Recently I started a new project using the ADXRS453 MEMS gyro, but the test results was far from the datasheet as described, especially null temperature specification. Here's the test conditions below：
1. The sample rate is 500Hz
2. The power supply is 3.3V and has 1% precision over temperature
3. The rate data is read from Rate (RATEx) Registers ( The RATEx registers contain the temperature compensated rate
output of the device, filtered to f0/200 (~77.5 Hz ).
4. The system containing ADXRS453 was placed in the temperature equipment with -30 to 70 .
Theoretically the null changes about 0.5deg/s from Ta=-40 to 125 ( ADXRS453.pdf page3 and figure13). However the test result we got is here blow (x:temperature y:gyro rate):
As the figure being showed, when temperature is higher than 50℃, gyro null drifts fast and null error increases.