I'm characterizing ADIS16251 gyroscope, particularly about its bias as function of time and temperature. The selected setting for both experiments was dynamic range ±20°/sec, number of taps 128, and sampling frequency 100Hz.
Firstly as a function of time, I did capturing the data for about 30mins (at gyro temperature 45-46°C). Below is the plot I get using Alavar 5.2 freeware:
My question is how to interpret this data? What is the meaning of the first increasing slope?
And what will the in-run bias stability be? At the minimum 0.002°/sec of integration time 0.01sec (i.e. the sampling period itself) ,or the other minimum point of 0.0134°/sec of integration time ~40sec, or at any other point?
Secondly as a function of temperature, I obtain below plot:
My problem is that I seem to obtain >0.2deg/sec difference in the same temperature, but different time of sampling. The colors in the plot indicate different days of data collection. This experiment was intended to obtain a fitting equation so I'll be able to predict the bias at any particular temperature. Due to the spread of data, I'm not sure if a fitting equation would be appropriate (it'll be with large error at some temperature range).
Do you have any suggestion on this matter?
My application will involve the gyroscope to be in a field without temperature control, thus its temperature may vary from minus to positive degrees.
Thank you in advance for your time in answering.