I am making a signal conditioner for an electret mic. I would like to have a 2 pole LP and a 2 pole HP filter both with a -3db point of ~2kHz as part of this.The upper limit is ~8kHz. Each would require an independant level control.

My first thought was to do it the analog way which I've done numerous times but it would need a bunch of parts and it would be untuneable. OTOH, it would not be very noisy.

I looked at SCF's, they're simpler to build, they're adjustable, and take up less space but OTOH may not be entirely suited for audio.

Does anyone know any problems with SCF's like an LTC1059/60? Any sucker traps in general?

Thanks,

Joe

FWIW- for years I was chastised about my 230# weight and how unhealthy it was. Now, due to stage IV cancer, I'm now at 275 and my BMI is normal for the first time since they invented the BMI chart.
What does the doctor do? He gives me stuff to gain weight!

 

SC filters have switching noise that can be a problem with low level signals. For example the LTC1059/60 has a 10mVpp typical clock noise. If this is undesirable then a continuous-time filter may be preferable.

You might consider a UAF42 filter building block since it simplifies the circuit and I believe allows single resistor adjustment of the rolloff frequency. The design equations look complicated, but this reference would indicate that they can be simplified by setting some resistors equal to each other (the capacitors already are equal in the IC). Even though one chip can generate the bandpass function, for ease of adjustment you may want to use two ICs, with one for the HP feeding one for the LP.

 

Thank you very much for this. It will work well and save space too. The formulae don't look that bad, they will keep me occupied.