# Tuning Infinite Gain Multiple Feedback Active band pass filter

#### sdarquea

Joined Dec 8, 2022
3
I'm trying to integrate this ultrasonic sensor into my design and create the necessary circuitry for it. I asked CUI, the manufacturer, for any reference designs/guides for it and they sent me a pdf with a circuit that is attached here.

Between that, running it through Spice sim in Eagle, and the "PC Services" breakdown about another ultrasonic sensor found here I've been trying to fully understand what is going on and feel like I mostly understand, except for calculation of the resistors and caps for the band pass filter, which I would like to verify to make sure the resonant frequency of the filter is 48kHz to match the CUI device.

The bulk of the circuits and values in each example are nearly identical to one another, except for the resistor on the output of the op amp that makes up the band pass stage, one is 7.5k and the other 10k, but I don't think that has much effect on the filter stage itself. Also, one uses a separate transducer for Tx and Rx, while the CUI uses the same transducer for both Tx and Rx (at least that's what I've come to understand).

I've followed along some videos about IGMF filters like the one found here as well as this tutorial, but I can't make sense of the values I calculate as a result and the ones given in the schematics mentioned above, and don't know if its even possible to analyze the band pass filter stage outside of the context of the rest of the amplifier stages. Additionally, I've run the circuit through Spice Sim in Eagle, one with almost the whole circuit and one with just the band pass filter and have attached the results.

As you can see from the IGMF filter spice sim attached, using the values given I got the resonant frequency to be about 18kHz. When I change R5 in that circuit to 10k instead of 75k I get it to center near 48kHz, but that doesn't make sense to me computationally. I found the formula R' = R(fc/fc')^2 for adjusting the value of R4 where f and R are values for a given functional circuit and and f' and r' are just translations of values for a different frequency.

Using this I calculated R4 in that same circuit to be 694Ω, but this doesn't match sim. Additionally, working with the schematic in the IGMF filter spice attachment and following along the YouTube video mentioned above, I calculated R3 = 31.8k, R4 = 170Ω, and R5=63.7Ω, but none of these results match any of the sims I've run.

Can anyone provide me some insight about this circuit and the proper way to calculate/verify the values needed for resonant frequency of 48k. Ideally I would like to use values as close to the ones in the example schematics.

Thanks!

P.S. That was perhaps a verbose and convoluted description of my problem so please let me know if you need any clarification of what I'm asking or the process I've used to try and understand things.

#### Attachments

• 160.7 KB Views: 4
• 141.7 KB Views: 7
• 162.6 KB Views: 7
Last edited by a moderator:

#### Ya’akov

Joined Jan 27, 2019
7,050
P.S. That was perhaps a verbose and convoluted description of my problem so please let me know if you need any clarification of what I'm asking or the process I've used to try and understand things.
Welcome to AAC.

The verbosity isn't nearly as much of a problem as the wall of text. I've broken the post into reasonably coherent paragraphs to make it easier—and more likely—to be read. You should also use more sentences (periods instead of commas).

I hope someone is able to help, and again welcome—it's good to have you with us.

#### crutschow

Joined Mar 14, 2008
31,581
Sorry, but there's way too much info for me to try to understand in your long, rambling discussion without any separations or paragraphs to help.

If you describe the exact problem you have having with the filter, I will try to help with that.

Last edited:

#### Ya’akov

Joined Jan 27, 2019
7,050
Sorry, but there's way to much info for me to try to understand in your long, rambling discussion without any separation or paragraphs to help.

If you describe the exact problem you have having with the filter, I will try to help with that.
I broke it up as you were reading it, for what it's worth.

#### sdarquea

Joined Dec 8, 2022
3
Sorry, but there's way to much info for me to try to understand in your long, rambling discussion without any separation or paragraphs to help.

If you describe the exact problem you have having with the filter, I will try to help with that.
I do apologize for the rambling, just wanted to get across that I've made a decent attempt at trying to understand things before reaching out.

What I'm trying to accomplish is to make the necessary adjustments to the component values (or understand how to verify if they are already correct) to the CUI Devices circuit that is attached in order to tune the band pass filter stage to 48kHz resonant frequency.

I've attached some circuits that I made in Eagle and ran through the Sim as well as made some attempts at calculating the values but it seems I cannot find any agreement between what I calculate, the simulations, and the given values.

#### sdarquea

Joined Dec 8, 2022
3
Welcome to AAC.

The verbosity isn't nearly as much of a problem as the wall of text. I've broken the post into reasonably coherent paragraphs to make it easier—and more likely—to be read. You should also use more sentences (periods instead of commas).

I hope someone is able to help, and again welcome—it's good to have you with us.
Thanks for making those adjustments and for the tips! I will keep those in mind.

#### crutschow

Joined Mar 14, 2008
31,581
to the CUI Devices circuit that is attached in order to tune the band pass filter stage to 48kHz resonant frequency.
Can you post a clear picture of just that part of the circuit with the filter?
What are the required characteristics of the filter?

#### Audioguru again

Joined Oct 21, 2019
5,598
LTspice shows the same results as your simulation, a peak at 19.9kHz:

#### Attachments

• 24.9 KB Views: 4