0
$\begingroup$

I want to predict the maximum output error from nonlinearity for the working range of a MEMS accelerometer. I am given in the accelerometer spec sheet a nonlinearity value of 1% FSR for a ±40 g range. How do I convert the %FSR value to mg/g^2?

Taking the spec sheet error of 1% (0.01) and multiplying it by a 40 g yields a 0.400 g (400 mg) of error from nonlinearity at a 40 g input. Since the nonlinearity is a second-order error, the max error curve can be fitted to the 0.400 g value at 40 g. Thus the coefficient of nonlinearity would be 400 mg/40g^2 or 0.25 mg/g^2 which can be multiplied against each input value of interest under 40 g to get the rest of the nonlinearity error curve values. Is this correct?

Improve this question
$\endgroup$
2
  • $\begingroup$ IMHO, if you are after the maximum output error from nonlinearity then you should take the full range value i.e. 0.4g. $\endgroup$
    – NMech ♦
    Oct 7, 2021 at 14:08
  • $\begingroup$ full scale span is 80g, not 40g ... so I read the "1% nonlinearity", as a deviation from best fit straight line, of up to 0.8g from the line. Without knowing about the fundamentals of the sensor technology, there's nothing to say the worst case non linearity wouldn't be at mid-span, i.e. at 0g (although the offset associated with it would presumably be calibrated away). $\endgroup$
    – Pete W
    Oct 7, 2021 at 14:42

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.