state-variable parametric questions

[Eddododo]

Based on http://sound.westhost.com/articles/st-var-f5.gif this schematic
on http://sound.westhost.com/articles/state-variable.htm#ref this page

The configuration of the freq pots- it [apparently] allows for a greater sweep range with a smaller pot value
1) Firstly- is this simply a resistor-noise consideration?
2) Is the loading of the op-amp output via the freq potentiometers critical/significant?
3) Is there an inherent DISADVANTAGE of using the more common variable resistor setup (instead of the voltage divider) for frequency
4) Is there any disadvantage of using a 50k pot value for the freq (in the shown voltage-divider configuration) other than too wide a sweep range?
5) I am assuming that, in EITHER configuration, that a LOG taper is preferable?

you may have deduced that I have a set of 50k dual-gang LOG tapers already handy :p , but I seek the information regardless

Using tapering resistors, I figure I could use a 100k concentric b-taper pot with a ... 15k taper resistor if the 10k (plus the series resistance following it) loading is optimal..

[Eddododo]

Also curious on how to adjust the frequency formula for that pot setup

[PBE6]

I built that state variable circuit a few months ago using the standard values for everything except for C2 and C3 which I changed to be more bass-friendly. It worked brilliantly. I did a quick simulation back then to see if a 500k log pot would work, and it didn't seem to bother the opamps one bit. That was only a simulation mind you, I never actually tried it in real life.

I would recommend using a dual log pot, the sweep was quite smooth on my build.

[Eddododo]

Yes I've been playing with it alot, solid circuit,  going to check it out on a higher supply voltage

[thelonious]

except for C2 and C3 which I changed to be more bass-friendly.

Do you mind sharing what cap values you settled on? I was looking at that circuit for bass as well. I'm certainly planning to breadboard it and play with different values, but it would be great to have a starting point for the caps.

[PBE6]

I used 22nF as suggested in the article, as I was interested in getting boosts in the 700 Hz range as well as being able to cut frequencies under 500 Hz down to 100 Hz. The *low* low end wasn't as much of a concern.

Increasing the dual pot value will allow the frequency to be swept lower. You may want to try reducing the caps even further to something like 15nF or 10nF and using a 100k dual log pot to increase the available frequency range both higher and lower.

[MrStab]

re. the pot: i was sure i remembered reading this somewhere, and sure enough, it's on the ESP page you posted:

Good tracking is an essential requirement for a tuning pot. This generally means that you'll use a linear pot, because they generally have better tracking and linearity than log or antilog pots at the same price-point.

but then, PBE6 had success with a log pot, so YMMV. in my head it makes more sense to have log. maybe the site's recommendation of linear pots has more to do with how in-sync the wafer of each pot is likely to be, as opposed to actual feel.

[PRR]

For Frequency, leave the resistor/pot values alone (they are picked for other reasons), change the capacitors (smaller for higher).

> using the more common variable resistor setup (instead of the voltage divider) for frequency

Resistor is resistor. The electrons don't care. Except:

> assuming that, in EITHER configuration, that a LOG taper

The plan shown only has a 12.2K/2.2K range, only 5.5:1. Audio Taper is wanted when range exceeds 10:1. An Audio taper here, hardly anything will happen from "4" to "0".

If you do want a wider range:

The stopper R12 R13 can't be much less than 2K or the opamp will strain. So the pot must go higher than 10K to get a range of >10:1. (Which does suggest 50K pots.)

The pot at places R9 R10 "wants to be" REVERSE-Audio to get the knob to feel right. Dual Rev Aud pots are rare. You can swap the wires and re-number the knob, but then you turn "down" to go "up" in frequency, which may feel wrong.

Rod's attenuator plan fixes this. Audio Taper will feel "right".

With 50K pots, the 10K at R9 R10 is a heavy loading and the design taper will sag. This can be of some use with Linear pots. However for 50K Audio you should raise 10K to 20K (22K). To keep the same frequency you must decrease C2 C3 to about half.

[lion]

I built that state variable circuit a few months ago using the standard values for everything except for C2 and C3 which I changed to be more bass-friendly. It worked brilliantly.

PBE6 – I'm wondering, did you use the state variable filter by itself, as shown in the westhost schem, or do you have any surrounding circuitry - like a pre- or buffer stage etc - in your build?

[PBE6]

For that particular build I put an Orange Squeezer compressor in front of the EQ in the same box.

[lion]

Thanks PBE6. I recon the state variable filter will benefit from a buffer in front.

I have build the circuit now and it works, but some fine tuning needed. Contrary to one of the points in this thread, I'd like to limit the control range set by the freq dual pot rather than expand it.

Probably more correct to start a new thread than completely highjack this one.