An introduction to pressuremeters

<< Pressuremeters

How to decide what pressuremeter to use >>

Pressuremeters are devices for carrying out insitu testing of soils and rocks for strength and stiffness parameters. They are generally cylindrical, long with respect to their diameter, part of this length being covered by a flexible membrane.

Pressuremeters enter the ground by pushing, by pre-boring a hole into which the probe is placed, or by self boring (fig.1) where the instrument makes its own hole. Once in the ground, increments of pressure are applied to the inside of the membrane forcing it to press against the material and so loading a cylindrical cavity. A test consists of a series of readings of pressure and the consequent displacement of the cavity wall (fig.2), and the loading curve so obtained may be analysed using rigorous solutions for cylindrical cavity expansion and contraction. It is the avoidance of empiricism that makes the pressuremeter test potentially so attractive.

The test is usually carried out in a vertical hole so the derived parameters are those appropriate to the horizontal plane.

Fig.1 A self boring pressuremeter, approximately 1.25m x 0.08m (click for full-size)

Inserting the pressuremeter

The interpretation of the pressuremeter test must take account of the disturbance caused by the method used to place the probe in the ground. The least disruptive of the methods is self boring where disturbance is often small enough to lie within the elastic range of the material and is therefore recoverable. This is the only technique with the potential to determine directly the insitu lateral stress, $\sigma_{ho}$, the major source of uncertainty when calculating the coefficient of earth pressure at rest, $k_o$. However all methods allow the confining stress to be inferred.

The disturbance caused by pre-boring and pushing is never recoverable. However for any pressuremeter test it is possible to erase the stress history of the loaded material by taking it to a significantly higher stress than it has previously seen, and then to reverse the direction of loading. The point of reversal is a new origin and the stress:strain response will be that due to the undisturbed properties of the material. In fig.2 the three types of test are shown.

Fig.2 Test curves for 3 types of probe in Gault clay at about 5mBGL (click for full-size)

The tests were carried out at the same location (a heavily over-consolidated Gault clay site) at similar depths and give similar results for strength and stiffness. Although the loading paths appear very different there are similarities in the unloading paths and whenever a small rebound cycle is taken. These cycles are of particular importance. No matter how disturbed the material prior to insertion all types of pressuremeter test have the potential to make repeatable measument of shear stiffness and the reduction of stiffness with increasing strain.

Pre-boring

A pocket is formed in the ground by conventional drilling tools and the instrument is subsequently placed in the pre-formed hole. The major defect in this method is the complete unloading of the cavity that takes place in the interval between removing the boring tool and pressurising the probe. The material must be capable of standing open and so the method is best suited to rock. As fig.2 indicates it is possible to make a test in stiff clay. However comparing the pre-bored curve to the self-bored shows how much further the cavity may have to be expanded before the influence of insertion disturbance can erased. The method can be used in dense sand if drilling muds are used to support the open borehole but it is unlikely to be suitable for loose sands. The Ménard pressuremeter widely used in France is an example of a pre-bored device. In the UK the High Pressure Dilatometer (the terms “dilatometer” and “pressuremeter” are interchangeable in this context) is available and is used in rocks, hostile materials such as boulder clay, and dense sands. See fig.3.

Fig.3 73mm & 95mm High Pressure Dilatometer (click for full-size)

A pre-bored operation will require the assistance of a drilling rig. Unlike the other insertion methods, if the hole is cored then it may be possible to make laboratory tests on material that is directly comparable to that being tested by the pressuremeter.

Pre-bored pressuremeter testing in a vertical hole has been carried out to depths greater than 500 metres and depths of 200 metres are routine.

Pushing

As the name suggests, pushed-in pressuremeters are forced into the ground so raising the state of stress in the surrounding soil. A special case of this approach is the Cone Pressuremeter (CPM) where a 15cm2 cone is connected to a pressuremeter unit of the same diameter. The disturbance caused to the material is total and the only parameter that can be obtained from the loading path is the limit pressure of the soil. The ‘pushed’ curve in fig.2 is an example of a CPM test and shows a clear plateau after the cavity has been expanded by about 15%. Strength parameters are derived from the contraction curve and stiffness parameters from the response of small rebound cycles. The method is fast and can make a test in any material into which a cone can be inserted. The coupling of the profiling capability of the cone with the ability to make direct measurements of strength and stiffness is especially attractive. However as fig.2 indicates the stresses required to make a satisfactory test are much higher than for the other methods, and at these levels of stress it is probable that crushing of the soil particles is taking place. This may be a significant factor especially for tests in sand. Also obtaining reaction for pushing the probe may present difficulties – a jacking force of 10 tonnes or more is not unusual.

Self boring

Figure 4 below (click for full-size) shows a schematic of the Cambridge self boring pressuremeter (SBP). The instrument is a miniature tunnelling machine that makes a pocket in the ground into which the device very exactly fits. The foot of the device is fitted with a sharp edged internally tapered cutting shoe. When boring, the instrument is jacked into the ground, and the material being cut by the shoe is sliced into small pieces by a rotating cutting device. The distance between the leading edge of the shoe and the start of the cutter is important and can be optimised for a particular material. If too close to the cutting edge the ground suffers stress relief before being sheared. If the cutter is too far behind the shoe edge then the instrument begins to resemble a close ended pile. In stiff materials the usual setting is flush with the cutting shoe edge. The cutting device takes many forms. In soft clays it is generally a small drag bit, in more brittle material a rock roller is often used.

The instrument is connected to the jacking system by a drill string. This is in two parts, an outer fixed casing to transmit the jacking force and an inner rotating rod to drive the cutter device. The drill string is extended in one metre lengths as necessary to allow continuous boring to take place. All the cut material is flushed back to the surface through the instrument annulus, there is no erosion of the cavity wall. Normally water is used but air and drilling muds have been applied with success.

Self boring is effective in materials from loose sands and soft clays to very stiff clays and weak rock. It will not operate in gravel and materials hard enough to damage the sharp cutting edge. In principle the probe can be made to enter the ground with negligible disturbance. In practice, self boring results in a small degree of disturbance that must be assessed before deciding a value for the insitu lateral stress. Experience has shown that the self boring disturbance is low enough to remain within the elastic range of the material.

The SBP requires a modest amount of reaction. On some soft clay sites it is possible for the self boring kit to operate without support from other drilling tools. The minimum interval between tests is one metre. Where tests are more widely spaced or in materials with occasional bands of hostile layers the SBP can be used in conjunction with a cable percussion system, or be driven by a rotary rig using special adaptors. Self boring in a vertical hole is routinely carried out to depths of 60 metres or more.

The self boring method is also used as a low disturbance insertion system for other devices such as load cells and permeameters.

Construction and calibration

There are many designs of pressuremeter in current use, some of which are of complex construction. Figure 5 on the right (click for full-size) is a view of the inside of a 6 arm Cambridge self boring pressuremeter. There are transducers for measuring the radial displacement of the membrane at 6 places and the total and effective pressure being applied to the cavity wall. The electronics for the signal conditioning including the conversion from analogue to digital is contained in the probe itself. Apart from supplying power, the output of the probe may be connected directly to the serial port of a small computer. This approach is necessary in order to obtain a high resolution free of noise. Pressuremeters with local instrumentation are able to resolve without difficulty displacements of 0.5 microns and pressure changes of 0.1kPa.

Pressuremeters can be expanded using air or a non-conducting fluid such as light transformer oil. There are automated systems for pressurising the equipment. Automation allows the expansion of the cavity to occur at a constant rate of strain. It is conventional to log the output of the pressuremeter on computer and to plot the loading curve in real time.

Meticulous calibration of the equipment is vital. The transducers must be calibrated regularly both for sensitivity and drift. Almost all pressuremeters suffer the defect that the output of the transducers is governed by the movements and pressure on the inside of the membrane, where what is required is the displacements and stresses acting on the cavity wall. The properties of the pressuremeter membrane can be a significant source of uncertainty. It requires an amount of work to make it move, and an additional component to keep it moving. This is relevant to tests in soft soils. The membrane contribution may be estimated by carrying out membrane expansion tests in free air.

The other major influence on the measurements is system compliance, or the contribution of the probe itself to the measured stiffness. This can be a significant source of error if the probe is used in very stiff soils or weak rock. This contribution may be estimated by inflating the instrument to full working load inside a metal sleeve of known elastic properties.

The importance of the various calibrations depends on the type of pressuremeter and where it is being used. For example the contribution of the hose supplying pressure to the probe is highly relevant if volume changes are being measured at the surface, but is of no importance at all for a probe with internal instrumentation, such as the Cambridge family of devices.

Advantages and limitations of the pressuremeter test

• A large number of fundamental soil properties are obtained from a single test.
• To derive these properties, no empirical correcting factors whatever are needed.
• Measurements are made insitu at the appropriate confining stress.
• A large volume of material is tested - a typical test loads a column of material 0.5 metres high and extending to more than 10 times the expanded cavity radius. This is the equivalent of at least 1000 triaxial tests on 38mm samples.
• Representative loads are applied – in the example shown in fig.2 about 12 tonnes is being applied to the cavity wall.
• Results can be obtained quickly as all the data logging and most of the analysis is carried out by automated systems.
• Commercial operation has shown that the instruments, though more complex than conventional site investigation equipment, are reliable.
• There are many materials whose properties can only be realistically determined by insitu measurement.
• The pressuremeter test is particularly appropriate for predicting the performance of laterally loaded piles.
• Pressuremeter tests are routinely used to calibrate finite element models of complex geotechnical problems.

Limitations

• The instrument will not penetrate gravels, claystones or the like, so generally pressuremeter testing requires support from conventional drilling techniques.
• Failure planes and deformation modes are not always appropriate to those occurring in the final design. An estimate of the anisotropy of the material will be required in order to derive vertical parameters from lateral values.
• Many familiar design rules and empirical factors are based on parameters obtained from traditional techniques. It is not always possible to use them with pressuremeter derived values, even if the insitu parameters more accurately represent the true state of the ground.
• Only two stress paths can in practice be followed, undrained and fully drained.
• The instruments and their associated equipment are complex by conventional site investigation standards and can only be operated by trained personnel.
• Use of an inappropriate analysis to interpret a pressuremeter test can result in seriously misleading parameters.

<< Pressuremeters

How to decide what pressuremeter to use >>