## BNO085 Accelerometer Scale Factor Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

BNO085 Accelerometer Scale Factor

What is the scale factor for Accelerometer?

I received raw accelerometer data but how can convert to g?

I received mostly -160,80,4032 accelerometer data as x,y,z when staying stable position; but how can convert it to calculated accelerometer data?

demir

Posts: 6
Joined: Sun May 30, 2021 9:50 am

Re: BNO085 Accelerometer Scale Factor

Which library are you using and it there a reason you are raw ADC values and not the translated values the Adafruit libraries provide?

An you using specific example code? If so, what?

millercommamatt

Posts: 653
Joined: Tue Jul 31, 2018 4:57 pm

Re: BNO085 Accelerometer Scale Factor

demir's question was cross-posted in this discussion:
viewtopic.php?f=19&t=179820&p=875668#p875668

gammaburst

Posts: 579
Joined: Thu Dec 31, 2015 12:06 pm

Re: BNO085 Accelerometer Scale Factor

If the accelerometer is held still, with any axis perfectly vertical, the acceleration is exactly +/- 1 g along that axis.

Use that fact and the raw data to figure out the scale factor and offset for each axis.

sj_remington

Posts: 84
Joined: Mon Jul 27, 2020 4:51 pm

Re: BNO085 Accelerometer Scale Factor

No, gravity varies by about 0.7% due to Earth's irregular shape and composition.
Also, Earth's rotation reduces your apparent gravity as you approach the equator.
The Sun's and Moon's positions also have a slight effect (think tides).

gammaburst

Posts: 579
Joined: Thu Dec 31, 2015 12:06 pm

Re: BNO085 Accelerometer Scale Factor

Correct, but unfortunately, the BNO085 cannot detect those slight variations.

Which is why most people assume exactly 1 g for the purpose of computing MEMS accelerometer scale and offset factors.

sj_remington

Posts: 84
Joined: Mon Jul 27, 2020 4:51 pm

Re: BNO085 Accelerometer Scale Factor

The BNO085 (and other similar inexpensive sensors) can easily detect variations of 0.7% of a g.

Take care using rigid words such as "exact", "cannot", and "fact".

A 0.7% error is acceptable for many applications, so your suggested inversion method of calibration would be adequate, and it's easy to do.

gammaburst

Posts: 579
Joined: Thu Dec 31, 2015 12:06 pm

Re: BNO085 Accelerometer Scale Factor

Agreed!

When using the BNO085 to measure g at the North or South pole and at the top of Mt. Everest, one would certainly want to take those calibration considerations under advisement.

sj_remington

Posts: 84
Joined: Mon Jul 27, 2020 4:51 pm

Please be positive and constructive with your questions and comments.