Frequently Asked Question

A MEMS accelerometer differs from integrated circuits in that a “proof mass” is machined into the silicon. Any displacement of the component causes this mass to move slightly according to Newton’s second law, and that change is detected by sensors. Usually the proof mass disturbs the capacitance of a nearby node; that change is measured and filtered.

(Source: https://www.digikey.com/en/articles/techzone/2012/jul/a-designers-guide-to-mems-sensors)

As a result, all acceleration MEMS have Cross tilt sensitivity, which means an input in one axis may be partly sensed in another. For instance, If we are measuring the X axis, then Y and Z axis are called cross axis. The output of X axis will be influenced when Y or Z axis also tilt.

The cross tilt sensitivity is often specified in the MEMS datasheet. For example a small, low-power 3-axis accelerometer ADXL335 has a cross-axis sensitivity of ±1%.

https://www.analog.com/media/en/technical-documentation/data-sheets/ADXL335.pdf


In our datasheet, we have specified that the cross tilt error can be calculated with a formule:
Cross tilt sensitivity error: < (0,12 * cross tilt angle)² % typ.
This is an indication that there is an extra error on top of the accuray specified in the datasheet.

For example, QG40N-KDXYh-090-AV-CM-UL,we use X axis as the primary sensing axis.
If the Y-axis tilt is 0° --> X-axis does not suffer from cross tilt error caused by the Y-axis. You will have an ideal output as

If the Y-axis tilt is 10° --> X-axis is suffering from cross tilt error (caused by the Y-axis) with an amount of (0.12*10)²% = ±1.44% from X-axis value. That means when X tilt 45°, the cross tilt error caused by Y axis will be 45°*1.44%=0.648°.


Here are more articles which explain the genernal concept.
https://www.mouser.com/pdfdocs/an32_crossaxis_compensation.PDF
http://cache.freescale.com/files/sensors/doc/support_info/SENSORTERMSPG.pdf
https://www.murata.com/en-global/support/faqs/products/sensor/accel/char/0008