TRANSPORT and ROAD RESEARCH LABORATORY Department of the Environment Department of Transport TRRL LABORATORY REPORT 901 THE STRENGTH OF CLAY SUBGRADES: ITS MEASUREMENT BY A PENETROMETER by W P M Black Any views expressed in this Report are not necessarily those of the Department of the Environment or of the Department of Transport Pavement Design Division Highways Department Transport and Road Research Laboratory Crowthorne, Berkshire 1979 ISSN 0305-1293
16
Embed
TRANSPORT and ROAD RESEARCH LABORATORY Department of … · The CBR of a soil is the ratio expressed as a percentage, at any value of penetration, of the force On the CBR plunger
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
TRANSPORT and ROAD RESEARCH LABORATORY
Department of the Environment Department of Transport
TRRL LABORATORY REPORT 901
THE STRENGTH OF CLAY SUBGRADES: ITS MEASUREMENT BY A PENETROMETER
by
W P M Black
Any views expressed in this Report are not necessarily those of the Department of the Environment or of the Department of Transport
Pavement Design Division Highways Department
Transport and Road Research Laboratory Crowthorne, Berkshire
1979 ISSN 0305-1293
Abstract
1.
2.
3.
4.
5.
.
7.
8.
C O N T E N T S
Introduction
The penetrometer
The relationship between CBR and undrained shear strength C u
The relationship between qc and CBR
Comparison of measured CBR values and values estimated by the penetrometer
commercial purposes, provMed the source is acknowledged
Ownership of the Transport Research Laboratory was transferred from the Department of Transport to a subsidiary of the Transport Research Foundation on 1 st April 1996.
This report has been reproduced by permission of the Controller of HMSO. Extracts from the text may be reproduced, except for commercial purposes, provided the source is acknowledged.
THE STRENGTH OF CLAY SUBGRADES: ITS MEASUREMENT BY A PENETROMETER
ABSTRACT
Improved methods of pavement design require that soil strength should be characterised in terms of fundamental parameters. Although the strength parameter CBR widely used in present methods of road design is empirical in origin, it is shown that it can be related to the undrained shear strength of cohesive soils by one of two linear relationships, one characterising soils in an undisturbed overconsolidated state and the other soils that are remoulded. After validating these relationships it is shown that a penetrometer with a small-diameter cone that is capable of measuring the shear strength of a soil can be used to estimate its CBR. The penetrometer can then be used to char- acterise soil strength in terms of either shear strength or CBR.
1. INTRODUCTION
The basic structural function of a road pavement is to reduce the stresses generated by traffic loading in the
soil over which the road passes to a level which the soil will accept without failure: at the same time, in
carrying out this function, the pavement itself should not be stressed sufficiently to bring about its failure.
A major input to all methods of design must therefore be a measure of the strength of thesoil, ie its ability
to resist the stresses imposed by traffic loading. The input is required for the design of new roads and for
the design of the total or partial reconstruction of damaged existing roads. Although not required in some
• methods of strengthening existing pavements by overlaying, knowledge of soil strength can often improve
the effectiveness of such methods.
Laboratory testing of remoulded soils under moisture conditions appropriate to those to be encountered
in service is normally adopted in designing new roads. This is dictated by the difficulty of carrying out in-situ
testing of the relevant soils under appropriate moisture conditions particularly in cut. On embankments yet
to be built, in-situ testing at the design stage is, in any case, an impossibility. When designing the reconstruction
or strengthening of existing roads it is obviously preferable to determine the strength of the subgrade, under
equilibrium with its long-term moisture regime, by in-situ testing.
The development of a structural approach to the design of new, strengthened and reconstructed pave-
ments, ie an approach based on the real stress-strain behaviour of pavements is required to provide the
flexibility necessary to accommodate changing technical and economic circumstances. The input characterising
soil strength of such a design method must therefore be in terms of fundamental parameters. A further
requirement of any design test is that it should be carried out with equipment that is either readily available
(or that is not unreasonably expensive to acquire) and that can be operated effectively by testing personnel
of normal quality. In order to minimise the considerable amount of testing inevitable on the varying soils
normally encountered on a road contract, the test should also have a good experimental reproducibility.
The strength of subgrades has been most commonly assessed by measuring the Californian Bearing
Ratio (CBR) of the soil. This arbitrarily conceived test has now been carried out both on remoulded soil
in the laboratory and in-situ, and used in many countries in correlative studies with pavement thickness.
In the present design recommendations for new roads 1 the design input is a test on a remoulded laboratory
sample and the great majority of experience is in terms of this type of test; in-situ values are however also
sometimes used, particularly in reconstruction problems. However the coefficient of variation of a large
number of CBR tests can be as much as 17 per cent and because of its indifferent reproducibility and the
size of sample required for a single laboratory test (5 kg), testing is time-consuming and the handling of
materials becomes a problem on a large job.
For these reasons it is not an ideal test for assessing the strength of soil under experimental pavements;
the removal of enough material to derive a good mean CBR value could be achieved only at the expense of a
great deal of excavation and damage to an existing road pavement. Equally the use of an in-situ CBR test
would require too many large holes to get enough useful information.
An alternative approach to direct CBR testing is to infer the CBR value from in-situ penetrometer
tests using a penetrometer of small diameter. The advantages of using such a penetrometer is that large
numbers of measurements can be made quickly. When testing an existing pavement a number of measure-
ments can be carried out through a conventional core hole without excessive damage to the pavement; •
also the strength variation with depth can be easily measured. The disadvantages at present are that the
commonly used penetrometer, described later, is empirically calibrated and has an undefined factor of
safety, ie the equipment reads strength values that are less than the real ones. Its calibration makes no
allowance for differences in the behaviour of remoulded and undisturbed soils.
This Report investigates the characteristics of a static penetrometer in use at the Laboratory; its
ability to measure undrained shear strength (the usual role of penetrometers in soil mechanics) is assessed
and related to the stress developed on the penetrometer cone, to the shear strength and to the CBR of the soil.
2. THE PENETROMETER
Historically the particular static penetrometer used at the Laboratory over a number of years was the Soil
Assessment Cone Penetrometer developed to measure the ability of soil to carry military vehicles. Itwas
comprised of a 30 °-included.angle cone 12.5 mm diameter at its base on the end of a 600 mm long, 9.5 mm
diameter steel shaft, and the CBR value was indicated directly by a pointer operated by a calibrated com- pression spring.
The range of the instrument was 0 - 1 5 per cent CBR, the rate of penetration being 25 mm/sec. The
force necessary to record 15 per cent CBR was 667 N for a spring compression of 25.4 mm. The instrument
has a non-linear CBR scale derived empirically with an unknown factor of safety. Figure 1 compares the
scale with the original linear scale used before production models were available.
From the spring calibration and the original linear CBR scale, the relation between CBR reading and the pressure on the penetrometer cone qc (in kPa), is
This calibration will be checked against experimental data.
5. COMPARISON OF MEASURED CBR VALUES AND VALUES ESTIMATED BY THE PENETROMETER
Two soil conditions are examined. If it is suspected that the particular soil under investigation does not fall
into either category, some calibration testing would be required.
5.1 Undisturbed overconsolidated soils
Although this section is headed 'undisturbed overconsolidated soils' the materials investigated were
soils that have a low strain to failure and these are most commonly the undisturbed overconsolidated soils.
However many compacted fills which wet up to a low equilibrium suction beneath roads, the usual British
condition, also come within this category; they do so because they take on aspects of overconsolidated
soils, one of which is that the act of shearing causes a negative change in pore water pressure; this results
in a small strain at failure.
Over the years parallel strength measurements have been made of in-situ CBR and penetrometer values
derived from the Soil Assessment Cone Penetrometer using the original linear scale. Some of the results have
been published 13; other unpublished work has been made on motorways and full-scale experiments; all the
results are given in Figure 4(a). The dotted line is the original penetrometer calibration given in Equation 1
and is not drawn as the best fit through the points, although it seems to fit well.
Most of the results in Figure 4(a) rely on single measurements of CBR; they are therefore associated
with a coefficient of variation of up to 17 per cent. It is likely that quite a lot of the scatter can be attributed
to this fact. An exception is the results from the M6 motorway, shown as filled triangles; these were based on
the mean of five CBR measurements for each plotted point; they all fall close to the calibration line. In
Key's work 13 the penetrometer readings were averaged over a depth of soil that was greater than that charact-
erised by the in-situ CBR test and he attributed some of the scatter in his results to the effect of changes of
soil strength with depth. Some areas tested had a dried crust and some had been locally wetted by rain. The
open circles in Figure 4(a) are those that Key considered to most likely be in error because of the effects of
strength gradients. The filled circles represent more reliable measurements. Below a CBR value of 10, the
measurements are closely grouped around the line of equality. Even better agreement is achieved if the
mean strength only of each section is plotted as shown by the + signs. This confirms that much of the scatter
in Figure 4 is the result of using too few measurements of CBR to produce reliable average values.
Allowing for the obvious causes of scatter Equation 1 provides a good calibration for the penetrometer
and the estimated values of CBR are of acceptable accuracy.
5.2 Remoulded soils
The open and filled circles on Figure 4(b) are unpublished results obtained by Sparks at TRRL on
remoulded Brickearth and London Clay statically compacted into a CBR mould. The data shown by triangles
were obtained by Gnanayutham 14 using a larger cone size to test London Clay; an experimental correlation
was used to adjust his measurements. The dotted line through the points in Figure 4(b) is the penetrometer
calibration given by Equation 8. It is a good fit up to a CBR of about 9.
Further confirmation that Equation 8 was appropriate for remoulded soils was obtained from Scala's
results 7. These measurements of CBR values and qc were made on compacted fill subgrades beneath roads
or in-situ soil subgrades that had been heavily rolled. An analysis of the 80 CBR test results showed that the
stress on the CBR plunger was on average 51 per cent of the stress on the plunger at a penetration of
12.5 mm; ie they behaved as remoulded soils (see Section 3 above). There was considerable scatter in Scala's
results but his mean correlation was qc ; this is very close to Equation 8. CBR =
221
6. CONCLUSIONS
1. On average, CBR values estimated from measurements of penetration resistance compare well with direct measurements of CBR.
. For pavement-design purposes, where high precision is not required in strength measurements, two
calibrations cover normal circumstances. For normal road subgrades of overconsolidated soils at
equilibrium with shallow water tables, the linear penetrometer calibration is based on
% CBR =
345
where qc is the ultimate bearing capacity of the soil measured by the penetrometer measured in kPa.
For remoulded soils the corresponding relation is
% CBR = _ _
230
3. The above relations are compatible with the assumption that the undrained shear strength of a soil
(C u) and its CBR are linearly related but the relation depends on the soil history. For undisturbed
overconsolidated soils the relation is
C u CBR = _ _ where C u is in kPa
and for remoulded soils
C u CBR =
J.3eSr
4. It is recommended that penetrometers should be used with linear calibrations of soil strength.
7. ACKNOWLEDGEMENTS
The work described in this report was carried out in the Pavement Design Division (Division Head:
Mr N W Lister) of the Highways Department of TRRL.
8. REFERENCES
1. DEPARTMENT OF THE ENVIRONMENT, ROAD RESEARCH LABORATORY. A guide to the
structural design of pavements for new roads. Road Note 29, Third Edition. London, 1970
(H M Stationery Office).
. BRITISH STANDARDS INSTITUTION. Britfsh Standard BS 1377: 1975. Methods of test for
soils for civil engineering purposes. London, 1975 (British Standards Institution).
. BLACK, W and N W LISTER. The strength of clay subgrades: its prediction in relation to road
performance. Conference on Clay Fills. London, 1978 (Institution of Civil Engineers).
. SKEMFrON, A W. The bearing capacity of clays. Building Research Congress 1951. (Building
Research Station). pp 180-190.
. GOLDER, H O. The relationship of runway thickness and undercarriage design to the properties of
the subsoil. Airport papers 4, 5. Institution of Civil Engineers, London, 1946.
7
.
.
.
.
DAVIES, E H. An investigation of defects in roads on the Isle of Wight using three methods of
flexible pavement design. Department o f Scientific and Industrial Research, Road Research Laboratory Note No. RN/1089/EHD. January, 1949 (unpublished).
SCALA, A J. Simple methods of flexible pavement design using cone penetrometer. Proc. 2nd
WISEMAN, G and J G ZEITLEN. A comparison between CBR and shear strength methods in the design
of flexible pavements. Proc 5th Int. Conf. Soils and Found. Eng, Paris 1961. Vol 2, pp 359-363.
GIBSON, R E. Discussion on the bearing capacity of screwed piles and Screwcrete cylinders. Journal Inst. Ovil Eng. 34, 1950.
10. MEYERHOF, G G. The ultimate bearing capacity of foundations. Gebtechnique 2, 1951, pp 301-332.
11.
12.
13.
14.
15.
SANGLERAT, G. The penetrometer and soil exploration. Amsterdam, London and New York, 1972.
(Elsevier Publishing CO).
MARSLAND, A. A comparison of the results from static perietrometer tests and large in-situ plant
tests in London Clay. Department o f the Environment Building Research Establishment Current Paper CP 87/74. Garston, 1974 (Building Research Establishment).
KEY, J W. An investigation to compare in-situ CBR values measured on a heavy clay soil with a
prototype cone penetrometer against those obtained from a standard apparatus. Department o f
Scientific and Industrial Research, Road Research Laboratory Note LN/682/JWK, 1964 (unpublished).
GNANAYUTHAM, I. A laboratory investigation of the relation between CBR/dry density/suction/
shear strength/elastic modulus for London Clay. MSc Thesis, University of Surrey.
BLACK, W. A method of estimating the California Bearing Ratio of cohesive soils from plasticity data. Gebtechnique, December 1962, pp 271-282.
8
1 . 0 - -
2.0
3 . 0 -
4.0,==
Current product ion model
5 . 0 - -
6.0 ̧ - -
7 . 0 - -
8.0 u
C. B. R. 0
9.0 ==
10.0 -
1110 - -
12.0 " -
13.0 - -
=-1
- - ~ 2
- ' 3
- - 4
- - , 5 ¸
- - 6 Models used up to 1964
- 7
- 8
- 9
- - 1 0
- 1 1
- - 1 2
- 1 3
14.0 14
C.I . 0
- - 2 0
- - 4 0
- - ' 60
-18o
-- 1 0 0
- - 120 Present cone index scale
C.B.R. =.C. I , - - 140 20
-- 1 6 0
- - 1 8 0
- - 200
-- 220
" ;240
- - 260
280
Fig. 1 COMPARISON OF NON-L INEAR SCALE WITH THE O R I G I N A L L INEAR SCALE OF THE CONE PENETROMETER
25mm dia rack 1300mm long
Depth scale
Stop watch
Operating handle
E E
Q LO
Dial gauge (100 divisions -- CBR 10)
Proving ring (7kN capacity)
Tripod with folding legs
9.5 mm dia. penetrometer shaft
Platform for dead load/operators can stand on platform)
L i 30 ° cone 12.5mm dia
i~ 800ram = I I
Fig. 2 DIAGRAiVI OF T.R.R.L. PENETROMETER
0
.C~ d " "
C~ ~ 0
E ~
0
0~
CD C~ 0~
C ) ¢D 0~
0 OO v--
0 ~D
0
0 0~ v--
C ) O
0 O0
¢D r.O
0
0 0~
0
(}.ua:) Jad) "El "8 ":D
"I -
+=A
>
E
l -
z l.l.i
I - -
..l-
a z
z LI.I UJ
W ,.n
Z 0 I-.
.=.I I.I.I
n "
0
0 I.I.I
0 0 n.- a .
i.i=
E
t -
O . > . t~
r r nt~
15
10
5
/
/ CBR= qc ~ ~ t _d ~/~ O
345 O
o Qoo/S . o O
• O
Oxford clay, Alconbury By-pass O Cut (doubtful results) Single CBR measurements • Cut (good results) Single CBR measurements A Cut. Mean values per test section 8 CBR
measurements • Fill. Single CBR measurements
• Fill. Weald clay, TRRL compaction bay
• cut and fili. Keuper I~ar[, M-6 M0torway
[] Cut. Gault Clay, Maidstone By-pass
I I I 5 10 15
CBR
a) UNDISTURBED OVER CONSOLIDATED SOILS
20
E S ¢.-
O .
n -
~ 3
10
5
S 0 0
O
O O CBR = qc O O
230 ~ O Z~ A
A A
A A
O Brickearth • London Clay
A London clay (results corrected for cone size)
I I 10 15
CBR
b) REMOULDED SOl LS STATICALLY COMPACTED IN CBR MOULD
i 20
Fig. 4 RELAT ION BETWEEN MEASURED CBR VALUES AND CBR VALUES FROM THE PENETROMETER
(1533) Dd0536361 1,500 8/79 HPLtdSo ' ton G1915 PRINTED IN ENGLAND
ABSTRACT
The strength of clay subgrades: its measurement by a penetrometer: W P M BLACK: Department of the Environment Department of Transport, T R R L Labora tory Repor t 901: Crowthorne, 1979 (Transport and Road Research Laboratory) . Improved methods o f pave- ment design require that soil strength should be characterised in terms o f fundamental para- meters. Although the strength parameter CBR widely used in present methods o f road design is empirical in origin, it is shown that it can be related to the undrained shear strength o f cohesive soils by one of two linear relationships, one characterising soils in an undis turbed overconsolidated state and the other soils that are remoulded. After validating these relation- ships it is shown that a penetrometer with a small-diameter cone that is capable of measuring the shear strength of a soil can be used to estimate its CBR. The pene t rometer can then be used to characterise soil strength in terms of either shear strength or CBR.
ISSN 0 3 0 5 - 1 2 9 3
ABSTRACT
The strength of clay subgrades: its measurement by a penetrometer: W P M BLACK: Department of the Environment Department of Transport, T R R L Labora tory Repor t 901 : Crowthorne, 1979 (Transport and Road Research Laboratory'). In,proved methods of pave- ment design require that soil strength should be characterised in terms o f fundamental para- meters. Although the strength parameter CBR widely used in present methods of road design is empirical in origin, it is shown that it can be related to the undrained shear strength o f cohesive soils by one of two linear relationships, one characterising soils in an undisturbed overconsolidated state and the other soils that are remoulded. After validating these relation- ships it is shown that a penetrometer with a small-diameter cone that is capable of measuring the shear strength of a soil can be used to estimate its CBR. The penet rometer can then be used to characterise soil strength in terms of either shear strength or CBR.