1
\$\begingroup\$

I have a 2nd order Sallen-Key Low-pass filter, cut-off frequency of 170Hz.

enter image description here

When attached to a perfect source, this is the bode plot. As expected, 40dB/decade response. enter image description here

The low-pass filter is used as part of a 20Hz to 170Hz band-pass filter. enter image description here

When attached to a perfect source, this is the bode plot. Again as expected, 40dB/decade response. enter image description here

However, I attached the band-pass filter to an instrumentation amplifier that acts as a source. enter image description here

This is the bode plot now. 20dB/decade response. enter image description here

I attempted to override the high-pass filter, and attached the instrumentation amplifier to the low-pass filter directly. This is the bode plot.

enter image description here

This post suggests the capacitor values being low as a factor, but the capacitor values here are normal. The cut-off frequency of 170Hz is well within the GBP of the op-amp I'm using, the ADA4091, which is 1.22MHz.

Please let me know if more information is required.

Additional information: The remainder of the circuit, simulating a strain gauge at 100Hz. enter image description here

AC analysis for remainder of circuit. enter image description here

PR10 is the low-pass, PR3 and 4 (only 4 visible) is outputs from Left and Right, which is output from a simulated strain gauge. PR8 is a point within the InAmp.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ What did you apply to Left and/or Right? \$\endgroup\$ Dec 20, 2023 at 21:27
  • \$\begingroup\$ Have to see the whole schematic, as connected, for the final graphs. Looks like no power supply somewhere or else no connection of the AC 1 input source in a way that impacts the output. Details matter. \$\endgroup\$ Dec 21, 2023 at 1:22
  • \$\begingroup\$ @RohatKılıç I have updated the question, thank you. \$\endgroup\$ Dec 22, 2023 at 12:06
  • \$\begingroup\$ It's worth mentioning that this is a good learning exercise, but if you want to make a hardware implementation of this, most IC in-amps will blow the 3-op-amp one out of the water in terms of specifications. You can also look at integrated switched capacitor filters, some have differential inputs. \$\endgroup\$ Dec 22, 2023 at 20:29

1 Answer 1

3
\$\begingroup\$

Here's a clue that might point at a bad model for the InAmp (for example): -

enter image description here

It tells me that you have no signal of any reasonable value entering your filter from the InAmp.

Or, it might point to a bad connection of the source to the InAmp inputs. I can't be sure because your source connection isn't shown.

Or maybe the 100 Ω resistor at the output of the InAmp got changed to something too high in value.

Maybe something else but, for sure, the dB scale is firmly in the noise.

\$\endgroup\$
4
  • \$\begingroup\$ Thank you, I will investigate from this direction. I've also updated the question to show the source connection. \$\endgroup\$ Dec 22, 2023 at 12:09
  • \$\begingroup\$ @HFOrangefish have you used the scope function to look at the signal levels going into the InAmp? No need to post new images in your question of course. \$\endgroup\$
    – Andy aka
    Dec 22, 2023 at 12:16
  • 1
    \$\begingroup\$ I just did. The inputs at "Left" and "Right" are also suffering from low dB (-116dB). Signal at output of InAmp is also low, and so is the point PR8 inside the InAmp. I'm getting your point that this might be an InAmp problem as this happens even with the low-pass filter disconnected. \$\endgroup\$ Dec 22, 2023 at 12:47
  • \$\begingroup\$ @HFOrangefish are we finished here? \$\endgroup\$
    – Andy aka
    Dec 26, 2023 at 18:00

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.