GDS Triaxial Automated System

The GDS Triaxial Automated System Hardware Handbook

© GDS Instruments Ltd

About this Manual

The GDSTAS users manual describes the GDS Triaxial Automated System hardware.  Please refer to the GDSLAB software manual for information about setting up and running software for your GDSTAS system.  This manual is divided into logical chapters.  Where necessary at the start of each chapter is a contents page showing in detail the contents of the chapter.

 About this Manual

System Overview

The study and measurement of soil properties in the Triaxial Test

The GDS PC-controlled Triaxial Automated System

Accuracy and resolution of measurement and control

GDSTAS Hardware

Hardware overview

ADVTAS System Overview

The Triaxial Cell

Setting up the Instrumentation for the triaxial cell

Serial Pad

Setting Up the System

System Overview

The study and measurement of soil properties in the Triaxial Test

Typically, soil comprises a skeleton of soil grains in frictional contact with each other forming an open-packed structure (loose/soft) or close-packed structure (dense/hard). The soil particles may be microscopic in the case of clays, just visible in the case of silts, and clearly visible in the case of sands and the larger particle sized gravels.  The soil skeleton, which can also be cemented, forms an interstitial system of connecting spaces or pores.  The pores in the soil will usually contain some moisture even in unsaturated soils.  The flow of pore water can be restricted by the small size of the pores thus giving rise to low permeability particularly in clays. During construction the change in load or total stress is shared between the soil structure and the pore pressure.  The time-dependent flow of water in soil under applied load is referred to as consolidation and is the means whereby the load is transferred from pore pressure to structural loading or so-called effective stress.  It is this time-dependency which gives clays their unique behaviour whereby they have a short term or undrained strength which is different from the long term or drained strength. The ability of the soil skeleton to support load is called the shear strength of the soil.  This strength depends on the frictional nature of the interparticle contact and is measured by the coefficient of friction or angle of shearing resistance which depends on normal effective stress, and by the constant, cohesion.  The deformability of the soil skeleton is measured by various elastic theory deformation moduli such as Young’s modulus and Poisson’s ratio.

               Soil is often geologically prestressed to a maximum past pressure or preconsolidation pressure. This prestress constitutes a yield point. At stresses less than yield the soil behaves like an elastic solid i.e. the strains are nearly recoverable.  At stresses more than yield the soil behaves like a plastic material i.e. the strains are not recoverable and the mathematical theory of plasticity is sometimes used to describe soil behaviour.  Stress distributions, however, can be generally described using the mathematical theory of elasticity.

 These properties and characteristics can be studied and measured in the triaxial test.

               In the triaxial test, a right cylindrical specimen of soil is sheathed in a rubber membrane and placed in a cell filled with pressurised water.  Drainage is provided between the pores and the outside of the cell via a porous stone interface and a pore water duct.  Through this duct pore pressure can be measured and pore water can flow during consolidation and swelling. Axial stress is applied by a piston moving through the top or base of the cell. The piston is usually actuated by an electric motor and gearbox turning a screw or sometimes by hydraulic means. 

               The New PC-Controlled Stress Path Triaxial Testing System enables the classic triaxial test to be computer-automated.  In addition, complex loadings can be applied which follow real geological, construction and in-service conditions where vertical and horizontal stresses both change at the same time.  This is important because soil is a loading-dependent material i.e. the deformation moduli are not unique material constants but are related to the actual loading pattern or stress path.

System Elements

The system comprises the following elements: 

• The GDS LF50 loading frame (50kN) with a USB interface. 

• A cell pressure controller which controls cell pressure. 

• A back pressure controller which controls back pressure and measures volume change. 

• An eight channel 16 bit data acquisition device.

• Transducers; an axial displacement transducer of the potentiometric type, a load cell (the user can choose the load cell range), and a pore pressure transducer. 

• controlling software (GDSLAB) for Windows XP and up.

Typical Test menu

The test menu is as follows:

               B CHECK

               SATURATION RAMPS

               ISOTROPIC CONSOLIDATION

               ANISOTROPIC CONSOLIDATION

               UNCONSOLIDATED-UNDRAINED

               CONSOLIDATED-UNDRAINED WITH PORE PRESSURE MEASUREMENT

               CONSOLIDATED-DRAINED WITH VOLUME CHANGE MEASUREMENT

               STRESS PATHS

               LOW SPEED CYCLIC LOADING

               VARIALBE TEMPERATURE TESTING

               Note: see GDSLAB software manual for further test module details

System features 

• Automatic area correction.  The system automatically uses volume change and axial displacement to compute the current average area of the test specimen.  This is used in all control calculations.  The average area is defined as the cross-sectional area of the volumetrically equivalent right cylinder of the same height.  The resolution of volume change is 1cu.mm. 

• Pore pressure is measured at the base pedestal using a rigid pore pressure sensor. 

• Volume change measurement is resolved to one cubic millimetre. 

• Axial displacement is measured directly by transducer. 

• The classic U-U, C-U and C-D tests can be both strain and stress controlled in both compression and extension.

• The system allows a single continuous linear stress paths to be programmed in stress space.  Any number of continuous linear paths can be made with user-intervention at the end of each path.

System advantages

The new GDS PC-controlled Automated Triaxial Testing System has many advantages including:

• Very simple to set up and use, where Cabling and piping is reduced to an absolute minimum. 

• The whole system including PC can be laid out on a lab bench area 1.5m by 0.9m. 

• The electrical and hydraulic inter-connections of the system are open and easy to see and not hidden behind wall panels. 

• Tests can be carried out manually to give "hands-on" experience, or under direct PC control for automated testing. 

• Ideal for teaching modern triaxial testing in colleges and universities.

• Designed for effective stress and stress path testing for commercial and public works laboratories.

Accuracy and resolution of measurement and control

In general, all accuracies within the GDSTAS are better than 0.5 % of full scale values.  For the data acquisition system all measured values have a resolution that is better than 0.1 % of full scale values. 

               The resolution of control is as follows: 

• All pressure control has a resolution of 1 kPa

• The loading frame is controlled by rate of deformation control.  All other types of control (target position, target load, rate of load are achieved by a special algorithm being carried out within the computer).

GDSTAS Hardware

Hardware overview

The system consists of a load frame, two computer controlled pressure sources, a data acquisition device (either parallel or serial) and a computer.  The computer controls the system allowing a simple interface providing full system control, data logging and calculation of required parameters.

Due to the flexible nature of GDSLAB, a GDSTAS system may be setup using either Standard Controllers, Enterprise Controllers or Advanced Controllers.

Using GDS Standard or Enterprise Pressure Controllers

Two GDS Pressure Controllers.  See GDS handbook entitled STDDPCv2 or ELDPC for specific controller information.
               Each pressure controller senses pressure from a pressure transducer inside the controllers pressure cylinder, converts this into digital form and then turns the stepper motor to increase or decrease the pressure as required. Diagram 2.1 below shows the system setup when using GDS STDDPCv2 controllers.

Using Advanced Pressure Controllers (ADVTAS)

Two GDS Advanced Pressure Controllers. See GDS handbook entitled Advanced Pressure Controllers for specific controller information.

Diagram 2.2 below shows the system setup when using GDS Advanced controllers connected via an RS232 cable.

The Triaxial Cell

At the center of the system is the GDS LF50 Load Frame (or similar). This provides a test chamber in which the test conditions may be accurately controlled.  The parameters that can be controlled/measured are: 

• Cell pressure

• Back Pressure

• Specimen volume change

• Pore pressure

• Axial Force

• Axial Displacement

               The axial force is provided by the load frame.

Setting up the Instrumentation for the triaxial cell

Introduction

You will normally have the following instruments attached to the triaxial cell: 

• Load Cell

• Displacement Transducer

• Pore Pressure Transducer 

                     Each of these devices is used for data acquisition by the computer.  The load cell could either be an internal submersible load cell, an external load cell or an external proving ring.  The manner in which the instruments are attached will be determined by the style of your load measurement system - this is described in the following sections.  For manual verification you may also choose to have some other indicators.  For example, if you are using a proving ring as the load measurement device you may have a second dial gauge to measure deformation of the proving ring and hence indicate load, and you also may have a second deformation dial gauge attached to the proving ring.

                     You must be very careful to ensure that the equipment attached to the top of the triaxial cell does not interfere with other equipment attached to the top of the triaxial cell, and also does not interfere with the triaxial cell itself.

The Load Measuring System

There are two common types of load measuring system in use.  These are:

• Internal Submersible Load Cell

• External Load Cell

                     The setup for each of these is described below:

Internal Submersible Load Cell

An internal submersible load cell is a load cell that is attached to the end of the loading ram on the inside of the triaxial cell.  The load cell in this condition is immersed in the cell water and is subjected to the changes of pressure inside the triaxial cell.  The advantage of the internal submersible load cell is that it measures deviator force directly - there is no compensation necessary for the cell pressure.  The disadvantages are that you require one internal submersible load cell per triaxial cell.  In addition, like all measurement transducers, the internal submersible load cell is not a perfect transducer in that it has a very small reaction to changes in cell pressure.  Theoretically changes in cell pressure should cause no effect on the measurement of axial force, however, in practice there is a small effect on the measurement of axial force when the cell pressure changes. 

                     With the internal load cell the transducer cable runs up the centre of the attached load ram and exits through a side connector at the top of the load ram.  The end of the load ram interfaces directly with the load button on the top cross beam of the triaxial machine.

The External Load Cell

The external load cell is attached to the underside of the top beam of the triaxial machine by a special swiveling ball joint connector.  The external load cell is attached to the frame of the machine, and therefore the load cell is associated with the triaxial machine and not with the triaxial cell.  In this case you therefore need only one load cell per triaxial machine, and not one load cell per triaxial cell as is the case for the internal submersible load cell.  Because the load cell rests on the top of the load ram of the triaxial cell, it measures the effect of cell pressure in the triaxial cell acting on the end of the loading ram which runs through the top of the triaxial cell.  It is therefore necessary to apply a correction related to the cross sectional area of the loading ram to enable the deviator force to be calculated.  For a computer controlled system, this does not represent a problem because all of these corrections can be made automatically as the test proceeds.  The finite size of the external load cell means that you need to be careful with the relationship between the load cell and the displacement transducer to ensure that these two devices do not interfere with each other.

Displacement Transducers

Introduction

A variety of displacement transducers can be used to measure specimen deformation.  For a normal soil test specimen, the range of the displacement transducer should be 20% of the specimen height.  For example, with a 50mm diameter test specimen, which is 100mm high, the range of the displacement transducer should be at least 20mm - this will normally mean a 25mm or 50mm displacement transducer is used.  Care must be taken with the physical relationship between the load measuring device and the transducer to measure specimen deformation.  The arrangement should be such that there is no possibility of the specimen deformation indicator interfering with either the load measuring device or parts of the triaxial cell - whatever the position of the loading ram. 

Linear Resistive Displacement Transducer

This is the normal type of displacement transducer supplied by GDS for use with the GDS TAS system.  The device has a normal range of 50mm which is set up as +/-25mm.  An aluminium clip attaches to the load ram of the triaxial cell and supports the displacement transducer so that the indicator of the transducer touches onto the top of the triaxial cell.  Relative movement between the load ram and the top of the triaxial cell is measured using this technique.  Care must be taken to ensure that at full stroke in either direction no parts of the transducer can come into contact with either the body of the triaxial cell or the load measuring device. 

Digital Indicator

In some situations it may be required to use a device with an electronic digital output, such as a digital dial gauge.  When devices with a digital output are used you will need to have some electronic input device as part of your data acquisition system to convert the digital output into a form that the computer can understand.  For GDS this will be a device called the Digital Indicator Multiplexer.

LVDTs

By special arrangement an LVDT can be used as the axial displacement measuring device.  Provided the LVDT has its own power supplies and signal conditioning unit that provides a DC voltage output in the range 100mV to 10V, then this device can be interfaced directly to GDS data acquisitions.

Pore Pressure Measurement

Pore pressure is measured by means of a block attached directly to the triaxial cell.  A narrow bore pipe connects to the base pedestal.  The pore pressure transducer is on the outside of the triaxial cell.  Provision is made by means of two valves to allow the pore pressure transducer to be de-aired and for the pipe-work from the pore pressure transducer to the base pedestal to be flushed with water at the start of a test.  The output of the pore pressure transducer feeds directly into the GDS data acquisition system. 

Serial Pad

8 channels of 16 bit data acquisition.  Each channel can be user defined with gain and range (span) settings.  +/- 5 Volts supply voltage is available individually for each transducer.

General Description serial pad

The serial pad connects via an RS232 cable either directly into the back of a PC with an RS232 port or into a USB to RS232 convertor. 

               A wide range of signals are accommodated by the inputs, which range from +/- 3mV to +/- 10V full scale. Sensor energising is provided as +/- 5V DC. 

               Power is derived from a standard IEC power cable.  The internal universal power supply accepts voltages in the range 85 Volts to 264 Volts AC.

Connecting Transducers for GDSTAS

The first three channels on the serial pad are dedicated channels which are associated with the software as follows:

                       Channel 0: Load Cell

                       Channel 1: Pore Pressure Transducer

                       Channel 2: Displacement Transducer

               Any further channels may be used for other user defined transducers but the first 3 channels must be connected as below (see the first figure below).

Dedicated channels for a serial pad used in a GDSTAS.

Transducer connections (as looking into the plug of the transducer).

Setting Up the System

Setting the Datum of Pressure Measurement

There are three identical pressure transducers in the GDS system.  There is one in each of the two pressure controllers as follows:

• Cell pressure controller

• Back pressure controller.

               In addition, there is a third transducer measuring pore pressure.  This measurement is routed through the parallel pad.

               Naturally there will be small differences between measurements of the same pressure made by these transducers.  This is because they have slightly different accuracies (their specified deviation from a standard value) and calibrations (actual relationships between a standard value and the read value). This is quite normal and should be taken into account when interpreting your results because you do not have any control over these inherent discrepancies. 

               You do, however, have control over setting the common zero or datum of pressure measurement.  This is so that all three pressure measuring systems (i.e. the transducers and their associated analogue-digital conversion) measure pressure from the same “base line”.  This is how you do it. 

               First you need to set up your datum of pressure measurement. Normally this will be an elevation equal to the mid-height of the triaxial test specimen.  Probably the best way of doing this is to connect a short length (say 300mm) of small bore nylon tubing to the back pressure connector of the cell.  This is the connection to the top cap drain.  Fill the cell with water.  You will not have a test specimen in place for this procedure.  Apply a small positive cell pressure using the cell pressure controller.  You can do this by setting a target pressure.  Open the valve to the back pressure line.  Water will flow out of the cell from your short tube. Stop pumping when the tube is full of water and water drips out of the open end. Fix the open end of the water-filled tube at an elevation corresponding to the mid-height of the test specimen (or the base of the test specimen if you prefer) 

               Now the water in the cell is at a pressure corresponding to this elevation head.  Connect your back pressure controller to the base pedestal pore water port and open the valve. Now the cell pressure controller, back pressure controller and pore pressure transducer all share the same pressure set by the external tube.  You can now zero the displays of these values.  Now all three displays of pressure are zeroed to the same datum of pressure measurement! 

NOTE:

               If the datum of pressure measurement for controllers is set at a different time to that of the pore pressure transducer it is possible to have a difference of up to 9 kPa between the back pressure controller and the pore pressure transducer when measuring the same pressure.  Consider the diagram below (figure 2.7).  If the zero offset for the pore pressure controller is applied when there is water in the cell the actual pressure at the pore pressure transducer will be 5 kPa (this is the head of water) but because we have just applied a zero offset this will be read as zero kPa.  Now if the controller soft zero is set when the controller is open to atmosphere the zero pressure datum for the controller will be the outlet of the controller.  Now when the controller is connected to the cell and the valve is opened the controller will now measure the complete head shown (5 + 4 kPa) but the pore pressure transducer will read zero because we have just zeroed it in this condition.  Therefore the two transducers will have a nine kPa difference due to their different datum of pressure measurement.              

               To overcome this you need to zero both devices at the same time and using the same datum as described in this section.

Common mistakes when setting datum of pressure measurement

Setting up the Triaxial Cell

In a GDS computer controlled system, it is equally adept at carrying out classic "standard" tests as well as advanced tests. 

               Normally, you will set up your system with the computer to your left, the GDS LF50 to your right, with the bank of two linking controllers length ways in the middle.  In this way, the interface cables from the computer are conveniently connected to the controllers, while the pressure connectors from the controllers are immediately adjacent to the test cell.

               One controller will be connected to the valve of the cell chamber itself (this becomes the "cell pressure controller"), while the second is connected to the valve to the pore water duct to the base pedestal (this becomes the "back-pressure controller").  When making connections, remember to flush air out of the connectors using a syringe of deaerated water or a pressure controller targeting a small pressure.

Area Correction for the Triaxial Test

The cross sectional area of the test specimen is continually corrected for the effects of change in volume (where back pressure is provided by a GDS Digital Controller) and axial deformation. Axial stress is therefore based on the average cross sectional area defined as the cross sectional area of the volumetrically equivalent right cylinder (Bishop and Henkel, 1962). 

Using the Triaxial Extension Device

The GDS triaxial extension device enables triaxial extension to be carried out as routinely as triaxial compression.  The device prevents cell pressure from acting vertically on the top cap resting on the test specimen.  This allows axial stress to be reduced below cell pressure.

               The setting up, docking and undocking procedure is as follows: 

·        Set up the test specimen in the usual way, using the top cap with plain ends

        (i.e.  without a metal hemispherical seating).

·        Fit the bell-mouthed flexible sleeve onto the top cap with the bell-mouth uppermost.

·        Lightly apply a thin coating of silicone grease to the inside of the bell-mouthed sleeve.

·        Fill and de-air the cell in the usual way.  When water runs out of the top vent tube, close it off with a straight connector and plug.

Carry out isotropic consolidations, B checks and saturation stages as required.

To dock the top cap and sleeve to the vented reaction head, the following procedures should be used:

· Shut the back-pressure valve on the cell.  As cell pressure is released during docking,  Shutting the back-pressure valve "locks in" the current state of isotropic effective stress for fully saturated soils.

· Shut the cell pressure valve and vent the cell to atmospheric pressure using the bleed valve on the top of the cell.

· Close the bleed valve on the cell and vent the top vent tube to atmospheric pressure by removing the plugged connector.

· Raise the open end of the water filled vent tube above the cell and secure in position.

· By turning the large knurled nut on the reaction head, lower the vented reaction head, displacing some water from the cell through the vent tube, while observing the pressure in the lower chamber as indicated on the lower chamber pressure controller.

· Continue to lower the reaction head until it locates in the flexible sleeve, and the lower chamber pressure registers a small permanent increase of pressure of, say, 1kPa.

· Now lower the open end of the water filled vent tube to below the cell.  This applies a small negative pressure to the sleeve and seals the top cap to the vented reaction head.

· Pressure to the required value in small increments of, say, 1kPa (or use the RAMP function).  During this procedure, continuously adjust the reaction head as required to maintain a pressure in the lower chamber equal to the cell pressure minus the friction correction plus, say, 1kPa. A seal between the top cap and the reaction head has been obtained if the cell pressure controller succeeds in restoring cell pressure. If the cell pressure cannot be achieved then the flexible sleeve has not sealed correctly. It is then necessary to repeat the procedure.

· Tighten the lock nut on the reaction head.

· Open the valve to back pressure.  Testing may now proceed.           

To undock at the end of the test, the following procedures should be used: 

· Turn off all valves to the cell and vent the cell pressure to atmospheric pressure using the bleed valve in the top of the cell.

· Raise the open end of the water filled vent tube above the cell and secure in position.  This applies a small positive pressure to the interface between the top cap and the reaction head.

· Release the lock nut on the reaction head and slowly raise the reaction head by turning the large knurled nut.

· Set the cell pressure to zero pressure.  When zero pressure is reached, disconnect the cell pressure line from the cell and empty the cell of water.  Dismantle the cell in the usual way.

Triaxial Test Hints

CO2 helps de-airing

De-airing pore water ducts, porous stones, connections and cohesionless test specimens prepared dry may be greatly facilitated by first purging with carbon dioxide CO2.  This heavier-than-air gas does not mix with air or with water at room temperature and pressure.  Complete purging may be detected by holding a lighted match over the outlet.  The lighted match extinguishes in CO2. Remember to purge the triaxial cell and use de-aired water there too.

Nold de-aerator 

Good quality de-aerated water may be obtained with the Nold de-aerator which utilises cavitation caused by a high speed rotated vane to release air from tap water.  This air is then removed by a vacuum pump.  De-aeration should be carried out immediately before use unless the de-aerated water is stored under a good vacuum. 

De-airing connectors

When making connections with the hydraulic connectors, ensure that the valves and connectors are filled with de-aired water right up to the outside orifice.  This may be facilitated by using a hypodermic needle. Remember to grind off the point of the needle for safety.

Side drains 

When testing clays in the triaxial cell, filter paper side drains may be used to facilitate the axial equalisation of pore pressures.  This reduces test time, gives more uniform stresses and strains, and means that the pore pressure measured at the base pedestal is more typical of the pore pressure regime in the test specimen.  For very soft clays, however, conventional vertical strip side drains may make a significant contribution to the rigidity of the test specimen. 

This problem may be overcome by using filter paper spiral drains cut to form a continuous helix. Of course, for triaxial extension, spiral side drains are essential. Remember to select a testing rate which allows equilibration of pore pressure.  If this is not done, you may end up testing a stiff outer shell of consolidated soil.

Membrane penetration 

When testing granular soils down to fine sands, membrane penetration causes false volume change measurements (unless cell pressure and back pressure do not change) which may be corrected for.

Connecting tubing

Polythene and PVC tube should be avoided because of their relatively high permeability to air. Remember to site the test cell close to the end of the pressure controllers and cut the nylon connecting tubing to minimum lengths.

A simple guide to the fitting of swagelock pressure fittings.

1. Ensure the end of the tube is cut straight using the supplied cutter. 

2. Place the Backing nut  with the screw towards the end to be attached. 

3. Place the Backing ring in front of the nut. 

4. Place the Olive  in front of the Backing ring.

5. Push the tube square into the connection and slide the nut to the connector. 

6. For the first operation tighten the nut by hand until "finger tight" and then apply three quarters (270 degrees ) of a turn using a spanner. 

7. For subsequent use during normal operation tighten the nut by hand and then apply one quarter (90 degrees) turn using a spanner.            

Temperature control 

High quality triaxial and consolidation testing may only be carried out in a rigorously temperature controlled environment.  Temperature variation causes changes in the volume of water in the soil and test cell, and changes in dimensions in the test cell which for strain controlled tests means changes in loading.  Accordingly, measurements of pore pressure, axial loading and axial deformation may be affected.

Back-pressure 

In spite of our best efforts in de-airing our test systems, this process must be imperfect! Accordingly, it is good practice to test at high back pressures wherever appropriate and thereby ensure the compression of any residual air with consequent improvement in saturation.  For undrained tests on dilatant soils, elevating pore pressure before a test also ensures pore water pressures do not become negative thus causing cavitation and invalidating the undrained condition. Remember to elevate cell pressure as well to keep effective stresses unchanged.