If a geotechnical instrument is not working perfectly before installation, it is not likely to work well afterwards. It may be faulty at the time of shipment to the user, and/or it may have been damaged in transit. The user should always make pre-installation acceptance tests.
The following guidelines for pre-installation acceptance testing apply to vibrating wire piezometers, with particular reference to installation in boreholes. Before finalising these guidelines, I asked for comments from some of the manufacturers of vibrating wire piezometers - Geokon, USA; Geotechnical Instruments, UK; Roctest, Canada; Slope Indicator Company, USA and Soil Instruments, UK - and have included all their suggestions.
The term 'temperature transient' refers to different parts of the instrument being at different temperatures, such that strains occur within the instrument and the reading changes until an equilibrium temperature is reached.Temperature transients cause false readings.
This is the reason why an inclinometer should be left at the bottom of the casing for about 10 minutes before starting readings.To observe this effect on a vibrating wire piezometer, hang it out of the window on a hot or cold day, and take a reading every minute.Alternatively, hold the piezometer lightly in the hand, and watch the reading change.
On a hot or a cold day, if a vibrating wire piezometer is lowered down a water-filled borehole, it will take between 20 minutes and two hours, depending on the manufacturer, to achieve a uniform temperature and hence to give a correct reading.
It is therefore unreasonable to try to check readings against the head of water in the installation borehole as the piezometer is lowered - nobody is going to wait long enough for temperature equilibrium. Reading errors caused by temperature transients cannot be corrected by using the manufacturer's temperature correction factors, because these assume uniform temperature. Almost all instruments are affected by temperature transients - this issue is by no means limited to vibrating wire piezometers.
Checking the zero reading
When a vibrating wire piezometer is in air it measures atmospheric pressure, just like a barometer.When the piezometer is submerged in water that is open to the atmosphere, it measures the sum of the water pressure and atmospheric pressure.
Manufacturers provide a pre-shipment 'zero' reading and corresponding barometric pressure with each piezometer, and a temperature correction factor. In my view it is essential that users make a high-quality zero check before installation, after allowing for temperature equilibrium, and this should be compared with the manufacturer's pre-shipment value, correcting for barometric pressure and temperature. If the two differ by more than an amount recommended by the manufacturer (or acceptable to the user), the manufacturer should be contacted.
In a recent specification for field installation I wrote the following wording for checking the zero reading:
1. A pre-installation acceptance test shall be performed on all vibrating wire piezometers, as specified herein, to check the zero reading. Throughout such testing, and at all other times prior to installation, the piezometer shall be not be dropped or otherwise mishandled.
2.The piezometer shall be placed in an indoor environment at a constant temperature of +/- 2degrees C, for a minimum period of one hour [this time is adequate for all piezometers manufactured in England, USA and Canada, but should be reviewed with manufacturers in other countries], to achieve thermal equilibrium.A piezometer reading shall then be made, together with a reading of temperature by using the internal temperature sensor [but note that some manufacturers do not fit an internal temperature sensor as standard, but include it at the customer's request, so the customer should specify its inclusion, and a reading of barometric pressure by using a barometer with a minimum accuracy equal to +/- 0.025 percent of the full-scale range of the piezometer.
3.The piezometer reading, when corrected for temperature and barometric pressure in accordance with the manufacturer's instructions, shall agree with the factory zero reading to within +/-0.5% of the full-scale range of the piezometer.
4.Piezometers that do not meet the above criterion shall be returned to the manufacturer.
5.After such pre-installation acceptance testing, vibrating wire piezometers shall be replaced in the protective material that the manufacturer used for shipping, and the protective material shall not be removed until immediately prior to installation.
Note that the above wording does not call for a check on the slope of the calibration plot. If changes have occurred between factory and site, they are much more likely to be changes in zero reading than changes in the slope of the calibration plot.
Checking the slope of the calibration plot
Although changes in slope of the calibration plot are unlikely, some users choose to verify that this has not occurred. Because it is often difficult to duplicate factory conditions in the field, checks on the slope of the plot should be regarded more as function checks than as calibration checks.
Two methods are possible, in priority order (trying to check the readings as the piezometer is lowered down the installation borehole is not a viable option, for the reasons given above):
1.Connect the piezometer to a known air pressure source, and check a few points within the range.
This can be done in a laboratory. Alternatively some manufacturers provide a small portable pressure chamber with a precise electrical pressure transducer for this purpose.
2.Lower the piezometer within a water-filled pipe, and check that a few readings agree with the level of water in the pipe. If an installed inclinometer casing is available, this can be used.Remember that the water level varies as the piezometer is lowered or raised, because of the volume of the cable below the water level. Note also that if a piezometer with a dry filter is placed in water, surface tension in the filter will affect the reading.Hence, for checking, either the filter should be removed or the filter and cavity should be completely saturated.The same problem can arise if the piezometer is placed in a sand-filled bag.
For both methods, it is essential that enough time is allowed for thermal equilibrium.
Pre-installation testing of vibrating wire piezometers by the user is strongly recommended. Testing should consist of checking the zero reading. Checking the slope of the calibration plot is possible, but is usually not necessary.
John Dunnicliffmay be contacted at firstname.lastname@example.org