I'm communicating with an ADIS16210 using a National Instruments USB-8451 in conjecture with LabVIEW.
I've encountered some problems with the unit. First of all, the datasheet states that the readings from the accelerometers are given as two's complement numbers.
This is apparently not the case for the y-axis data. The accelerometer readings are 1 g when the unit is orientated with the y-axis vertically aligned. However, when gradually rotating the unit 180 degrees, one would expect a decrease, but an increase is observed. The increase is continuing until the accelerometer readings are wrapped around, reaching -1.7 g. and continuing to increase until -1 g i reached from below. The x and z-accelerometer readings are behaving more normal, but they're biased by approximately 0.5 g.
A factory reset (bit 3 at address 3E) has been tested, but it didn't improve anything. The bias correction terms have also been adjusted, without improving anything.
Is there a possiblity that the data are not stored as two's complement numbers? Or that the unit is (possibly accidentially) configured wrong?
Test readings of the Production identification number was successful and correct (indication that the convertion from read 1-d array of bytes into a signed integer is correct).
The unit's serial number (located at address 0x58) is 441.
Anyone with tips or similar issues?