Are the wipers of R10 & R11 adjusted separately or are they mechanically connected (2-gang potentiometer)? Do you or they have a simulation file for this design?

It's a dual gang, single shaft, yes I know the schematic is missing the dashed lines between the 2 pots. No I haven't figured out how to simulate the circuit in Eagle yet. But the HPF is based on this design.

 

I thought Eagle was a schematic capture and layout package. I was not aware that it could do simulation.

 

I thought Eagle was a schematic capture and layout package. I was not aware that it could do simulation.

It has SPICE models in it, but you may be right. I'm not sure, haven't figured it out.

 

This might be feasible for a low pass application with an analog anti-aliasing filter. This is a high pass application where all the frequencies above the Nyquist limit will be folded back into the baseband.

Agreed. You still need a hardware/analog anti-aliasing filter.

 

Agreed. You still need a hardware/analog anti-aliasing filter.

I'm afraid this is above my knowledge level. I thought anti-aliasing was only needed in analog-digital conversions and not needed in DC applications.

So is another filter beyond the 2nd order HPF needed in my application?

 

I decided to do an LTspice version of your Adjustable HPF. I may or may not have replicated the circuit accurately. If I did, I don't think it is doing what you want it to do. Rather than moving the corner frequency it is shifting the shape and the attenuation in the passband. IMHO this is NOT what you intend to do. The initial gain stage is fine, but the rest is (fill in your own pejorative term).

The following traces show the gain stage V(O1) and the final output V(Vo) at the extremes of adjustment, [0.15,,0.95].

 

You can look up switched capacitor filter, for example,

https://www.analog.com/media/en/technical-documentation/application-notes/an40f.pdf

 

Check out the following Application note from TI:

Filter Design in Thirty Seconds -- SLOA93, December 2001​

​

For additional ideas on filter design, I think this might be closer to what you have in mind. Notice that over the range of interest the passband gain is constant and identical for the simulated settings. Bigger values for resistor RK move the corner lower.

 

Hmmmm, sounds like you're onto something.
So R1=2*R2. I wonder why this is. Is it possible to use the same dual gang pot? If I put a 50k across the 50k pot, that's close to 1/2 of R through most of it's range. It would be fairly accurate at the lower end of the HPF... but then the ratio would get further from 1/2.
I also need to use a voltage divider arrangement to get 4.5V off of one leg of the pot, and it connects back to the opamp a little different. Can I still use the 2nd side of a TL072?

 

Hmmmm, sounds like you're onto something.
So R1=2*R2. I wonder why this is. Is it possible to use the same dual gang pot? If I put a 50k across the 50k pot, that's close to 1/2 of R through most of it's range. It would be fairly accurate at the lower end of the HPF... but then the ratio would get further from 1/2.
I also need to use a voltage divider arrangement to get 4.5V off of one leg of the pot, and it connects back to the opamp a little different. Can I still use the 2nd side of a TL072?

I think you may have drawn the same conclusion that I did from this simulation I did 2.5 years ago. I reworked it to cover your frequency range of interest, used a TL072, and replaced the V+/2 voltage source with a standard voltage divider.

 

If you are prepared to accept non-unity gain, then you can achieve a Butterworth filter with R1=R2
A resistor across the pot won't work.

 

Some amount of non-unity gain could be acceptable. If R1=R2, do you still need the 4.5V?

 

Some amount of non-unity gain could be acceptable. If R1=R2, do you still need the 4.5V?

Yes. Although I'm not a fan of single supply circuits, I'd always use a ±15V supply for audio.
If you use a gain of 1.577, then the frequency setting resistors are equal for a Q of 0.7

 

Most (all) of guitar pedal power supplies are single supply, typically 9, 12, 18V.

 

The 2nd-order Sallen-Key filter has a Butterworth response (flat until a sharp corner) when the opamp gain is 1 and R1=2*R2.
If you make the opamp gain about 1.6 times then R1= R2 for a Butterworth response.

The datasheet for the UAF42 does not show a frequency response for us to see if its highpass is 1st-order or 2nd-order,
Maybe the response of a UAF42 is a 2nd-order highpass.

 

Some amount of non-unity gain could be acceptable. If R1=R2, do you still need the 4.5V?

Of course the 4.5V bias voltage is needed for the opamp with a gain of one and is powered from 9V. The "half the supply voltage" bias allows the output of the opamp to produce a maximum up and down output voltage swing.

Since an opamp input draws a very low current then the 4.5V can be two 22k resistors in series producing 4.5V and a 22uF capacitor to ground making a very low impedance.

 

Why must the filter have an adjustable cutoff frequency?
It would be easy to make a switched 4th-order Butterworth highpass filter but it is not two 2nd-order circuits in series that would produce a droopy Bessel response (not a sharp-corner Butterworth response). You must lookup the resistor ratios for it to be Butterworth..

 

Why must the filter have an adjustable cutoff frequency?

A variable HPF is an extremely useful knob for acoustic guitarists to have. Low end feedback is always a concern, so depending on the stage conditions you want more or less low end, more or less roll off.

The Grace Designs Felix I mentioned a few posts up is variable HPF from 40-1000 hz. It is designed for more than just guitar. Practically speaking, most acoustic guitarists would set it between 80-400 hz.

 

Why must the filter have an adjustable cutoff frequency?
It would be easy to make a switched 4th-order Butterworth highpass filter but it is not two 2nd-order circuits in series that would produce a droopy Bessel response (not a sharp-corner Butterworth response). You must lookup the resistor ratios for it to be Butterworth..

Two cascaded 2nd order Butterworths is a 4th order Linkwitz-Riley response, and it used in professional audio for crossovers, because, for two coherent sources, the outputs of high-pass and low-pass add to form a flat response.

 

You can still get 4-gang reverse log pots at a reasonable price
https://www.mouser.co.uk/datasheet/2/54/ptd90-777943.pdf

For a 4th order Butterworth, the cut-off frequencies of the two 2nd order sections are identical but the Q values are 0.618 and 1.618 (Golden ratio!) for the two sections
see figure 8:26
https://www.analog.com/media/en/training-seminars/design-handbooks/Basic-Linear-Design/Chapter8.pdf