1
\$\begingroup\$

I'm designing a circuit with an ICF (indirect current feedback) amplifier, and am trying to calculate the gain of the resultant circuit.

Using the AD8237 I followed the example here when designing with one modification. Instead of using a voltage follower, I used a resistor divider as the datasheet said I could (the ICF amplifier does not require a low impedance Vref like other 3 amp instrumentation amplifiers). Doing this does affect the gain, as can be seen from the snapshot from the datasheet below:

Amplifier with Vref voltage divider

The gain when Vref = 0 is G = (1 + R2/R1). I tried dusting off some Thevenin equivalent circuit knowledge (it has been quite a while) and calculated the resistance with R1 removed and the voltage sources and calculated it as (R2 + R3||R4). Replacing R2 in the gain equation with this gives G = (1 + (R2 + R3||R4)/R1) which is the same as what is in the datasheet. So far so good!

The challenge comes when adding the next part of the circuit - an offset voltage Vdac and resistor Ra as seen in the circuit below.

Full circuit

Doing the same as before, I calculated the resistance with R1 removed and all voltage sources shorted. From this I calculated it as (R3||R4 + R2||Ra) which when replacing R2 in the gain equation results in a gain of G = (1 + (R3||R4 + R2||Ra)/R1).

Is this the correct way of calculating this? Is the resultant gain I've calculated correct?

Edit: I don't think my math/understanding here is correct. I tried to derive equation (5) from the example as a check on my understanding and didn't have any luck. Can someone help me understand how that gain equation is being derived?

Many thanks!

\$\endgroup\$

1 Answer 1

0
\$\begingroup\$

Is this the correct way of approaching this?

In my opinion you might be making things too complicated. You have R3 and R4 that set the voltage at the reference pin so, instead of trying to drive your DAC offset control voltage to the FB pin via \$R_A\$, just drive the reference pin with your DAC output. Use a small value DAC output resistor (maybe 100 ohms) followed by a 10 uF decoupler to stop DAC clock noise then feed the ref pin.

If you want to restrict the offset range control from your DAC to give you less granularity then you can attenuate it.

\$\endgroup\$
1
  • \$\begingroup\$ Part of the reason for going to an ICF amplifier is to remove the sensor offset before amplification so that the amplification stages are not saturated before applying the Vref shift (which was happening in previous 3-amp Instrumentation Amplifier iterations of this design). Wouldn't driving the Vref directly with the DAC result in the same problem? Adding a cap to the output of the DAC is a good idea, thanks. \$\endgroup\$
    – Bales
    Mar 17, 2018 at 23:19

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.