3
\$\begingroup\$

I'm using an instrumentation amplifier (AD8421 from analog device) to amplify very small signals in magnitude from 20uV to 500uV max. the problem I have with this is that the signals might have some offset voltage (depends on noises) of up to 100mV. Now because I need high gain to get the signals, these offset voltage can easily saturate my output. My circuit

In order to resolve this issue, I used an Op amp in reference pin to adjust the reference according to signal. The pic above shows my circuit. This works fine, but I need to change the value of R9 (in U3 LT1028) from around 100k to 1M. How can I do this according to my input signal? I thought of using a FET to change this but I can't figure out how to do this.

Some basic explanations: VG1 is my input signal, U1 Op amp is a HPF which is not really my question here, and U3 is my reference op amp amplifier. Also, R1 to R4 (on the left side of U1) are supposed to suppress these offsets for me but for some reason they don't do it. Also, the device is used for signal acquisition from rats, so the offset really depends on grounding of the animal body and that is something that may not happen always.

\$\endgroup\$

2 Answers 2

1
\$\begingroup\$

I'd use an integrator between the output of the In-amp and its reference pin. This is the tried and tested way to remove rogue dc offsets from the output and it works automatically because of negative feedback.

However, if the dc input level is so high it swamps the In-amp input stage then ac couple the sensor signal to the input.

\$\endgroup\$
2
  • \$\begingroup\$ So you'rs saying that U3 should be integrator? I'll see what happens. \$\endgroup\$
    – jimiweber
    Sep 25, 2016 at 7:11
  • \$\begingroup\$ Sorry, I was a little busy with some stuff, I tested your suggestion and it seems to be working. Though I haven't worked on it thoroughly. \$\endgroup\$
    – jimiweber
    Sep 26, 2016 at 19:00
1
\$\begingroup\$

In biomedical applications, galvanic electrode voltage and noise will degrade signal! and induce galvanic offsets, so for example, this is why "active guarding" is used to derive the common mode voltage then buffered and fed back to "right leg" (arbitrary). (hint look up)

Otherwise in general applications a select on test, DC null adjust is required, but if not possible a HPF such as @Andy's integrated DC error negative feedback to input offset control. This can be designed for any low frequency cutoff of HPF.

Here's an example of a precision INA with extra op amp for offset adjustments with an example of 10V common mode and 10mV dc offset with gain x1000 enter image description here

ref http://support.technologicalarts.ca/docs/Components/amp04.pdf

\$\endgroup\$
3
  • \$\begingroup\$ about what you mentioned for DC null adjust, can you elaborate more \$\endgroup\$
    – jimiweber
    Sep 26, 2016 at 7:05
  • \$\begingroup\$ This AMP04 that you mentioned uses a different topology than the instrumentation amplifier I'm currently using, but AD8421 actually has much less offset voltage (an order of magnitude smaller). Do you think that AMP04 topology works better because of its topology? \$\endgroup\$
    – jimiweber
    Sep 26, 2016 at 8:49
  • \$\begingroup\$ only if you need a stable offset correction by trimpot. \$\endgroup\$ Sep 26, 2016 at 15:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.