Department of Civil Engineering Soil Mechanics Laboratory

Laboratory Exercise 7: Direct Shear Test

References and Sources:
ASTM D3080: Standard Test Method for Direct Shear Test of Soils under ConsolidatedDrained Conditions. Bardet, Jean-Pierre. (1997). Experimental Soil Mechanics. Prentice-Hill. Holtz, R.D. Kovacs, William D.

Introduction
The laboratory handouts for this week include a discussion of shear tests in general, and a specific discussion of the Karol-Wamer shear testing apparatus (which you will be using). These handouts provide a sufficient discussion of the direct shear test.

In addition to the direct shear test, other tests exist for the determination of shear strength of soils. These tests include: Triaxial test Simple shear tests: Roscoe or Cambridge Norwegian Geotechnical Institute (NGI) Swedish Geotechnical Institute (SGI)

Hollow cylinder triaxial test Directional shear test

The direct shear test is a strain-controlled test: the rate at which the soil will be strained is controlled. A specimen of soil will be placed into a shear box, and consolidated under an applied normal load. The shear box is made of two separate halves, an upper and a lower. After the application of the normal load, these two halves of the be moved relative to one another, shearing the soil specimen on the plane that is the separation of the two halves. The
1

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

direct-shear test imposes stress conditions on the soil that force the failure plane to occur at a predetermined location (on the plane that separates the two halves of the box). On this plane there are two forces (or stresses) acting - a normal stress, n, due to an applied vertical load Pv and a shearing stress, , due to the applied horizontal load Ph. computed as: n = Pv/A = Ph/A where A is the nominal area of the specimen (or of the shear box). It is usually not corrected for the change in sample area caused by the lateral displacement of the sample under the shear load Ph. These stresses should satisfy Coulomb's equation: = c + n tan As there are two unknown quantitie s (c and ) in the above equation, two values, as a minimum, of normal stress and shear stress will be required to obtain a solution. These stresses are simply

Since the shear stress and normal stress have the same significance as when used in a Mohr's circle construction, rather than solving a series of simultaneous equations for c and tan , one may plot on a set of coordinate axes the values of t versus n from several tests (generally with on the ordinate), draw a line through the resulting locus of points, or the average locus of points, and establish the slope of the line as tan and the -axis intercept as the c parameter. This is commonly known as the Mohr-Coulomb Failure Envelope.

For cohesionless soils, the intercept is usually negligible, and Coulombs equation becomes = n tan Test inaccuracies and surface-tension effects of damp cohesionless materials may give a small value of c, called the "apparent" cohesion. This should be neglected unless it is more than 1 or 2 psi. If the c value is large and the soil is a cohesionless material, the reason for the large value should be investigated.

The direct shear test was formerly quite popular, but with the development of the triaxial test which is much more flexible, it has become less popular in recent years. The advantages of the direct shear test are:
2

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

1. Cheap, fast and simple - especially for sands. 2. Failure occurs along a single surface, which approximates observed slips or shear type failures in natural soils. Disadvantages of the test include: 1. Difficult or impossible to control drainage, especially for fine-grained soils. 2. Failure plane is forced--may not be the weakest or most critical plane in the field 3. Non-uniform stress conditions exist in the specimen. 4. The principal stresses rotate during shear, and the rotation cannot be controlled. Principal stresses are not directly measured. Because the drainage conditions during all stages of the test markedly influence the shear strength of soils, the direct shear test is only applicable for relatively clean sands which are free draining during shear. For clay soils, some unknown amount of consolidation could occur during shear, which would give a larger shear strength than actual. Therefore the test is not generally recommended for cohesive soils.

Procedure
Each group will perform one direct shear test, each at a specified (unique) normal load. The three groups in each section will then share their data, in order to plot the Mohr-Coulomb Failure Envelope (there will thus be three points on this plot). Because the three groups in each section will be sharing data, it is imperative that each group set up its specimen at the same density as all other groups. This density will be specified by the instructor. 1. Become familiar with the apparatus, see figure 1 below.

Figure 1 Apparatus and equipment 3

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

In particular:

How

to

apply

normal loads (the pressure the indicator, calibration

curve, the belofram valve, regulator).

the

The

function

of

each screw in the shear box.

The

measurement

the dial indicators are taking (whats it measuring, what are the units, which will it
Figure 2 a) motor, b) motor control panel

direction move?)

The direction of travel of all parts of the apparatus The electric motor (on/off switch, forward/reverse, speed), determine which direction is forward and reverse on the motor before assembling specimen, see figure 2 a and b above.

2. Determine the volume of the specimen to be constructed.

4 Figure 3 Locked halves of the shear box

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

Using the fixing screws, lock the two halves of the shear box together see figure 3 above.

Measure the cross-sectional area of the interior of the shear box. Measure the depth of the shear box. Determine the thickness of the bottom frictional plate and the top porous stone.

When the specimen is setup, the frictional plate will be at the bottom of the shear box. Sand will then be placed on top of this, and on top of the sand will be a porous stone. When this is all set up properly, the top of the porous stone should be flush with the top of the shear box. Knowing this, and with the above dimensions measured, the volume of the prepared specimen may be calculated. 3. Knowing the density desired and the volume of the specimen, calculate the mass of sand required to set up the specimen. 4. Obtain the required mass of sand. 5. Set up the specimen.

Place the shear box in the direct shear device. Make sure the fixing screws (screws that locks the shear box together) are locked in place and the set screws (screws that adjust the gap space between the shear box halves) are at the bottom of their travel.

Place the bottom frictional plate into the bottom of the shear box with the rough side up.

Pour the sand in rather loosely. Place the porous stone on top of the sand. It should be sticking up out of the shear box at this point, see figure 4a below.

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

Tap on the shear box to densify the sand until the top porous stone becomes flush with the top of the shear box. You now have a specimen of sand placed at a known density.

This may take several attempts to get it just right. Be patient.

6. Place the loading cap and ball bearing on top of the porous stone. 7. Lower the reaction bar of the pneumatic loading frame until it just touches the ball bearing. Use the nuts to lock the reaction bar into place, see figure 4b above. 8. Remove the fixing screws, see figure 4a above. 9. Remove the fixing screws, see figure 4a above 10. Adjust all dial gages. Take initial readings at this time. 11. Use the calibration curve to determine the applied air pressure to achieve the desired normal pressure. 12. Close the belofram check valve and the blow-off valve, see figure 5 below.

Figure 5 Regulator and valve locations, check valve in closed position.

13. Adjust the regulator to the desired air pressure. Do not exceed the indicated limit. 14. Open the check valve. 15. Record the new vertical displacement dial reading. It will have moved, and the specimen will thus have densified. This combined with the initial dial reading you already obtained will allow you to calculate the actual density of the specimen (it will be slightly denser than planned).

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

16. Adjust the set screws. Starting with the set screws at the bottom, turn each turn clockwise. This will separate the two halve of the shear box. This separation should be slightly smaller than the largest grains of soil in the sample. 17. Shear the specimen.

Start the horizontal (shear) loading and take readings of the load, shear displacement, and vertical displacement dials. Take readings every 0.01 inches of shear (horizontal) deflectionthis corresponds to every 10 units on the dial.

The shearing is complete when load is determined to be dropping (the residual strength of the soil has been reached).

18. Back off the applied normal pressure, dismantle and clean the shear box.

Calculations
1. Compute the shear stress as = Ph / A

2. Compute the normal stress as n = Pv / A 3. Compute the horizontal displacement, h, and the vertical displacement, v, for each observed value.

4. Plot shearing stress against horizontal displacement and obtain the maximum value of shearing stress, max.

5. Plot a graph of normal displacement vs. shear displacement. This should be on the same page (same horizontal scale) as the shearing stress vs. shear displacement plot.

6. Using data from all three groups, plot shearing stress against normal stress and show the angle of internal friction, , and the intercept, c.

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

Report:
1. Memo-format report. 2. Visual classification. 3. Plots of shear stress v. horizontal deflection and vertical deflection v. horizontal deflection on the same page to the same scale. 4. Mohr-Coulomb Failure Envelope. 5. State internal angle of friction and cohesion. 6. Remarks stating any errors associated with the test, specific errors witnessed, and specifically the manner in which these errors could have affected the results.

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

The Karol-Wamer Direct Shear Testing Machine Discussion and Procedure

Prepared by William D. Kovacs The use of this piece of equipment is to measure the effective stress parameters of c' and '. The direct shear test can be run on any type of soil sample, but it is particularly adapted to cohesionless materials. This machine can also be used to establish the undrained shear strength of saturated highly plastic cohesive soils under certain conditions.

The direct shear test is normally run as a strain-controlled test, however, under certain circumstances tests can be run under stress controlled conditions. In order to obtain the effective stress parameters of any soil, excluding the most pervious medium and fine sands, it is necessary to use a slower rate of shearing in order for induced pore pressures to dissipate. The remainder of this report presents a discussion on the sample preparation for testing, equipment preparation, and running the direct shear test.

Undisturbed samples will be selected for strength testing in the usual manner. Once extruded, the samples should be trimmed carefully down to the internal diameter of the direct shear rings, in a similar manner as for consolidation test samples. When dry sand samples are to be tested, (undisturbed samples are almost impossible to obtain) the void ratio should be duplicated in the laboratory. The object is to get at least 1 inch of soil sample sandwiched between a layer of filter paper and a porous stone on either side of the sample. The porous stones provide drainage and are used to distribute the pressure during the test.

Place two porous stones on top of the soil sample to be extruded. Extrude the soil sample into the locked pair of shear rings until the top of the two stones become flush with the top of the direct shear rings. An additional 1/8 inch to 1/4 inch of soil is further extruded and the direct shear sample is separated from the field sample. After final trimming of the sample, filter paper and porous stone are placed on each side of the soil sample, centering it within the shear rings. The water content is determined from the remainder of the field sample. Additional data is taken as to the number of the shear ring, the soil description, and the water content. If the soil

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

sample is to be inundated, it is first necessary to saturate the porous stone by boiling them for at least five minutes and letting them cool under deaired water.

Equipment Preparation

Prior to testing, the direct shear machine should be checked for cleanliness and proper mechanical condition. Special care should be taken to see that the roller bearings underneath the direct shear plate are properly cleaned and contain a slight amount of oil. The main loading shaft should be returned to its original starting position.

Specimen Setup in Machine

The following steps are now presented in order in which a specimen would be prepared in the direct shear machine for testing. At this time the hardened direct shear rings would be held together by three bolts and the soil specimen would be centered inside the rings. The outside faces of the soil specimen should be covered with a piece of filter paper cut to the appropriate diameter followed by a porous stone. The shear rings with soil specimen and stones are then carefully placed inside the direct shear box. The top yoke is placed over the rings, and the entire box pushed forward towards the load ring. The restraining screw is attached finger tight. The loading shaft i positioned through the loading collar and the appropriate nut attached with a s small seating load. The top transverse bar is now placed over the sample such that the bar will just rest upon the loading ball with base centered over the specimen. 'The top tightening screws are brought down until they are just finger tight. The bottom tightening screws are then brought up and tightened securely by hand. At this time a small normal seating load in the order of 100 lbs/sq. ft is placed on the soil specimen. The vertical dial indicator is positioned and set to read some arbitrary initial reading. Next, the horizontal dial indicator is attached and set at some initial reading. This sets the stage for the next phase, consolidation of the specimen with the application of the desired overburden pressure or some other normal pressure.

Consolidation of the Test Specimen

10

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

From the boring logs and the initial moisture-density determinations, the insitu effective overburden pressure at the sample is computed. To establish the appropriate Mohr failure envelope it is desirable to have at least three points on the failure envelope. These are usually established at normal pressure of at least the overburden pressure, the overburden pressure plus the footing load, and finally, the overburden pressure plus at least two times the footing load.

With the belofram switch off, the air pressure regulator is adjusted to the desired normal load. (This is proceeded by the calibration of this particular test gauge.) The vertical dial indicator is read and noted on the data sheet. At the same time the air switch is released, a stop watch is started in order to record the appropriate time deformation characteristics of the soil sample undergoing consolidation. Consolidation data i taken at the usual time increments. The data s will either be plotted on a square root of time graph or on a logarithm of time graph. The purpose of recording the time deformation characteristics of the soil sample is to establish the rate of shear testing of the soil. For clean sands and soils with high permeability, it is not necessary to go through this procedure. Under these circumstances, 25 revolution of the loading handle per minute is the maximum rate of shear to use for these conditions. For all other types of soils it is necessary to establish the time deformation characteristics and estimate t50 or t90. If the square root of time graph is used, t90 will be established, while if a logarithm base, t50 will be established. As a conservative estimate of the failure of a direct shear test, 50t50 or 12t90 is ordinarily used. An estimate of the failure displacement is taken to be 0.2 of an inch, in the horizontal direction. The shear rate is then calculated and will be shown in an example further on in this discussion. In the event the testing time is very long, a stress controlled test may be run in place of a strain controlled test. During the direct shear test, the deflection of the load ring is ignored. The true deflection of the sample undergoing shear is measured by the horizontal dial indicator. Once we have established the rate of testing we are now ready to perform the test.

Test Procedure

The final dial reading after consolidation is noted on the data sheet. Adjust the anchoring nut to provide a small seating load on the sample. The horizontal dial indicator is adjusted to some

11

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

convenient zero point. Until this time, the two hardened shear rings have been held together by three screws, these three connector screws are now removed.

The separating screws are now turned to force the shear rings apart at a prescribed distance depending on the soil type. For sands of a fine to medium size, 3/4 of a revolution of the set screw is used. For coarser sands, 1 revolution is used. For fine grain soils and clays, 1/3 revolution is used. These values are not absolute and can be altered accordingly. Initial readings are taken at this time. The test is now started by the counter clockwise rotation of the loading device at the prescribed rate of shear either manually or by motor. Unless automatically recorded, the load dial and vertical deflection dial indicator will be read at various prescribed horizontal deflection readings such as every ten divisions of the three place dial until failure. Usually the test is carried out past failure to determine the residual strength which may be used in design. At this point the test is completed and the machine is dismantled.

Specimen Removal and Cleanup

The shear loading is stopped, put into reverse, and reduced until no shear stress is on the sample. The normal pressure is now reduced. The apparatus is taken down in reverse order as assembled. The sample is extracted and weighed in order to determine the dry density and water content at failure. The machine is cleaned and the bearings are checked for dirt and are oiled and cleaned if necessary.

Computations and Report

The shear stress on the sample is easily computed by dividing the shear force as measured by the load ring or load cell by the area of the sample, uncorrected for deflection. To evaluate for the shear test, graphs of shear stress versus horizontal deflection, vertical deflection versus shear (horizontal) displacement, and a Mohr envelope are plotted. From the Mohr envelope, the effective stress parameters c' and ' are established.

Discussion:

12

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

For running direct shear tests on clean sands, the rate of shear is limited to 25 revolutions per minute of the loading shaft. When soils of lesser permeability are tested (and therefore require a longer time for pore pressure dissipation), the shear rate can be reduced to the appropriate level for a strain controlled test. If this is not desirable, as in the case of a more impervious soil, a stress controlled test may be run in order to establish the effective stress parameters.

As in the triaxial testing, it is possible to run multiphase direct shear tests on one sample. This would be accomplished by stressing a soil until failure, at which time the normal pressure would be immediately increased and the soil would be allowed to consolidate; another direct shear test similar to the first would be run on the sample. After failure was reached in the second phase, the normal force would again be increased to the desired level and the sample would be left to consolidate. The process would be repeated for the third stage and the resulting Mohr envelope would be made from three points of shear stress at three different normal pressures.

The direct shear test can be used to measure the effective stress parameters of any type of soil as long as the pore pressure induced by the normal force and the shear force can dissipate with time. In the case of clean sands, this is no problem as the pore pressure dissipates readily; however, in the case of highly plastic clays, it is merely necessary to have a suitable strain rate so that the pore pressure can dissipate with time.

If an undisturbed sample has been obtained under the ground water table, it is absolutely necessary to submerge the sample during the test so as to simulate the field conditions. It is necessary to inundate the sample in order to relieve any surface tension that may be present in the sample.

The direct shear test can also be used to determine the undrained shear strength of saturated plastic clays, with qualification. The test must be run rapidly within 5 to 10 minutes to avoid consolidation under normal force during testing, and hence, an increase in shear strength.

Similarly, it is desirable to maintain the levels of induced pore pressure within the sample and not to allow it to dissipate, giving an increase in strength. Although the top and bottom face of the soil sample are in contact with filter paper and porous stones providing drainage access, the low
13

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

permeability of the plastic clay reduces the rate of pore pressure decrease with time. As a reasonable approximation, it is assumed that the pore pressure across the shear plane is very close to the original pore pressure induced by the normal load. Therefore, as a reasonable approximation, the test measures the undrained shear strength of the soil.

In cases where the soil sample is saturated insitu regardless of grain size, soil type, the sample should be inundated to relieve any surface tension that may develop between the soil particles, inducing tensile forces within the sample. These tensile forces would tend to consolidate the sample three dimensionally.

14

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

Direct Shear Test

Analyst Name:_____________________________ Project Information: Project Name:____________________________ Project Location:________________________ Boring No.:_______________ Sample Description: Source:_____________________________ Condition:_________________________________ Vis. Class. & USCS Symbol:_______________________________________________________ Structure:_______________ Consistency:__________________ Specific Gravity:__________ Specimen Density:__________ Normal Load:______________ Shear Box Dimensions: Depth:___________ Diameter:____________ Area:______________ Stone thickness:___________ Frict. Plate Thickness:______________ Specimen Dimensions: Height:_______ Area:________ Volume:________ Mass:_________ Loading Information: Elapsed Time (sec) Air Pressure:______ Stone+Cap+Ball______ Prov. Ring Calib.______ Sample No.:________________ Date:____________ Test No._________

15

University of Washington

Department of Civil Engineering

Soil Mechanics Laboratory

16