MPXV7002 warm-up drift

[r3user]

Hello,

We are using pressure sensor MPXV7002 in our application where it needs to sample at 100Hz (also averaging multiple samples to reduce the noise). The device needs to start sampling as soon as it is turned on (few hundred ms is OK). We are observing a noticeable zero pressure DC shift after power on. It lasts for about 8-10 seconds before settling down. The change is not significant but significant enough to hurt the tiny range where our device operates. The effect is very similar to what this article describes. https://www.allsensors.com/engineering-resources/white-papers/warm-up-drift 

Has anyone seen something similar? If this is normal, what are the best compensation techniques? Should we do what the article describes (power modulation needing hardware change)? If the dynamics of the shifting remains constant of over time, then it should be possible to fix it in software. Any suggestions?

This is an unrelated question but will this sensor be supported/manufactured by NXP in coming years? We hope it is not going obsolete soon.

Thanks..

G

[david_diaz]

Hello,

The errors of the NXP integrated pressure sensors, are minimized by calibration and compensation done after assembly, and before the gel fill operation. However, because of slight variations in fabrication, every device is produced with a unique temperature characteristic. Both the magnitude and the sign of the temperature coefficient are variable from unit to unit. The best way to get higher accuracy is to calibrate each sensor at multiple temperatures.

What is the temperature condition of your device? 

In this case I recommend implementing an Auto-Zero function.

Sources of offset errors are due to device to device offset variation (trim errors), mechanical stresses (mounting stresses), shifts due to temperature and aging. Performing auto-zero will greatly reduce these errors. The amount of error correction is limited by the resolution of the A/D. In pressure sensing applications where a zero-pressure reference condition can exist, auto-zero can be implemented easily when an integrated pressure sensor is interfaced to an MCU.

You may find more information here​.

In regard with your last question, I cannot guarantee how long this part will be available, since it is not in the NXP Product Longevity program. The life cycle description of the product is Maturity/Saturation. I hope it will be available for longer period and it depends on market/customer's demands.

Please let me know if this information is useful for you.

If I misunderstood your question, feel free to let me know.  I will be glad to help.

Have a great day.

David Diaz.

Note: If this post answers your question, please click the Correct Answer button. Thank you!

[r3user]

Hello David,

Thanks for the detailed response. We do perform auto-zero for the offset calibration. The zero-offset I described happens at the sensor startup due to warmup (first 8-10 seconds). This is our hypothesis and the article I shared seems to support it.  Look at this sample data we collected. Is there a recommended auto-zero calibration method to compensate for such drift. In our application, it is not possible to wait for 8-10 seconds.

May the first question I should have asked is, is such behavior normal with Freescale pressure sensors? If not, we need to probe our A2D and power supply.

Let me know if I am not clear.

Thanks!