Hi Matt,

I am working with the ADIS16488 for an application that is similar to the one mentioned above.

Our application is integrating the gyros over time. I am trying to compensate for the effect of the Earth's rotation on the sensor.

I have tried various sample sizes to attempt to eliminate the bias error.

The data that we are getting from stationary tests show that the uncompensated gyro drift is approximatly 30-40 degrees per hour. From the datasheet I was expecting about 6 or 7 degrees per hour.

Is there an optimum bias sample window to eliminate the bias error?

Thanks,

Rich

Hello Rich,

I branched this into a new discussion, since it was related to the ADIS16488, instead of the ADXRS450. The 6.25 deg/hour number (In-run Bias Stability) comes from the minimum point in the Allan Variance curve, which is associated with an integration time of ~50 to 200 seconds. The ADIS16488 provides a "continuous bias estimator" that can simplify this process. This function continuously averages the sensor outputs (gyroscope and accelerometer), so that it can update the user correction registers on demand. During the averaging period, the following conditions can impact accuracy:

NOTE: continuous bias estimator is synchronous with internal data production.

REFERENCES:

I hope that helps! Let us know what you find out.

Best,

NevadaMark