I'm designing an audio mixer in which I'd like to have a variable 2-pole high pass filter on each channel. For stereo channels I want to use one control for both channels. I've come up with a few options but I'd like to know of any I haven't thought of. I'm shooting for variable frequency range between 5-6 octaves and Butterworth characteristic, and good distortion performance. For my own use.

1. VCA-based filter using THAT 2162. This is the best distortion-wise but would be pretty expensive to use one of the dual-VCA chips in each channel; expensive enough that the filter components might end up costing more than the rest combined (not including enclosure).

2. OTA-based filter using LM13700. Higher distortion but lower cost.

3. State-variable design with variable feedback resistor. This isn't the best for variability over the range I want, and would be subject to the accuracy of dual pots for accuracy between channels of a stereo pair.


So what other options are there that I haven't thought of? I'd like to go with the VCA solution but I'm balking at paying so much for a part of the design that won't even be used all the time.

 

So what other options are there that I haven't thought of?

Switched capacitor filter?

 

The THAT2162 would be my first choice, as well; THAT Corp.'s line of Blackmer Cell devices are pretty nifty and can do some amazing things.

There is another transconductance amplifier you might consider, and that is Linear Tech's LT1228 (http://www.linear.com/product/LT1228). Page 18 of the data sheet shows a design for a voltage-controlled state variable filter. I've no idea what kind of performance you might get from this part, but it looks interesting.

 

I'm designing an audio mixer in which I'd like to have a variable 2-pole high pass filter on each channel. For stereo channels I want to use one control for both channels. I've come up with a few options but I'd like to know of any I haven't thought of. I'm shooting for variable frequency range between 5-6 octaves and Butterworth characteristic, and good distortion performance. For my own use.

1. VCA-based filter using THAT 2162. This is the best distortion-wise but would be pretty expensive to use one of the dual-VCA chips in each channel; expensive enough that the filter components might end up costing more than the rest combined (not including enclosure).

2. OTA-based filter using LM13700. Higher distortion but lower cost.

3. State-variable design with variable feedback resistor. This isn't the best for variability over the range I want, and would be subject to the accuracy of dual pots for accuracy between channels of a stereo pair.


So what other options are there that I haven't thought of? I'd like to go with the VCA solution but I'm balking at paying so much for a part of the design that won't even be used all the time.

Used to be various consumer ICs that did stereo tone controls in a single chip, very often with DC controls - that is; the volume was controlled by a DC potential applied to one of the pins. Some had DC controlled tone functions as well.

The European consumer ICs often had a TDA prefix, but Japanese makes exist, with most manufacturers having their own unique prefix.

 

There are parametric equalizer modules available at retail prices but I don't think you want to throw money at this instead of working it out yourself. Still, the words, "parametric equalizer" lead into a field of audio work that might give you ideas.

 

I don’t know your audio preferences or sensibilities, but you might want to consider a DSP solution.

You may change any audio characteristics you want, without changing hardware.

 

To calibrate your expectations, multiple, multiple-pole, tracking filters of any kind will be relatively expensive. More expensive than any other function block in the mixer? Absolutely. A technique from all-analog audio boards of a past era - photoresistive cells. Light bulb (nowadays an LED) and cadmium sulphide light-dependent resistor (LDR) close coupled in a light-tight package. Similar to an optocoupler in function, but with near-perfect audio performance. Getting several of them to track took a few adjustments or sorting through lotsa parts for matched sets. A 2-pole Sallen-Key LPF has two frequency setting resistors, so tracking stereo channels would take 4 LDRs.

http://www.calculatoredge.com/electronics/equal comp high pass.htm

ak

 

Thanks for the replies.

I've looked into switched capacitor filters before but, at least in the case of the LMF100, it doesn't look possible to get a cutoff frequency as low as 20Hz and have the switching frequency be beyond the audio range.

The LT1228 looks to have marginally better distortion than LM13700 but is even more expensive than the VCA option.

The tone control/equalizer circuits I've seen don't seem to have high pass functions. Low frequency shelving, but not high pass.

Converting to digital then back to analog just for a high pass filter seems needlessly complicated.

Maybe I'll try out the LDR idea. Otherwise I suppose I'll just accept the distortion from the OTA.