Post Go back to editing

[AD8237] Strain gauge output offset

Hi team,

I am considering the AD8237 as a strain gauge gain amplifier.

The reason is that the natural offset of the strain gauge is large, about ± 500 mV.

With a general instrumentation amplifier, this offset will saturate the output.

I am thinking about signal gain of about 100 to 200 times.

At this time, I plan to use the DAC to adjust the output common mode.

Is this usage okay?

In the above formula, Vin(max) is very large at 500 mV, so RA becomes large.

RA has a big effect on the gain, but is there any other problem?

I think it is possible in theory, but I would like to know if there is any problem in actually using it in this way.

Best regards,

T-yoshi

Parents
  • Hi T-yoshi, 

    It looks like you are referencing the following article: https://www.analog.com/en/analog-dialogue/articles/how-to-stay-out-of-deep-water.html.  With an indirect current feedback architecture like the AD8237 uses, it allows the use of high gain even with larger offsets.  The spec to be mindful of is the differential input operating voltage.  Even at a supply of 1.8V, the AD8237 has can support +/-750mV range which is plenty of room if your max offset expected is 500mV.  Depending on VA (max) in equation coming from DAC, Ra should not get too large if say you are starting at a gain of 100 with R2=100kohm and R1=1kohm, the parallel combination of this is only 990ohm multiplied by Va(max)/500mV.  If the resulting gain is too high, you can always adjust the resistor choices to scale it to your target gain.    I believe this circuit should work well for you, but please let us know if you run into any additional problems.

    Thanks,

    Dave

  • Hi David-san,

    Thank you for your reply.

    Yes, I'm referencing this article.

    I was relieved to find that the expected input was adjustable in this way.

    I understand that the error can be adjusted in principle, but when actually calibrating using this method, is there anything I should be concerned about due to transient operation?

    Best regards,

    T-yoshi

  • Hi T-yoshi,

    You are welcome.  

    The larger the gain or input offset change, the more sensitive the output will be to changes in that offset. 

    I assume you will do a one time calibration, with no force applied to the strain gauge?  If you do not know the initial starting offset, then in a high gain setting it will likely start with the output of the AD8237 railed until you start to make a correction with the DAC to bring the AD8237 Vout-Vref to 0V. This is a DC based correction so it should adjust rather quickly, however you would still have to wait until the output has settled after each DAC code change.

    Will you need to calibrate periodically due to temperature change or other?  If that is the case, it would be good to know the amount of change in the +/-500mV offset expected to help with deciding the best gain choice and expected sensitivity.  

    Thanks,

    Dave

Reply
  • Hi T-yoshi,

    You are welcome.  

    The larger the gain or input offset change, the more sensitive the output will be to changes in that offset. 

    I assume you will do a one time calibration, with no force applied to the strain gauge?  If you do not know the initial starting offset, then in a high gain setting it will likely start with the output of the AD8237 railed until you start to make a correction with the DAC to bring the AD8237 Vout-Vref to 0V. This is a DC based correction so it should adjust rather quickly, however you would still have to wait until the output has settled after each DAC code change.

    Will you need to calibrate periodically due to temperature change or other?  If that is the case, it would be good to know the amount of change in the +/-500mV offset expected to help with deciding the best gain choice and expected sensitivity.  

    Thanks,

    Dave

Children
No Data