Checking and Evaluating Inclinometer Data Measurement

Inclinometers provide a measurement of time dependant displacement in a near vertical borehole. They may be installed to check that actual movements correlate to those predicted during the design phase. Also, measurement results of in place inclinometer could be use in early warning system. They are probably the most valuable tool available to a landslide monitoring. Two types of accelerometers are now being used in inclinometer probes: the servo-accelerometer and recently introduced MEMS accelerometer. The servo- accelerometer probes, have the highest resolution of the available inclinometers on the market. A maximum system accuracy of 1, 2 mm in 30 m or 1:24000 is possible with this instrument, but is ordinarily closer to 1:4000 without corrections for systematic errors. The resolution is nearly linear and constant at inclinations between ± 30° from vertical. MEMS technology has recently been introduced for inclinometer probes and in-place inclinometers. The primary advantages of the MEMS type are low power consumption, durability, wireless transmission and low cost. Maximum system accuracy can not be reached in field measurements. Slope Indicator Company, on the basis of many field measurements, state that the system field accuracy is ±7.6 mm per 30 m. This accuracy includes a combination of random and systematic errors. Random errors can occur within the sensors, limiting the precision of the probe. Systematic errors occur due to human actions that affect the condition of the sensors–probe and the data collection procedure. Random error is less influential error and it tends to remain constant during the measurements. But, systematic error tends to vary with each measurement. Systematic errors can be corrected using strict procedures but random errors can not be corrected. Random error can be minimized in better installations and with more precise reading procedures. Very often inclinometer accuracy have not been fully explained by inclinometers manufacturer. This paper describes these two kind of errors and how to eliminate them prior to presenting the results.