O N T E N T - G E N E R A L , a n d I V ,
SFE~IFIC G R A V I T Y ) .
IX -4
C O R R E C T E D
-----v--e -C -
OP- Y-
EM iiiO-2-1906
Appendix IX
30 Nov 70
3. PRgPARATION OF SPECIMEN. A sample sufficient to provide a

minjmum of three identical specimens is required. Specimens shall be
prepared jn a humjd room to prevent evaporation of moisture. The speCj-
men is generally prepared by progressive trimming in front Of the speck-
men cutter. IIowevcr, satisfactory specimens of hard soils often may be
obtained by cutting and trimming wjthout using the specimen cutter.
Specimens of very soft, sensitive soils may be obtained more c.onviskntly
by pushing the cutter into the sample without preliminary trimming. Ex-
treme cart shall be taken jn preparjng undisturbed specimens of sensitive
soils to prevent disturbance of their natural structure. Preferably, spec-
imens of compacted soil should be trimmed from samples compacted in a
compaction mold, using a pressing or kneading action of a tamper having
an area less than one-sjxth the area of the sample. A less desirable pro-
cedurc is to compact the soil to the desired density and water content di-
rectly in the shear box, in a single layer using a similar kneading action.
The procedure usually used in preparing specimens by progressive trim-
ming follows :
z!* Cut a sample of soil approximately i-f/4 in. high and 4- i/2 in.
in diameter from the sample to be tested.
!?a Place the sample of soil on a;glass plate and center the speci-
men cutter on top of the sample. Push the cutter vertically into the sam-
ple not more than I/4 in. and carefully trjm the soil from the edge of the
Cutter. Repeat the operation until the specimen protrudes above the top of
the cutter. care should bc taken to insure that no voids are formed bC_
tween the cutter and specimen.
C.
_ Remove the portion of the specimen protruding above the cut-
ter, using a wire saw for soft specimens, and a straightedge, knives, or
other convenient tools for harder specimens. Trim the specimen flush
with the top of the cutter. If a pebble or other prot,rusion is encountered
on the surface, .rcmove it and fill the void with soil.
!& Place a previously weighed glass plate on the surface of the
IX-5
E M iiiO-2-1906
Appendix IX
30 Nov 70
specimen. M,any soils will adhere to glass; consequently, it is advisable

to use waxed paper or similar material between the specimen and glass
plate. 1nvLi.t the specimen, trim the bottom as described in step 2, and on
this surface place another weighed glass plate.
-e. From the soil trimmings obtain 200 g of lnaterial for water
content and s,pecific gravity determinations (see Appendixes I, WATER
C O N T E N T - #GENERAL, and IV, SPECIFlC GRAVITY).
-f. Repeat the procedures outlined above t,o produce two additional
specimens.
4. P R O C E D U R E . a-. Preliminary. The procedure for setting up the
test specimen shall consist of thz following:
(1) Record all identifying information for the specimen, such
as project, boring number , .and other pertinent data, on the data sheet
(Plate IX-1 is a suggested form); note any difficulties encountered in
preparation of the specimen. Measure the inside area and height of the
-
shear bo? and record as the initial dimensions of the specimen on the data
sheet. Weigh and record the weight of specimen plus tare (specimen cut-
ter and glass plates).
(2) Assemble the shear box with the upper frame held in
alignment with the lower frame by means of the alignment pins or screws.
Place a previously saturated porous stone, rough side up, on the .baseplate
of the shear box in the bottom of the lower frame.
(3) Insert the specimen cutter, sharpened edge first, into the
upper frame of the shear box until it is wedged firmly and is parallel with
the top of the upper frame (the inside edge of the upper frame should be
beveled slight.ly to accept the cutter). Lay a piece of waxed paper, slightly
smaller than ,the specimen, on the surface of the specimen, and with a
smooth, continuous press of the transferring plate, force the specimen
from the cutter and into firm contact with the porous stone. While press-
ing the specimen from the cutter, care must be exercised to prevent
tilting or otherwise disturbing th& specimen. Withdraw the transferring
IX-6
EM 11iO-2-i906
Appendix IX
30 Nov 70
plate, remove the specimen cutter, and peel the waxed paper from the
specimen. Place a previously saturated porous stone, rough side down,
on top of the specimen and lower the piston onto this porous stone.
(4) Place the shear box in position on the loading apparatus.
At this stage of the test the upper and lower frames are in contact. As-
semble the loading equipment, and mount the two dial indicators to be
used for measuring vertical and horizontal deformation. The dial indica-
tor measuring vertical deformation should be set so that it can measure
movement in either direction.
b .
Consolidation. The procedure for consolidating the specimen
shall consist of the following:
(i) Apply the desired normal load gently to the specimen.
The range of normal loads for the three specimens will depend on the
loadings expected in the field. The maximum normal load should be at
least equal to the maximum normal load expected in the field in order that
the shear strength data need not be extrapolated for use in design analysis.
Generally, normal loads less than about 3 tons per sq ft may be applied in
a single increment, whereas greater normal loads should be applied in
several increments to prevent the soil from squeezing out of the box, F.or
very soft soils it is usually necessary to apply even the relatively lighter
normal loads in increments.
(2) As soon as possible after applying the normal load, fill
the water reservoir with distilled or demineralized water to a point above
the top of the specimen. Maintain this water level during the consolidation
and subsequent shear phases so that the specimen is at all times effec-
tively submerged.
(3) Allow the specimen to drain and consolidate under the
desired normal load or increments thereof prior to shearing. During the
consolidation process, record on the data sheet (see Plate VIII-2 of Ap-
pendix VIII, CONSOLIDATION TEST) the vertical dial readings after
various elapsed times. Readings at O.i, 0.2, 0.5, 4, 2, 4, 8, i5, and 30 min,
IX-7
EMIllO-2-1906
Appendix IX
30 Nov 70
and 1, 2, 4, 8, .and 24 hr for each increment of normal load are usually

satisfactory. During the course of the test, plot the dial readings versus
elapsed tin-_; on a semilogarithmic plot (see Fig. 5 of Appendix VIII, CON-
SOLIDATION TEST). Allow each load increment to remain on the speci-
men until it is determined that primary consolidation is complete (see
Appendix VIII, CONSOLIDATION TEST).
C.
- siear T e s t . The procedure for shearing the specimen after
consolidation shall consist of the following:
(1) Raise the upper frame of the shear box about l/l6 in. by
turning the el.evating screws. The amount of clearance between the upper
and lower fra.mes should be sufficient to prevent the Wo frames from
coming in contact during the shear test, yet not ::ermit the soil to extrude
between the frames. Lock the upper frame to the loading piston by means
of the horizontal locking screws. In raising the upper frame, the applied
load on the specimen is increased by an amount equal to the weight of the
-
upper frame. Adjust vertical load by reducing applied load by this
amount. Retract.the elevating screws.
(2) Remove the alignment pins.

(3) Shear the specimen at a relatively slow rate so that a
fully drained condition (no excess pore pressures) exists at failure. The
following equation shall be used as a guide in determining the minimum
time required from start of test to shear failure:
tf = 5ot50
where
= total. elapsed time to failure in minutes
tf
= time in minutes required for the specimen to achieve 50 percent
t50
IX-8
EM iiiO-2-1906
Appendix IX
30 Nov 70 -
consolidation sequence of the test may be used for determining if 95 per-

cent consolidation has been achieved. When 50 to 70 percent of the esti-
mated failure load has been applied to the specimen (depending on the
shape of the stress-deformation curve), reduce the size of the increments
to one-half the initial size. As failure is approached, use a series of in-
crements equal to one-fourth of the initial load increment to accurately
define the failure load.
(b) Controlled-strain test. The rate of strain may be
determined approximately by dividing the estimated horizontal deforma-
tion at maxi:mum shear stress by the computed time to failure, The
tfw
test shall be continued until the shear stress becomes essentially constant,
as shown in Figure 3b, or until a horizontal deformation of 0.5 in. has
been reached.
(5) The horizontal and vertical deformations and the applied
shear force shall be observed at convenient intervals. Plate IX-2 is a
IX-i0
EM lllO-2-i906
Appendix IX -
30 Nov 70
suggested form for recording the observations.

(6) Remove the specimen from the shear box, blot any
excess moisture, and trim away a minimum of i/i6 in. from all sides of
the specimen to form a rectangular block. Determine the water content
of the specimen (see Appendix I, WATER CONTENT - GENERAL). The
dry weight of the specimen should be computed using the water content
based on specimen trimmings taken at the start of the test.
a. From the recorded data compute and record on the data sheet
-
(see Plate IX-l) the initial and final water contents (see Appendix I,
WATER CONTENT - GENERAL). Compute also the void ratio before
test, after consolidation, and after test, and the initial and final degrees
of saturation, and dry density before test, using the following equations:t
V
- vs
Void ratio before test, e = oG -
o
S
Void ratio after consolidation, ec =

vc v-vs
S
vf - vs
Void ratio after test, ef = v
S
t Equations in brackets are based on units of measurement shown on the

following page in explanation of symbols.
IX- 12 .-
EMiilO-2-i906
Appendix IX
30 Nov 70
Units of measurement are those commonly used in computations for the

direct shear tests.
& Complete the data sheet, Plate IX-2.
C. The shear stress, T, in tons per square foot may be calcu-
lated from the following equation:
7= gx 0.465
where
F = applied shear force, lb
A = horizontal cross-sectional area of the specimen, sq cm (assumed
to be constant and equal to the initial area for routine testing)
6. PRESENTATION OF RESULTS. The results of the direct shear test
shall be shown on the report form, Plate IX-3. The shear stress and
vertical defyormation during shear shall be plotted versus the horizontal
deformation. As shown in Figure 3, the maximum shear stress, ~~~~~
is either the actual maximum or peak shear stress or, if the shear stress
increases continuously during the test, the shear stress at O.S-in. hori-
zontal deformation. When the shear stress decreases after reaching a
maximum value, the minimum shear stress attained before 0.5-in.8hori-
zontal deformation is considered to be the ultimate shear stress, T
ult
as shown in Figures 3c and 3d. The time to failure, tf, is the elapsed
time between the start of shear and the maximum shear stress. The
maximum shear stress shall be plotted against the normal stress, as
shown in Figure 4, and the strength envelope drawn to determine the
drained or effective angle of internal friction, o . In normally consoli-
dated soils (see Fig. 4a), the strength envelope is based on normal stresses
greater than any past or existing overburden pressure.
A brief.description of undisturbed specimens should be given on the
report form under Remarks. The description should include color, ap-
proximate consistency, and any unusual features (such as stratification,
IX- 14
E M iiio-2-19.06
Appendix IX,
30 Nov 70
STRENGTN E N V E L O P E
t-
IA
: CZO ,AV
:
ul
. / I A
z
*
a. NORMALLY CONSOLIDATED SOIL
s
ii
k
ul
az
<
t&l
2 0 ,
\/
S TRENG TM E N V E L O P E
NORMAL STRESS, U, T/SO F T
b. O V E R - C O N S O L I D A T E D O R C O M P A C T E D S O I L
Figure 4. Examples of strength envelope s
f i s s u r e s , r o o t s , shells, sand pockets, etc.). For compacted specimens,

give method of compaction used and the relation to maximum density and
optimum water content.
7. SIMPLIFIED PROCEDURE FOR DRAINED (S) DIRECT SHEAR
TESTS. In problems involving the long-term stability of fine-grained
soils, thje c o h e s i o n i n t e r c e p t , c , due to over -consolidation generally is
ignored. The drained direct shear test may be performed by shearing
duplicate specimens under the maximum normal stress expected in the
IX-15
E M iilO-2-1906
Appendix IX
30 Nov 70
field. The strength envelope for determining the effective angle of inter-
n a l f r i c t i o n , + , shall be drawn as a straight line from the origin to the
average value of the maximum shear stresses under this normal stress.
Regardless of the magnitude of the expected normal stress, the normal
stress used in this simplified procedure for testing compacted specimens
must be at least 3 tons per sq ft to avoid the effects of prestressing
caused by the compaction.
Although the testing of duplicate specimens under the same normal
stress is satisfactory for relatively pervious or normally consolidated
soils, it may not be conservative for preconsolidated clays. For the latter
soils, the duplicate specimens should be sheared under different normal
stresses, both of which are known to exceed the preconsolidation pressure,
as illustrated by points A and B in Figure 4b.
a. POSSIBLE ERRORS: Following are possible errors that would
cause inaccurate determinations of strength and stress-deformation
-
characteristics:
w cimen drsturbed Ghile t r i m m i n g . The trimming of speci-
a. ,Spe
-
mens must be done in the humid room with every care taken to minimize
disturbance of the natural soil structure or change in the natural water
content. As a rule, the effect of trimming disturbance is inversely pro-
portional to the size of the specimen.
b.
- Specimen disturbed while fitting into shear box. The speci-
men must exactly fit the inside of the shear box to insure complete
lateral confinement, yet a pretrimmed specimen must be inserted with-
out flexing or compressing. The specimen cutter must have the
identical inside dimensions as those of the shear box.
-
C. Galvanic action in shear box. To prevent any change in the
strength or stress-deformation characteristics due to galvanic currents
in tests of long duration, all metal parts of the shear box should be of
the same noncorrosiye material.
d. l?ermeability of porous stones too low. Unless the porous
IX-16
EM ii10-2-1906
Appendix IX
30 Nov 70
stones are frequently cleaned, they may become clogged by soil particles
and full drainage of the specimen inhibited.
e. Slippage between porous stone and specimen. When testing
undisturbed firm or stiff clays, particularly under low normal loads, it
may not be possible to transfer the required shear force to the specimen
by means of the standard porous stone. In such a case, slippage of the
porous stone will result and a portion of the shear force will be applied to
the specimen by the rear edge of the upper frame. The slippage may be
marked by tilting -of the upper frame and the development of an inclined
shear plane through the upper rear corner rather than one through the
midheight of the specimen. The use of dentated porous stones or of wire
cloth or abrasive grit between the stone and the specimen may be necessary
to effect the transfer of shear stress.
f. Rate of strain too fast. The time to failure in the drained (S)
direct shear test must be long enough to achieve essentially complete
dissipation of excess pore pressure at failure. The criterion given in
paragraph 45(3) should be considered as no more than an approximate guide
to the minimum time to failure; twice this time may be necessary for some
soils. In general, it is safer to shear too slowly.
IX-i7
EM IilO-2-1906
30 Nov 70
APPENDIX IXA:
DRAINED (S) REPEATED DIRECT SHEAR TEST
1. INTRODUCTION. The repeated direct shear test is used to measure

the residual shear strength of a soil (primarily clay shale material) under
drained conditions. In this test, a square soil specimen acted on by a
normal stress is repeatedly sheared by reversal of the direction of shear
until a minimum, i.e. residual, shear stress is determined. The concept
of the test is that a specimen of material is acted upon by normal and
shear stresses until, after large shear deformation, disaggregation and
increase in parallel orientation of soil particles in the direction of shear-
ing occur and a surface or a thin zone of remolded material is formed.
Thus a minimum drained shearing resistance is offered. The shear sur-
face or zone can be formed by (a) shearing an intact specimen, (b) cutting
a plane surface through an intact specimen with a fine wire, or (c) shear-
ing a completely remolded specimen. Practical laboratory considerations
favor the use of a precut shear plane as it may not be possible to induce a
plane surface by shearing an intact stiff-to-hard specimen. Any irregu-
larities of the shear surface would introduce an added resistance that
would not be a measure of the shear strength of the material.
2. APPARATUS. The apparatus may be similar to that described in
Appendix IX, DRAINED DIRECT SHEAR TEST. A controlled strain de-
vice in which the direction of shear is capable of being reversed should be
used to apply the horizontal shear force to the specimen.
3. SPECIMEN PREPARATION. -a. Specimens should be prepared
using procedures similar to those described in Appendix IX, DRAINED
DIRECT SHEAR TEST; special saws may be required for trimming stiff-
to-hard materials. Special care mustbe exercised to ensure that the
specimen is not subjected to air-drying during or after trimming
ope rations.
EMiiiO-2-i906
Appendix IXA
30 Nov 70 -
-b. The specimen should consist of two pieces of intact material

trimmed to fill the inside of the shear box or confining ring. The two
pieces should be of approximately equal height and have a total height
not in excess of 1 in. (preferably, the total height should be 0.5 in., but
this is often not practical for stiff, fissured materials). The top and bot-
tom surfaces of each piece should be plane and parallel. A close fit of
each piece to the inside of the shear box is necessary. Stiff-to-hard ma-
terials may be cut to shape with a bandsaw or, in the case of very hard
materials, with a diamond wheel.
-
C. From the soil trimings obtain about 200 g of material for
water content and specific gravity determinations (see Appendix I, WATER
CONTENT - GENERAL, and Appendix IV, SPECIFIC GRAVITY).
4. P R O C E D U R E , 5. Preliminary. The procedure for setting up the
test specimen shall consist of the following:
(i) Record all identifying information for the specimen, such
as project title, boring number, sample number, and other pertinent data, -
on the data sheet (Plate IX-l is a suggested form); note any difficulties
encountered in preparation of the specimen. Determine the dimensions of
the specimen and record as the initial dimensions of the specimen on the
data sheet. Weigh and record the weight of specimen (plus tare, if used).
(2) The lower half of the specimen should be firmly seated
against a saturated porous plate in the lower half of the shear box. A
O.OlO- to 0.020-in. projection of the lower half of the specimen above the
lip of the box is desirable; certainly the top of this half of the specimen
should not be initially below the lip. Then the upper half of the specimen
should be placed in the shear box, the upper porous plate (saturated)
added, and the remainder of the shear apparatus assembled.
A specimen of softer material may be precut inside the
shear box. In this case, an intact specimen is firmly seated between sat-
urated porous plates in the apparatus ; then, a plane should be cut with a
small-diameter (0.008- to 0.044-in. diameter) steel wire through the
IXA- 2
EM 1110-2-1906
Appendix IXA
30 Nov 70
specimen at the separation between the upper and lower halves of the box.
After cutting, the two halves of the specimen should be separated, and the
cut surfaces inspected for planeness. Any irregularities should be re-
moved with a straightedge.
(3) Place the shear box in position on the loading apparatus.
At this time the upper and lower frames are in contact. Assemble the
loading equipment and mount the two dial indicators or other deformation
measuring apparatus to be used for measuring vertical and horizontal de-
formation. Both indicators should be set so that they can measure defor-
mation in either direction.
b
-* Consolidation, The procedure for consolidating the specimen
shall consist of the following:
(I) Apply the normal stress to the specimen without impact.
A single value of normal stress may be used. A standard value of about
6 tons per sq ft is recommended as being high enough to prevent the
swelling of most clay shale material yet low enough to minimize the prob-
lem of soil extruding from between the two halves of the box during shear.
In addition, tests under higher normal stresses may be used to determine
whether the strength envelope is a straight line.
(2) As soon as possible after applying the normal stress, fill
the water reservoir with distilled or demineralized water to a point above
the top of the specimen. Maintain this water level during the consolida-
tion (or swell) and subsequent shear phases so that the specimen is at all
times effectively submerged.
(3) The specimen should be allowed to consolidate or swell to
an essentially equilibrium condition under the normal stress; a minimum
period of 16 hr should be allowed before shear.
-
C. Repeated Shear Test. The procedure for shearing the speci-
men after consolidation shall consist of the following:
(1) A gap should be formed between the two halves of the box
to ensure that normal and shear stresses are borne only by soil. This
IXA- 3
E M 1110-2-1906
Appendix IXA
30 Nov 70
gap should be kept between 0.015 to 0.025 in. to minimize extrusion of

remolded soil from the shearing surface. Periodically during the test,
the gap should be checked by inserting thickness gages, and adjusted as
needed.
(2) Remove the alignment pins.
(3) Shearing displacement should be initiated at a controlled
rate not in excess of 0.5 in. per day (about 0.00035 in. per min). Shear
movement under constant normal stress should be continued with a re-
versal of direction after about 0.25-in:displacement to each side of the
starting position until a minimum shearing resistance is attained. A
semilogarithmic plot of shear stress (arithmetic scale) versus cumulative
shear displacement (logarithmic scale) should be maintained during the
test to show when a minimum value has been reached; only the shear
stress measured at the midpoint of each shearing stroke (that is, when the
two halves of the shear box are aligned vertically) should be plotted. -,
(4) Observations of vertical and horizontal deformations and
the applied shear force sufficient to define the stress deformation curve
for each shearing stroke shall be made.
(5) If, after the standard test is complete, the effects of
increased normal stress and decreased displacement rate are to be
studied, this information should also be obtained according to uniform
procedures.
First, the normal stress should be doubled while the two
halves of the shear box are vertically aligned, and the specimen should be
permitted to come to equilibrium (minimum of 16 hr). Shear displacement
should be initi.ated at a rate of about 0.5 in. per day and continued until a
minimum shearing resistance is reached.
Second, the effect of decreased rate of displacement
should be determined as follows. After the minimum shearing resistance
is reached under the high normal stress (approximately 12 tons per sq ft)
and at a displacement rate of about 0.5 in. per day, the rate of displacement
IXA- 4
EM iilO-2-1904
Appendix IXA
30 Nov 70
should be reduced to a tenth of the standard rate (that is, to about 0.05 in.
per day) without any change in the normal stress. The rate of displace-
ment should be reduced soon after the upper half of the shear box has
passed through the initial, vertically aligned position. Shearing with re-
peated reversals of direction as described above should be continued
until a minimum shearing resistance is reached.
(6). Remove the specimen from the shear box, blot any excess
moisture, and determine the water content of the speciment (see Appendix
I, WATER CONTENT - GENERAL). The dry weight of the specimen
should be computed using the water content based on specimen trimmings
taken during specimen preparation.
5. COMPUTATIONS. See Appendix IX, DRAINED (S) DIRECT SHEAR
TEST.
6. PRESENTATION OF RESULTS. Report forms have not been stand-
ardized; the method of presentation of results will be determined by the
project (design) engineer requesting the tests.
7. POSSIBLE ERRORS. The following are possible errors that may
cause inaccurate determinations of strength and stress-deformation
characteristics:
a. Air -Drying of Specimen During Preparation. The trimming
-
of specimens should be done in a humid room with every precaution taken
to prevent change in natural water content. Air-drying may cause the
specimen to slake readily when inundated with water, and thus change the
strength or stress-deformation characteristics.
b. Top and Bottom Surfaces of Each Half of a Specimen Not
-
Plane and Parallel. Irregular surfaces may introduce a geometric com-
ponent to the measured shearing resistance.
t If considerable remolded material exists in the shear zone, it should

be removed and its water content should be determined in addition to
that of the remainder of the specimen.
IXA-5
EM iilO-2-1906
Appendix IXA
30 Nov 70
C. Too Large a Gap. Maintaining too large a gap between the

upper and lower frames of the shear box may result in excessive extru-
sion of the specimen.
d. Absence of Gap. A gap must be maintained throughout the
-
test to prevent the normal load from being borne by the lower frame of
the shear box.
e. Inaccurate Measurement of Shear Stress. Because of the
very small shear resistances offered by some clay shale materials, meas-
urement of shear stress must be very precise.
L* Permeability of Porous Stones Too Low. Unless the porous
stones are frequently cleaned, they may become clogged by soil particles
and ingress or egress of water to or from the specimen may be inhibited.
g* Galvanic Action in Shear Box. To prevent any change in the
strength or stress-deformation characteristics due to galvanic currents
in tests of long duration, all metal parts of the shear box should be con-
-
structed of the same noncorrosive material.
h. Stopping Test Too Soon. The test must be carried to a cumu-
-
lative shear deformation sufficient to establish that the minimum shear
resistance offered by a specimen under a given normal stress has indeed
been determined. A semilogarithmic plot of shear stress (arithmetic
scale) versus cumulative shear displacement (logarithmic scale) is es-
sential in making this determination.
IXA-6 -
E M i410-2-i906
30 Nov 70
APPENDlX X:
TRIAXIAL COMPRESSION TESTS
1. PRINCIPLES OF THE TRIAXIAL COMPRESSION TEST. The triaxial

compression test is used to measure the shear strength of a soil under
controlled drainage conditions. In the basic triaxial test, a cylindrical
specimen of soil encased in a rubber membrane is placed in a triaxial
compression chamber, subjected to a confining fluid pressure, and then
loaded axially to failure. Connections at the ends of the specimen permit
controlled drainage of pore water from the specimen. The procedures
presented herein apply only to the
P
basic test conducted with limited P
- = DEVIATOR STRESS
t A
drainage conditions, and do not in-
clude special types or variants of this
, test. In general, a minimum of three
specimens, each under a different
confining pressure, are tested to es- -
tablish the relation between shear
-
SOIL
strength and normal stress. The test SPECIMEN -
is called triaxial because the three
principal stresses are known and
controlled. Prior to shear, the three
principal stresses are equal to the
chamber fluid pressure. During
shear, the major principal stress,
uls is equal to the applied axial

stress (P/A) plus the chamber
pressure, ~3 ( s e e F i g . 1). The ap-
plied axial stress, y1 - u3, is F i g u r e i. Diagram showing
stresses during triaxial com-
termed the d e v i a t o r s t r e s s . T h e pression test
EM 1140-2-1906
Appendix X
--
30 Nov 70
intermediate principal stress, u2, and the minor principal stress, u3*
are identical in the test, and are equal to the confining or chamber pres-
sure hereafter referred to as u3.
A soil mass may be considered as a compressible skeleton of solid
particles. In saturated soils the void spaces are completely filled with
water; in partially saturated soils the void spaces are filled with both
water and air. Shear stresses are carried only by the skeleton of solid
particles, whereas the normal stress on any plane is carried by both the
solid particles and the pore water. In a triaxial test, the shear strength
is determined in terms of the total stress (intergranular stress plus pore
water pressure), unless (a) complete drainage is provided during the test
so that the pore pressure is equal to zero at failure, or (o) m e a s u r e m e n t s
of pore pressure are made during the test. When the pore pressure at
failure is known, the shear strength can be computed in terms of the stress
carried by the soil particles (termed effective or intergranular stress).
-
In recent years, significant advances have been made in the techniques of
measuring pore pressures in the triaxial test and in the interpretation of
the data obtained; however, difficulties still exist in this respect. Pore
pressure measurements during shear are seldom required in routine
investigations, as the basic triaxial tests are sufficient to furnish shear
strengths for the limiting conditions of drainage. Procedures for measuring
pore pressures in the triaxial test during shear are discussed elsewheret
and are beyond the scope of this manual.
2. TYPES OF TESTS. The three types of basic triaxial compression
tests are unconsolidated-undrained, consolidated-undrained, and
consolidated-drained, subsequently referred to as the Q, R, and S tests,
respectively. As these names imply, they are derived from the drainage
conditions allowed to prevail during the test. The type of test is selected
t A. W. L3ishop and D. J. Kenkel, The Measurement of Soil Properties in

the Triaxial Test, 2nd ed. (London, Edward Arnold L,td., 1962).
X-2
E M 1410-2-1906
Appendix X
30 Nov 70
to closely simulate, or to bracket, the conditions anticipated in the field.

In the basic tests, the initial principal stresses are equal; that is, no
attempt is made to duplicate stress systems in the field in which the prin-
cipal stresses are not equal.
a.
- Q Test. In the Q test the water content of the test specimen
is not permitted to change during the application of the confining pressure
or during the loading of the specimen to failure by increasing the deviator
stress. The Q test is usually applicable only to soils which are not free-
draining, that is, to soils having a permeability less than 10 x i0 - c m
per sec.
b.
- R Test. In the R test, complete consolidation of the test
specimen is permitted under the confining pressure. Then, with the hater
content ~held constant, the specimen is loaded to failure by increasing the
deviator stress. Specimens must as a general rule be completely satu-
rated before application of the deviator stress.
c.
- S Test. In the S test, complete consolidation of the test
specimen is permitted under the confining pressure and during the loading
of the specimen to failure by increasing the deviator stress. Consequently,
no excess pore pressures exist at the time of failure.
3. A P P A R A T U S . a-. Loading Devices. Various devices may be used to
apply axial load to the specimen. These devices can be classified as either
apparatus in which axial loads are measured outside the triaxial chamber
or apparatus in which axial loads are measured inside the triaxial chamber
by using a proving ring or frame, an electrical transducer, or a pressure
capsule. Any- equipment used should be calibrated to permit determination
of loads actually applied to the soil specimen.
Loading devices can be further grouped under controlled-strain
or controlled-stress types. In controlled-strain tests, the specimen is
strained axially at a predetermined rate; in controlled-stress tests, pre-
determined increments of load are applied to the specimen at fixed inter -
vals of time. Controlled-strain loading devices, such as commercial
X-3
EM iiIO-2-i906
Appendix X
30 Nov 70
testing machines, are preferred for short-duration tests using piston-type

test apparatus. If available, an automatic stress-strain recorder may be
used to measure and record applied axial loads and strains.
!L* Triaxial Compression Chamber. The triaxial compression

chamber consists primarily of a headplate and a baseplate separated by a
transparent plastic cylinder.* A drawing of a typical triaxial compression
chamber for i.4-in. -diameter specimens is shown in Figure 2. Chamber
DETAILS OF DftAlNAGE
CONNECTIONS 1HROLGH BASE
Figure 2. Details of typical triaxial compression chamber
* Adequate safety precautions should be .taken, or the transparent plastic

cylinder should be replaced by a metal cylinder, if chamber pressures
in excess of 100 psi are used.
X-4
EM WO-2:4906
Appendix X
30 Nov 70
dimensions and type will vary dependiiig on the size of specimen tested
and on pressure and load requirements. The baseplate has one inlet
through which the pressure liquid is supplied to the chamber and two
inlets leading to the specimen base and cap to permit saturation and
drainage of the specimen when required, The headplate has a vent valve
so that air can be forced out of the chamber as it is filled with the pres-
sure fluid. The cylinder is held tightly against rubber gaskets by bolts or
tie rods connecting the headplate and baseplate.
In piston-type test apparatus in which the axial, loads are mea-
sured outside the triaxial compression chamber, piston friction can have
a significant effect on the indicated applied load, and measures should be
taken to reduce friction to tolerable limits. Pistons generally should con-
sist of ground and polished case-hardened steel rods with diameters be-
tween I/4 and I/2 in. for testing 1.4-in.-diameter specimens; heavier
pistons are required for larger specimens. The following measures have
been found to reduce piston friction to tolerable amounts.
(i) The use of linear ball bushings as shown in Figure 2. The
unique design of these bushings
permits unlimited axial move-
ment of the piston with a mini-
mum of friction. Leakage around
the piston is reduced by means of
Q-rings, Quad-rings, flexible
diaphragms, or other devices. A
seal incorporating Q-rings is
shown in Figure 2. The benefi -
cial effects of using linear ball
bushings in comparison with
steel bushings are demonstrated
Figure 3. Effect of lateral force on
by the data shown in Figure 3. piston friction in triaxial compres-
The amount of lateral force sion apparatus
X-5
.EM 1110-2-1906
,Appendix X
30 Nov 70
transmitted to the piston, if the specimen cap tends to tilt during a test,
cannot be determined; however, the data shown in Figure 3 indicate that
the resulting piston friction would be negligible even for relatively large
lateral forces.
(2) Rotation of the piston within the bushings during the appli-
cation of the deviator stress. (Commercial devices are available to rotate
the piston during the test.) This method is very effective in reducing fric-
tion; however, a more complex design of the specimen cap is necessary,
and unless the piston is rotated continuously, appreciable friction would
still develop during longtime tests. When linear ball bushings are used,
the piston should never be rotated except under special conditions desig-
nated by the manufacturer.
Although these measures will reduce piston friction to
negligible amounts during the course of the test, it is always preferable
to measure the actual piston friction before the start of the test. This
can readily be done by starting the axial load application with the bottom -
of the piston raised slightly above the top of the specimen cap. Thus any
starting friction or residual friction, as indicated by the load necessary
to move the piston down into contact with the cap, can be subtracted from
the measured load.
c. Specimen Caps and Bases.
- Specimen caps and bases should
be constructed of a lightweight noncorrosive material and should be of
the same diameter as the test specimen in order to avoid entrapment of
air at the contact faces. Solid caps and bases should be used for the Q
test to prevent drainage of the specimens. Caps and bases with porous
metal or porous stone inserts and drainage connections, as shown in
Figure 4, should be used for the R and S tests. The porous inserts
should be more pervious than the soil being tested to permit effective
drainage. For routine testing, stones of medium porosity are satisfac-
tory. The specimen cap should be designed to permit slight tilting with
the piston in contact position, as shown in Figure 4.
X-6
E M IiIO-2-1906
Appendix X
30 Nov 70
completely impervious, the use of special membranes or chalnber fluidsI_

may sometimes be necessary, such as during periods of undrained shear
that exceed a few hours. The membrane is sealed against the cap and
base by rubber O-rings or rubber bands. Leakage around the ends of the
membrane, where it is sealed against the cap and the base, as well as
through fittings, valves, etc., can develop unless close attention is given
to details in the manufacture and use of the apparatus.t
e.
- Equipment for Preparing Specimens. (i) C o h e s i v e s o i l s . A
specimen trimming frame is recommended for preparing specimens of
most cohesive soils. The specimen is held in a vertical position between
two circular plates containing pins which press into the ends of the speci-
men to prevent movement during trimming. The edges of the trimming
frame act as vertical guides for the cutting equipment and control the final
diameter of the specimen. Details of a typical trimming frame for 1.4-in.-
diameter specimens are shown in Figure 5. Wire saws and knives of
various sizes and types are used with the trimmer (see Fig. 7, p. i2).
Split or solid cylinders with a beveled cutting edge can also be used to
trim specimens. The use of a motorized soil lathe may be advantageous
in reducing the time required for preparing specimens of certain types of
soils. A miter box or cradle (see Fig. 8, p. 13) is required to trim the
specimen to a fixed length and to insure that the ends of the specimen are
parallel with each other and perpendicular to the axis of the specimen.
(2) Cohesionless soils. A forming jacket consisting of a split
mold which incloses a rubber membrane is required for cohesionless soils.
The inside diameter of the mold minus the double thickness of the mem-
brane is equal to the diameter of the specimen required. A funnel or spe-
cial spoon (see Fig. 5 of Appendix VII, PERMEABILITY TESTS) for
placing the material inside the jacket and a tamping hammer or vibratory
t S. J. Poulos, Report on Control of Leakage in the Triaxial Test<Soil

M e c h a n i c s S e r i e s N o . 71, Harvard [Jniversity (Cambridge, M a s s . ,
March 1964).
X-8
EM 1110-2-1906
Appendix X
30 Nov 70
ALTERNATE LOWFR
S?FCl~
f%um
Figure 5. Details of a trimming frame for preparing

f .4-in.-diameter specimens
X- 9
E M iiiO-2-4906
Appendix X
30 Nov 70
regulator should be capable of controlling pressures to within *i/2 percent,

though more precise methods of controlling and maintaining chamber pres -
sures are required for tests of long duration.
(3) Measuring equipment, such as dial indicators and calipers.
Precise instruments should be used for measuring the dimensions of a
specimen with the desired accuracy.
(4) Deaired water, distilled or demineralized.
(5) Vacuum and air pressure supply.
(6) Bourdon gages of various sizes and capacities.
(7) A timing device, either a watch or clock with second hand.
( 8 ) B a l a n c e s , sensitive to O.Oi g and,to 0.i g.
(9) Apparatus, necessary to determine water content and
specific gravity (see Appendices I, W A T E R C O N T E N T - G E N E R A L , a n d
IV, SPECIFIC GRAVITY).
4. PREPARATION OF SPECIMENS. Specimens shall have an initial
height of not less than 2. i times the initial diameter, though the minimum
initial height of a specimen must be 2.25 times the diameter if the soil
contains particles retained on the No. 4 sieve. The maximum particle
size permitted in any specimen shall be no greater than one-sixth of the
specimen diameter. Triaxial specimens i.4, 2.8, 4, 6, i2, and 45 in. in
diameter are most commonly used.
a. Cohesive Soils Containing Negligible Amounts of Gravel.
Specimens i.4 in. in diameter are generally satisfactory for testing cohe-
sive soils containing a negligible amount of gravel, while specimens of
larger diameter may be advisable for undisturbed soils having marked
stratification, fissures, or other discontinuities. Depending on the type of
sample, specimens shall be prepared by either of the following procedures:
(I) T r i m m i n g s p e c i m e n s o f c o h e s i v e s o i l . A sample that is
uniform in character and sufficient in amount to provide a minimum of t!lree
specimens is required. For undisturbed soils, samples about 5 in. in diam-
eter are preferred for triaxial tests using 4.4-in.-diameter specimens.
x-ii
EM iiiO-2-1906 -..
Appendix X
30 Nov 70
Specimens shall be prepared in a humid room and tested as soon as possible

thereafter to prevent evaporation of moisture. Extreme care shall be taken
in preparing the specimens to preclude the least possible disturbance to the
structure of the soil. The specimens shall be prepared as follows:
(a) Cut a section of suitable length from the sample. As a
rule, the specimens should be cut with the long axes parallel to the long axis
of the sample; any influence of stratification is commonly disregarded.
However, comparative tests can be made, if necessary, to determine the
effects of stratification. When a 5-in.-diameter
undisturbed sample is to be used for 1.4-in.-
diameter specimens, cut the sample axially into
quadrants using a wire saw or other convenient
cutting tool. Use three of the quadrants for
specimens; seal the fourth quadrant in wax and
preserve it for a possible check test.
( b ) C a r e f u l l y t r i m e a c h speci-
men to the required diameter, using a trim-
ming frame or similar equipment (see Fig. 7).
Use one side of the trimming frame for
Figure 7. Prepared triaxial s p e c i m e n , t r i m m i n g f r a m e ,

and cutting tools
X-12
E M iilO-2-1906
Appendix X
30 Nov 70
preliminary cutting, and the other side for final trimming. A specimen
after trimming is also shown in Figure 7. Ordinarily, the specimen is
trimmed by pressing the wire saw or trimming knife against the edges
of the frame and cutting from top to bottom. In trimming stiff or varved
clays, move the wire saw from the top and bottom toward the middle of
the specimen to prevent breaking off pieces at the ends. Remove any
small shells or pebbles
encountered during the
trimming operations.
Carefully fill voids on
t h e s u r f a c e of the s p e c i -
men with remolded soil
obtained from the trim-
mings. Cut specimen to
the required length
(usually 3 to 3-1/2 in.
for 1.4-in. - d i a m e t e r
specimens and 6 to 7 in.
for 2.8-in. - d i a m e t e r
specimens) using a mi-
ter box, as shown in
Figure 8
(c) F r o m F i g u r e 8. Squaring ends of specimen with c
miter box
the soil trimmings, ob-
tain 200 g of material for specific gravity and water content determi-
nations (see Appendixes I, W A T E R C O N T E N T - GENERAL, and IV,
SPECIFIC GRAVITY).
(d) Weigh the specimen to an accuracy of iO.01 g for i.4-
in. -diameter specimens and *0.1 g for 2.8-in.-diameter s p e c i m e n s .
(e) Measure the height and diameter of the specimen to an
accuracy of *O.Oi in. Specimen dimensions based on measurements of the
X-13
E M 1.liO-2-1906
Appendix X
30 Nov 70
stored in an airtight container for at least i6 hr. The desired density may
be produced by either (f) kneading or tamping each layer until the accumu-
lative weight of soil placed in the mold is compacted to a knw,n volume 1~
(2) adjusting the number of layers, the number of tamps per layer, and
the force per tamp. For the latter method of control, special constant-
force tampers (such as the Harvard miniature compactor for 1.4-in. -
d i a m e t e r specimenst or similar compactors for 2.8-in. - d i a m e t e r a n d
l a r g e r specimentst) a r e n e c e s s a r y . After each specimen compacted to
finished dimensions has been removed from the m o l d , proceed in accor-
dance with steps (c) through (e) of paragraph 4$(i).
k.Cohesionless Soils Containing Negligible--~Amounts of Gravel.
Soils which possess little or no cohesion are difficult if not impossible to
trim into a specimen. If undisturbed samples of such materials are avail-
able in sampling tubes, satisfactory specjmens can usually be obtained by
freezing the sample to permit cutting out s u i t a b l e s p e c i m e n s . S a m p l e s
should be drained before freezing. The frozen specimens are placed in
the triaxial chamber, allowed to thaw after application of the chamber
pressure, and then tested as desired. Some slight disturbance probably
occurs as a result of the freezing, but the natural stratification and
structure of the material are retained. In most cases, however, it is
permissible to test cohesionless soils in the remolded state by forming
the specimen at the desired density or at a series of densities which will
permit interpolation to the desired density. Specimens prepared in this
t A. Casagrande, J. M. Corso, and S. D. Wilson, Report to Waterways

E x p e r i m e n t S t a t i o n on the 1949-1950 Program of Investigation of
Effect of Long-Time Loading on the Strength of Clays and Shales at
Constant Water Content, Harvard University (Cambridge, Mass., July
1950).
$ A.,Casagrande and R. C. Hirschfeld, S e c o n d P r o g r e s s R e p o r t on I n v e s -
tigation of Stress-Deformation and Strength haracteristics of Com-
~~~~~~~~~~~~~~.~~~~~~6~~~ies No. 65, Harvard U n i v e r s i t y
/-
X-15
EM IliO-2-i906
Appendix X --
30 Nov 70
manner should generally be 2.8 in. in diameter or larger, depending on

the maximum particle size. The procedure for forming the test specimen
shall consist of the following steps:
(i) Oven-dry and weigh an amount of material sufficient to
provide somewhat more than the desired volume of specimen.
(2) Place the forming jacket, with the membrane inside, over
the specimen base of the triaxial compression device.
(3) Evacuate the air between the membrane and the inside
face of the forming jacket.
(4) After mixing the dried material to avoid segregation,
place the specimen, by means of a funnel or the special spoon, inside the
forming jacket in equal layers. For 2.8-in.-diameter specimens, 40 l a y e r s
of equal thickness are adequate. Starting with the bottom layer, compact
each layer by blows with a tamping hammer, increasing the number of
blows per layer linearly with the height the layer above the bottom
1ayer.t The total number of blows required for a specimen of a given
material will depend on the density desired. Considerable experience is
usually required to establish the proper procedure for compacting a mate-
rial to a desired uniform density by this method. A specimen formed
properly in the above-specified manner, when confined and axially loaded,
will deform symmetrically with respect to its midheight, indicating that a
uniform density has been obtained along the height of the specimen.
(5) As an alternate procedure, the entire specimen may be
placed in a loose condition by means of a funnel or special spoon. The
desired density may then be achieved by vibrating the specimen in the
forming jacket to obtain a specimen of predetermined height and corre-
sponding density. A specimen formed properly in this manner, when
- ._
t L i a n g - S h e n g Chen, An investigation of stress strain and strength
characteristics of cohesionless soils by triaxial compression tests,
Proceedings, Second International Conference on Soil Mechanics and
F o u n d a t i o n Engineering8 vol. V (Rotterdam, i948), pp. 35-4?.
X-i6
EM ii10-2-i906
Appendix X
30 Nov 70
confined and axially loaded, will deform symmetrically with respect to its
midheight.
(6) Subtract weight of unused material from original weight
of the sample to obtain weight of material in the specimen.
(7) After the forming jacket is filled to the desired height,
place the specimen cap on the top of the specimen, roll the ends of the,
membrane over the specimen cap and base, and fasten the ends with
rubber bands or O-rings. Apply a low vacuum to the s~~~:rrnen through
the base and remove the forming jacket.
(8) Measure height and diameter as Gpecified in paragraph
4&(i J(e).
c. Soils Containing Gravel. The size.of specimens containing appre-
-
ciable amounts of gravel is governed by the requirements of paragraph4. If the
material to be tested is in an undisturbed state, the specimens shall be prepared
according to the applicable requirements of paragraphs 4a and 4b, with the size
of specimen based on anestimate of the largest particle size. In testing com-
pacted soils, the largest particle size is usuallyknown, and the entire sample
should be tested, whenever possible, without removing any of the coarser parti-
cles. However, it may be necessaryto remove the particles larger than a cer-
tain size to comply with the requirements for specimen size, though such practice
will result inlower measured values of the shear strength and should be avoided
if possible. Oversize particles should be removed and, if comprising more
than i0 percent byweight of the sample,be replaced by anequalpercentage by
weight of material retained on the No. 4 sieve and passing the maximum allow-
able sieve size. The percentage of material finer than the No. 4 sieve thus re-
mains constant (see paragraph 2h of Appendix VI, COMPACTION TESTS).
It will generally be necessary to prepare compacted samples of material con-
taining gravel inside a forming jacket placed on the specimen base. If the
material is cohesionless, it should be oven-dried and compacted in layers
inside the membrane and forming jacket using the procedure in paragraph 4h
as a guide. When specimens of very high density are required, the
x-17
EM iiiO-2-4906
Appendix X
30 Nov 70
semples should be compacted preferably by vibration to avoid rupturing the

membrane. The use of two membranes will provide additional insurance
against possible leakage during the test as a result of membrane rupture. If
the sample contains a significant amount of fine-grained material, the soil
usually must possess the proper water content before it can be compacted to
the desired density. Then, a
special split compaction mold is
used for forming the specimen.
The inside dimensions of the
mold are equal to the dimensions
of the triaxial specimen desired.
No membrane is used inside the
mold, as the membrane can be
readily placed over the compacted
specimen after it is removed
from the split mold. The speci-
men should be compacted to the
desired density in accordance
with paragraph 4&(2).
5. Q T E S T . 5. P r o c e d u r e .
The procedure for the Q test shall
consist of the following steps:
(i) Record all identi-
fying information for the sample
project number or name, boring
number, and other pertinent data,
on a data sheet (see Plate X-I).
(2) Place one of the
prepared specimens on the base.
Figure 9. Placing rubber membrane (3) Place a rubber
over a 2.8-in. -diameter specimen
using a membrane stretcher membrane (see Fig. 9) in the
X-18
E M IflO-2-i906
Appendix X
30 Nov 70
membrane stretcher, turn both ends of the membrane over the ends of the
stretcher, and apply a vacuum to the stretcher. Carefully lower the
stretcher and membrane over the specimen as shown in Figure 9. Place
the specimen cap on the top of the specimen and release the vacuum on the
membrane stretcher. Turn
the ends of the membrane
down around the base and up
around the specimen cap and
fasten the ends with O-rings
or rubber bands. With i.4-
in. -diameter specimens of
relatively insensitive soils,
it is easier to roll the mem-
brane over the specimen as
shown in Figure IO.
(4) Assemble
the triaxial chamber and
place it in position in the
loading device. Connect
the tube from the pres-
sure reservoir to the base Figure IO. Rolling rubber membrane
of the triaxial chamber. over a *.4-in.-diameter s p e c i m e n
With valve C (see Figure 11)

on the pressure reservoir closed and valves A and B open, increase the
pressure inside the reservoir and allow the pressure fluid to fill the
triaxial chamber. Allow a few drops of the pressure fluid to escape
through the vent valve (valve B) to insure complete filling of the chamber
with fluid. Close valve A and the vent valve.
(5) With valves A and C closed, adjust the pressure regu-
lator to preset the desired chamber pressure. The range of chamber
pressures for the three specimens will depend on the loadings expected
x-i9
E M i440-2-1906
Appendix X
30 Nov 70
Figure Ii. Schematic diagram of triaxial compression apparatus

for Q test
in the field. The maximum confining pressure should be at least equal to

the maximum normal load expected in the field in order that the shear
strength data need not be extrapolated for use in design analysis. Record
the chamber pressure on data sheets (Plates X-i and X-2). Now open
valve A and apply the preset pressure to the chamber. Application of the
chamber pressure will force the piston upward into contact with the ram of
the loading device. This upward,force is equal to the chamber pressure
acting on the cross-sectional area of the piston minus the weight of the
piston minus piston friction.
x-z0
E M 4iiO-2-4906
Appendix X
30 Nov 70
(6) Start the test with the piston approximately 0.i in. above
the specimen cap. This allows compensation for the effects of piston
friction, exclusive of that which may later develop as a result of lateral
f0rce.s. Set the load indicator to zero when the piston comes into contact
with the specimen cap. In this rnanner the upward thrust of the chamber
pressure on the piston is also eliminated from further consideration.
Contact of the piston with the specimen cap is indicated by a slight move-
ment of the load indicator. Set the strain indicator and r e c o r d on the
data sheet (Plate X-2) the initial dial reading at c ontact . Axially strain
the specimen at a rate of about 1 percent per minute (for plastic mate-
rials) and about 0.3 percent per minute (for brittle materials that achieve
maximum deviator stress at about 3 to 6 percent strain); at these rates
the elapsed time to reach maximum deviator stress would be about
15 to 20 min.
(7) Observe and record the resulting load at every 0.3 per-
cent strain for about the first 3 percent and, thereafter, at every 1 per-
cent, or for large strains, at every 2 percent strain; sufficient readings
should be taken to completely define the shape of the stress-strain curve so
frequent readings may be necessary as failure is approached. Continue the
test until an axial strain of i5 percent has been reached, as shown in
F i g u r e s i2a, i2b, and i2d; h o w e v e r , when
the deviator stress decreases
after attaining a maximum value and is continuing to decrease at i5 percent
strain (Fig. i2c), the test shall be continued to 20 percent strain.
(8) For brittle soils (i.e., those in which maximum deviator
stress is reached at 6 percent axial strain or less), tests should be per-
formed at rates of strain sufficient to produce times to failure as set
forth in paragraph 5a(6) above; however, when the maximum deviator
stress has been clearly defined, the rate may be increased such that the
remainder of the test is completed in the same length of time as that taken
to reach maximum deviator stress. However, for each group of tests
about 20 percent of the samples should be tested at the rates set forth
x-2j
EM iIIO-2-4906
Appendix X
30 Nov 70
0 5 IO I5 20
!z
0 5 IO I5 2c 0 5 IO I5 20
A X I A L . S T R A I N , -70 A X I A L 5TRAlN, -70
C.
- &
Figure i2. Examples of stress-strain curves
X-22
EM iIIO-2-4906
Appendix X
30 Nov 70
in paragraphs 5a(6) and 5a(7) above.

(9) Upon completion of axial loading, release the chamber
pressure by shutting off the air supply with the regulator and opening
valve C. Open valve B and draw the pressure fluid back into the pres-
sure reservoir by applying a low vacuum at valve C. Dismantle the tri-
axial chamber. Make a sketch of the specimen, showing the mode of
failure.
(40) Remove the membrane from the specimen For 4.4-
in.-diameter specimens, carefully blot any excess moisture from the
surface of the specimen and determine the water content of the whole
EM 11iO-2-i906
Appendix X
30 Nov 70
specimen (see Appendix I, WATER CONTENT - GENERAL). For 2,8-in.-

diameter or larger specimens, it is permissible to use a representative
portion of the specimen for the water content determination. It is essen-
tial that the final water content be determined accurately, and weighings
should be verified, preferably by a different technician.
(33) Repeat the test on the two remaining specimens at differ-
ent chamber pressures, though using the same rate of strain.
b. Computations. The computations shall consist of the following
steps:
(1) From the observed data, compute and record on the data
sheet (Plate X-i) the initial water content (see Appendix I, WATER CON-
TENT - GENERAL)s volume af solids, initialvoid ratio, initial degree of
saturation, and initial dry density, using the formulas given in Appendix II,
UNIT WEIGHTS, VOID RATIO, POROSITY, AND DEGREE OF SATURATION.
(2,) Compute and record on thL9 data sheet (Plate X-2) the axial
strain, the corrected area, and the deviator stress at each increment of
-
strain, using the following formulas:
AH
Axial strain, E = -
HO
AO
Corrected area of specimen, A sq cm = -
corr 2 - E
P
Deviator stress, tons per sq ft = - x 0.465
A corr
where AH = change in height of specimen during test, cm

Ho = initial height of specimen, cm, (Where a significant de-
crease in specimen volume occurs upon application of the
chamber pressure, as in partially saturated soils, the
height of the specimen after application of the chamber
pressure should be used rather than the initial height.)
A0 = initial area of specimen, sq cm
x-24
EM iiiO-2-i906
Appendix X
30 Nov 70
P = net applied axial load, lb (the actual load applied to speci-

men after correction for piston friction and for the upward
thrust of the fluid pressure in the triaxial chamber)
(3) Record the time to failure on the data sheet (Plate X-2).
(4) The rubber membrane increases the apparent strength of
the specimen. Investigationst with specimens I.5 in. in diameter and
membranes 0.008 in. thick, for instance, indicate the increase in deviator
stress to be 0.6 psi at 15 percent axial strain. The correction, cr, to be
made to the measured deviator stress for the effect of the rubber mem-
brane is computed as follows:
rDoM ~(1 - E)
err =
A0
w h e r e Do = initial diameter of specimen

M = compression modulus of the rubber membrane
fI = axial strain
A0 = initial cross-sectional area of the specimen
The c.ompression modulus may, without great error, be assumed to be

equal to that measured in extension. An apparatus for determining the
extension modulus of rubber is described in another w0rk.t In tests of
very soft soils the membrane effect may be significant, and in these tests
it is advisable to compute or estimate the correction and deduct it from
the maximum deviator stress. For most soils tested using membranes of
standard thickness, the correction is insignificant and can be ignored.
c. Presentation of Results. The results of the Q test shall be
recorded on the report form shown as Plate X-3. Enter pertinent infor-
mation regarding the condition of the specimen or method of preparing the
t Bishop and Henkel, op. cit., pp. i67-i7i.
X-25
E M iiiO-2-4906
Appendix X
30 Nov 70
specimen under Remarks. Plot the deviator stress versus the axial
strain for each of the specimens as shown in Figure i2. The peak or
m a x i m u m devia.tor stress represents failure of the specimen; when the
deviator stress increases continuously during the test, the deviator stress
at 15 percent axial strain shall be considered the maximum deviator
stress. When the deviator stress decreases after reaching a maximum,
the minimum deviator stress attained before 15 percent axial strain shall
be considered the ultimate deviator stress, as shown in Figures i2c and
i2d. Construct Mohr stress circles on an arithmetic plot with shear
stresses as ordinates and normal stresses as abscissas. As shown in
F i g u r e 13, the applied principal stresses, uI and cr3, are plotted on the
abscissa, and the Mohr circles are drawn with radii of one-half the maxi-
mum deviator stresses and with their centers at values equal

(rl ; u3)
to one-half the sums of the major and minor principal stresses
-
Plot a Mohr circle, or a sufficient segment thereof, for each specimen in
____q-
I
L- - _ _ J_--
I
--CT -__-_+
!,
I o-
F i g u r e i3. Construction of Mohrs circle of stress
X-26
E M itto-2-i906
Appendix X
30 Nov 70
the graph in the upper right corner of report form. A sketch of each
specimen after failure should be shown above the Mohr circles (Plate X- 3).
The following procedures should be followed in drawing strength envelopes:
(4) U n d i s t u r b e d s p e c i m e n s . For undisturbed specimens, the
strength envelope should be drawn tangent to the Mohr circles. Q tests of
saturated soils usually indicate a strength envelope that is parallel to the
abscissa as shown in Figure 14a, so the angle of internal friction is usually
equal to zero. Strength envelopes indicated by Q tests on partially satu-
rated soils are usually curved as shown in Figure +4b, particularly for the
lower normal stre_s se s. When the curvature is pronounced, the shear
s t r e n g t h p a r a m e t e r s + and c are not constants.
(2) Compacted specimens. For compacted specimens, the
strength envelope should be drawn through points on the Mohr circles rep-
resenting stresses on the failure plane as shown in Figure 14~.
6. Q TEST WITH BACK-PRESSURE SATURATION. In cases where a
foundationsoil exists that is partially saturated during exploration but
which will become completely saturated without significant volume change
either before or during construction, it is necessary to sat.urate Q test
specimens, using back pressure, before they are sheared. Such field con-
ditions may occur when, due to heavy rains or other reasons, the water
table is raised above the level that existed during initial sampling: Con-
struction of cofferdams, river diversions, and closure sections, or perco-
lation of rainwaters can also create conditions that increase foundation
water contents after exploration but before embankment construction and
subsequent consolidation of the foundation.
For the Q test with back-pressure saturation, the apparatus should
be set up similar to that shown in Figure 46 (page X-30). Filter strips
should not be used and as little volume change as possible should be per-
mitted during the test. After completing the steps outlined in paragraphs
4 and 5~(1) through 5~(4) (note that the procedures for attaching the mem-
brane to the cap and base and for deairing the drainage lines are similar
X-27
EM iiiO-2-i906
Appendix X
30 Nov 70
v
[
S TREUG T/i ENK!7L OPE - @=o
u. UNDISTURBED SOIL, COMPLETELY SATURATED
b. UNDISTURBED SOIL, f+RTlALLY S A T U R A T E D
E#vELOPE DRAWN THROUGH POfNTS OAI CIRCLES

REPRESGVThVG STRESSES ON FAILURE PLA AM AS
mmm 8~ a WHERE a= 45 * + I
2\
NORMAL STRESS G, T/SQ FT
c. COMPACTED SOIL
Figure 14. Examples of strength envelopes

for Q tests
X-28
EM iiiO-2-i906
Appendix X
30 Nov 70
to those used in the R test), apply 3-psi chamber pressure to the speci-
men with all drainage valves closed. Allow a minimum of 30 min for
stabilization of the specimen pore water pressure, measure AH, and
begin back-pressure procedures as given in paragraphs 7b(2) through
7l~(5). After verification of saturation, and remeasurement of AH, close
all drainage lines leading to the back pressure and pore water measure-
ment apparatus. Holding the maximum applied back pressure constant,
increase the chamber pressure until the difference between the chamber
pressure and the back pressure equals the desired effective confining
pressure (see paragraph 5&(5)). Then proceed as outlined in paragraphs
5~(6) through 5~( 1 i).
7. R TEST. All specimens must be completely saturated before appli-
cation of the deviator stress in the R test. A degree of saturation over
98 percent can be considered to represent a condition of essentially com-
plete saturation; if pore water pressures are to be measured during shear,
however, the specimens must be iO0 percent saturated. Computations of
X-29
E M IiiO-Z-i906
Appendix X
30 Nov 70
the degree ef saturation based on changes of volume and water content

are often imprecise, so complete saturation of a specimen should be
assumed only when an increase of the chamber fluid pressure will cause
an immediate and equal increase of pressure in the pore water of the
specimen. In general, it is preferable to saturate the soil after the speci-
mens have been prepared, encased in membranes, and placed within the
compression chamber, using back pressure. A back pressure is an
artificial increase of the pore water pressure which will increase the
degree of saturation of a specimen by forcing pockets of air into solution
in the pore water. The back pressure is applied to the pore water simul-
taneously with an equal increase of the chamber pressure so that the
effective stress acting on the soil skeleton is not changed. In other words,
the pressure differential across the membrane remains constant during
the back pressure saturation phase. Thus, when the back,pressure is in-
creased sufficiently slowly to avoid an excessive pressure differential
within the specimen itself, the degree of saturation will be increased
while the volume of the specimen is maintained essentially constant.
.: Figure 15 gives the back pressure theoretically required to produce a
desired increase in saturation if there is no change in specimen vol-
d ume. It is important to note that the relation shown in Figure 45 is
based on an assumption that the water entering the specimen contains
no dissolved air.
a.
- Apparatus. In addition to the apparatus described in para-
- through 3g, the following equipment are necessary for R tests

graphs 3a
utilizing back pressure for saturation:
(i) Air reservoir and regulator for controlling the back
p r e s s u r e , similar to those used to control the chamber pressure.
(2) Bourdon gage attached to the back pressure reservoir to
measure the applied back pressure. As relatively large back pressures
and chamber pressures are sometimes required, it is essential that these
two pressures be measured accurately to insure that the precise
X-30
EM iiiO-2-i906
Appendix X
30 Nov 70
Figure i5. Back pressure required to attain various

degrees of saturation
difference between them is known. A differential pressure gage1 will

permit this difference to be measured directly.
(3) Calibrated burette or standpipe capable of measuring
volume changes to within 0.i cc for 1,.4-in.-diameter specimens, 0.5 cc
for 2.8-in.-diameter specimens, and 1 cc for 6-in.-diameter specimens.
This burette is connected in the back pressure line leading to the top of
t John Lowe, III, and Thaddeus C. Johnson, Use of back pressure to in-
crease degree of saturation of triaxial test specimens, ASCE Research
Conference on Shear Strength of Cohesive Soils, University of Colorado
IBoulder, Cola., June i960).
x-31
EM iiiO-2-i906
Appendix X
30 Nov 70
the specimen to measure the volume of water added to the specimen dur-
ing saturation and volume changes of the specimen during consolidation.
If the water added to the specimen becomes saturated with air, a higher
back pressure will be required than that given in Figure i5. Therefore,
precautions should be taken to minimize aeration of the saturation water
by reducing the area of the air-water interface or by separating the air
and water with a rolling rubber diaphragm.* A relatively long (over
6-foot) length of thick-walled, small-bore tubing between the burette and
the specimen will also reduce the amount of air entering the specimen.
Adequate safety precuations should be taken against breakage of the burette
under high pressures.
(4) Electrical pressure transducer or no-flow indicator with
which the pressure of the pore water at the bottom of the specimen can be
measured without allowing a significant flow of water from the specimen.
This is an extremely difficult measurement to make since even a minute
flow of water will reduce the pressure in the pore water; yet the measuring
device must be sensitive enough to detect small changes in pressure.
Ellectrical pressure transducers, while relatively expensive, offer almost
complete protection against flow, are simple to operate, and lend them-
selves to the automatic recording of test data. Several types of manually
balanced pressure-measuring systems employing a no-flow indicator are
being used successfully,$ though a full discussion of their relative merits
* H. B. Seed, J. K. Mitchell, and C. K. Ghan, The strength of compacted

cohesive soils, ASCE Research Conference on Shear Strength of Cohe-
sive Soils, University of Colorado (Boulder, Colo., June i960).
$ Bishop and Henkel, op. cit., pp. 52-63, 206-207.
A. Andersen, L. Bjerrum, E. DiBiagio, and B. Kjaernsli, Triaxial
Equipment Developed at the Norwegian Geotechnical Institute, Publica-
tion No. 2i, Norwegian Geotechnical Institute (Oslo, i957).
A. Casagrande and R, C, Hirschfeld, First Progress Report on Investi-
gation of Stress-Deformation and Strength Characteristics of Compacted
Clays, Soil Mechanics Series No. 61, Harvard University (Cambridge,
Mass., May 1960).
X-32
EM iiiO-2-i906
Appendix X
30 Nov 70
and shortcomings is not possible here.

b. -*
- Procedure. The procedure for the R test utilizing back
pressure for saturation shall consist of the following steps:
(i) Proceed as outlined in paragraphs !&(i) through 5aJ4),
with the exception that specimen bases and caps with porous inserts and
drainage connections should be used and back pressure equipment should
be included as shown in Figure 16. Saturated strips of filter paper (such
as Wha.tmans No. 54) placed beneath the membrane a:-Id extending from
the base along three-fourths of the specimen lerigth will reduce the time
required for saturation and consolidation. These strips must neither
overlap and form a continuous circumferential coverage of the specimen
Figure i6. Schematic diagram of typical triaxial compression

apparatus for R and S tests
X-33
EM iilO-2-i906
Appendix X
30 Nov 70
nor form a continuous path between the base and the cap. Place saturated
filter paper disks having the same diameter as that of the specimen be-
tween the specimen and the base and cap; these disks will also facilitate rr
moval of the specimen after the test. The drainage lines and the porous
inserts should be completely saturated with deaired water. The drainage
lines should be as short as possible and made of thick-walled, small--bore
tubing to insure minimum elastic changes in volume due to changes in
pressure. Valves in the drainage lines (valves E, F, and G in Figure i6)
should preferably be of a type which will cause no discernible change of
internal volume when operated (such as the Teflon-packed ball valve made
by the Whitey Research Tool Co.). While mounting the specimen in the
compression chamber, care should be exercised to avoid entrapping any
air beneath the membrane or between the specimen and the base and cap.
(2) Estimate the magnitude of the required back pressure by
reference to Figure i5 or other theoretical relations. Specimens should
be completely saturated before any appreciable consolidation is per-
mitted, for ease and uniformity of saturation as well as to allow volume
changes during consolidation to be measured with the burette; there-
fore, the difference between the chamber pressure and the back pres-
sure should not exceed 5 psi during the saturation phase. To insure that
a specimen is not prestressed during the saturation phase, the back pres-
sure must be applied in small increments, with adequate time between
increments to permit equalization of pore water pressure throughout the
specimen.
(3) With all valves closed, adjust the pressure regulators to
a chamber pressure of about 7 psi and a back pressure of about 2 psi.
Record these pressures on the data sheet (Plate X-4). Now open valve A
to apply the preset pressure to the chamber fluid and simultaneously
open valve F to appIy the back pressure through the specimen cap. Im-
mediately open valve G and read and record the pore pressure at the
specimen base. When the measured pore pressure becomes essentially
x-34
EM iiiO-2-i906
Appendix X
30 Nov 70
constant, close valves F and Gt and record the burette reading.

(4) Using the technique described in step (3), increase the
chamber pressure and the back pressure in increments, maintaining the
back pressure at about 5 psi less than the chamber pressure. The size of
each increment might be 5, IO, or even 20 psi, depending on the compres-
sibility of the soil specimen and the magnitude of the desired consolidation
pressure. open valve G and measure the pore pressure at the base im-
mediately upon application of each increment of back pressure and observe
the pore pressure until it becomes essentially constant. The time required
for stabilization of the pore pressure may range from a few minutes to
several hours depending on the permeability of the soil. Continue adding
increments of chamber pressure and back pressure until, under any incre-
ment, the pore pressure reading equals the applied back pressure im-
mediately upon opening valve G.
(5) Verify the completeness of saturation by closing valve F
and increasing the chamber pressure by about 5 psi. The specimen shall
not be considered completely saturated unless the increase in pore pres-
sure immediately equals the increase in chamber pressure.
(6) When the specimen is completely saturated, hold the max-
imum applied back pressure constant and increase the chamber pressure
until the difference between the chamber pressure and the back pressure
equals the desired consolidation pressure. Open valve F and permit the
specimen to consolidate (or swell.) under the consolidation pressure.
Valve E may be opened to allow drainage from both ends of the specimen.
At increasing intervals of elapsed time (O.i, 0.2, 0.5, 1, 2, 4, 8, i5, and
30 min, 1, 2, 4, and 8 hr, etc.), observe and record (Plate X-5) the burette
readings and, if practicable, the dial indicator readings (it may be neces-
sary to force the piston down into contact with the specimen cap for each
7 If an electrical pressure transducer is used to measure the pore pres-

sure, valve G may be safely left open during the entire saturation
procedure.
X-35
EM 1iiO-2-1906
Appendix X
30 Nov 70
reading). Plot the burette readings (and dial indicator readings, if taken)
versus the logarithm of elapsed time, as shown in Figure 5 of Appendix
VIII, CONSOLIDATION TEST. Allow consolidation to continue until a
marked reduction in slope of the curve shows that 100 percent primary
consolidation has been achieved.
(7) Close valve G, unless pore pressure measurements are
to be made during shear , and valves E and F, and proceed according to
paragraphs 5a(6) through 5a(iO), except use a rate of strain for the R
test of about 0.5 percent per minute (for plastic materials) and about
0.3 percent per minute or less for brittle materials that achieve a maxi-
mum deviator stress at about 3 to 6 percent strain; the strain rate used
should result in a time to maximum deviator stress of approximately
30 min. Relatively pervious soils may be sheared in i5 min. These
rates of strain do not permit equalization of induced pore pressure
throughout the specimen and are too high to &llcw satisfactory pore pres-
sure measurements to be made at the specimen ends during shear.t
Therefore, these rates of strain are applicable only to R tests in which
no pore pressure measurements are made during shear. Where pore
pressure measurements are made at the ends of the specimens as in R
tests, the time to reach maximum deviator stress should generally be at
least 120 min; considerably longer time may be required for materials
of low permeability. For brittle soils (i.e., those in which the maximum
deviator stress is reached at 6 percent axial strain or less), after the
maximum deviator stress has been clearly defined, the rate of strain
may be increased so that the remainder of the test is completed in the
same length of time as that taken to reach maximum deviator stress.
However, for each group of tests in a given test program, at least
20 percent of the samples should be tested to final axial strain at rates
of strain outlined in the first sentence of this paragraph.
c. Computations. The computations shall consist of the follow-
ing steps:
(1) From the observed data, compute and record on the data
sheet (Plate X-l) the initial water content, volume of solids, initial void
t Bishop and Henkel, op. cit., pp. i92-204.
X-36
EM IiiO-2-i906
Appendix X
30 Nov 70
a. UNDISTURBED SOIL, NORMALLY CONSOLIDATED
be UNDISTURBED SOIL, OVER-CONS0LlDATE.D
NORMAL STRESS G, T/S2 FT
c. COMPACTED SOIL
Figure i7. Examples of strength envelopes
for R tests
X-39
EM iiiO-2-i906
Appendix X
30 Nov 70 -
routine laboratory work. Therefore, S tests of fine-grained impervious

materials should normally be performed with direct shear equipment (see
Appendix I X DRAINED (S) DIRECT SHEAR TEST). However, if scheduling
permits, it may be desirable to perform companion S triaxial tests of
impervious soils to compare the results with those from S direct shear
tests. If the soil to be tested is relatively impervious and contains gravel
which would preclude the use of direct shear equipment, consideration
should. be given to using triaxial equipment with pore pressure measure-
ments in order to obtain the drained shear strength parameters within a
reasonable length of time. All specimens must be completely saturated
before application of the deviator stress in the S test.
a. Apparatus. The apparatus used to perform the R test, as
illustrated in Figure 46, will usually be satisfactory for the S test,
though the equipment for saturation by back pressure will not be necessary
for relatively pervious soils. In general, controlled-strain testing should
-
be used for relatively pervious soils, and controlled-stress testing should
be used for relatively impervi<,us soils.
b. Procedure. The procedure for the S test shall consist of the
following steps:
(I) For soils requiring saturation by back pressure, proceed
as outlined in paragraphs 6b(i) through 6b(6). For pervious soils which
can be effectively saturated by seepage, that is, by water percolating
through the specimen under a small hydraulic head, omit the back pres-
sure equipment described in paragraph 6a and proceed as follows:
(a) Proceed as outlined in paragraphs 5a(i) through
5a(4), with the exception that specimen bases and caps with porous inserts
and drainage connections should be used and the apparatus should include
a water supply container and a calibrated burette with a vacuum connection
as shown in Figure 16. For specimens of cohesionless soil prepared as
described in paragraph 4k, the porous inserts and drainage lines (including
the burette) should be dry, and a low vacuum (less than 5 psi) should be
X-40
EM IiiO-2-i906
Appendix X
30 Nov 70
maintained at both the specimen base and cap (with valves D and G closed)
to support the specimen while assembling and filling the triaxial chamber.
(b) Keeping valves A and C closed, adjust the pressure
regulator for a chamber pressure of about 5 psi and then open valve A to
apply this pressure to the chamber.
(c) With valves E and G closed, maintain a low vacuum
through the burette to the specimen cap. Then open valve D and elevate
the water supply container so that a hydrostatic head of 1 to 2 ft is applied
to the base of the specimen.
(d) When the saturation water rises into the burette,
disconnect the vacuum from the burette. Permit seepage under the small
head to continue until the rate of flow into the burette is constant, and then
close valve D.
(2) With valves E and F open (see Fig. i6), lowpr the piston
.nto contact with the specimen cap and increase the axial load at a rela-
tively slow rate so that a fully drained condition exists at failure with
controlled-strain loading or after each increment of load with controlled-
stress application. As for the direct shear test, considerable experience
and judgment are generally required in determining the proper rate of axial
load application (see Appendix IX, DRAINED (S) DIRECT SHEAR TEST).
Theoretical formulas are also availablet for estimating the time required
for failure in S tests. Special precautions may be necessary for tests
requiring an axial loading duration in excess of a few hours to insure that
the chamber pressure (as well as the back pressure, if used) is main-
tained constant, that temperature fluctuations are minimized, and that
evaporation or aeration of the water in the burette is reduced as much as
possible. Placing about 1 cc of oil or dyed kerosene over the water sur-
face in the burette will minimize evaporation.
(3) Record the dial indicator and burette readings at
t Bishop and Henkel, op. cit., pp. i24-d27, 204-206.
x-4i
EM 1110-2-i906
Appendix X
30 Nov 70
increasing intervals of elapsed time under each increment of load. For

relatively impervious soils, plot either or. both of these readings versus
the logarithm of elapsed time, as shown in Figure 5 of Appendix VLLL,
CONSOLIDATION TEST, to establish when primary consolidation has been
essentially completed for each increment of load. Record the final dial
indicator and burette readings for each axial load increment on a form
similar to Plate ~-6 prior to applying the next increment. With controlled-
strain loading, periodically observe and record (Plate X-6) the resulting
load and the dial indicator and burette readings; sufficient readings should
be taken to completely define the shape of the stress-strain curve, Con-
tinue the test until an axial strain of i5 percent has been reached; however,
when the deviator stress decreases after attaining a maximum value and
is continuing to decrease at 15 percent strain, the test shall be continued
to 20 percent strain (see .Fig. 12).
(4) Upon completion of axial loading, close valves E and F
and proceed as outlined in paragraphs 5a(8) through 5a(iO), except measure
the specimen diameter, as described in paragraph 4$t)(e), after the com-
pression chamber has been dismantled. While considerable difficulty may
be encountered in measuring the diameter of the specimen after the test,
such measurements will permit the most reliable computations of the
specimen properties at failure.
C. Computations. The computations shall consist of the follow-
ing steps:
(i) From the observed data, compute and record on the data
sheet (Plate X-i) the initial water content, volume of soilds, initial void
ratio, initial degree of saturation, and initial dry density using the for-
mulas previously presented.
(2) Compute the cross-sectional area of the specimen, AC,
after completion of consolidation using the formulas presented in para-
graph 6~( 2).
(3) Using the dimensions of the specimen after consolidation
X-42
EM IIIO-2;1906
Appendix X
30 Nov 70
(5) Record the time to failure on the data sheet Plate X-6).
\
(6) Correct the maximum deviator stress, if necbssary, for
the effect of membrane restraint (see paragraph 5lJ4)).
d. Presentation of Results. The results of the S test shall be
presented on the report form shown as Plate X-3, as described in para-
griph 52, If volume changes of the specimens during shear were mea-
sured, plot the volumetric strain versus axial strain for each specimen
below the stress-strain curves.
cause inaccurate determinations of strength and stress -deformation
characteristics :
a. Apparatus. (i) Leakage of chamber fluid into specimen.
Such leakage might occur through or around the ends of the membrane or
through the drainage connections and it would decrease the effective stress
in a specimen during undrained shear, Very little leakage is needed to
-
cause a very large change in effective stress, and the longer the period of
undrained shear, the greater the amount of leakage. (Leakage will not
influence the effective stress during periods of specimen drainage, but it
will introduce errors in volume change measurements,)
(2) Leakage of pore water out of specimen. This leakage
might occur through fittings or valves and it would increase the effective
stress in a specimen during undrained shear.
(3) Permeability of porous inserts too low.
(4) Restraint caused by membrane and filter paper strips.
(5) Piston friction.
!z* Preparation of Specimens. (i) Specimen disturbed while
trimming. Disturbance of the natural soil structure does not always
result in strength measuremerits which are too low, that is, on the safe
side; disturbed specimens will consolidate more under the effective con-
solidation pressure in R or S tests and the measured strengths will be
too high.
X-44
EM IllO-2-1906
Appendix X
30 Nov 70
(2) Specimen disturbed while enclosing with membrane. The

techniques of placing the membrane around the specimen illustrated in
Figures 9 and 10 may not be satisfactory for sensitive undisturbed soils
since the specimen would .tend to be flexed while binding the membrane to
the unsupported cap. Alternatively, the specimen can be set upon an
inverted cap clamped to a ringstand and the membrane placed over the
specimen and bound to the cap; then the specimen and cap can be inverted
onto the base and the lower end of the membrane secured.
(3) Specimen dimensions not measured precisely. Dial gages
or micrometers are helpful in obtaining precise measurements. When the
specimen diameter is measured after being enclosed by the membrane,
twice the thickness of the membrane must be subtracted from the measure-
ment. The cross-section area of large specimens may be determined
most satisfactorily from circumference measurements.
c. Q Test. (i) Changes in specimen dimensions upon applica-
tion of chamber pressure. Partially saturated specimens will compress
under the chamber pressure so the change in height, AHo, due to the
application of chamber height should be recorded. When this change in
height is significant, the area of specimen before shear, A should be
C
computed according to the formula:
- 2AHo
HO
AC =Ao H
0
as given in paragraph 6~(2).

(2) Rate of strain too fast.
(3) Water content determination after test not representative.
Friction between the soil and the cap and base restrains the radial defor-
mation at the ends of the specimen and this end restraint induces a non-
uniform pore pressure distribution which, in turn, causes pore water
migration within the specimen. For relatively impervious soils, a
X-45
EM iiiO-2-i906
Appendix X
30 Nov 70
significant migration of pore water could occur only in a test of long dura-
tion (such as S and some R tests); however, for more pervious .soils, an
appreciable redistribution of water content can occur within the shor.t
duration of a Q test. Therefore, it may be desirable to determine the
water content of the end sections (about i/6 of the height at each end)
separately from the middle portion. Correlations of strength with water
content should be based on the water content of the middle portion, though
the dry weight of the entire specimen is needed to compute the initial soil
properties.1
20 R Test. (1) Back pressure increments too large in relation
to effective consolidation pressure.
(2) Back pressure increments applied too rapidly.
(3) Chamber and back pressures not precisely maintained
during consolidation phase. Variations in either or both of these pres-
sures (often much larger than the difference between them) can result in
overconsolidation of the specimen.
(4) Specimen not completely consolidated before shearing.
(5) Rate of strain too fast.
(6) Excessive variations in temperature during shear. An
increase in temperature will decrease the effective stress in a specimen
during undrained shear. This danger, obviously, increases with the dura-
tion of the test.
(7) Specimen absorbed water from porous inserts at end of
test. As in a consolidation test or a direct shear test, the specimen will
absorb water from the porous inserts and drainage lines at the end of the
R or S test no matter how rapidly the apparatus is disassembled and the
specimen removed. To obtain an accurate water content determination at
1 A. Casagrande and S. J. Poulos, Fourth Report on Investigation of

Stress-Deformation and Strength Characteristics of Compacted Clays,
Soil Mechanics Series No. 74, Harvard University (Cambridge, Mass,
October i964).
X-46
EM iiiO-2-1906
Appendix X
30 Nov 70
the end of the test, the specimen should be allowed to swell completely
under a small (2 or 3 psi) chamber pressure and the increase in volume
measured by means of the burette. This volume change can then be used
to correct the water content measured after the test.
e. S Test. (i) Rate of strain or rate of loading too fast.
(2) Inaccurate volume change measurements during shear.
Where volume changes are measured using a burette, inaccuracies may
result from incomplete saturation of the specimen, leakage, evaporation,
or temperature fluctuations.
x-47
E M 1110-2-1906
Change 1
1 my 80
APPENDIX XA:
CYCLIC TRIAXIAL TESTS
1. P R I N C I P L E S O F C Y C L I C T R I A X I A L T E S T I N G . T h e stress-
controlled cyclic triaxial test is used to evaluate the liquefaction
potential and strength of a soil under shear stresses representative of
those induced by an earthquake. For horizontal soil deposits, the tri-
axial specimen is consolidated isotropically, and the cyclic shear
stresses generated by an earthquake are simulated in the laboratory by
cycling shear stresses along the 45-deg plane of a triaxial compression
specimen under undrained conditions. For sloping ground surfaces, the
triaxial specimen is consolidated anisotropically. Hence, the test is
comparable to an R test with pore pressure measurements (see
Appendix X, TRIAXIAL COMPRESSION TESTS) with the notable exception
that the load is cyclically applied to the specimen such that the specimen
is subjected to alternating cycles of vertical compression and extension
about some ambient stress state, which produce corresponding cyclic
shear stresses on the 45-deg plane, as illustrated in Figure 1.t Cyclic
shear stresses in the test on anisotropically consolidated specimens are
depicted in Figure 2.t-t Thus, the major difference between cyclic tri-
axial and conventional triaxial equipment is a rigid fixed connection
between the specimen cap and the loading piston.
The consequence of cyclic loading under undrained conditions is
generally an incre.a.se in the pore water pressure, which causes the
t M . L . S i l v e r , Laboratory Triaxial Testing Procedures to Deter-

mine the Cyclic Strength of Soils, Report No. NUREG-31, 1976,
U. S. Nuclear Regulatory Commission, Washington, D. C.
tt M. L. Silver et al., Cyclic Triaxial Strength of Standard Test
S a n d , Journal of Geotechnical Engineering, ASCE, Vol 102,
GT5, May 1976.
XA- 1
E M ~~~0-2-~906
Appendix XA
Change 1
1 May &Y
r
1 ,
Zc = CONSOLIDATION
STRESS
Aqc = CYCLIC DEVIATOR
STRESS
STRESS PATH
w
z @
z. I
5 \
2
NORMAL STRESS
Figure I. Mohrs circle of total stress representation for a cyclic

triaxial strength test for an isotropically consolidated specimen
(after Silver, 1976)
NOTE:
+3c~~3c~ ;y 1 iiir&jr;;
zlC
CONSOLIDATION
cdc = CYCLIC DEVIATOR
STRESS
Tcyc,ic = CYCLIC SHEAR STRESS,+
z
ii
iii
z
g
NORMALSTRESS
Figure 2. Mohrs circle of total stress representation for a cyclic

triaxial test for an anisotropically consolidated specimen (after
Silver et al., 1976)
XA- 2
-
E M 1110-2-1906
Appendix XA
Change 1
1 May 80
effective stress to decrease and the cyclic deformations of the specimen
to increase. In this test, initial liquefaction is defined to occur when the
pore water pressure first equals the confining pressure, i.e., a condi-
tion where the effective stress is zero, with failure defined in terms of
a limiting cyclic strain, generally either 5 or IO percent from peak to
peak. Peak-to-peak strain is usually referred to as double amplitude
strain and is the total strain that the specimen undergoes on consecu-
tive peaks of the strain-time trace. For isotropically consolidated
specimens, these strains will be alternatively compressional and ex-
tensional; however, in the case of anisotropically consolidated speci-
mens, these strains may be alternative compression-extension or
permanent compression strain only.
The applied cyclic axial stress, uI - 0-3 or P/A (see Fig. 1 of
Appendix X, TRIAXIAL COMPRESSION TESTS), is termed the cyclic
d e v i a t o r s t r e s s , udc , which is alternatively positive and negative
about some ambient stress state. For isotropically consolidated speci-
mens, test results are expressed in terms of the cyclic stress ratio,
which is the cyclic shear stress, qdc/2 9 normalized by
udc2f13c
the confining pressure ~3~ . For anisotropically consolidated speci-
mens, test results may be expressed in terms T where
c y c l i c lCfc 9
is the cyclic shear stress on the failure plane (45 t +I /2)
Tcyclic
a n d cfc is the normal stress on this plane during consolidation. The
value of c$ can be estimated or determined from static tests. Re-
searcht has shown that the cyclic strength depends on density, confining
t J. P. Mulilis, R. C. Horz, and F. C. Townsend, The Effect of

Cyclic Triaxial Testing Techniques on the Liquefaction Behavior
of Monterey No. 0 Sand, M i s c e l l a n e o u s P a p e r M P S - 7 6 - 6 ,
Apr f976, U. S. Army Engineer Waterways Experiment Station,
CE, Vicksburg, Miss.
F. C. Townsend, A Review of Factors Affecting Cyclic Triaxial
T e s t s , Dynamic Geotechnical Testing, ASTM STP 654, Jun 1977.
XA- 3
E M 11iO-2-1906
Appendix XA
Change 1
1my 80
p r e s s u r e , applied cyclic shear stress, stress history, specimen prepara-
tion procedure, and uniformity and shape of cyclic wave form; hence,
close attention must be given to testing details and equipment.
2. SPECIMENS. In most cases, only high-quality undisturbed samples
should be used for testing since sand fabric (particle orientation) has a
considerable effect on cyclic strength. Presently, no techniques are
available for duplication of the in situ fabric by laboratory reconstitution.
If there is an insufficient number of undisturbed specimens available for
a testing program, then correlations between undisturbed and reconsti-
tuted specimens are necessary to provide data over the entire range of
conditions desired.
3. APPARATUS. All the equipment listed in Appendix X, TRIAXIAL
COMPRESSION TESTS, for performing triaxial compression R tests
with pore pressure measurements, is required for cyclic triaxial tests
in addition to the following special equipment: (1) cyclic loading equip-
ment, (2) electronic transducers and high-speed recorders for data
acquisition, and (3) provisions for fixing the specimen cap to the loading
piston for the extension portion of the loading cycle.
-a. Loading Devices. Dynamic loading equipment used for stress-

controlled cyclic triaxial tests must be capable of applying a uniform
sinusoidal load at a frequency of 1 Hz or greater. The equipment may
be either (1) a closed-loop electrohydraulic system or (2) a pneumatic
system.T Both systems basically consist of a pressurized fluid (oil or
air) whose pressure into a loading actuator is varied by a servovalve
or regulator commanded by a control unit. The system should be capa-
ble of maintaining the cyclic deviator load constant throughout the test,
i.e., must provide sufficient hydraulic oil or air to permit the loading
piston to follow the sudden and rapid specimen deformation at
7 C. K. Chan and J. P. Mulilis, A Pneumatic Sinusoidal Loading

S y s t e m , Journal of Geotechnical Engineering, ASCE, Vol 102,
No. GT3, Mar 1976.
XA-4
E M 4140-2-1906
Appendix XA
Change 1
1 May 80
liquefaction. The loading device must maintain uniform cyclic peak
loadings throughout the test. Unsymmetrical compression-extension
load peaks, nonuniformity of pulse duration, ringing, or overshoot
must not exceed tolerances illustrated in Figure 3.t In these cases,
differences in peak compressive and extension loads or durations
greater than 10 percent are unacceptable. Both ringing and over-
shoot can induce abnormally rapid pore pressure rises affecting
specimen failure.
A problem common to most cyclic loading equipment is the re-
duction in cyclic load at the onset of large specimen deformation. For
the test to be meaningful, the load must be symmetrical in extension
and compression up to peak-to-peak strains of 20 percent; and the peak-
to-peak load should not decrease by more than 20 percent from the ini-
tially applied values until the specimen,peak-to-peak strains exceed 10
percent. The equipment must also be able to apply the cyclic load about
an initial static load on the loading piston. This static load counteracts
the uplift pressure, which results, from the chamber pressure acting on
the reduced area of the specimen cap because of the fixed loading piston
connection. Further, in the case of anisotropically consolidated speci-
mens, the static load applies some of the consolidation stress.
h. Specimen Cap. The specimen cap should be of a lightweight
noncorrosive material equipped with porous metal or porous stone
inserts and drainage connections. The cap can be similar to those in
paragraph 3~ of Appendix X, TRIAXIAL COMPRESSION TESTS, except
(1) provisions for fixing the loading piston rigidly to the cap must be
provided and (2) specimen diameter should be 2.8 in. or larger. The
most common connection is simply straight threads backed by a
shoulder on the piston, which tightens up against the loading piston.
c. Triaxial Compression Chamber. The triaxial compression
-
t Silver, op. cit.
XA-5
E M II,iO-2-3906
Appendix XA
Change 1 -.
1 May 80
a. CORRECT SHAPE b. UNSYMMETRICAL PEAKS
NOT ACCEPTABLE
c. UNSYMMETRICAL PULSES d. LOAD RINGING
NOTE: Pc = COMPRESSIVE LOAD

tc = DURATION OF COMPRESSION
LOAD
te = DURATION OF EXTENSION
LOAD
NOT ACCEPTABLE
e. OVERSHOOT
Figure 3. Examples of acceptable and unacceptable

loading wave forms for cyclic triaxial strength
tests (after Silver, 1976)
XA- 6
-
E M IIIO-2-1906
Appendix XA
Change 1
1 May 80
A. HYOROSTATICSEALAIRSUPPLY PORT
B. VENT
C. TOPORAINAGEPORT
D. CHAMBER FLU10 ORAIN
E. TRANSDUCERCABLECONNECTOR
F. BDTTDMDRAINAGEPDRT
PART NO. NAME MATERIAL QUANTITY

1 CHAMBERFOOT ALUMINUMCHANNEL 2
2 CHAMBERBASE ALUMINUMPLATE 1
3 SPECIMENBASE ALUMINUM 1
4 POROUS PLATE STAINLESSSTEEL 2
5 SPECIMENMEMBRANE RUBBER 1
6 SPECIMENCAP ALUMINUM 1
7 SPECIMENCAPCONNECTOR BRASS 1
8 BEARINGSUPPORT ROD STAINLESSSTEEL 2
9 CYLINDER CLEARACRYLIC 1
10 CHAMBER CAP ALUMINUM 1
II CYLINDER SEALING RING ALUMINUM 1
12 LOCKING KNOBS STEEL 3
13 AIR BEARING ASSEMBLY ALUMINUM 1
14 PISTON LOCK BRASS 1
15 LOADING PISTDN STAINLESS STEEL 1
16 SHOULDERSCREW STEEL 2
17 BALL BUSHING STAINLESSSTEEL 2
18 O-RING RUBBER 1
19 O-RING RUBBER 1
20 O-RING RUBBER 1
21 O-RING RUBBER 2
1 _
22 O-Ring RUBBER 1
23 O-RING RUBBER 1
SCALE IN INCHES
1 0 12 3
TRIAXIAL COMPRESSION CHAMBER
Figure 4. Triaxial compression chamber used for cyclic triaxial tests

(working drawings for design shown available from Soil Mechanics
Division, Geotechnical Laboratory, U. S. Army Engineer Waterways
Experiment Station)
XA- 8
E M 1140-2-4906
Appendix XA
Change 1
1 May 80
TO TO
LOADER LOADER
BOLT
/9=L/T NUT
UN/ON NUT
CLAMP
Lkik
/_OAD/lVG f/STON
l_OAD/hJG PETON
a. U N I O N N U T b. SPLIT NUT
TO
LOADER
LOADING P/5TON
c. S P H E R I C A L C L A M P
Figure 5. Examples of successful connections between triaxial cell

loading piston and cyclic loading actuator (after Silver, 1976)
XA- 9
E M 1iio-2-19oT
Appendix XA
Change 1
1 May 80
e . Ret ording Equipment.
- Specimen behavior in stress-controlled
cyclic triaxial tests is evaluated from continuous time records of applied
load, specimen deformation, and change in pore water pressure.
Commonly, these parameters are recorded on a multichannel strip-
chart recorder, as illustrated in Figure 6. Analog to digital data acqui-
sition systems may be used provided that data can be converted later
into a convenient form for data analysis and interpretation. Fast
recording system response is essential if accurate specimen perfor-
mance is to be monitored when failure conditions are approached. It
is recommended that the response characteristics in Table 1 be
satisfied.
For analog strip-chart recording equipment, the magnitude
of the load, deformation, and pore water pressure (chamber pressure
recording is optional) recorder trace must be of sufficient amplitude
and time resolution to enable accurate data reduction. Resolution of
each variable should be better than 2 percent of the maximum value -
being measured. To take advantage of recorder accuracy and for sub-
sequent data analysis, a recorder speed of 2-4 cycles per in. of record-
ing paper is acceptable. The clarity of the trace with respect to the
background should provide sufficient contrast and minimum trace width,
so that the minimum resolution of 2 percent of the maximum value of
the recorded parameter is maintained, and the trace may be included
in reports.
-f. Measurement Transducers. Load, displacement, and pore

water pressure transducers are required to monitor specimen behavior
during stress-controlled cyclic triaxial tests; provisions for monitoring
the chamber pressure are optional. Each of these transducers must
have the required capacity to ensure that the full range of specimen
characteristics are monitored; at the same time, they must have the
required sensitivity to ensure that small changes in specimen behavior
are properly measured and recorded.
XA- IO
EM 1110-2-4906
Appendix XA
Change 1
1 May 80
910s = NOISIAIO 1
a1'33l03
LLllLuG
ISd 01 = NOISIAIO 1
E M 1110-2-3906
Appendix XA
Change 1 -
1 May 80
Table 1
Data Acquisition
Minimum Response Characteristics for
Cvclic Triaxial Strength Tests
1. A n a l o g R e c o r d e r s
Recording Speeds: 0.5 to 50 cm/set (0.2 to 20 in./sec)
System Accuracy (include linearity and hysteresis): 0.5% (I)
F r e q u e n c y R e s p o n s e : 100 Hz
2. Measurement Transducers
D i s p l a c e m e n t (2) Pore
Load Cell Transducer (LVDT) Pressure
Minimum sensitivity, mv/v 2 0.2 mv/O.O25 m m / v 2
(AC LVDT)
5 mv/O.O25 m m / v
(DC LVDT)
Nonlinearity, % full scale i 0.25 +
- 0.25 + 0.5
-
H y s t e r e s i s , $J full scale 2 0.25 0.0 2 0.5
Repeatability, $ full scale 2 0.10 2 0.04 +- 0 . 5
Thermal effects on zero
shift or sensitivity, L 0.005 + 0.02
% of full scale/C(oF) (i 0.025) (2 0.01)
Maximum deflection at full 0.425
rated value in mm (in.) (0.005)
Volume change character-
istics (cu in./psi) 4.0 x lo-4
Note: (1) S y s t e m f r e q u e n c y r e s p o n s e , sensitivity, and linearity are functions
of the electronic system interfacing, the performance of the signal
conditioning system used, and other factors. It is therefore a
necessity to check and calibrate the above parameters as a total
system and not on a component basis.
(2) LVDTs, unlike strain gauges, cannot be supplied with meaningful
calibration data. System sensitivity is a function of excitation
frequency, cable loading, amplifier phase characteristics, and
other factors. It is necessary to calibrate each LVDT-cable-
instrument system after installation, using a known input standard.
XA-42
E M 1~10-2-1906
Appendix XA
Change 1
1 May 80
(4) L o a d t r a n s d u c e r s . . The total load capacity of the load
transducer (load cell) should be of the proper order of magnitude with
respect to the maximum total loads to be applied to the test specimen.
Generally its capacity should be no greater than five times the total
maximum load applied to the test specimen to ensure that the necessary
measurement accuracy is achieved. The minimum performance char-
acteristics of the load cell are presented in Table 1.
A rigid load cell is required to avoid resonance problems
that can develop with closed-loop electrohydraulic loading systems.
Miniature lightweight load cells are used to prevent iner-
tia effects during cycling. The load cell is normally placed outside the
triaxial chamber. If the load cell is located inside the triaxial chamber,
special provisions must be provided to either pressurize the load cell
or normalize the effects of cell pressure to ensure that load readings
are not affected by the cell pressure. In all cases, the response and
performance of a load cell located inside the pressure chamber must
be documented.
(2) Deformation measurement. The stress-controlled cyclic
triaxial test requires a deformation transducer with a high resolution
and small range during the initial portion of the test and less resolution
but large range during the final portion of the test. The linear variable
differential transformer (LVDT) is generally considered to be the most
suitable deformation transducer for the test, and its minimum specifi-
cations will be discussed. However, other displacement measuring
devices, such as eddy current sensors or optical methods, may be used
if they meet the required performance criteria. Potentiometer type
deformation transducers are not recommended because they are easily
damaged if their maximum travel is exceeded.
The displacement transducer must have a range of at least
20-30 percent of specimen height but should not exceed 60 percent of
specimen height. The specifications for this transducer, representing
XA-43
E M 1110-2-1906
Appendix XA
Change 1
1May 80
levels on nonlinearity, hysteresis, and repeatability, are also presented
in Table 1.
Accurate deformation measurements in dynamic triaxial
tests require that the LVDT be properly mounted to avoid excessive
mechanical system compression between the load frame, the triaxial
cell, the load cell, and the loading piston. Thus, it is recommended that
the LVDT be located either on the actuator rod or on the loading piston
attached to the test specimen.
(3) Pore water pressure measurements. Pore water pres-
sures may be individually measured in the drainage line(s) leading to
either (or both) the specimen cap or base. However, more reliable
measurements may be achieved if pore water pressure is measured at
both the cap and base simultaneously by having a connection in the
drainage lines between the specimen and the pressure transducer. The
use of a differential pressure transducer will facilitate data reduction.
A rigid pore water pressure measuring system is re-
quired. To achieve reliable measurements, it is recommended that pore
-4
water pressure transducer volume change not exceed 1 Xi0 cu in.
per psi and that the transducers have the sensitivity and performance
characteristics summarized in Table 1. Moreover, the rigidity of all
the assembled components of the pore water pressure measurement
-4
system should not exceed 1 Xi0 cu in. per psi. In general, stiff,
small-diameter tubing of metal or Saran, short-tubing runs, and high-
quality ball valves are required to meet this specification. In all cases,
the rigidity of the entire pore water pressure system, including the
transducer, must be ascertained and documented. Methods for measur-
ing the rigidity of the entire pore water pressure system are presented
XA-14
E M 1110-2-1906
Appendix XA
Change 1
I MaY !$o
by Wissa (1969)t and Bishop and Henkel (1962).tt
g. Saturation Equipment. The apparatus for back-pressure
saturation described in Appendix X is adequate provided the optional
vacuum source with a vacuum regulator is included in the setup. The
vacuum source must have a means of regulation for those instances
where the effective consolidation stress to be applied to specimens is
less than the pressure difference suppliable by the vacuum. Carbon
dioxide gas (C02) has been found very helpful in obtaining complete
saturation in those instances where a full vacuum cannot be applied.
The CO2 is permitted to seep up through the specimen prior to seepage
saturation.
h. Tamping Rod for Moist Tamping Specimen Preparation.
(Optional) The standard moist tamping specimen preparation proce-
dure uses a tamping foot, whose diameter is one-half that of the speci-
men. Figure 7 shows a sketch of a tamping foot for a 2.8-in.-diameter
specimen. The tamping rod consists of a 3/4-in.-diameter steel rod 12
in. long with a 1.4-in .-diameter steel tamping foot attached to the end.
4. TESTING PROCEDURES. Specimen preparation, back-pressure
saturation, and consolidation procedures specified in paragraphs 4h
and 7k(1) through 7b(7),
- Appendix X, TRIAXIAL C O M P R E S S I O N T E S T S ,
are adequate and pertinent. However, preferred techniques for handling
undisturbed samples of cohesionless soils, preparing remolded speci-
mens for a standard test calibration procedure,* and anisotropic con-
solidation are as follows:
2. Reconstituted Specimen Preparation (Moist Tamping). The
procedure is as follows:
t A. E. 2. W i s s a , Pore Pressure Measurement in Saturated Stiff

S o i l s , Journal of Soil Mechanics and Foundations Engineering,
ASCE, Vol 95, No. SM4, July 1969.
tt A. W. Bishop and D. J. Henkel, The Measurement of Soil Properties
in the Triaxial Test, 2nd ed., London, Edward Arnold Ltd., 1962.
* Mulilis et al., op. cit.; and Silver, op. cit.
XA-15
E M 1110-2-4906
Appendix XA
Change 1
1 May 80
DFOOT
- l l/Z
hOLO
Figure 7. Tamping rod
(I) Adjust the water content of the material, so, that the
initial degree of saturation of the compacted material will be between
20 and 70 percent.
(2) Place the forming jacket, with membrane inside, over
the specimen base of the triaxial compression chamber.
(3) Evacuate the air between the membrane and inside face
of the forming jacket, so that the membrane is taut and flush against
the face.
(4) Determine the inside diameter and height of the mold to
within 0.001 in. for 2.8-in.-diameter specimens, and calculate the
XA-16
E M 1140-2-3906
Appendix XA
Change 1
1May 80
volume based on these measurements.
(5) Select the number of layers to be used in the preparation
of the specimen. The maximum thickness should not exceed 1.0 in. for
specimens having diameters less than 4.0 in.
(6) Starting with the bottom layer, compact each layer to the
desired density using the tamping rod. The locking collar on the rod
must be adjusted to produce the desired height to within 0.001 in. for
each layer. This is most easily done by adjusting the travel of the foot
for each layer using a set of gage blocks.
(7) A procedure of variable compaction, such as that de-
scribed in paragraph 4h(4), Appendix X, TRIAXIAL COMPRESSION
TESTS, may be used to construct uniform specimens. An alternative
and similar procedure consists of building the specimen in equal
thickness layers but placing each successive layer in a slightly denser
state. For example, an increase of one percentage point of relative
density per layer (the top layer placed to the desired average .relative
density for the entire specimen) has been found satisfactory for medium
dense sand.
(8) After the forming jacket is filled to the desired height,
place the specimen cap on top of the specimen, pull the membrane end
over the cap, and fasten with O-rings. Carefully apply a vacuum to the
interior of the specimen while observing the response of the specimen
pressure transducer. Adjust the vacuum regulator as necessary to
assure that the differential pressure supplied by the vacuum is not
greater than the intended effective consolidation stress. A vacuum
gage located between the vacuum regulator and valve K (see Fig. 16
of Appendix X) simplifies this procedure by indicating the differential
supplied by the vacuum while valve K is still closed. With a vacuum
on the interior of the specimen, remove the forming jacket and measure
the specimen dimensions.
h. Undisturbed Specimen. Preparation. Sampling, transporting,
E M 1110-2-1906
Appendix XA
Change 1 --
1 JQy 80
and testing undisturbed specimens of loose free-draining cohesionless
materials so as to minimize disturbance is an extremely difficult task.
However, the current practice is as follows:
(1) Obtain samples using a fixed-piston sampler and
drilling mud.
(2) Place perforated packers at both ends of the sampling
tube and allow the tube to drain in a vertical position. Drainage of the
tube should be monitored and, if insufficient, can be aided by the appli-
cation of a low vacuum, i.e. 5 in. (2.5 psi) of mercury (Hg), to the
bottom packer.
(3) After drainage, the sample may be carefully transported
to the laboratory, or alternatively frozen in the field using dry ice or
liquid nitrogen and transported to the laboratory in a frozen condition.
F r e e - d r a i n i n g s a m p l e s , which contain no silt or clay layers that would
form ice lenses upon freezing, should be transported frozen. It is
important to monitor the sample length at all stages (after sampling,
after drainage, after freezing, after transportation, and again before .-
testing). Samples should be stored vertically, and storage time at the

laboratory minimized to prevent corrosion of the tubes.
(4) (Optional) Keeping the sample frozen, X-ray the tube to
observe possible layering of the sample, voids, gravels, etc., in the
tubet and to facilitate selection of 7-in. sections for testing.

(5) If a milling machine is available, the tube may be cut
lengthwise at two diametrically opposite places, using a rapid feed,
and then cut into sections with an electric hacksaw. If a milling machine
is not used, the desired section is cut with an electric hacksaw or a tube
cutter with stiffening collars. The cut ends of the tube are then cleaned
t W. F. Marcuson III and E. L. Krinitzky, Dynamic Analysis of F t .

P e c k D a m , Technical Report S- 76- 1, U. S. Army Engineer Water-
ways Experiment Station, Nov 1976, Vicksburg, Miss.
XA-18
E M 1110-2-1906
Appendix XA
Change 1
1 May 80
of -burrs, and the specimen is pushed from the tube after having been
permitted to thaw slightly.
(6) The ends of the specimen should be trimmed smooth and
perpendicular to their length using a mitre box or other trimming
device. The specimen is then placed in the triaxial chamber and en-
closed in a rubber membrane; then, a small vacuum (always less than
T3) is applied to hold it firmly while thawing. It is usually necessary
to place the sample in a freezer at times during the cutting and trimming
process to ensure that the specimen remains frozen until enclosed in
the membrane.
(7) Dimensions of the specimen are,taken to calculate the
initial volume. If frozen, measurements are checked after the speci-
men has thawed. There should be no significant volume change; other-
wise, freezing or transportation have caused sample disturbance.
-c. Specimen Measurement. Inasmuch as density greatly in-

fluences the cyclic triaxial strength, it is imperative that accurate
density determinations and volume change measurements be made
during saturation and consolidation. For this reason, diameter mea-
surements using a circumferential tapet to the nearest 0.001 in., height
measurements to the nearest 0.01 in. at four locations, and weights to the
nearest 0.1 g are recommended for 2.8-in-diameter s p e c i m e n s .
-d. Saturation. Saturation generally includes a stage during which

water is allowed to seep into the specimen while its interior is under
vacuum, followed by back pressuring to drive the remaining air in the
specimen into solution with the pore water. A procedure found to be
effective is as follows:
(i) After assembling the triaxial chamber, apply a small
chamber pressure, 1-3 psi, to the specimen, and increase the vacuum to
the interior, noting that any vacuum on the interior of the specimen
t Commercially available from PI Tape, Box 398, Lemon Grove,

Calif. 9 2 0 4 5 .
XA-19
E M 4310-2-1906
Appendix XA
Chanke 1
1 May. 80
contributes to the effective stress and that the difference between the
vacuum and chamber pressures must not exceed the effective consoli-
dation stress.
(2) After allowing the specimen to remain under vacuum for
a period of approximately 5-15 min, permit de-aired water to slowly
seep up through the specimen from the bottom. The upward movement
of water must be slow enough to minimize entrapment of air pockets.
The use of de-aired water further promotes the removal of air by
solution. Flushing of water through the specimen should continue until
gas bubbles no longer come from the upper drainage line.
(3) After seepage saturation, reduce the vacuum and increase
the chamber pressure simultaneously until the specimen is at atmo-
spheric pressure. Back pressure the specimen in steps and evaluate
the degree of saturation at appropriate intervals using Skemptons pore
pressure parameter B . To do this, close the drainage lines from the
s p e c i m e n , increase the chamber pressure 5 or 40 psi, and adjust the
axial load to match the increase in chamber pressure. Observe the -
pore pressure increase, Au , and calculate B , using the equation:
The specimen shall not be considered adequately saturated unless the

B value is greater than 0.95.
In some cases, a satisfactory degree of saturation cannot
be achieved by ordinary vacuum seepage and back-pressure saturation
techniques. This may be due to a weak vacuum source or to the need
to keep the vacuum low to prevent prestressing the specimen. In the
latter case, it is possible to apply a sufficient vacuum to the chamber
so that there is no prestressing of the specimen when a full, i.e.
14.7 psi, vacuum is applied to the interior of the specimen.
XA-20
E M 11iO-2-i906
Appendix XA
Change 1
1 May 80
In either of the cases where the vacuum is limited, CO21
can be allowed to slowly seep upward from the bottom of the specimen,
while the specimen is being formed or after it has been confined in the
triaxial chamber. The CO2 will displace the air in the specimen and,
being much more soluble in water than air, will enable subsequent
saturation steps to be carried out successfully.
During saturation, the change in height of the specimen
should be measured to the nearest 0.001 in. In addition, during satura-
tion and consolidation, an axial load must be applied to the loading
piston, which is screwed into the specimen cap, to compensate for the
uplift force on the loading piston. This static load will be expressed as
Ps = F A weight of apparatus below the load cell

3 r-
where
Ps = static load applied to specimen
= chamber pressure
u3
Ar = area of the loading piston
s. C o n s o l i d a t i o n . Isotropic consolidation is defined as
K
c
z%=l
-
u3c
where
= consolidation ratio
KC
-
vertical effective consolidation stress
ulc =
-
= horizontal effective stress
u3c
To consolidate the specimen isotropically, maintain the applied back-
pressure constant and increase the chamber pressure until the
t Mulilis et al., op. cit.
XA-21
E M 1110-2-1906
Appendix XA
Change 1
1 May 80
difference between the chamber pressure and the back pressure equals
the desired consolidation pressure. An axial load to counterbalance up-
lift due to increasing the chamber pressure must be applied. This may
require incrementally applying the consolidation pressure to provide
sufficient time to apply and adjust the counterbalancing uplift load.
Changes in specimen height during consolidation should be measured to
the nearest 0.001 in., and the change in specimen volume to the nearest
0.01 cc. A plot of burette and/or height readings versus logarithm of
elapsed time, as shown in Figure 5 of Appendix VIII, CONSOLIDATION
TEST, is optional.
Anisotropic consolidation is defined as
-
%
K =--#I
c -
u3c
This consolidation condition may be achieved by incrementally increas-

ing the chamber pressure, G3c , and by increasing the axial load to the
required ylc , plus uplift counterbalancing value, until the final T
3c
value is achieved. As stated previously, changes in specimen height
and volume should be measured during consolidation.
Following consolidation, the drainage lines should be
closed and the pore water pressure observed for a period of time to
verify that no leaks in the membrane or pore water pressure system
have occurred. If the time for consolidation exceeds 8 hr, the B value
should be redetermined prior to cyclic loading.
-f . Cyclic Loading. If it does not already exist, a large air pocket

should be formed at the top of the triaxial chamber by draining water
from the chamber but leaving sufficient water to cover the top of the
specimen. The air pocket is required, so that the large, rapid piston
movements in and out of the chamber at the onset of failure do not
XA-22
E M 1110-2-1906
Appendix XA
Change 1
1 May 80
create chamber pressure fluctuations. Compressed air instead of water
may be used as the confining fluid provided saturation and consolidation
do not exceed 8 hr. The procedure is as follows:
(1) Record test number and specimen identification on
recorder trace.
(2) Zero the recorder and transducer outputs, and record
calibration steps and scale factors.
(3) Close valve F (D should have been already closed) as
shown in Figure 16 of Appendix X, TRIAXIAL COMPRESSION TESTS.
(4) Record the consolidation pressure, T , and estimate
C
the magnitude of cyclic load to be applied for the desired stress ratio,
SR , with the equation:
PC = 2 X ;3c X SR X A
C
where
P = estimated cyclic load to be applied to the specimen, lb
- C
= consolidation pressure (chamber pressure-back pressure), psi
u3c
SR = d e s i r e d s t r e s s r a t i o
= area of specimen after consolidation, sq in.
AC
(5) Initiate cyclic loading with the first half cycle in com-
pression using a 1- to 2-Hz sinusoidal load form. During cyclic loading,
the chamber pressure is maintained constant, and the axial load, axial
deformation, and change in pore water pressure (recording of chamber
pressure is optional) are recorded with time. The load is cycled until
either (a) the cyclic double amplitude strain exceeds 20 percent: (b)
500-load cycles or the maximum number required in the program is
s u r p a s s e d ; (c) n e c k i n g of the specimen is observed; or (d) the load
wave form deteriorates beyond acceptable values.
In cases of anisotropic consolidation, where
XA-23
E M 1iiO-2-1906
Appendix XA
Change 1
1 May,80
K - i
udc < c
-
2
2 u3c
there will be no reversal in the direction of shear stress application,

and the net axial stress will always be the major principal stress. For
these conditions, the pore pressure does not usually increase suffi-
ciently to produce initial liquefaction, rather the specimen tends to
deform progressively. Conversely, where
K - 1
udc > C
- 2
2u3c
there will be a stress reversal and conditions of initial liquefaction

usually occur. I
g. S p e c i m e n R e m o v a l . Following cyclic testing, the specimen

should be carefully removed from the triaxial cell in order that no
particles are lost, then dried and weighed for dry unit weight
calculations.
5. COMPUTATIONS. a. -From the initial specimen data recorded on
Plate XA-3, compute the initial water content, initial void ratio, initial
dry density, and if required, the initial relative density, using equations
presented in Appendixes I, WATER CONTENT - GENERAL, and II,
UNIT WEIGHTS, VOID RATIO, POROSITY, AND DEGREE OF
SATURATION.
b. From saturation and consolidation data, compute the values
- of
Skemptons pore pressure parameter B , and record any specimen
dimension changes during saturation, the static piston loads required
to counterbalance uplift pressures, and the magnitude of applied back
pressure (Plate XA-1 (Contd)).
XA- 24
E M iIIO-2-1906
Appendix XA
Change 1
3 MAY 80
s. From specimen dimension and volume changes during
saturation and consolidation, compute the height, area, and dry density
of the specimen after consolidation, using the following formulas:
= +$ (Ho - 2AHs)
AS
= As (Ho - AHs)
vS
H H - AHs - AHc
C 0
- AVw
vS
A= H
C
%
X 62.4
dc = AcH
C
where
A = area of specimen after saturation, sq cm
As z initial area of specimen, sq cm
Ho = initial height of specimen, cm
ALs = change in height during saturation, cm

vs = volume of specimen after saturation, cc
H = height of specimen after consolidation, cm
ALc = change in height during consolidation, cm
A = area of specimen after consolidation, sq cm
AGw = burette change during consolidation, cc
= dry density after consolidation, pcf
dc
= weight of solids, g
wS
d. After testing, use the data from the strip chart recorder and
-
the relationships and definitions presented in Figure 8 to determine the
cyclic stress, stress ratio, double amplitude strain (all based on con-
solidated specimen dimensions), and the induced pore water pressure;
record these data on Plate XA-3. The number of cycles required to
XA-25
EMIllO-2-4906
Appendix XA
Change 1
1 May 80
CYCLE 1 , CYCLE 2 , CYCLE3
AXIAL LOAD J!&!J)&
AXIAL DEFORMATION 0
CONFINING PRESSURE, Zsc

INITIAL LIQUEFACTION
DEFINITION OF CYCLIC STRESS, STRESS RATIO
AND DOUBLE AMPLITUDE STRAIN
Apt t Ape
CYCLIC STRESS, trdc =
2Ac
CYCLIC STRESS RATIO, SR = *g

3c
A$ + A+
DOUBLE AMPLITUDE STRAIN, 6da, % = H hz 100
c
WHERE:
PEAK CYCLIC LOAD IN COMPRESSION
PEAK CYCLIC LOAD IN EXTENSION
CYCLIC DEVIATOR STRESS
CYCLIC SHEAR STRESS
CYCLIC DEFORMATION IN COMPRESSION
CYCLIC DEFORMATION IN EXTENSION
DOUBLE AMPLITUDE AXIAL STRAIN, PERCENT
AREA AND HEIGHT AFTER CONSDLIDATION, RESPECTIVELY
Figure 8. Definitions and equations for computing cyclic stress,

stress ratio, and double amplitude strain
XA-26 -
E M 3110-2-1906
Appendix XA
Change 1
1 May 80
achieve initial liquefaction (defined as 100 percent pore pressure re-
sponse) and various strain amplitudes are also recorded in Plate XA-2.
The uniformity of the load trace into the failure state should
be evaluated to ensure that the load uniformity criteria (see paragraph
3a)
- are achieved.
6. PRESENTATION OF RESULTS. -
a. Results of individual tests on
specimens, which have been isotropically consolidated, are generally
presented by plotting various percent (e.g. 5, 40, and 20) double ampli-
tude strain versus number of cycles required to reach those strains.
Since anisotropically consolidated specimens often reach failure strains
in compression only, the various percentages of zero-to-peak compres-
sive strain can be plotted versus number of cycles to reach those strains.
In addition, pore pressure increase, Au/ g3c , versus number of cycles
to reach.those percentages of increase (e.g. 25, 50, 75, and 100) can be
plotted; the number of cycles to reach the given increase is denoted
as N respectively.
25i N5oi 9 N75i 9 a--d Ni 9
h. In analyzing the results of a series of isotropically consoli-
dated cyclic triaxial strength tests, the number of cycles required to
achieve initial liquefaction and various values of double amplitude
strain are plotted (see Plate XA-2 and Fig. 9) versus stress ratio,
SR , that is defined as
udc
SR=t-
- -
2 u3c
In the case of anisotropically consolidated cyclic triaxial

strength tests, the definition of stress ratio is not appropriate, as G3c
does not represent the total consolidation stress. Hence only 2udc
which is the cyclic deviator stress, versus log number of cycles for
various values of strain (double amplitude and/or zero to peak com-
pressional) are plotted on Plate XA-2.
For test series involving both anisotropically and isotropically
XA-27
EM 3110-2-i906
Appendix XA
Change 1
1. May 80
0.6
10% DOUBLE AMPLITUDE STRAIN
INITIAL LIQUEFACTION
SUMMARYCURVE
MONTEREY SAND NO.0
WETTAMPINGCOMPACTION
DR=6OPERCENT
I I I I lllll I I I Ill1
0
1 10 100
NUMBEROFCYCLES
Figure 9. Range of cyclic triaxial strength values for initial
liquefaction and IO percent double amplitude strain
consolidated cyclic triaxial specimens, a summary figure of cyclic

shear stress, on the failure plane (45O t $/2), required to
Tcyclic
produce a specified strain in a specified number of cycles versus the
effective normal stress on the failure plane after consolidation,
ufc
for various K ratios , as shown in Figure IO, is desirable. The +
C
used in establishing the failure plane can be derived from static tests
or may be assumed at some representative value.
c. To estimate the cyclic triaxial strength of undisturbed
- speci-
m e n s , a density correction to account for differences between in situ
densities and specimen density after consolidation can be made using
XA-28
E M 1140-2-1906
Appendix XA
Change 1
1 I4ay 80
= INITIAL SHEAR STRESS ON POTENTIAL FAILURE SURFACE

Tfc
= CYCLIC SHEAR STRESS DEVELOPED ON POTENTIAL FAILURE SURFACE
1 cos @
7
cyclic
udc
= -y-- C O S @ OR cd,,. - Dac (Kc - 1) -
2
= ,N,T,AL N O R M A L S T R E S S O N P O T E N T I A L FAILUE S U R F A C E
Ufc
I
0
=L + 1) -(Kc - 11 S I N +
2
z 0 I I I I I
0 1 2 3 4 5
N O RM A L STRESS ON FAILURE PLANE AFTER CONSOLIDATION, cfc
Figure IO. c y c l i c s h e a r s t r e s s o n failUrt3 phx!, Tcyclic ,
versus normal stress on failure plane after consolidation,

for 5 percent strain in 5 cycles
Vfc
XA-29
E M 1110-2-1906
Appendix XA
Change 1
1 May 82
the following empirical relationship,t provided Dd (laboratory) is near
50 percent:
Dd field
mdc udc
2 cr = 2 u x Dd l a b
o(Dd field) o(Dd lab)
7. CALIBRATION OF EQUIPMENT AND TESTING PROCEDURES.

Adherence to apparatus specifications and testing procedures are
necessary but does not ensure that the cyclic triaxial test results from
one laboratory will be comparable to those of other laboratories.
Accordingly, a performance calibration has been developed using a
standard test sand, Monterey No. O,tt and standard test procedures.$
It is recommended that each laboratory verify its performance by cali-
bration using this standard test sand and techniques as follows:
-
-a. Standard Test Sand - Monterey No. 0. Figure 11 p r e s e n t s
the grain-size distribution and pertinent physical properties.
b. Specimen Preparation. Specimens, 2.8 in. in diameter by
-
6-7 in. high, are to be prepared to an initial density of 98.5 pcf (60 per-
cent relative density) using the moist tamping procedure and moist
tamping rod described in paragraphs 3h
- and 4a.
-
c.
- Effective Confining Pressure. The effective confining
t H. B. Seed, and I. M. Idriss, A Simplified Procedure for Evaluat-

ing Soil Liquefaction Potential, Journal of Soil Mechanics and
Foundations Engineering, ASCE, Vol 97, No. SM9, Sep 71.
tt Available from Lone Star Company, 9315 San Leandro St., Oakland,
Calif. 94603. Small quantities available from C. K. Chan,
Richmond Field Station, University of California, Berkeley,
Richmond, Calif., while supplies last. An alternative sand and
corresponding calibration curve available from WES Soils Research
Center, Vicksburg, Miss. 39180.
t Mulilis et al., op. cit.; Silver, op. cit.; and Silver et al., op cit.
XA- 30 .-
E M 1410-2-1906
Appendix XA
ange 1
Y May 80
pressure to be used is 14.50 psi. Sufficient back pressure to achieve a
B value greater than 0.95 is required.
d. Testing Frequency. 1 Hz.
-
e . T e s t R e s u l t s . Figure 9 presents the range in acceptable
-
values for stress ratio versus number of cycles to initial liquefaction
and 40 percent double amplitude strain.
8. POSSIBLE ERRORS. In addition to those described in paragraph 9,
Appendix X, TRIAXIAL COMPRESSION TESTS, the following are
possible errors that would cause inaccurate cyclic triaxial strength
determinations :
a. Apparatus.
- (f) Loading wave form. Nonuniform, eccentric
load wave forms may result from excessive piston friction, improper
gain setting, or operation of the closed-loop servo valve. Insufficient
air or hydraulic fluid at failure conditions will cause unacceptable
load reduction. Misalignment between the loading piston and load
actuator may also cause unacceptable loading wave forms.
(2) Electronic transducers and recording equipment. Im-
proper calibration or sensitivity of the electronic transducers, in-
correct balancing of amplifiers or zero settings, or improper range
settings will result in inaccurate recording of actual loads, deforma-
tions, and pressures occurring during the test. It is also essential
that the recorder response be rapid enough to follow all changes in the
transducer output.
b. Specimen Preparation and Testing.
- (1) S p e c i m e n d i m e n s i o n s
not measured precisely or density improperly calculated. A circum-
ferential tape for measuring specimen diameter is recommended for
obtaining precise measurements. Twice the thickness of the membrane
must be subtracted for measurements of single membrane-encased
specimens. An improperly calibrated burette will lead to incorrect
volume change measurements during consolidation with resulting
errors in specimen density computations.
XA-32
EM 4110-2-1906
Appendix XA
Change 1
1 May 80
(2) Percent undercompaction in lower specimen layers
improper for achieving uniform density.
(3) Incomplete saturation resulting in low B values. A
variety of problems can cause low B values: (a) use of insufficiently
de-aired water may prevent dissolving of air in the specimen without
resorting to extremely high back pressures; (b) incomplete de-airing
or saturation of pore pressure transducer and drainage lines (can be
avoided by applying a vacuum); and (c) system leaks due to punctured
membrane, poor membrane sealing to cap and base, loose fittings, or
improperly designed O-ring grooves (can be detected by using a bubble
chamber while applying vacuum to the system).
(4) Incorrect application of consolidation stresses.
Whenever the loading piston is fixed to the specimen cap, the static
uplift load equal to the area of the piston rod multiplied by the chamber
pressure must be accounted for when applying pressures, whether
during back-pressure saturation, B-value checks, or consolidation.
(5) Prestraining specimens. By allowing the effective con-
fining stress to vary, specimen strength can be greatly increased.t
(6) Scale factor for recorder traces not in agreement with
actual data measurements. Reduction of data from recorder traces
where incorrect scale factor, i.e., inches on recorder trace per pound
(load), per inch (deformation), or per pounds per square inch (pressure),
is used will result in data reduction error. The calibration steps should
always be recorded on the recorder trace prior to and after testing.
t Townsend, op. cit.
XA- 33
E M 1110-2-1906
Change 1
1 May 80
APPENDIX XB:
DETERMINATION OF CRITICAL VOID RATIO
i. DEFINITION OF CRITICAL VOID RATIO. Critical void ratio is

that void ratio at which a cohesionless soil can undergo deformation
or actual flow without volume change. In general, it is a function of
confining pressure, i.e., an increase in confining pressure yields a
lower value of critical void ratio. Theoretically, a soil with a void
ratio above the critical value for the confining pressure is subject to
flow failure or liquefaction if it undergoes sufficient undrained stress,
whether that stress is cyclic or monotonic (steadily increasing). How-
ever, if the void ratio is lower than the critical value for the given
confining pressure, the soil will not liquefy under monotonic stress
increase. Liquefaction resulting from monotonic stress increase and
initial liquefaction occurring under cyclic stresses are not the
same phenomenon. For the purposes of this appendix, liquefaction
refers to the behavior of a mass of cohesionless soil during flow
slides. For monotonic loading conditions, failure is caused by a sub-
,
stantial reduction in shear strength due to large increases in pore
pressure and the consequent great reduction of effective stress.
2. PRINCIPLES OF MONOTONIC TRIAXIAL TESTING. The mono-
tonic stress-controlled triaxial compression test is used to determine
the approximate values of critical void ratio over an applicable range
of confining pressure. The test is comparable to an R test with pore
pressure measurements (see Appendix X, TRIAXIAL COMPRESSION
TESTS) with the exception that load is applied to the specimen in
increments. The increments are arbitrary but should be chosen to
produce several points on the stress-strain curve prior to peak devi-
ator stress or the tendency to dilate, depending on the response of the
XB-4
E M 1110-2-1906
Appendix XB
Change 1
1 May 80
specimen as outlined below. The ability to execute an efficient test
program to establish critical void ratio over a range of confining
pressures requires recognition of the nature of specimen behavior.
In general, three types of response from specimens can be antici-
pated, as illustrated in Figure 4.
I I C-DILATANT FA,,?E
B-,M,TED LIOUEFACTION FA,LRE
, . /_A~,OEFACT!ON FA,LRE
------____---
Figure 1. Typical stress-strain and pore

pressure response, monotonic triaxial E
tests on sands
-a. Complete
- Liquefaction. If the void ratio after consolidation
is sufficiently higher than the critical value for the given confining
p r e s s u r e , the specimen will bear several load increments with minor
axial strain (usually less than 2 percent) while exhibiting steadily
XB-2
-
E M 1410-2-1906
Appendix XB
Change 1
1 May 80
accelerating pore pressure increases until failure by liquefaction oc-
curs catastrophically at peak deviator stress. During the collapse of
the specimen, which occurs over a fraction of a second, the deviator
stress actually borne by the specimen declines rapidly while the pore
pressure approaches, but not necessarily equals, the confining pressure.
-b. Partial or Limited Liquefaction. If the void ratio after con-

solidation is slightly higher than or very nearly equal to the critical
value for the given confining pressure, the specimen will tolerate
several load increments with small axial strain until at peak deviator
stress sustained deformation will occur. The deviator stress will
essentially remain constant or suffer minor decline during the
deformation. In some cases, deformation will cease, but the addition
of small increments of load that counter the effects of area increase
due to strain will maintain deformation. During the deformation, the
pore pressure will rise to a value somewhat less than the confining
pressure. In the latter stages of the test, the specimen response may
become dilative, i.e., it will bear additional deviator stress, and the
pore pressure will begin to decrease.
-c. Dilatant Failure. If the void ratio after consolidation is lower

than the critical value for the given confining pressure, the specimen
will tend to dilate, i.e. continue to accept loading, and the pore pressure
decreases.
3. APPARATUS. All the equipment is listed in Appendix X, T R I A X I A L
COMPRESSION TESTS, for performing triaxial compression R tests
with pore pressure measurements plus (a) electronic transducers and
high-speed recorders for data acquisition, and (b) stress-controlled
loading capacity.
a. Loading Devices. The equipment may be either (1) a pneu-
-
matic system, (2) a closed-loop electrohydraulic system, or (3) a
deadweight system. A deadweight system is practical only for i.4-
in.-diameter specimens. Such small specimens may be used, but
XB-3
E M 1110-2-1906
Appendix XB -
Change 1
1May 80
they are undesirable from the standpoint of specimen preparation
(see Appendi.x X, TRIAXIAL COMPRESSION TESTS, pp X-35 and
X-16). Figure 2 shows a satisfactory arrangement for deadweight
loading, which consists of a horizontal loading bar bearing on the load
rod (piston) and connected to a weight hanger suspended beneath the
triaxial chamber. A stopblock must be provided beneath the weight
hanger that will permit deformation of the specimen equivalent to not
less than 30 percent axial strain but will prevent contact of the
horizontal lo,ading bar with the top of the triaxial chamber. The
loading assembly should be designed to support 300 lb and constructed
of lightweight material. Pneumatic or closed-loop electrohydraulic -
systems must be capable of maintaining the axial load, i.e., supply
sufficient air or oil to allow the piston to follow the sudden and rapid
deformation during liquefaction.
b. e:cimen Cap. The specimen cap should be of a lightweight
-
noncorrosive material equipped with porous .metal or porous stone -
inserts and drainage connections. The cap can be similar to those in

Figure 4 of Appendix X, TRIAXIAL COMPRESSION TESTS, but
lubricated end platens similar to those used by Castrot are more
desirable . Preferable specimen diameter is 2.8 in. or larger.
c . T-
- r i a x i a l C o m p r e s s i o n C h a m b e r . The chamber is the same
as that in paragraph 3b - of Appendix X, TRIAXIAL COMPRESSION
TESTS, or the newer design shown in paragraph 3c
- of Appendix XA,
CYCLIC TRIAXIAL TESTS, except that the rigid connection of the
loading piston to specimen cap is not required.
-d. Loading
- Piston to Loading Equipment Connection. See
paragraph 3d- of Appendix XA, CYCLIC TRIAXIAL TESTS.
-e. Recording
- Equipment. See paragraph 3e - of Appendix XA,
CYCLIC TRIAXIAL TESTS.
t G . C a s t r o , Liquefaction of Sands, Harvard Soil Mechanics Series

No. 81, Jan 1969, Cambridge, Mass.
XB-4
-
E M 1110-2-1906
Appendix X B
Change 1
1 WY 80
J LOADING BAR
7 -kJ-
TRIAXIAL C H A M B E R
i NARROWWORKBENCH
OR CANTILEVERED
PLATFORM
I
A-
LOADHANGER
NOTE C L E A R A N C E B MST B E
EQUIVALENT TO NOT LESS
T H A N 30 P E R C E N T A X I A L
S T R A I N A N D L E S S THAN
C L E A R A N C E A.
a
f
STOP BLOCK
Figure 2. Deadweight loading system
XB- 5
E M 1110-2-1906
Appendix XB
Change 1
J_s May 80
- f. M e-a s u r e m e n t T r a n s d u c e r s . See paragraph 3f- of Appendix
XA, CYCLIC TRIAXIAL TESTS.
g. Back-Pressure Saturation Equipment. See paragraph 3g
of Appendz<A, C Y C L I C T R I A X I A L T E S T S .
-h. Tamping
- Rod for Moist Tamping Specimen Preparation.
(Optional) See paragraph 3& of Appendix XA, CYCLIC TRIAXIAL
TESTS.
4. TESTING PROCEDURES. Specimen preparation, back-pressure
saturation, and consolidation procedures specified in paragraphs 4b-
and 7b(l)
- through 7b(7)
- of Appendix X, TRIAXIAL COMPRESSION
TESTS, are adequate and pertinent. However, preferred techniques
for handling undisturbed samples of cohesionless soils, preparing
remolded specimens, and anisotropic consolidation are as follows:
2. Reconstituted Specimen Preparation (Moist Tamping).
The procedure is as outlined in paragraph 4a- of Appendix XA,
CYCLIC TRIAXIAL TESTS, except that the initial water content of -
the material should be near the bulking water content, i.e., 5- 10

percent. This water content permits molding of specimens in the
vicinity of the critical void ratio, which is typically a relatively high
value, i.e ., usually between 10 and 40 percent relative density,
depending on the confining pressure.
b. Specimen Preparation (Undisturbeg). See paragraph 4b- of
-
Appendix XA, CYCLIC TRIAXIAL TESTS.
2. Specimen
- Measurement. Inasmuch as void ratio is a
principle of the test results, it is imperative that accurate initial
specimen dimension measurements and volume change determinations
during saturation and consolidation be made. Diameter measurements
should be made to the nearest 0.001 in. using a circumferential tape7
t Commercially available from PI Tape, Box 398, Lemon Grove,

Calif. 9 2 0 4 5 .
XB-6 -
E M 1110-2-3906
Appendix X B
Change 1
.l May 80
at the top, midheight, and bottom of the specimen or dial-gage
calipers at the same locations, except two readings should be taken
at each position by rotating the dial-gage calipers 90 deg to produce
a total of six diameter readings. Care must be taken not to deform
or disturb the relatively loose specimen during diameter measure-
ments. Height measurements to the nearest 0.001 in. at four locations
and weights to the nearest 0.1 g are recommended for 2.8-in.-diameter
specimens. Record the data on the data sheet. (Plate XB-1).
d. Preliminary Seepage Saturation. See paragraph 4d-of
-
Appendix XA, CYCLIC TRIAXIAL TESTS.
2. C o n s o l i d a t i o n . See paragraph 45 of Appendix XA, CYCLIC
TRIAXIAL TESTS.
-f. Monotonic Loading. If it does not already exist, a large air

pocket should be formed at the top of the triaxial chamber by drain-
ing water-t from the chamber but leaving enough water to cover the
top of the specimen. The air pocket is required, so that the large
rapid piston movements into the chamber at the onset of failure do
not create chamber pressure fluctuations. The procedure is as
follows:
(1) Record test number and specimen identification on
recorder trace.
(2) Zero the recorder and transducer outputs, and record
calibration steps and scale factors.
(3) Close valve F (D should have already been closed) as
shown in Figure 16 of Appendix X, TRIAXIAL COMPRESSION TESTS.
(4) Initiate incremental axial loading using an increment
t Compressed air instead of water may be used as the confining

medium provided saturation and consolidation procedures do not
exceed 8 hr.
XB-7
EM 1410-2-4 906
Appendix XB
Change 1 -
1 May 80
determined as described in (5) below. During loading, the chamber
pressure is :maintained constant, and the axial load, axial deformation,
and change in pore water pressure are recorded with time. Slow
recorder speeds can be used as long as the specimen is deforming
slowly; manual recording of load, pore water pressure, and deforma-
tion can be made as long as specimen behavior permits (this practice
can save timte in data reduction in that recorder traces are trouble-
some to read). Loading is continued on approximately l-min intervals
until either (a) the specimen liquefies or partially liquefies, or (b) the
specimen tends to dilate. A complete stress-strain record, such as
shown for liquefaction or limited liquefaction in Figure 1, is difficult to
obtain and requires experience- However, it is achieved by increasing
the recorder speed to 10 to 20 in. of record per second just as the spec-
imen indicates impending failure. The clues to approaching rapid fail-
ure are an obvious increase in deformation rate and steadily accelera-
ting increases in pore water pressure. If the specimen deforms
-
steadily but not catastrophically, additional load increments may be
applied as the straining tends to cease until the pore pressure begins
to decline.
(5) The magnitude of the monotonic load increments is
arbitrary but should be selected to provide several points on the
stress-strain curve prior to peak deviator stress. For a confining
p r e s s u r e , T3c, of 0.5 tsf, a reasonable increment would be equivalent

to about 4.0 :psi of deviator stress based on the initial area of the
specimen. If the confining pressure is 5.0 tsf, an increment equiva-
lent to about 5.0 psi is more practical. In any case, it is necessary to
reduce the load increment as peak deviator stress is approached in
order to prevent the application of the last increment from significantly
exceeding the true peak value. Confident use of such a procedure must
be learned through experience with initial tests on a given soil.
&. Z&ecimen R e m o v a l . Following testing, the specimen should
XB-8
E M 1110-2-i906
Appendix XB
Change 1 -
I May 80
or limited liquefaction, F
is taken as the difference between the
3f
consolidation confining pressure, T and the maximum pore water
3c
pressure generated during the test (see Figs. 1 and 3). Figure 4 illu-
strates construction of the critical void ratio, which is the approximate
relationship between e and ; 3f for the range of u3c The most
C
desirable test series includes two to three tests at each of several
pertinent confining pressures, u3c ; e.g., 0.5, 4.0, 2.0, 4.0, and 8.0
tsf. The objective is to build the specimens near the critical void ratio,
but slightly above it, to obtain complete and limited liquefaction in fail-
ure. This can be accomplished by trial. A trial procedure is not
wasteful since all tests are of value in establishing the critical void
ratio relatifonship. It is suggested that a pilot test be performed with
a specimen built to about 20 percent relative density and by applying a
moderate consolidation stress (about 2 tsf). The behavior of this speci-
men should indicate the approximate position of critical void ratio.
Smaller confining pressures would produce higher critical void ratios;
and larger confining pressures, lower critical void ratios. Those
specimens that tend to dilate upon loading serve to help fix a lower
-
bound to the position of the e versus the s-3f r e l a t i o n s h i p .
7. POSSIBLE ERRORS. In addition to those described in paragraph 9.
of Appendix X, TRIAXIAL COMPRESSION TESTS, the following are
possible errors that would cause inaccuracies in the determination
of critical void ratio.
a. A
- - p p a r a t u s . (I) L o a d i n g s y s t e m . Insufficient air or hydraulic
fluid at failure conditions will cause unacceptable load reduction. Mis-
alignment between the loading rod piston and load actuator or air piston
may also cause unacceptable loading conditions.
(2) Electronic transducers and recording equipment. Im-
proper calibration or sensitivity of the electronic transducers, incorrect
balancing of amplifiers or zero settings, or improper range settings
will result in inaccurate recording of actual loads, deformations, and
XB-40
E M ~110-2-~I.906
Appendix XB
Change 1 -
1 May 80
C O N S O L I D A T I O N P R E S S U R E m3c, TSF
1 .o 5.0 10.0
I I ljllll ll I pIIll
\,ASXJMED c~fnc,4L vo1~3 R A T I O
I
w
I
-\ RANGE IN INDICATED POSITION
OF CRITICAL VOID RATIO
I -
STRESS
-
-
4
I I
0.77 FAILURE RESPONSE
0 3c af
6 -6-7 LIQUEFACTION
>
0 l LIMITED LIQUEFACTION
A DILATIVE
0.75 I 1 1 I I IllIll I I IllIll

0.01 0.05 0.1 0.5 1 .o 5.0 10.0
EFFECTIVE CONFINING PRESSURE AT FAILURE E3f, T S F
Figure 4. Example of the determination of approximate

critical void ratio
pressures occurring during the test. It is also essential that the re-
corder response be rapid enough to follow all changes in the trans-
ducer output.
b_. Specimen Preparation and Testing. (I) S p e c i m e n d i m e n s i o n s
not measured precisely or density improperly calculated. A circum-
ferential tape for measuring specimen diameter is recommended for
obtaining precise measurements. Twice the thickness of the membrane
must be subtracted for measurements of single membrane-encased
specimens. An improperly calibrated burette will lead to incorrect
volume cha.nge measurements during consolidation with resulting errors
in specimen void ratio computations.
(2) Percent undercompaction in lower specimen layers im-
proper for achieving desired average density.
XB-12
EM gdjO-2-1906
Appendix XB
Change 1
1 May 80
(Uniform density in a reconstituted specimen is essential).
(3) Incomplete saturation resulting in low B values. The
following problems can cause low B values: (a) use of insufficiently
de-aired water may prevent dissolving of air in the specimen without
resorting to extremely high back pressures; (b) incomplete de-airing
or saturation of pore pressure transducer and drainage lines (can be
avoided by applying a vacuum); and (c) system leaks due to punctured
membrane, poor membrane sealing to cap and base, loose fittings, or
improperly designed O-ring grooves (can be detected by using a bubble
chamber while applying vacuum to the system).
(4) Assuming that the loading piston is not connected to the
specimen cap, the static uplift load equal to the area of the loading
piston multiplied by the cha.mber pressure must be accounted for when
applying axial stresses.
(5) Scale factor for recorder traces not in agreement with
actual data measurements. Reduction of data from recorder traces
where scale factor, i.e., inches on recorder trace per pound (load),
per inch (deformation), or per pounds per square inch (pressure), will
result in data reduction error. The calibration steps should always
be recorded on the recorder trace prior to and after testing.
XB-13
E M 1110-2-i906
30 Nov 70
APPENDIX XI:
UNCONFINED COMPRESSION TEST
1. INTRODUCTION. The unconfined compression test is used to meas-

ure the unconfined compressive strength of a cohesive soil. The uncon-
fined compression test is applicable only to coherent materials such as
saturated clays or cemented soils that retain intrinsic strength after re-
moval of confining pressure; it is not a substitute for the Q test. Dry or
crumbly soils, fissured or varved materials, silts, and sands cannot be
tested meaningfully in unconfined compression. in this test, a laterally
unsupported cylindrical specimen is subjected.to a gradually increased
axial compression load until failure occurs. The unconfined compression
test is a form of triaxial test in which the major principal stress is equal
to the applied axial stress, and the intermediate and minor principal
stresses are equal to zero. The unconfined compressive strength, q
U
is defined as the maximum unit axial compressive stress at failure or at

15 percent strain, whichever occurs first. The undrained shear strength,
S is assumed to be equal to one-half the unconfined compressive
U
strength. The axial load may be applied to the specimen either by the con-
trolled strain procedure, in which the stress is applied to produce a pre-
determined rate of strain, or by the controlled stress procedure, in which
the stress is applied in predetermined increments of load.
2. APPARATUS. The apparatus consists of the following:
a. Equipment for Preparing Specimen. A trimming frame as de-
scribed% paragraph 3~ of Appendix X, TRIAXIAL COMPRESSION TESTS,
or a trimming cylinder with beveled cutting edges may be used for trim-
ming specimens. The equipment should include wire saws and knives of
various sizes and types for use with the trimming frame. A motorized
soil lathe may be used advantageously under certain circumstances. A.
miter box or cradle is required to trim the specimen to a fixed length and
to ensure that the ends of the specimen are parallel with each other and
perpendicular to the vertical axis of the specimen.
b. Loa.ding Device . A number of commercially available
controll<d-strain or controlled-stress types of loading devices are suit-
able for applying the axial loads in the unconfined compression test. In
XI-1
EM iiiO-2-1906
Appendix XI
30 Nov 70
general, controlled- strain

type loading devices are
preferable, and the proce-
dures described herein are
based on the use of this type
of equipment. If available,
an automatic stress-strain
recorder may be used to
measure and record applied
axial loads and displace-
ments. A typical loading
device is shown in Figure 1.
Any equipment used should
be calibrated s o that the
loads actually applied to the
soil specimen can be deter-
mined. The required sensi-
tivity of stress-measuring
equipment for both controlled-
stress and controlled-strain
testing will vary with the
strength characteristics of
the soil. For relatively weak
soils (compressive strengths
less than 1.0 ton per sq ft),
i+gure 1. Typical unconfined compres- the unit load should be mea-
sion test apparatus
surable to within 0.01 ton per
sq ft. For soils with compressive strengths of 1.0 ton per sq ft or greater,
the loads should be measurable to the nearest 0.05 ton per sq ft.
c. Measuring equipment, such as dial indicators and calipers,
-
suitable for measuring the dimensions and axial deformation of a specimen
XI-2
E M lilO-2-i906
Appendix XI
30 Nov 70
to the nearest 0.001 in.

d.
- Timing device, either a watch or clock with second hand.
e.
- Balances, sensitive to 0.1 g.
-f. Other. Apparatus necessary to determine water content and

specific gravity (see Appendixes I, WATER CONTENT - GENERAL, and
IV, SPECIFIC GRAVITY).
3. PREPARATION OF SPECIMENS. 5. S p e c i m e n S i z e . U n c o n f i n e d
compression, specimens shall have a minimum diameter of 1.0 in. (prefer-
ably i.4 in.), and the largest particle in any test s p e c i m e n will be no
greater than one-sixth the specimen diameter. The height-to-diameter
ratio shall be not less than 2.i. Commonly used diameters of unconfined
compression specimens are 1.4 and 2.8 in. Specimens of 1.4-in. diameter
are generally used for testing cohesive soils which contain a negligible
amount of gravel.
b.
- Undisturbed Specimens. Generally, undisturbed specimens
are prepared from undisturbed tube or chunk samples of a larger size
than the test specimen. Core or thin-wall tube samples of relatively small
diameter may be tested without further trimming except for squaring the
ends, if the condition of the soil requires this procedure. Specimens must
be handled carefully to prevent remolding, changes in cross section, or
loss of moisture. To minimize disturbance caused by skin friction between
samples and metal sampling tubes, the tubes should be cut into short
lengths before ejecting the sam,.les. Sample ejection should be accom-
plished with a smooth continuou,s, and fairly rapid motion in the same
direction that the sample entered the tube. All specimens shall be pre-
pared in a humid room to prevent evaporation of moisture. The specimen
shall be prepared as follows:
(I) From the undisturbed sample cut a section somewhat
larger in length and diameter than the desired specimen size.
XI-3
E M iliO-2-1906
Appendix X I
30 Nov 70
It is generally desirable to prepare duplicate specimens for unconfined

compression testing, and selection of material for testing should be made
with this in mind.
(2) Carefully trim the specimen to the required diameter
using a trimming frame and various trimming tools (see Fig. 7, Appendix
X, TRIAXIAL COMPRESSION TESTS). Remove any small shells or
pebbles encountered during the trimming operations. Carefully fill voids
on the surface of the specimen with remolded soil obtained from the trim-
mings. Cut the specimen to the required length, using a miter box (see
Fig. 8, Appendix X, TRIAXIAL COMPRESSION TESTS). Where the pres-
ence of pebbles or crumbling results in excessive irregularity at the ends,
cap the specimens with a minimum thickness of plaster of Paris, hydro-
stone, or other support material. Care must be taken to insure that the
ends of the specimen are parallel with each other and perpendicular to the
vertical axis of the specimen.
-
(3) From the soil trimmings obtain 200 g of material for
specific gravity and water content determinations (see Appendixes I,
W A T E R C O N T E N T - GENERAL, and IV, SPECIFIC GRAVITY).
(4) Weigh the specimen to an accuracy of *0.01 g for 4.4-in.-
diameter specimens and *0.1 g for 2.8-in.-diameter specimens. If speci-
mens are to.be capped, they should be weighed before capping.
(5) Measure the height of the specimen with calipers or a
scale and the diameter with calipers or circumference measuring devices.
If the specimen is cut to a fixed length in a miter box, the length of the
miter box can be taken as the height of specimen for routine tests, and
additional height measurements are not usually necessary. It is always
advisable to measure the diameter of the specimen after trimming, even
though specimens are cut to a nominal diameter in a trimming frame.
Make all measurements to the nearest *0.01 in. Determine the average
initial diameter , D of the specimen using the diameters measured at
0
t h e top, Dt, c e n t e r , DC, a n d b o t t o m , Db, o f t h e s p e c i m e n , a s f o l l o w s :
XI-4
E M iilO-2-1906
Appendix XI
30 Nov 70
(6) If the specimen is not tested immediately after preparation,

precautions must be taken to prevent drying and consequent development of
capillary stresses. When drying before or during the test is anticipated,
the specimen may be covered with a thin coating of grease such as petro-
latum. This coating cannot be used if the specimen is to be used in a sub-
sequent remolded test.
c. Remolded Specimens. Remolded specimens usually are pre-
pared in conjunction with tests made on undisturbed specimens after the
latter has been tested to failure. The remolded specimens are tested to
determine the effects of remolding on the shear strength of the soil. The
remolded specimen should have the same water content as the undisturbed
specimen in order to permit a comparison of the results of the tests on
the two specimens. The remolded specimen shall be prepared as follows:
(i) Place the failed undisturbed specimen in a rubber mem-
brane and knead it thoroughly with the fingers to assure complete remold-
ing of the specimen. Take reasonable care to avoid entrapping air in the
specimen and to obtain a uniform density.
(2) Remove the soil from the membrane and compact it in a
cylindrical mold with inside dimensions identical with those of. the undis-
turbed specimen. The compaction effort is not critical since the water
contents of soils subjected to remolded tests are always considerably
wetter than optimum. Care must be taken, however, to insure uniform
density throughout the specimen. A thin coat of petrolatum on the inside
of the molding cylinder will assist in the removal of the specimen after
compaction.
(3) Carefully remove the specimen from the mold, preferably
by means of a close fitting piston, and plane off the top of the specimen.
T h e s p e c i m e n is then ready for testing.
XI-5
EM iiiO-2-i906
Appendix XI
30 Nov 70
(4) Follow the steps outlined in paragraphs 3h(4) and 3k(5).

a. Record all identifying information for the sample such as
project, boring number, visual classification, and other pertinent data on
the data sheet (see Plate XI-i which is a suggested form). The data sheet
is also used for recording test observations described below.
b. Place the specimen in the loading device so that it is centered
on the bottom platen; then adjust the loading device carefully so that the
loading ram or upper platen barely is in contact with the specimen. If a
proving ring is used for determining the axial load, contact of the platen
and specimen is indicated by a slight deflection of the proving ring dial.
Attach a dial indicator, sensitive to 0.001 in., to the loading ram to mea -
sure vertical deformation of the specimen. Record the initial reading of
the dial indicator on the data sheet (Plate XI-1). Test the specimen at an
axial strain rate of about 1 percent per minute. For very stiff or brittle
materials which exhibit small deformations at failure, it may be desirable -
to test the specimen at a slower rate of strain.. Observe and record the
resulting load corresponding to increments of 0.3 percent strain for the
first 3 percent of strain and in increments of 1 or 2 percent of strain
thereafter. Stop the test when the axial load remains constant or when
20 percent axial strain has been produced.
C. Record the duration of the test, in minutes, to peak strength
(time to failure), type of failure (shear or bulge), and a sketch of speci-
men after failure on the data sheet (Plate XI-2).
d* After the test, place the entire specimen or a representative
portion thereof in a container of known weight and determine the water
content of the specimen in accordance w.ith Appendix I, WATER CONTENT
- GENERAL.
5. COMPUTATIONS. The computations consist of the following steps:
a. From the observed data, compute and record on the data sheet
(Plate XI-i) the water content, volume of solids, void ratio, degree of
XI-6
EM 1110-2-1906
Appendix XI
30 Nov 70
Where the unconfined compress ive strength of a specimen is a lso ob-

tained after remolding, the selisitiv!ty ratio, St, s h a l l a l s o b e c a l c u l a t e d
and reported. The sensitivity ratio is defined as follows:
qu (undisturbed)
St = -----.-.-*
qC, (reniolded)
-7. POSSIBLE ERRORS. Following are possible errors that would c a u s e

inaccurate determinations of uncoilfirled c<)mpressive s t r e n g t h :
a.
- Test not appropriate to type of soil.
b.
- S p e c i m e n d i s t u r b e d \:,hile trilriming.
-
C.
- Loss of initial water content. A small change i.n water content
can cause a larger change in the strength of a clay, so it is essential that
every care be taken to protect the specimen against evaporation while
trimming and measuring, during the test, and when remolding a specimen
to determine the sensitivity.
d.
- Rate of strain or rate of loading too fast.
a. U S E O F O T H E R T Y P E S O F EQUI.PMENT F O R U N D R A I N E D S H E A R
STRENGTH DETERMINATIONS. Various other types of laboratory equip-
ment, such as cone penetrometers and vane shear apparatus, may be used
advantageously in the laboratory as a supplement to the basic unconfined
compression test equipment for determining the undrained shear strength
of cohesive soils. The use of these testing devices generally results in
savings in cost and time. However, the devices should b,e used with cau-
tion until sufficient data and procedural details are established to assure
their successful application. Use of such testing apparatus, as a rule,
XI -a
EMiiiO-Z-i906
30 Nov 70
APPENDIX XII:
RELATIVE DENSITY
4. GENERAL. Relative density expresses the degree of compactness of

a cohesionless soil with respect to the loosest and the densest conditions
of the soil that can be attained by specified laboratory procedures; a soil
in the loosest state would have a relative density of 0 percent and in the
densest state, of iO0 percent. The dry unit weight of a cohesionless soil
does not, by itself, reveal whether the soil is loose or dense, due to the
influence of particle shape and gradation on this property. Only when
viewed against the possible range of variation, in terms of relative den-
sity, can the dry unit weight be related to the compaction effort used to
place the soil in an embankment or indicate the volume-change tendency
of the soil when subjected to shear stresses. Originally, ielative density,
DdD was defined by the equation
e e
Dd, p e r c e n t = e maT e x ioo
max min
where
e Z void ratio of the soil in the loosest state which can be attained
max
in the laboratory
e = void ratio of the soil in place
e Z void ratio of the soil in the densest state which can be attained
min
in the laboratory
though, for ease of computation, relative density may be expressed in
terms of dry unit weight by the equation
Yd - yd min x Yd max
Dd, percent = x ioo
Yd max - Yd min Yd
E M lliO-2-i906
Appen? ix XII
30 Nov 70
where
= dry unit weight of the soil in place, called in-place detbsili
yd
= dry unit weight of the soil in the loosest state which caI> ho
Yd min
attai.ned in the laboratory, called minirr~um density
= dry unit weight of the soil in the densest state which ~911 t:r
d max
attained in the laboratory, called ma~imurn density.
The in-place density of a soil may be determined by various methods

in the field or from undisturbed samples brought into the laboratory; its
determination will not be considered in this test procedure which covers
only the methods for determining the minimum and maximum densities.
There are few difficulties in determining minimum clensity by a standard
method, but restrictions imposed by the availability of special apparatus
under certain conditions or the character of the soil may not permit the
determination of maximum density by a single, standard method to the e>:-
clusLon of all others. However, the method described herein, employing a
vibratory table, shc.Oil be prefe ed to all other methods;t the alternative --
method for attaining the max:: _&rn denseness of a soil by hammering on
the r ,bld (as described in App...ldix XIIA, M O D I F I E D P R O V I D E N C E
VIB1 .TED DENSITY TEST) should be followed only when the use of the
vibl, t,Jry table is not feasible. The method used for determining the
ma .l:num density must be reported when presenting relative density data.
2. APPARAWS. The apparatus for determining the minimum and
mil.ximum densities of a cohesionless soil shall consist of the following:
a. Cylindrical molds or measures of O.l-cu-ft capacity (6-in. in-
side diameter) and 0.5..cu-ft capacity (ii-in. inside diameter), as shown
in Figure i.$ The molds shouid be cast of silicon aluminum alloy
AS TM - ,:G7 OA.
b. Surcha.. ge assembl: e s , ~~ fit each size mold, as shown in
t Other methods are currently being investigated to overcome the prob-
lem of segregation encountered in some soils using the vibratory table.
$ These molds, as well as the surcharge assemblies and dial indicator
holder shown in Figure 2, are available commercially fxorn Stebbins
Mfg. & Supply co., i733 Blake St., Denver, Colo.
XII-2
E M 1110-2-1906
Appendix XII
30 Nov 70
Figure 2. Each assembly shall include a surcharge weight (equivalent to

2 psi), a surcharge baseplate with handle and a guide sleeve with clamp
assemblies.
c. Dial indicator, having 0.001-in. graduations and a 2-in. range.
d
-- Holder, for dial indicator, with collar, as shown in Figure 2.
e. Calibration bar, metal, 3 in. by 4/8 in. by 12 in. long.
-f. Vibratory table, as shown in Figure 2, w i t h a cushioned steel

vibrating deck about 30 by 30 in., actuated by
an electromagnetic vibrator and mounted to a
concrete floor or slab of large mass. The
vibrator should be a seminoiseless type with
a net weight over 100 lb. The vibrator shall
have a frequency of 3600 vibrations per rnin-
ute and variable vibrator amplitudes to a
maximum of at least O.Oi5 in. under a 250-lb
load, and be suitable for use with 230-volt
alternating current.t
& Hoist, having a capacity of at

least 300 lb, for handling the heavier sur-
charge weight.
h. Pouring device, as shown in
Figure 3, which has metal funnels with i/2-
and i-in.-diameter cylindrical spouts, each
attachable to a metal can 6 in. in diameter
by 12 in. high.
1. Hand scoop, large, metal.
i. Mixing pans, large, m&al.
k* Sample splitter or riffle.

Figure 3. Pouring
1. Straightedge, i5-in., steel.
device, with i-in.-
m.
- Platform scales, having a capacity diameter spout
1 The Syntron V P - 8 0 a n d V P - 2 4 0 V i b r a t o r y P a c k e r s , m a n u f a c t u r e d
by Syntron Co., Homer City, Pa., have proven satisfactory. (The
VP-80 has been replaced by a later model, VP-86, with the same
characteristics.)
XII-5
E M 11iO-2-1906
Appendix XII
30 Nov 70
of at least iO0 lb and sensitive to 0.01 lb.

n.
- Oven (see Appendix I, WATER CONTENT - GENERAL).
3. CALIB_.ATION O F E Q U I P M E N T . Each mold must be calibrated as
follows :
a.
- Determine the weight, of each mold to the nearest 0.03 lb.
W
m
b.
- Determine the capacity, or total inside volume, V of the
m
0.1 -cu-ft mold to the nearest 0.0001 cu ft and of the 0.5 -cu-ft mold to the
nearest 0.001 cu ft.
C. Determine the inside cross-section area, A
at the open end
m
of the 0.1-cu-ft mold to the nearest 0.0001 sq ft and of the 0.5-cu-ft mold
to the nearest O.OOi sq ft.
d* Determine the initial dial reading, ho, for each mold and sur-
charge baseplate combination in the following manner:
(i) Measure the thickness of the calibration bar, t and of
C
the surcharge baseplate, ts, to the nearest 0.001 in.
( 2 ) L a y t h e calibrati+n bar across the top of the mold along
the axis defined by the bracket.s for the dial indicator holder, as shown in
Figure 4a, see page 9.
(3) Insert the dial indicator holder into the brackets on one
side of the mold so that the dial indicator stem rests on the calibration
bar, and note the dial reading. Then insert the dial indicator holder into
the brackets on the opposite side of the mold and note the dial reading.
The dial indicator holder should be placed in the same position in the
guide brackets for each reading by means of matchmarks on the holder
and on the brackets.
(4) Compute the average, hr, of the two dial readings.
(5) Compute the initial dial reading, ho, by the equation
h = hr + t - t
o c S
4. PREPARATION OF SAMPLE. The soil to be t<Bst.cd lnu:<: !)<a oven-<lriktcl
XII-6
E M ,l1iO-2-1>06
Appendi>. XII
30 Nov 70
and then permitted to cool in an airtight container. Aggregations of

fine particles shall be thoroughly broken ar 1 a represeittative sample
removed from the soil using a sample splitter or riffle. The repre-
sentative sample should weigh at least 25 lb if the maximum particle
size is less than 1 -I/2 in. and at least 100 lb if the m a x i m u m p a r t i c l e
size is between I -I/2 and 3 in. If the sample contains more than
10 percent by weight particles larger than 3 in. the determination of
the maximum Lnd minimum density becomes a test of a research
nature.
5. PROCEDURE. Every precaution must be observed while handling the
sample to prevent segregation and to preserve the oven-dried condition.
The minimum and maximum densities shall be determined using the same
sample, by first placing the soil into a mold in the loosest possible state
to attain the minimum density and then vibrating it into the densest possi-
ble state to attain the maximum density. The O.f-cu-ft mold shall be used
for the determinations if the maximum particle size is less than f-i/2 in.
and the 0.5-cu-ft mold shall be used if the maximum particle size is be-
tween f-f/2 and 3 in.
-a. Minimum Density Determination. The procedure for deter-

mining the minimum density shall consist of the following steps:
(I) Record all identifying information for the sample. such as
project, boring number, etc., on a data sheet (Plate XII-1 is a suggested
form).
(2) Record the weight, inside volume, and end area of the
mold on the data sheet.
(3) Carefully place the oven-dried soil into the mold in the
loosest possible condition.7 Fill the mold in layers, using the pouring
device for material having maximum particle sizes less than 3/8 in. 11:
the hand scoop for material having larger particle sizes. Exercise tl1.e
.
greatest care at all times to avoid jarring the mold or otherwise disturb-
ing the previously placed layers. When using the pouring device (wi:h the
t If the maximum density of the oven-dried sample is to be determined

also, the mold may be attached to the deck of the vibratory table before
filling (see paragraph 5h( 2)).
XII- 7
mvfiiiO-Z-i906
Appenc%x XII
30 Nov 70
4/2 -in. -diameter spout if the maximum particle size passes the No. 4 sieve
and the $-in. -diameter spout if the maximum particle size is between the
NO. 4 sieve and 3/8-in. sieve), adjust the height of the spout to maintain a
free fall of the soil of about 1 in. With a steady flow of soil from the spout,
move the pouring device in a spiral path from the outside to the center of
the mold to form each layer of uniform thickness without segregation.*
When the maximum particle size of the sample exceeds 3/8 in., place the
soil into the mold by means of the scoop held as closely as possible to the
previously placed layer so the soil slides but does not fall from the scoop;
restrain the larger particles with the hand where necessary to prevent their
rolling from the scoop. Continue filling the mold until the soil rises slightly
above the top of the mold, with care that no large particles which project
above the top of the mold are placed in the final layer. Using the straight-
edge, carefully trim the soil surface level with the top of the mold.
(4) If the maximum density of the oven-dried sample is not to
-
be determined,$ weigh the mold and soil to the nearest 0.04 lb and record
the weight on the data sheet; ait?rnatively, the contents of the mold may be
emptied into a mixing pan and weighed.
(5) Steps (3) and (4) should be repeated until consistent re,sults
(within 1 percent) are attained.
-b.
-Maximum Density Determination With Oven-Dried Sample. - The
procedure for determining the maximum density shall consist of the fol-
lowing steps:
(i) Proceed in accordance with paragraphs 5&(1) through 5&(3).
(2) Attach the mold to the deck of the vibratory table, if this
* Static electricity in dry sand can cause bulking similar to that produced
by a trace of moisture on the particles; a static -eliminating balance
brush can be applied to the equipment in contact with the sand when this
effect becomes bothersome.
$ Sf the maximum density of the oven-dried sample is to be determined
a 1 so9 proceed in accordance with paragraphs 5k(2) through 5b(7).
XII-8
E M 1i10-2-3906
Appendix XII
30 Nov 70
had not been done prior to filling the mold with soil (see paragraph 5~(3)).
(3) Place the guide sleeve on the top of the mold and clamp it
f i r m l y to the mold.* Lower the s u r c h a r g e b a s e p l a t e onto the surface of
the soil and remove the handl;. Using the hoist if necessary, lower the
surcharge weight onto the surcharge baseplate.
(4) It has been determined that for a particular vibrating
table, mold, and surcharge assembly, the maximum dry density of a spec-
imen may be obtained at a displacement amplitude (rheostat setting) less
than the maximum amplitude of which the apparatus is capable; i.e., dry
density may increase with increase in rheostat setting to a setting, beyond
which the dry density decreases. Therefore each laboratory should
determine for its apparatus the rheostat setting at which maximum den-
sity is produced and use this setting for subsequent maximum density
testing.* A clean, durable, subrounded to rounded material should be
used in making these determinations (in both 0.1- and 0.5-cu-ft molds).
The test should be performed as given in this appendix, except that the
rheostat setting should be increased from zero to 100 in increments of i0
and measurements taken after each period of vibration; the material
should be vibrated for a period of 8 min at each rheostat setting and not
removed until after the last determination has been made (100 percent
rheostat setting). The particle size distribution of the specimen should
be determined before and after the test to assess the extent of degrada-
tion, if any.
(5) Remove the surcharge weight and guide sleeve from the
mold, and obtain dial indicator readings on opposite sides of the surcharge
* The inside surface of the guide sleeve must align with the inside sur-
face of the mold, so two of the three clamping bolts should be provided
with lock nuts, as noted in Figure 2. By properly adjusting and locking
these two bolts, the guide sleeve will be drawn automatically into cor-
rect alignment when the third bolt is tightened.
i It may be desirable to redetermine the optimum rheostat setting at the
inception of testing for each major project.
XII-9
E M 11iO-2-i906
Appendix XII
30 Nov 70
(3) Loss of material from mold before weighing. To pre-

vent spilling any material before the mold and contents can be weighed,
rap the side of the mold carefully to settle the contents.
c.
- Maximum Density Determination. (1) Insufficient amplitude
of vibratory table under load. Measurements should be made at least
once to verify that the requirements given in paragraph 2f
- are satisfied.
(2) (Rescinded)
(3) Loss of fine material from mold during vibration.

Fine particles may escape from the mold during vibration as dust.
(4) Misalignment of guide sleeve with mold. The adjust-

-
ment of the clamping bolts must be checked periodically to ascertain
that misalignment of the guide sleeve will not cause binding of the sur-
charge weight.
XII-14
EM 1iiO-2-1906
30 Nov.70
APPENDIX XUA:
MODIFIED PROVIDENCE VIBRATED DENSITY TEST
I. INTRODUCTION. The modified Providence vibrated density test is

used to determine the maximum density of a cohesionless soil when the
use of a vibratory table (as described in Appendix XII, RELATIVE
.DENSITY) is not feasible, especially in field laboratories. The apparatus
and procedure are similar to those used for the Providence vibrated den-
sity test,t except that the surcharge is ,applied with a 38.5-lb weight
(equivalent to 1 psi) instead of the iOOO-lb calibrated spring (equivalent to
26 psi). In this test, a sample of oven-dried soil is placed in a heavy
steel mold, compressed under the surcharge, and vibrated to maximum
density by repeatedly striking the side of the mold with a hammer.
2. APPARATUS. The apparatus shall consist of the following:
a. Cylindrical steel mold having a capacity of about 0.16 cu ft and
an inside diameter of 7 in., as shown in Figure i.
b. Surcharge assembly, as shown in Figure 2, consisting of a sur-
-
charge weight with guide rod, a surcharge baseplate with lifting rod, and a
guide yoke with clamping bolts.
C. Depth gage consisting of a dial indicator, having 0.004 in.
graduations and a 2-in. range, fitted with a depth gage base and an extra
long stem to reach from the top of the guide yoke to the top of the sur-
charge baseplate.
d.
- Calibration bar, metal, 2 in. by i/8 in. by 7-i/2 in. long.
e.
- Hamme r, 2 - i/2 -lb ball peen.
4 Pouring device (as shown in Figure 3 of Appendix XII,
--
7 U. S. Army Engineer Division, New England, CE, New England Division
Laboratory Procedure for Performing the Providence Vibrated Density
Test to Determine the Maximum Density of Cohesionle ss Soils
maltham, Mass., August i957).
XIIA - 1
EM iIIO-2-1906
Appendix XHA
30 Nov 70
-8otxaE
EXTRA STROMZ
STEEl. PI&?
SCREWS, EwLr
YALF SECTiON SPACED
Figure 50 Details of mold
XL4 - 2
EM li40-2-i906
Appendix XIIA
30 Nov 70
L_ 4
ELEVATION
1
~WAFGE WEIGHT
&LEVATON
SURCHARGE EIASEPLATE
GUIOE ROQ
Figure 2. Eq ulp ment for maximum density determination
XIIA-3
E_M iiio-2-i906
Appendix XIIA
30 Nov 70
RELATIVE DENSITY) having metal funnels with i/2 - and 1 -in. -diameter
cylindrical spouts, each attachable to a metal can 6 in. in diameter by 42
in. high*
& Hand scoop large, metal.
t ;at;;e yDn;;tJ;r:r;;f;;l*
L . . .
i- Straightedge9 l5-in., steel.
& Balance or scale of at least 25-lb capacity sensitive to O.Oi lb.
L.- Oven (see Appendix I, WATER CONTENT - GENERAL).

3. CALIBRATION QF EQUIP&IENT. The mold shall be calibrated as
follows:
ae Determine the capacity9 or total inside volume, Vm, of the
mold to the nearest 0.0004 cu ft.t
!z* Determine the inside cross-section area, Am, at the open end
of the mold to the nearest O* OOO* sq ft.
C. Determine the initial dial reading, ho, in the following
manne r :
(1) Measure the thickness of the calibration bar, t and of
C
the surcharge baseplate, t , to the nearest 0.001 in.
(2) Attach the wide yoke to the mold, tightening the clamping
bolts firmly.
{3) Place the calibration bar inside the guide yoke so that it
lies on top of. the mold in alignment with the top of the yoke.
(4) Place the depth gage on the top of the yoke with the dial
indicator stem passing through one of the depth gage holes to rest on top
of the calibration bar, and note the dial reading. Then insert the stem
through the other hole and note the dial reading.
t Since t h e weight, WmB of this heavy mold, if included as a tare, would

detract from the accuracy of weighing the soil in the mold, the soil
should be removed from the mold for weighing.
XSIA-4 .-
EMiiiO-2-i904
Appendix XIIA
30 Nov 70
Compute the average, h of the two dial readings.

r
Compute the initial dial reading, h 0 by the equation
=hrtt t
hO C S
4. PREPARATION OF SAMPLE. The sample shall be prepared as de-

scribed in paragraph 4 of Appendix XII, RELATIVE DlZNSITY, except that
the maximum particle size of the sample must be iess than 2-4/2 in. The
sample should weigh at least 25 lb.
5. PROCEDURE. + Minimum Density Determination. The proce-
dure for determining the minimum density shall consist of the following
steps:
(4) Proceed in accordance with paragraphs 5$4) through 5$3)
of Appendix XII, RELATIVE DENSITY.
(2) If the maximum density of the oven-dried sample is not to
be determined,t empty the contents of the mold into a mixing pan and weigh
the pan and soil to the nearest O.Oi lb. Record the weight on the data sheet.
(3) Paragraph 5&(3) of Appendix XII, RELATIVE DENSITY,
and step (2) above should be repeated until consistent results (within 1 per-
cent) are attained.
-b Maximum Density Determination. The procedure f<,r deter-

mining the maximum density shall consist of the following steps:
(1) P.roceed in accordance with paragraphs 5&(i) through
5&(3) of Appendix XII, RELATIVE DENSITY.
(2) Rap the side of the mold gently with the hammer until the
surface of the soil settles at least I/8 in. below the top of the mold, taking
care that the surface of the soil remains parallel with the top of the mold.
(3) Lower the surcharge baseplate onto the surface of the soil
t If the maximum density of the oven-dried sample is to be determined

also, proceed in accordance with paragraphs 5h(2) through 5h(8).
XIIA-5
EM ii10-2-i906
Appendix XKA -
30 Nov 70
and remove the lifting rod. Place the surcharge weight on the surcharge
baseplate so that it is centered on the plate. Lower the guide yoke onto
the top of the mold, threading the guide rod through the center hole in the
top of the yoke, and clamp the yoke firmly to the mold.
(4) Strike the side of the mold between the flanges with sharp,
forceful blows of the ball-peen hammer, rotating the mold as each blow is
delivered. The blows*shall be uniformly distributed over the height and
circumference of the mold at a rate not exceeding 40 blows per minute
(5) When the vertical movement of the guide rod becomes
imperceptible, place the depth gage on the top of the yoke with the fil in-
dicator stem passing through one of the depth gage holes to rr /Dn top of
the surcharge baseplate, and note the dial reading. Then continue striking
the mold in cycles of about 25 blows, reinserting the depth gage into the
same hole after each cycle and noting the dial reading. The sample shall
be considered to have attained the maximum density when the change in
-
dial reading for any 25-blow cycle is less than 0.005 in.
(6) Obtain final dial readings with the depth gage inserted
through each of the holes i n the yoke. Record the dial readings on the
data sheet.
(7) Remove the guide yoke, surcharge weight, and surcharge
baseplate from the mold.
(8) Empty the contents of the mold into a mixing pan and
weigh the pan and soil to the nearest O.Oi lb. Record the weight on the
data sheet.
6. COMPUTATIONS. The computations shall be made as described in
paragraph 6 of Appendix XII, RELATIVE DENSITY.
7. POSSIBLE ERRORS. Besides the errors given in paragraphs 72
and %_ of Appendix XII, RELATIVE DENSITY, following are possible
errors that would cause inaccurate determinations of maximum density:
a. Loosening of yoke during vibration. The clamping bolts
should be checked before each dial reading to ascertain that the yoke is
XIIA - 6
EM j4iO-2--1906
Appendix XIIA
30 Nov 7C
tightly secured to the mold.

k Nonuniform distribution of hammer blows over the height and
circumference of mold.
5 Vibration of mold stopped too soon.
L+ Mold used as tare when determining weight of contents.
XIIA-7
E M 1110-2-1906
Appendix XIII
-
Change 1
1 May 80
known field performance of the soils.
2 . I D E N T I F I C A T I O N O F D I S P E R S I V E C L A Y S . a-. I n t r o d u c t i o n . Iden-
tification of dispersive clays may be required for earth structures not
yet constructed, for existing earth structures, and for natural soil
deposits. Positive identification of dispersive clays is by observed
performance of the soil in the field.* Dispersive clays cannot be iden-
tified by conventional index tests such as particle size distribution,
Atterberg limits, and compaction characteristics.
b. Laboratory Tests Used to Identify Dispersive Clays. Four
-
laboratory tests commonly used to identify dispersive clays are the
Crumb test, Soil Conservation Service (SGS) dispersion test, soil pore
water chemistry correlation, and the pinhole erosion test.yt
(1) The Crumb test (procedure given in Plate XIII-I) is often
used as an adjunct to other tests for identifying dispersive clays. HOW-
ever, the Crumb test is a useful indicator only in one direction. If the
Crumb test indicates dispersion (Crumb reading 3 or 4), the soil is
-
probably dispersive; however many dispersive soils, particularly
kaolinitic soils, do not react to the Crumb test (i.e., give Crumb read-
ings of I or 2).$
(2) The SCS dispersion test has been used to identify
* P . F o r s y t h e , Experience in Identification and Treatment of Dis-

persive Clays in Mississippi Dams, Dispersive Clays, Related
Piping, and Erosion in Geotechnical Projects, ASTM Special
Technical Publication No. 623, American Society for Testing and
Materials, May 1977, pp 135-f55.
tt J. L. Sherard et al., Pinhole Test for Identifying Dispersive Soils ,
Journal of the Geotechnical Engineering Division, Vol 102, No.
GT1, Jan 3976, pp 69-85.
Sherard, Dunnigan, and Decker, op. cit.
$ G. G. S. Holmgren, and C. P. Flanagan, Factors Affecting Spon-
taneous Dispersion of Soil Materials as Evidenced by the Crumb
T e s t , Dispersive Clays, Related Piping, and Erosion in Geotech-
nical Projects, ASTM Special Technical Publication No. 623,
American Society for Testing and Materials, May 1977, pp 218-239.
XIII - 2
E M j110-2-1906
Appendix XIII
Change 1
1 May 80
dispersive clays. Available results indicate for soils with SCS disper-
s ion < 35 percent, dispersive erosion will not be a problem; for soils
with SCS dispersion from 35 to 50 percent, dispersive erosion may or
m a y not occur; and for soils with SCS dispersion > 50 percent, disper-
sive erosion will be a problem.t,tt,$ The SCS dispersion test has about
85 percent reliance in predicting dispersive performance (about 85 per-
cent of dispersive soils show more than 35 percent SCS dispersion).tt
(3) Sherard et al.,%% have obtained a relationship between
dispersion and soil pore water chemistry based on pinhole erosion
tests and observed dispersion erosion in nature, as shown in Figure 1,
The soil pore water correlation has about 85 percent reliance in pre-
dicting dispersive performance.tt
(4) The pinhole erosion test is the most reliable test for
identifying dispersive soils.t,$$ Figure 2 shows a schematic repre-
sentation of the pinhole erosion test. In conducting the test, distilled
water under a low hydraulic head is caused to flow through a small
diameter hole in the soil specimen. For dispersive soils, the flow
emerging from the soil specimen is cloudy and the hole rapidly en-
larges. For nondispersive soils, the flow is clear and the hole does
not enlarge. The pinhole erosion test was developed for the purpose
.
of identifying dispersive soils and is not intended to be a geometri-
cally scaled model of an actual prototype structure. Since the theory
of similitude was not used in the design of the test, quantitative data
t Sherard, Dunnigan, and Decker, op. cit.

tt R. S. Decker and L. P. Dunnigan, Development and Use of the SCS
Dispersion,Test, Dispersive Clays, Related Piping, and Erosion
in Geotechnical Engineering Projects, ASTM Special Technical
Publication No. 623, American Society for Testing and Materials,
May 1977, pp 94-109.
$ N. L. Ryker, op. cit.
$$ J. L. Sherard et al., op. cit.
XIII- 3
E M 1110-2-1906
Appendix XIII
Change 1
1 Nay 80
K0
iI
1
. ii-
,.
-5
E M 1jiO-2-1906
Appendix XIII
Change 1
1 I%ay 80
are not obtained on the quantity of flow through the pinhole, amount of
soil erosion, or rate of soil erosion. Two limitations of the pinhole
erosion test for identifying dispersive soils have been observed. Un-
distrubed soil samples of high sensitivity (ratio of the peak undrained
strength of the soil in a natural state to the peak undrained strength
after it has been remolded without change in water content) may be
classified as dispersive from the pinhole erosion test, while in nature
the soil may be resistant to erosi0n.T Apparently, the natural structure
of the soil is destroyed by punching the pinhole in the undisturbed soil
specimen and a reaction analogous to dispersion is obtained in the pin-
hole erosion test. Soils with high sodium (> 80 percent ) and low total
dissolved solids (< 0.4 meq/!) in the soil pore water may show nondis-
persive in the pinhole erosion test, while the soil may exhibit dispersive
performance in the field.tt This may occur because a decrease in the
concentration gradient between the soil pore water and eroding fluid
(distilled water 2 0.0 meq/! for pinhole erosion test) results in a de- --
crease in the erosion rate for soils.* However, available data from case
histories indicate very few soils with total dissolved solids ci.0 meq/!
for which dispersive performance has been observed in the field.**
2. Field Tests Used to Identify Dispersive Clays. Four field
tests that have been used to identify dispersive clays are the Crumb
test (also used as a laboratory test), the ultraviolet light, the modified
0. Dascal, G. Pouliot, and J. Hurtubise, Erodibility Tests on a

Sensitive, Cemented Marine Clay (Champlain Clay), Dispersive
Clays, Related Piping, and Erosion in Geotechnical Projects, ASTM
Special Technical Publication No. 623, American Society for Testing
and Materials, May i977, pp 74-93.
Forsythe, op. cit.
A. Sargunam, Influence of Mineralogy, Pore Fluid Composition and
Structure on the Erosion of Cohesive Soils, Ph.D. Dissertation,
June 1973, University of California, Davis, Calif.
Sherard, Dunnigan, and Decker, op. cit.
XIII- 6
E M 1110-2-1906
Appendix XIII
;hE;eS;
hydrometer or Dilution-Turbidity test, and determination of soil pore

water chemistry by sodium electrode and chemical reagents or Wheat-
stone bridge.
(1) The Crumb test, which may be used in the laboratory or
field, has been described previously.
(2) The ultraviolet light has been used to indicate the pres-
ence of sodium in the soil.* Uranyl acid is mixed with the soil and the
intensity and amount of fluorescence under ultraviolet light is observed
The ultraviolet light has shown about 40 percent reliance in predicting
dispersive performance of soils in Mississippi.
(3) The modified hydrometer or Dilution-Turbidity test has
been used to identify dispersive clays.t,tt Prior to using the modified
hydrometer test to identify dispersive clays in a particular area, the
test results must be correlated with laboratory tests to establish a
range of values. For example, test data from the Bluff Hills region of
Mississippi indicate for a turbidity ratio < 4, dispersive erosion will
be a problem; for a turbidity ratio 4-9, dispersive erosion may or may
not occur; and for a turbidity ratio > 9, dispersive erosion will not be
a problem.*
(4) Two methods have been developed for determination of
soil pore water chemistry in the field to use with the correlation shown
in Figure 1 and to identify dispersive c1ays.t The first method in-
volves a sodium electrode and chemical reagents to determine the
percent sodium and total dissolved cations (calcium, magnesium, and
* Forsythe, op. cit.

77 Ryker, op. cit.
$ C. P. Flanagan and G. G. S. Holmgren, Field Methods for Deter-
mination of Soluble Salts and -Percent Sodium From Extract for
Identifying Dispersive Clay Soils, Dispersive Clays, Related
Piping, and Erosion on Geotechnical Projects, ASTM Special
Technical Publication No. -623, American Society for Testing and
M a t e r i a l s , M a y 1977, pp 321-134.
XIII-7
E M 1110-2-1906
Appendix XIII
Change 1
1May 80
sodium). This method does not determine potassium, which exists in
small quantities (5 1 .O meq/ 1) in most soils. The second method uses
a Wheatstone bridge to determine the percent sodium and total dis-
solved cations (calcium, magnesium, sodium, and potassium).
-d. Recommended Procedure for Identification of Dispersive

S
Col a
i lyss .w i t h t h e f r a c t i o n f i n e r t h a n 0 . 0 0 5 m m 5 12 p e r c e n t a n d
C 4 generally do not contain sufficient colloids
with a plasticity index -
to support dispersive erosion. The pinhole erosion test is the recom-
mended laboratory test for identifying dispersive clays. A Crumb test
(procedure given in Plate XIII-I) should be run on each soil tested in
the pinhole erosion apparatus. There is no generally reliable field
test for identifying dispersive clays.
(1) To identify dispersive clays for earth structures not yet
constructed, pinhole erosion tests will be conducted on compacted
soil samples taken from proposed borrow areas.
(2) To identify dispersive clays in existing earth structures
or natural soil deposits, pinhole erosion tests will be conducted on
undisturbed soil samples.
3. WES PINHOLE EROSION APPARATUS. Detailed drawings for the
U. S. Army Engineer Waterways Experiment Station (WES) pinhole
erosion apparatus are given in Plates XIII-2 to XIII-4. The apparatus,
shown schematically in Figure 2, is a simplified version of a laboratory
erosion test apparatus constructed previously at WES.t The apparatus
is designed to accommodate three,different size specimens: (a) com-
pacted specimens 4.312 in. in diameter by 2.846 in. high (Harvard
miniature compaction mold), (b) compacted specimens 4.00 in. in diam-
eter by 4.59 in. high (standard compaction mold), and (c) undisturbed
t E . B . P e r r y , Piping in Earth Dams Constructed of Dispersive

Clay; Literature Review and Design of Laboratory Tests, Techni-
cal Report S-75-i 5, fiov i975, U. S. Army Engineer Waterways
Experiment Station, CE, Vicksburg, Miss.
XIII- 8
EM ii:O-2-1906
Append ix XIII
Change 1
1 May 80
specimens encased in Shelby-tube 3.i25 in. in outside diameter by 4.60
in. high. Accessory equipment needed includes: (a) de-aired distilled
water; (b) graduated cylinders of iO-, 25-, 50-, iOO-, and 250-mI capac-
ity; (c) stopwatch reading to 0.i set; (d) centigrade thermometer with
range 0 to 5OC and accurate to O.iC; (e) modeling clay,t plus i/8- and
f/4-in. circular hardware cloths cut to fit inside the specimen tube;
(f) pea gravel, i/4- to 3/8-in. size; and (g) i/i6-in.-diameter steel
welding rod.
4. CALIBRATION OF PINHOLE EROSION APPARATUS. a. Concept.
To interpret the results of the pinhole erosion test and to develop a
classification system containing intermediate grades between dispersive
and nondispersive clays, it is necessary to determine the relationship
between quantity of flow and initial hydraulic head as a function of the
size of the pinhole, as shown in Figure 3. This relationship is deter-
mined by substituting aluminum cylinders with varying pinhole diameter
(Plate XIII-4) for the soil specimen and measuring the quantity of flow
for various hydraulic heads. Two calibrations are required: (1) 4.60-
in.-long aluminum cylinders with varying pinhole diameter, and (2) 2.82-
in.-long aluminum cylinders with varying pinhole diameters. These cal-
ibrations should be conducted for each pinhole erosion apparatus used.
b. Assembly of Apparatus. To calibrate the pinhole erosion
apparatus, the temperature of the distilled water is recorded both
before and after the test. The 4.6O-in.-long aluminum cylinder with the
i/i 6-in. -diameter pinhole is pushed into the specimen tube and
positioned longitudinally in the center of the specimen tube as shown
(for a soil specimen) in Plate XIII-2. The i/4-in. hardware cloth is
placed next to the exit end of the aluminum cylinder (away from the
distilled water entrance) and positioned so that the pinhole is centered
in an opening in the hardware cloth. The 1/8-in. hardware cloth is
t Clayola modeling clay, Binney and Smith, Inc., 380 Madison Avenue,
New York, N. Y. iOO17.
XIII-9
E M i1iO-2-1906
Appendix XIII
Change 2
20 Aug 86
*
30
0 2 4 6 8 IO I2 I4
HYORAULIC H E A D A C R O S S C Y L I N D E R , H, IN.
Figure 3. Typical calibration curve for pinhole apparatus using

4.6-in.-long s p e c i m e n
placed next to the 1/4-in. hardware cloth. Modeling clay is placed
around the perimeter of the hardware cloths to hold them in place. Pea
g r a v e l (i/4- to 3/8 -in. size) is placed next to the hardware cloths, and
the drain plate (Plate XIII-3) is attached to the esit end of the specimen
tube. The */4-in. hardware cloth is placed next to the entrance end of
the aluminum cylinder and pusitioned so that the pinhole is centered in
an opening in the hardware cloth. The 1/8-in. hardware cloth is placed
next to the i/4-in. hardware cloth. Modeling cla!? is placed around the
perimeter of the hardware cloths to hold them in place and to provide
a watertight seal between the beveled edge of the aluminum cylinder
XIII-IO
EM lilO-2-1906
Appendix XIII
Change 1
1 May 80
and the inside of the specimen tube. P e a g r a v e l (1/4- to 3/8-in. s.izej
is placed next to the hardware cloths and the pressure plate (Plate
XIII-31 is attached to the entrance end of the specimen tube, completing
the assembly of the pinhole erosion test apparatus.
c. Test Procedure for Calibration. Distilled water from the
-
constant head tank is caused to flow through the pinhole in the
aluminum cylinder under a hydraulic head of 1 in. for a period of 2 to
4min (until the quantity of flow becomes constant). The quantity of
flow is measured continuously with a stop watch (nearest 0.i set) and
g r a d u a t e d c y l i n d e r s (i0, 25, 50, 100, or 250 ml) and recorded on the
data sheet (Plate XIII-5). The quantity of flow is measured for hydrau-
lic heads of i, 2, 3, 5, 7, 9, ii, 13, and 15 in. This process is then
repeated using the aluminum cylinders with 1/8-, 3/16-, and 1/4-in.-
diameter pinholes. The hydraulic capacity (maximum quantity of
water which will flow through the pinhole erosion apparatus for a
given hydraulic head and pinhole diameter) of the pinhole erosion
apparatus may be reached with the i/4-in.-diameter pinhole at less
than 15 in. of hydraulic head. This completes the calibration for the
4.60-in.-long aluminum cylinders. A similar calibration will then be
conducted using the 2.8i6-in.- long aluminum cylinders with 1/16-,
f/8-, 3/16-, and i/4-in.-diameter p i n h o l e s .
g. Presentation of Calibration Results. The results of the
calibration test for the 4.6-in.-long aluminum cylinders are plotted in
a manner similar to that shown in Figure 3. The measured quantities
of flow from the calibration test are used to prepare a classification of
test results for the pinhole erosion test using 4.6-in.-long s p e c i m e n s
(see Fig. 4). The classification shown in Figure 4 is depicted
graphically in Figure 5.
5. PREPARATION OF COMPACTED SPECIMENS FOR PINHOLE
EROSION TESTING. To identify dispersive clays for earth structures
not yet constructed, pinhole erosion tests will be conductfd on
XIII- 11
EM 1110-2-1906
Appendix XIII
Change 2
?? Aug 86
Hi = 2 in.
IO min
I
I I I
DARK FLOW CLOUDYFLOW 1 ~SLIGHTL? CLOUDY FLOW 1 ICLEAR FLOW
,
I
I
Hi = 15 in.
10 min
I
NOTE: INITIAL PINHOLE DIAMETER = Di = l/l6 in.

FINAL PINHOLE DIAMETER = Df
ALLQIJANTITIESOF FLOWWEREOBTAINED FROM
THE CALIBRATION TESTS IN FIG.3
Figure 40 Typical sequence of testing and classification

of test results for pinhole erosion test using 406-in.L
long specimens
XIII- 12
-
E M IIIG-2-i9GL
Appendix XIII
Change 1
1 May 8 0
compacted specimens from soil samples taken from proposed borrow
areas. The soil sample shall be maintained at natural water content
prior to testing.t If the soil sample contains gravel-size particles,
they will be removed by pushing the material through the No. 4 sieve
(0.187-in. Opening). The soil specimen tested in the pinhole erosion
apparatus shall be compacted at the same water content and dry
density as those proposed for the earth structure in the field. If the
natural water content is greater than that proposed for construction,
the soil shall be air-dried to the construction water content. If the
natural water content is less than the proposed construction water con-
tent, distilled water shall be added to bring the soil to the construction
water content. The type of compaction (kneading, dynamic, or other)
used in the laboratory to prepare the soil specimens for pinhole
erosion testing shall simulate the field compaction. Where the soil
is to be compacted in the field with a sheepsfoot roller, the soil
specimens for pinhole erosion testing shall be compacted with a
laboratory kneading compaction apparatus (see Appendix X, TRIAXIAL
COMPRESSION TESTS, pp X-14 and X-i5).tt When the soil is to be
compacted in the field with hand-operated tampers, such as in the
vicinity of an outlet conduit, the soil specimens for pinhole erosion
testing shall be compacted with a laboratory dynamic (impact) com-
paction apparatus (see Appendix VI, COMPACTION TESTS, p VI-i3).
Following compaction, the soil specimen for pinhole erosion testing
shall be tested immediately (the soil specimens shall not be allowed
to cure following compaction) in the pinhole erosion apparatus.
6. PREPARATION OF UNDISTURBED SPECIMENS FOR PINHOLE
t Shzrard, et al., op. cit.

11 S . D . W i l s o n , Suggested Method of Test for Moisture-Density
Relations of Soils Using Harvard Compaction Apparatus,
Proceedings for Testing Soils, American Society for Testing and
Materials, 4th Edition, Dee 1944, .pp 460-i62.
XIII- 14
-
E M iiiO-2-1906
Appendix XIII
Change 1
1 May 80
EROSION TESTING. To identify dispersive clays in natural deposits
or existing compacted earth structures, the pinhole erosion test will
be conducted on undisturbed soil samples. The soil sample is not
removed from the Shelby tube. A 4.6-in.-long section is cut from the
Shelby tube using a horizontal band saw, and the soil specimen en-
cased in the Shelby tube is placed in the pinhole erosion apparatus
for testing. If undisturbed block samples are available, a 4.6-in.-long
section of Shelby tube with a sharp cutting edge can be pushed into
the block sample to obtain a soil specimen for pinhole erosion testing.
7 . P I N H O L E E R O S I O N T E S T P R O C E D U R E . a-. G e n e r a l . T h e iden-
tifying information for the soil specimen is recorded on the data sheet
(Plate XIII-5 is a suggested form). The temperature of the eroding
fluid (distilled water) is recorded both before and after the test. The
curing time (time lapse between compaction and pinhole erosion test-
ing) is recorded for remolded specimens. Although zero curing time
is specified for the pinhole erosion test, some time will be consumed
in assembling the apparatus, punching the pinhole in the soil speci-
men, etc.
b. Assembly of Apparatus.
- To assemble the pinhole erosion
apparatus, the soi1 specimen is positioned longitudinally in the center
of the specimen tube as shown in Plate XIII-2 for a compacted speci-
men 4.00 in. in diameter by 4.60 in. high. The same procedure is
employed when testing an undisturbed soil specimen or a compacted
soil specimen 1.3125 in. in diameter by 2.816 in. high (these soil
specimens encased in adapters (Plate XIII-3) are positioned longitudi-
nally in the center of the specimen tube). The pinhole punch guide
assembly (Plate XIII-4) is attached to the exit end of the specimen tube
(away from the distilled water entrance), and a i/16-in.-diameter steel
welding rod is pushed through the pinhole punch guide until the rod
just touches (marks) the exit end of the soil specimen. The pinhole
punch guide is removed from the exit end of the specimen tube, and
XIII- 15
E M l.410-2-1906
Appendix XIII
Change 1
1 May 80
the 1/4-in. hardware cloth is placed next to the exit end of the soil
specimen and positioned so that the pinhole mark is centered in an
opening in the hardware cloth. The l/8-in. hardware cloth is placed
next to the 1/4-in. hardware cloth, and modeling clay is placed around
the perimeters of the hardware cloths to hold them in place. Pea
g r a v e l (4/4- to 3/8 -in. size) is placed next to the hardware cloths and
the drain plate (Plate XIII-3) is attached to the exit end of the specimen
tube. The pinhole punch guide assembly is attached to the entrance
end of the specimen tube, and a l/j6-in.-diameter steel welding rod
is used to pu:nch a pinhole longitudinally through the center of the soil
specimen. The pinhole punch guide assembly is removed, and the
1/4-in. hardware cloth is placed next to the entrance end of the soil
specimen and positioned so that the pinhole is centered in an opening
in the hardware cloth. The 1/8-in. hardware cloth is placed next to
the 1/4-in. hardware cloth, and modeling clay is placed around the
perimeter of the hardware cloths to hold them in place and to provide
a watertight seal between the soil specimen and the inside of the
specimen, tube, as shown in Plate XIII- 2. When the soil specimen is
encased in an adapter, modeling clay is useo to provide a watertight
seal between the beveled edge of the adapter (Plate XIII-3) and the in-
side of the specimen tube. Pea gravel (1/4- to 3/8-in. size) is placed
next to the hardware cloths, and the pressure plate (Plate XIII-3) is
attached to th.e entrance end of the specimen tube completing the
assembly of the pinhole erosion test apparatus.
c . T e s t P r o c e d u r e . The 1/4-in. pipe plug, located on top of the
-
specimen tube next to the manometer (Plate XIII-2) is removed, and
the entrance end of the specimen tube containing the pea gravel is
filled with distilled water. The i/4-in. pipe plug is placed back into
position, and distilled water from the constant head tank is caused to
flow through the pinhole in the soil specimen under a hydraulic head
of 2 in. for 10 min (if no flow occurs under the 2-in. hydraulic head,
XIII- 16
E M 4110-2-1904
Appendix XIII
Change 1
1 May 80
remove the pressure plate, pea gravel, and hardware cloths, attach
the pinhole punch guide assembly to the entrance end of the specimen
tube, and repunch the pinhole in the soil specimen). The quantity of
flow is measured continuously with a stopwatch (nearest 0.1 see) and
graduated cylinders (10, 25, or 50 ml) and recorded on the data sheet
(Plate XIII- 5). The color of the water in the flask (clear, cloudy, or
dark) is observed and recorded on the data sheet. At the end of the
$O-min flow under the 2-in. hydraulic head, the quantity of flow (ml/set)
is computed. If the quantity of flow under 2-in. hydraulic head in-
creases with time showing a cloudy or dark color and after a IO-min
flow is greater than the measured quantity of flow obtained when the
pinhole erosion apparatus was calibrated using the aluminum cylinder
with the 4/16-in.-diameter pinhole, i.e., Df > Di , the test is concluded.
If the quantity of flow under 2-in. hydraulic head does not increase
with time and remains clear and after a IO-min flow is approximately
equal to or less than the measured quantity of flow obtained when the
pinhole erosion apparatus was calibrated using the aluminum cylinder
with the $/16- in.-diameter pinhole, i.e., Df 5 D , the hydraulic head
i
is raised to 7 in., and the quantity of flow is measured continuously for
10 min. If the quantity of flow under a 7-in. hydraulic head increases
with time showing a cloudy color and after a 30-min flow is greater
than the measured quantity of flow obtained when the pinhole erosion
apparatus was calibrated using the aluminum cylinder with the 1/46-in.-
diameter pinhole, i.e., Df > D , the test is concluded. If the quantity
i
of flow under 7-in. hydraulic head does not increase with time and
remains clear and after a IO-min flow is approximately equal to or
less than the measured quantity of flow obtained when the pinhole
erosion apparatus was calibrated using the aluminum cylinder with
the i/j 6- in. -diameter pinhole, i.e., Df 5 Di , the hydraulic head is
raised to 15 in., and the quantity of flow is measured continuously for
40 min; then the test is concluded.
XIII- 17
E M 1110-2-1906
Appendix XIII
Change 1
1 May 80
8. PRESENTATION AND ANALYSIS OF PINHOLE EROSION TEST
R E S U L T S . 2. Presentation of Test Results. Plates XIII-6 to XIII-9
present the test results from the pinhole erosion test. These plates
are obtained from the calibration of the pinhole erosion apparatus
(see Fig. 3). Typical test results for a dispersive clay and a nondis-
persive clay are also shown in these plates.
b. Analysis of Test Results.
- The primary differentiation
between dispersive and nondispersive clays is given by the pinhole
erosion test results under a hydraulic head of 2 in.? For dispersive
clays, the quantity of flow under a hydraulic head of 2 in. continuously
increases and may reach a maximum value that is limited by the
hydraulic capacity of the pinhole erosion apparatus, in less than i0
min. For nondispersive clays, under a hydraulic head of 2 in. the
flow will be clear or slightly cloudy. As shown in Figure 4, the pin-
hole erosion test results are classified as Di , D2 , ND4 , ND3 ,
ND2 , o r ND1 .f Dis persive clays are classified as D or D
1 2
intermediate clays as ND or ND and nondispersive clays as
4 3
or ND
ND2 1 -
9. POSSIBLE ERRORS WITH PINHOLE EROSION TEST. As discussed
previously, undisturbed soil specimens of high sensitivity (ratio of the
peak undrained strength of the soil in a natural state to the peak un-
drained strength of the soil after it has been remolded without change
in water content), may be classified as dispersive from the pinhole
erosion test, while in nature the soil may be resistant to erosion.tt
Soils with high sodium (> 80 percent) and low total dissolved solids
(< 0.4 meq/1) in the soil pore water may show nondispersive in the
pinhole erosion test while the soil may exhibit dispersive performance
t Sherard et al., op. cit.

tt Dascal, Pouliot, and Hurtubise, op. cit.
XIII- 18
E M iiiO-2-1906
Appendix XIII
Change 1
1 MY 80
in the field.? H o w e v e r , available data from case histories indicate
very few soils with total dissolved solids < 1.0 meq/1 for which dis-
persive performance has been observed in the fie1d.t-t
t Forsythe, op. cit.

Tt Sherard, Dunnigan, and Decker, op. cit.
VIII- 19
E M 1110-2-1906
Appendix XIII
Change 1
1 MfZly 80
Crumb Test
I. The Crumb test is run using distilled demineralized water. Small

(5-10 g) crumbs of soil at natural water content are carefully placed
on the bottom of a iOO-ml clear glass beaker filled with distilled
demineralized water. The behavior of the soil crumb is observed and
readings are taken after 10 and 30 min.
2. Dispersion is detected by the formation of a colloidal cloud, which
appears as a fine misty halo around the soil crumb. The Crumb test
is rated for reaction or colloidal cloud formation as follows:
4 = no sign of cloudy water caused by colloidal suspension.
2 = bare hint of colloidal cloud formation at surface or soil crumb.
3 = easily recognized colloidal cloud covering one fourth to one
half of the bottom of the glass container.
4 = strong reaction with colloidal cloud covering most of the
bottom of the glass container. -
3. Since the Crumb test involves a small quantity of soil, several
tests should be run on each soil sample before making an evaluation.
The Crumb test may be used as an indicator of field performance of
dispersive soils using the following evaluation of soil crumb reaction:
No dispersion problem = 1
Possible dispersion problem = 2
Definite dispersion problem = 3 or 4
4. The Crumb test is a useful indicator only in one direction. If the

Crumb test indicates dispersion (Crumb reading 3 or 4), the soil is
probably dispersive; however, many dispersive soils, particularly
kaolinitic soils, do not react to the Crumb test (i.e., give Crumb
reading of 1 or 2).
PLATE XIII-1
XIII- 20

Laboratory Soils Testing

Uploaded by

Copyright:

Available Formats

Laboratory Soils Testing

Uploaded by

Document Information

Original Description:

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

Laboratory Soils Testing

Uploaded by

Copyright:

Available Formats

This document downloaded from All of the information, data and computer software

Dont forget to visit our companion site http://www.vulcanhammer.org

ENGINEER MANUAL E M 1110-2-1906 _

Change 2 dated 20 August 1986

ENGINEERING AND DESIGN

LABORATORY SOILS TESTING

HEADGUARTERS, DEPARTMENT OF THE ARMY

Thin manual wao prepared at the U. S. Army Engineer

Engineering and Design

1. This change to EM 1110-2-1906, 30 November 1970, provides the

a. Appendix VIII A; Swell and Swell Pressure Tests

b. Appendix XA: Cyclic Triaxial Tests

c. Appendix XB: Determination of Critical Void Ratio

d. Appendix XIII: Pinhole Erosion Test for Identification of

2. Substitute the attached pages as shown below:

3. Add Appendices VIII A, XA, XB, and XIII.

4. File this change sheet in front of the publication for future

FOR THE GRIEF OF ENGINEERS:

FORREST T. GAY, II:

DEPARTMENT OF THE ARMY E M iiiO-2-i906

Engineer Manual 30 November 1970

ENGINEERING AND DESIGN

LABORATORY SOILS TESTING

* 1. DEFINITION .............................................. l-l

This regulation supersedes EM 1110-2-1906, 10 May 65.

9. PREPARING CLAY SHALE MATERIAL

TYPES OF*SPECIFIC GRAVITY .............. IV-1

SHALE MATERIAL ................... V-24

1. DARCYS LAW FOR FLOW OF WATER THROUGH

2. APPARATUS, CALIBRATION OF EQUIPMENT,

6. POSSIBLE ERRORS ..................... V I I I A - 8 st

3. PREPARATION OF SPECIMEN ............. IX-5

5. COMPUTATIONS ...................... IX-12

6. PRESENTATION OF RESULTS ............. IX-i4

7. SIMPLIFIED PROCEDURE FOR DRAINED (S)

* 8. POSSIBLE ERRORS ..................... IX-16

1. PRINCIPLES OF THE TRIAXIAL COMPRESSION

4. PRINCIPLES OF CYCLIC TRIAXIAL TESTING l . l

* 5. COMPUTATIONS ...................... XA-24

i. DEFINITION OF CRITICAL VOID RATIO ....... XB-1

* 1. INTRODUCTION . . . . . . . . . . . . . . . . .e. . . . . XIIA- 1

i. CHARACTERISTICS OF DISPERSIVE CLAYS. ... XIII-:

* 6. PREPARATION OF UNDISTURBED SPECIMENS

AH Change in height during consolidation or before shear

* T Cyclic shear stress

Manual 30 November i970

LABORATORY SOILS TESTING

1. PURPOSE. This manual presents recommended testing procedures for

7 Glossary of terms and definitions in soil mechanics, Proceedings&

Office, Chief of Engineers, from the experience of the U. S. Army Engi-

An even greater variety in testing procedures exists for measuring

such as for viewing by designers or contractors or for research, soil samples

however, is the judgment and awareness necessary to adjust the tech-

+ M. J. Hvorslev, Subsurface Exploration and Sampling of Soils for

FOR THE CHIEF OF ENGINEERS:

WATER CONTENT - GENERAL

i. D EFINITION. Water content, w, is defined as the ratio, expressed

-a. Oven,- preferably of the forced-draft t y p e automatically con-

weight of 1000 g is recommended for material having a maximum particle