Hi.

I think you just closed the thread, ADXRS 453 null bias stability . and i wanna ask here if you don't mind.

The null value is so critical when we want to obtain "Angular position" from "angular velocity" by integrating over time.

Small Null value (for example 0.5 degree per sec that mentioned in datasheet in Null Accuracy section) will cause 1 degree error in 2 sec !!! And my research shows there is no way except integrating over time for obtaining angular position from MEMS gyroscopes.

Please let me know if you suggest any method for solving this problem.

Looking forward to hearing from you.

Joe

Hello Joe,

Thank you for your post! You are correct in that bias error in the gyroscopes will dictate the drift rate in your angular displacement estimate, when that estimation comes from integrating the gyroscope outputs over time. When you say "solve this problem," I am not sure that I understand what your specific goal is. It is not possible to make this contribution go to zero, but it is possible to reduces its effect by observing the bias and subtracting that bias error from each gyroscope measurement, before it goes into your integrator. Is that what you mean?

Best,

Mark