FROM VOLTAGE DIVIDER TO VCA

[carlozsulca]

Hi guys, after a long busy time, I find myself with a new project, I want to change a volume potentiometer for a voltage controlled amplifier, but there is something that is not clear to me, it is a capacitor that bypasses the high frequencies, I don't know How to implement all this with the vca, I took the vca from the lm13600 datasheet, I hope you help me

in the first image it is the divider that I want to replace with the vca the second one is an idea that I had but I don't think it will have the same result as the original circuit

[ElectricDruid]

Why do you want to keep the 560p anyway? It seems like a weird choice in the original circuit.

I did a quick sim of the situation, and it shows that as the volume is reduced, very high frequencies are allowed through. So reducing the volume by -10dB doesn't reduce the volume of the treble at 10KHz at all. Increasing the volume drop pushes the HPF cuoff so high we won't hear it, although there is less cut for everything above 1KHz or so. Since it lets more "fizz" and hiss through even when the volume is turned down, I don't see the point of it.

You can probably add a cap to the 13700 VCA much like you've done to get a similar effect, but you'd have to adjust the cap value - the VCA won't act like a 250K pot. Plus the VCA is linear, whereas the pot is probably log, so the response will be different too. You could fix that by using a log VCA like the '2164.

[PRR]

Model it with source impedance and also 100pFd load impedance. Like it was a vacuum-tube amp.

With 47(56?)pFd to 220pFd pot-shunt, the shunt cap semi-balances the treble loss of a half-up pot.

There may be no equivalent "problem" in the OTA plan. We can NOT just "translate" a pot-plan to an amplifier scheme without understanding.

[Rob Strand]

Model it with source impedance and also 100pFd load impedance. Like it was a vacuum-tube amp.

We can NOT just "translate" a pot-plan to an amplifier scheme without understanding.

Indeed.

You could try to build a literal circuit translation with one floating resistor equivalent and then another grounded resistor equivalent.   It would be a messy and impractical circuit with 3xOTAs - at 250k ohm probably noisy as well.   Once you step off that bus you need to model the whole thing.

[ElectricDruid]

With 47(56?)pFd to 220pFd pot-shunt, the shunt cap semi-balances the treble loss of a half-up pot.

I see! That's the idea, is it? I'd never come across that before.

Is it actually any good? I mean, this poor circuit can't know what cap is in the front of my amp, so it's compensating for something that's not tightly defined.

[Rob Strand]

With 47(56?)pFd to 220pFd pot-shunt, the shunt cap semi-balances the treble loss of a half-up pot.

I see! That's the idea, is it? I'd never come across that before.

Is it actually any good? I mean, this poor circuit can't know what cap is in the front of my amp, so it's compensating for something that's not tightly defined.

It's hard to know where the Fender bright switch starts and the Marshall fixed voicing (one channel brightened and the other "dulled") ends.
https://cdn.shopify.com/s/files/1/2284/6767/products/pro_ab763_schem_580x@2x.png?v=1553778099
http://cdn.tonegeek.com/wp-content/uploads/1987u.gif

At the end of the day it becomes part of the voicing of the amp.  Brighter at low levels and less bright when pushed.

It is what it is, which puts an end to any precise theoretical "compensation" based arguments.

[carlozsulca]

You can probably add a cap to the 13700 VCA much like you've done to get a similar effect, but you'd have to adjust the cap value - the VCA won't act like a 250K pot. Plus the VCA is linear, whereas the pot is probably log, so the response will be different too. You could fix that by using a log VCA like the '2164.

Of course I take into account the behavior of the original potentiometer, my idea is to use the 10k logarithmic pot as vca control

[carlozsulca]

I was talking with some users of the original circuit and most of them tell me that they use it not too far from the middle, so I used this kind of filter simulating that the potentiometer is in the middle, we will have to do the necessary tests and modifications anyway because there are things that I still do not take into account since I am working with tubes
I know that due to the logarithmic behavior of the potentiometer it is not exactly 125k, doing the tests will arrive at the exact value