Opinions About Boost Design

[spacecommandant]

Hi all, I've put together this volume/boost design meant to be used in a volume pedal enclosure.
It sounds good to my ears but I would love to hear any advice about improving the design in any way.
The things I want to keep the same:

The Volume treadle pot is A250k
There is a Max pot that sets the volume ceiling when Volume pot is fully clockwise
There is a decent amount of clean boost available

Thanks

[FSFX]

Hi all, I've put together this volume/boost design

A couple of comments.

1.  If you are using a charge pump then having a 47 ohm series resistor and a 10 or 12 volt Zener diode will provide better reverse polarity and overvoltage protection for the charge pump.

2. If you are using a charge pump with an input voltage of 9 volts to get 18 volts for the op amps then why have a voltage divider on the 18v to give a 9v Vref when you could just use the 9 volts power with maybe an RC filter to keep any noise away from the op amp inputs.

[spacecommandant]

Hi all, I've put together this volume/boost design

A couple of comments.

1.  If you are using a charge pump then having a 47 ohm series resistor and a 10 or 12 volt Zener diode will provide better reverse polarity and overvoltage protection for the charge pump.

2. If you are using a charge pump with an input voltage of 9 volts to get 18 volts for the op amps then why have a voltage divider on the 18v to give a 9v Vref when you could just use the 9 volts power with maybe an RC filter to keep any noise away from the op amp inputs.

Great points, thank you.
Something like this?

[BubbaFet]

It looks like a fine design. Now the trick is to start eliminating stuff and simplifying down to it's essence.
Do you really need it to run at 18V? Does it really need two dual-op amps? Et cetera.. etc... etc...

[FSFX]

Something like this?

I would just add a resistor and capacitor to filter the power to the op amp inputs.

[spacecommandant]

It looks like a fine design. Now the trick is to start eliminating stuff and simplifying down to it's essence.
Do you really need it to run at 18V? Does it really need two dual-op amps? Et cetera.. etc... etc...

I figured the 18V would allow for greater headroom/less chance of clipping...
And, no, I don't know that it does need both op amps, though I was under the impression that there could be situations where the output buffer would help, so I decided to add it. Overkill?

[spacecommandant]

Something like this?

I would just add a resistor and capacitor to filter the power to the op amp inputs.

Okay, thanks.
Do you think the output buffer is worth adding? Are there situations where it would help?

[ElectricDruid]

You could keep the output buffer and still get it down to a single dual op-amp chip if you changed the inverting gain stage for a non-inverting one. That would be able to provide the same high input impedance you have (in fact the input components would be identical up to the +ve input) but would save you one op-amp. Then the second op-amp can be your output buffer.
There's one other advantage to this change, which is that it makes the circuit non-inverting overall.

Is the output buffer strictly necessary? I don't know, but it certainly won't *hurt* and it might help, so I'd leave it in.

[FSFX]

Do you think the output buffer is worth adding? Are there situations where it would help?

I didn't want to completely redesign it for you. However, it is true that you could get all of the gain that you require from a single op amp. You could even use a BJT or JFET buffer as some other designs do.

You can look at other boosters and decide what you want in terms of gain, frequency response, linearity/distortion, noise performance (SNR) and dynamic range.

But, ultimately, it is your design so feel free to do it as you please. That is the way you learn.

[antonis]

Funny that I can't follow squared op-amp symbols but I think output signal is inverted..
(not bad by its own but a potential issue for mixing..)

P.S.
R5 could be an order of magnitude lower or even omitted..

[spacecommandant]

You could keep the output buffer and still get it down to a single dual op-amp chip if you changed the inverting gain stage for a non-inverting one. That would be able to provide the same high input impedance you have (in fact the input components would be identical up to the +ve input) but would save you one op-amp. Then the second op-amp can be your output buffer.
There's one other advantage to this change, which is that it makes the circuit non-inverting overall.

Is the output buffer strictly necessary? I don't know, but it certainly won't *hurt* and it might help, so I'd leave it in.

The reason I used an inverting stage was to be able to completely kill the signal when the Volume pot was CCW (foot pedal in heel position), which isn't possible with a non-inverting stage, if I understand that correctly.

[spacecommandant]

Funny that I can't follow squared op-amp symbols but I think output signal is inverted..
(not bad by its own but a potential issue for mixing..)

P.S.
R5 could be an order of magnitude lower or even omitted..

Yeah, I prefer the triangular symbols but I couldn't find them in Eagle.
Thanks for the tip on R5..

[FSFX]

Thanks for the tip on R5..

R5 does provide a useful function and a value of 10k is fine as the input impedance beyond it is so very high (1M), so I would leave it.
It will act to protect the op amp input from any voltage transients and, in conjunction with the input capacitance of the op amp it will provide some EMI filtering. If you reduced its value then it would be sensible to put a low value capacitor (about 100p) beyond it from the input of the op amp to ground. This is all about the EMC stuff we were discussing elsewhere in the forum.

[FSFX]

In fact I would recommend to keep R5 as 10k and to add a 100pF capacitor from IC2 pin 3 to ground (i.e. across R6) anyway.
That will provide a high end cutoff at about 10kHz when you drive the circuit from a 100k source impedance.
Normally you would likely be driving it from a lower impedance so the cutoff would be up to 100kHz.
That should provide quite adequate RF filtering at the input.

[spacecommandant]

In fact I would recommend to keep R5 as 10k and to add a 100pF capacitor from IC2 pin 3 to ground (i.e. across R6) anyway.
That will provide a high end cutoff at about 10kHz when you drive the circuit from a 100k source impedance.
Normally you would likely be driving it from a lower impedance so the cutoff would be up to 100kHz.
That should provide quite adequate RF filtering at the input.

That's helpful, thank you.
And I suppose the 1M pulldown resistors R4 and R9 can be removed if there's no on/off switch.. does that require a change of any of the cap values?

[FSFX]

And I suppose the 1M pulldown resistors R4 and R9 can be removed if there's no on/off switch.. does that require a change of any of the cap values?

I would still keep them as they provide a DC path to ground to keep the charge on C9 and C13 so they don't create any transients when an input or output device is plugged in. As they are just DC 'bleed' type of resistors then R4 could be higher in value (say 4M7 or 10M) if required to minimise the effect on the input impedance.

[spacecommandant]

And if I were to add a Tuner Out Jack, would splitting the signal on the output of the first buffer be a good spot to put it?

[merlinb]

The reason I used an inverting stage was to be able to completely kill the signal when the Volume pot was CCW (foot pedal in heel position),

So swap the pot positions, so the footpedal is a simple volume control to ground?

[antonis]

The reason I used an inverting stage was to be able to completely kill the signal when the Volume pot was CCW (foot pedal in heel position)

You know, of course, that there aren't pots of zero absolute minimum resistance..

[PRR]

You know, of course, that there aren't pots of zero absolute minimum resistance..

And if you really want to go to ZERO, why even have a guitar?

20dB, 30dB, maybe 40dB is plenty of cut for most musical situations. So 1% even 10% of end resistance doesn't hurt the music (it can offend the buyer who thought he could buy silence).