I have been trying to develop a set-up that could test the input voltage range and output swing of an op-amp.

To measure the input voltage range, the output should swing within the limitations mentioned in datasheet. The input ramp is chosen slightly beyond the input range limits, but still stays within absolute maximum ratings. For example: for an OPA735, with a ±2.5 V supply, the input voltage range is -2.6 V ~ 1 V and the output swing is -2.45 ~ 2.45 V.

If I have a ramp from -2.45 ~ 1.1 V and with a slope that doesn't cause slew limitations in a buffer configuration, I expect to see nonlinearity at 1 V output voltage because the ramp is not within input voltage range.

input voltage range test

Am I correct so far?

I have come across an op-amp: BH7673G. This one has a rail-to-rail input but an output swing of 3.8 Vp-p. If I ramp up the input, sweeping the entire input voltage range, the output would go non-linear because of output swing limitation, but I want it to get distorted because of input common-mode voltage range limitations to bench-test its input stage limitations by monitoring the output stage behavior.

My configuration seems to work only for op-amps with larger swing range compared to input voltage range. How can one test thd input voltage range of an op-amp like the BH7673G?

  • 1
    \$\begingroup\$ Can you link a datasheet? BH7673G does not appear to be an op-amp at all. The feedback is internal. \$\endgroup\$ Commented Feb 10 at 5:36
  • \$\begingroup\$ @SpehroPefhany I added the datasheet link at the bottom of the question. Yes the feedback is internal. So, is there a way to measure the input common mode range of this part? \$\endgroup\$
    – Fateme
    Commented Feb 10 at 5:49
  • \$\begingroup\$ electronics.stackexchange.com/questions/432775/… \$\endgroup\$
    – G36
    Commented Feb 10 at 9:21
  • \$\begingroup\$ @G36 Thank you but in the op- amps that were tested in the link you posted, input common mode voltage range and maximum output swing are equal. What I'm asking is for op-amps that have a larger ICMVR compared to maximum output swing. \$\endgroup\$
    – Fateme
    Commented Feb 10 at 12:50

1 Answer 1


Build a follower, but provide an additional stage with significant voltage gain and an output that is able to extend beyond the op-amp's own supply potentials:


simulate this circuit – Schematic created using CircuitLab

Note: Feedback is now to the op-amp's non-inverting input. This is necessary because MOSFET M1 inverts.

The potential at A will still follow input IN, but isn't as constrained as the op-amp's own output. The extra gain provided by M1 ensures that the output of OA1 is always well within its limits. Here's a plot of \$V_B\$ and \$V_A\$ during a sweep of \$V_{IN}\$:

enter image description here

I used -1V and +6V supplies here only to illustrate how MOSFET M1 can be configured to produce any output voltage in that range, at node A, and is far less constrained than the op-amp's output is, with its smaller supply range of only 0V to +5V.

However, you could use the same supplies for the MOSFET as for the op-amp, because the goal of this circuit is merely to prevent the op-amp output from saturating high or low, staying well within its range of achievable potentials, allowing you to focus on behaviour when the inputs stray beyond acceptable limits. As you can see from the graph, that goal is achieved: op-amp output \$V_B\$ is more-or-less constant at some value that the op-amp is perfectly capable of producing.

It's still a voltage follower, but this system will only fail to "follow" when OA1's input violates the op-amp's input constraints.

  • \$\begingroup\$ Or stick a Zener diode bypassed by a cap into the feedback path :) \$\endgroup\$ Commented Feb 10 at 14:52
  • \$\begingroup\$ @Kubahasn'tforgottenMonica Can you elaborate? You got me curious, but I confess I don't know what you mean \$\endgroup\$ Commented Feb 10 at 16:55
  • \$\begingroup\$ @SimonFitch why isn't the potential at A as constrained as the op-amp's own output? \$\endgroup\$
    – Fateme
    Commented Feb 12 at 3:57
  • 1
    \$\begingroup\$ @Fateme With MOSFET M1 fully on, \$V_A=-1V\$, when it's fully off, \$V_A=+6V\$. Node A can swing across the full range -1V to +6V. The op-amp output, by contrast, is at best constrained to be within its own supplies, 0V and +5V. \$\endgroup\$ Commented Feb 12 at 4:15
  • 1
    \$\begingroup\$ @Fateme, you don't have to use -1V and 6V, you can use the same supplies as the op-amp. I used -1V and +6V to illustrate how A is theoretically unconstrained. The goal, though, is to prevent OA1's output approaching the supplies, which will happen with 0V and +5V supplies for the MOSFET too. \$\endgroup\$ Commented Feb 12 at 4:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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