0
\$\begingroup\$

I designed a bandpass filter with the TI filter designer tool but I initially designed with a gain of 40 dB. However, I'd like to reduce the gain to 20dB but I'm having issues doing trying to figure out which resistors to change for the gain ratio. It's supposedly a Sallen-Key Topology. My electronics knowledge is very basic and I'm mainly trying to design a pre-amplifier for a hydrophone to be passed on to an ADC but the gain on my output suffers from two issues:

  • Amplitude of the AC signal is 7.8V pp but my ADC can only take up 3.3V.
  • ADC is also unipolar but I have negative voltages in my signal for this I'm thinking of using a summing amplifier.

enter image description here

Can I adjust the gain by changing resistor values, or is it a bit more complicated than that?

\$\endgroup\$
  • \$\begingroup\$ 20 dB on an input signal of 7.8 V will give you an output of 78 V. Since the maximum supply voltage of OPA659 is +/- 6 V, this is not going to work. \$\endgroup\$ – The Photon May 24 at 17:45
  • \$\begingroup\$ Its more complicated, as the gain of the op amps here affects the Q of the poles as well as the DC gain. \$\endgroup\$ – user69795 May 24 at 17:58
  • \$\begingroup\$ @ThePhoton I think the OP means that with 40 dB gain the output is 7.8 V and so they want only 20 dB so the output will be less. \$\endgroup\$ – jonk May 24 at 18:07
  • \$\begingroup\$ @jonk is right, with a 40 dB gain, the output is 7.8V. my input signal is only about 100mV peak to peak \$\endgroup\$ – MaskedAfrican May 24 at 18:25
0
\$\begingroup\$

It's more complicated than that. The resistors & capacitors more or less set the gain and filter characteristics simultaneously. The easiest thing to do would be to go back to the TI tool and enter in your new numbers.

To solve the problem of your ADC range being from 0-3.3V, you can use a "pseudo-ground" that's right in the middle of your ADC range (I'm assuming 1.75V). There are a lot of ways to skin this cat -- in your case, a simple RC circuit should do, as pictured below. There's an almost endless number of fancier ways, but if your circuit's going to be always on, and you don't care about a bit of DC bias in the signal going into your signal processing, this should be OK.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$

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.