This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TS5A4624: Component selection for multiplexed strain gauge interface

Part Number: TS5A4624

Hi all,

I'm using R+ and R- combined strain gauge, 3 wires like this :

 Résultat de recherche d'images pour "3 wire strain gauge schematic"

I connect the common point (RED) to GND and then use TS5A4624 to multiplexe White and Black into a voltage divider and then to an analog to digital conversion block:

It's working quite good but now I try to make it great. Main issue for now is the temperature sensibility. 

In my application V+ is 3.0V so the current is about 1.5mA in the multiplexer and my LSB is about 3mohm so about 2nA variation in sensor. So maybe the leakage current variation with temperature is becoming an issue ? Maybe also the discrepancy between R_on on NC versus NO way is getting worse with temperature?

Do you think there is a better component (pin to pin compatible) to keep good matching characteristics between NC and NO ways in temperature (and over time)?

I don't care about variations on both ways as I subtract the results of both measurement to get my useful result. So I just need to avoid mismatch between channels characteristics. Higher R_on ( <50 ohms) are also possible as long as the discrepancy between channel characteristics is better.

Any comments / idea are welcome,

Have a good day,

Thomas

  • Hi Thomas,

    That is a fun strain gauge to work with as I have used that part as well!

    Based on your selection criteria you have chosen the best part for your system! While low leakage is nice in this application - the variation in On resistance mismatch will be more limiting that from leakage currents. Higher Ron devices typically have a larger variation in resistance mismatch so that doesn't seem like the best route for your system.

    If you are concerned about temperature characteristics you can make sure to do a multi-temperature calibration to help resolve the smallest signal. Additionally, it may be good for you system to have ratio-metric supply voltages for all of your system components. 

  • Hi Dakotah,

    On resistance mismatch between channels is not a concern as we only work in relative way so initial offset is not an issue.

    But will Higher Ron devices exhibit higher Ron variations from one channel to the other in temperature (I've never seen those curves in datasheet) ?

    multi-temperature calibration was already the back-up plan (due to higher cost I was trying to avoid it ^^).

    Thanks for help !

    Regards,

    Thomas

    P.S. I upload again the figure of original post as the link seems to be broken:

  • Hi Thomas,

    Thank you for the image.

    You are correct - a device with higher on resistance can have a larger variation from channel to channel over temperature.

    I definitely understand working to avoid the cost of multi-temperature calibration. Just wanted to make sure the option was mentioned :)