2023-10-11 11:36 PM
Hello
We use a LIS2DW12 accelerometer to mesure tilt over a temperature range from -25°C to +70°C.
We set the full scale to 2g and chose High Performance mode with Low Noise enabled.
We want an accuracy of 0.5°, but we're observing a drift of the order of 40 mg on measured accelerations as a function of temperature (from 15° to 75°C), resulting in a drift of more than 2° on the tilt measurement.
We want to correct the accelerations as a function of temperature, and so we have the following questions.
- Is the drift the same on all 3 axes?
- The data sheet states that the 0g offset drift is 0.2mg/°C, but according to our measurements this is not linear. Is it normal ?
- Does the drift depend on the orientation of the sensor ?
- What law should be applied to correct the accelerations measured according to temperature ?
You will find in the enclosed file, the shape of the deltas of accelerations versus the temperature (the reference is 15° to calculate the deltas of accelerations)
Thank you by advance
Anne-Marie
Solved! Go to Solution.
2023-10-12 11:34 PM
Hi @AnneMarieBONNET ,
Welcome to ST Community!
That behavior is very strange and unexpected.
I confirm that the typical drift is 0.2mg/°C, so in your range you should see a drift of 12 mg if you have Vdd=1.8V. Here my answers to your questions:
- Is the drift the same on all 3 axes? Yes, it is pretty the same.
- The data sheet states that the 0g offset drift is 0.2mg/°C, but according to our measurements this is not linear. Is it normal ? No, it's not. In all our measurements we see a linear behaviour.
- Does the drift depend on the orientation of the sensor ? No, it doesn't but you need to keep the device in a steady state condition and in a mechanically controlled environment during the test.
- What law should be applied to correct the accelerations measured according to temperature ? It could be related to environmental noise or caused by soldering that induces mechanical stresses so check this.
If my reply answered your question, please click on Accept as Solution at the bottom of this post. This will help other users with the same issue to find the answer faster!
2023-10-12 11:34 PM
Hi @AnneMarieBONNET ,
Welcome to ST Community!
That behavior is very strange and unexpected.
I confirm that the typical drift is 0.2mg/°C, so in your range you should see a drift of 12 mg if you have Vdd=1.8V. Here my answers to your questions:
- Is the drift the same on all 3 axes? Yes, it is pretty the same.
- The data sheet states that the 0g offset drift is 0.2mg/°C, but according to our measurements this is not linear. Is it normal ? No, it's not. In all our measurements we see a linear behaviour.
- Does the drift depend on the orientation of the sensor ? No, it doesn't but you need to keep the device in a steady state condition and in a mechanically controlled environment during the test.
- What law should be applied to correct the accelerations measured according to temperature ? It could be related to environmental noise or caused by soldering that induces mechanical stresses so check this.
If my reply answered your question, please click on Accept as Solution at the bottom of this post. This will help other users with the same issue to find the answer faster!