Voltage Divider VS Variable Resistor

[ferusph]

Hello! I've been reading a lot in here for a long time but this will be my first post in the forums!

I just want to ask why volume and gain pots are usually wired as a voltage divider and not as a variable resistor (to ground).

I tested the variable resistor wiring on my guitar by simply shorting the 1st and 2nd lugs (see image below) of the pot and didn't hear or feel any difference. Are there other things I am missing here?

The reason I am asking is because I am currently planning on building a drive pedal with a switchable gain and volume pot. Wiring the pots as a variable resistor will make the switching easier by simply wiring the grounds of each volume/gain pairs to each side of one pole on a 3pdt, saving other poles for other functions.  But I want to be sure there would be no consequences of doing so.

[Rob Strand]

I tested the variable resistor wiring on my guitar by simply shorting the 1st and 2nd lugs (see image below) of the pot and didn't hear or feel any difference. Are there other things I am missing here?

Sometimes it's not a good example to use a guitar as the signal source.   Imagine a 100W amplifier and you short the pot to ground at the output - it's going to blow.

The thing that's missing in the guitar example is the impedance of the guitar.   If you look at a simple voltage divider made of resistors,



The output voltage is   Vout = Vin R2 / (R1 + R2).   When you place the pot to ground you are varying R2.   However notice that if R1 is zero Vout = Vin.   You need the R1 to be present in order to get variable output level with R2.   For the guitar case the "R1" is the impedance of the pickup so that's why you experiment works.

Another thing to notice is the size of R2 required to get a drop in volume.   Suppose you want to drop the volume by half.    That means R2 needs to be the same as R1.    For a 100W solid state power amplifier, R1 could be anything from 0.05 to 0.5 ohm.  To get half volume R2 would need to be a value equal to that.   That's means putting a load of 0.05ohm to 0.5 ohm on the amp.   That only works in theory.  With such a low impedance load it will blow-up the amp as the minimum load is 2, 4 or 8 ohm etc.

[ferusph]

Does that mean that the issue is the current rating of the pots? If that's the case would it be safe for pedals? To be specific, what I'm doing is a distortion circuit that uses a LM386.

[Rob Strand]

Does that mean that the issue is the current rating of the pots? If that's the case would it be safe for pedals? To be specific, what I'm doing is a distortion circuit that uses a LM386.

It's not so much to do with the rating of the pots, although they could be damage,  but more about loading down the thing you are shorting out ie. the thing before the pot.   If you shorted out the *output* of LM386 it wouldn't be good for it or the pot.   One option is to add you own R1 in series with the output.   You need to choose R1 high enough not to load down the thing driving it.

You still can get it issues with the relative values of R1 vs the pot.  If your pot is 100k  and R1 is 1k that means you have to turn the put to 99% of rotation (or 1% depending on how you look at it) before the level drops to half.   That will make it unusable in practice.   If R1 is 33k or 50k then at least it sort of works.    However, when the pot is max out you will now get a bit of volume drop since the pot means R2 can go higher than 100k and when you have a divider R2=100k and  R1 at 50k  the level is 2/3 the input.

What's the motivation for using that configuration?  It opens up a lot of problems you normally don't have to deal with when you use the voltage divider configuration.

[ferusph]

It's not so much to do with the rating of the pots, although they could be damage,  but more about loading down the thing you are shorting out ie. the thing before the pot.   If you shorted out the *output* of LM386 it wouldn't be good for it or the pot.   One option is to add you own R1 in series with the output.   You need to choose R1 high enough not to load down the thing driving it.

Okay thanks for the explanation! That clears a lot of things for me.

If R1 is 33k or 50k then at least it sort of works.    However, when the pot is max out you will now get a bit of volume drop since the pot means R2 can go higher than 100k and when you have a divider R2=100k and  R1 at 50k  the level is 2/3 the input.

That's what I was thinking at first, to put a resistor in place of the half of the potentiometer. I think I have to go with this method. The volume drop is a consequence I am willing to accept. The full output volume is too loud anyway.

What's the motivation for using that configuration?  It opens up a lot of problems you normally don't have to deal with when you use the voltage divider configuration.

Basically to switch between presets of gain and volume. Similar to a 2-channel preamp but without the need to repeat the entire circuit for a 2nd channel. The voltage divider method would require me to use a 4pdt footswitch which is hard to find, and also leaving no room for the LED.

[PRR]

......usually wired as a voltage divider....

Everything is a voltage divider!!

Sometimes the other part of the divider is hidden, but it is there.

uses a LM386.

The output resistance of a LM386 is about 0.2 Ohms. Although actually about 0.5 Amps maximum current. A variable resistor on the output would do about nothing infinity to 10 Ohms (logical, cuz 8 Ohms is no great strain). At this point it could be dumping 0.5 Amps and 5 Volts or 2.5 Watts in the variable resistor. Most are rated 0.25 Watts. It's like braking a truck with shoe leather.

[Rob Strand]

That's what I was thinking at first, to put a resistor in place of the half of the potentiometer. I think I have to go with this method. The volume drop is a consequence I am willing to accept. The full output volume is too loud anyway.

Basically to switch between presets of gain and volume. Similar to a 2-channel preamp but without the need to repeat the entire circuit for a 2nd channel. The voltage divider method would require me to use a 4pdt footswitch which is hard to find, and also leaving no room for the LED.

That should work out OK.

[FiveseveN]

The voltage divider method would require me to use a 4pdt footswitch

It does not require more poles, all you need to do is switch which wiper (pin 2) connects to the output.

[BJM]

Hi,

If you want two different settings on your pedal you could also double the potentiometer (s), as for example in this schematic:

It doesn't need a 4PDT switch and you can switch two pots.

Interesting replies btw, which for me brings up another question. I'm busy building a splitter, more or less like this design.

http://www.muzique.com/lab/splitter.htm

I'm not using JFETs but opamps (works OK) but I added the mute switches as described half way down the page. I'm guessing a potmeter fully CCW before the output is the same as the mute switch? The mute switch in the AMZ schematic goes in parallel with the 100K resistor, is that the same idea as "One option is to add you own R1 in series with the output"? How do you wire this R1?

[FiveseveN]

I'm guessing a potmeter fully CCW before the output is the same as the mute switch?

No, the pot provides a resistive load (the pot's nominal value), while a switch is a short to ground. I wouldn't ask an anonymous op amp buffer to try driving 1uF to ground, so use those series resistors. Here's a breakdown: https://www.analog.com/en/analog-dialogue/articles/ask-the-applications-engineer-25.html

The mute switch in the AMZ schematic goes in parallel with the 100K resistor, is that the same idea as "One option is to add you own R1 in series with the output"?

No, in parallel is the opposite of in series. The resistor comes between the output and the coupling cap, as described in the article I linked above.

[iainpunk]

hey, welcome to the forum.

the 386's gain can be controlled by a variable resistor between pins 1 and 8 which is easy to switch.
if you want different volume settings, just put 2 pots in there with pins 1 and 3 connected together, and have pins 2 connected to 2 sides of a switch. the 386 can drive almost every load above 8 ohm, so putting 2 pots in parallel really isn't something to worry about here.

cheers

[ferusph]

hey, welcome to the forum.

the 386's gain can be controlled by a variable resistor between pins 1 and 8 which is easy to switch.
if you want different volume settings, just put 2 pots in there with pins 1 and 3 connected together, and have pins 2 connected to 2 sides of a switch. the 386 can drive almost every load above 8 ohm, so putting 2 pots in parallel really isn't something to worry about here.

cheers

Thanks! Very informative! Would this also work with a J201?

[iainpunk]

Thanks! Very informative! Would this also work with a J201?

that really depends on the surrounding circuit! it could work quite well, but there are some topology's where it could cause promelms, like with high source impedance and/or low input impedance after the pots.
> post chematic

cheers