Hello,
I am working to diagnose an unexplained offset voltage in our strain gauge measurements with the ADS1220. It appears that if the measurement board heats up independently of the strain gauges, we get about 200 uV (on this order of magnitude) input-referred offset in our measurements, after removing the temperature sensitivity of the gauges themselves. The board is anywhere from 50-80C when these measurements are made, while the gauges are 30-40C.
Some relevant details:
- Excitation voltage 2.5V
- ratiometric measurements
- full bridge gauge configuration, no additional resistors
- gauges are quite close to the board - about 3cm of magnet wire
- IC is close less than 1cm from edge of board where the gauge connection is
The theories we currently have:
1. Thermal EMF - it does appear the board, and thus the magnet wires termination at the board, are at a higher temperature than the gauges (the other end of the magnet wire).
2. ADC internal offset - the datasheet Offset drift only accounts for about 15uV. Is there something else that might contribute?
If the issue is not the chip, or not entirely the chip, could you recommend any sensing techniques that may cancel out or reduce this effect (depends on the root cause, of course)? We would like to reduce the error by about an order of magnitude, down to 20uV over the 50C range.
Thanks very much for your time.
