Pot size for use as ‘blending’ voltage divider

[Sifunkle]

Question is more for to aid understanding, as 'try it and see' definitely works:

What governs the pot size to select as a variable voltage divider when the specific resistance doesn't matter as much as the ratio between sides of the wiper?

As an illustrative example, using the pot as a 'blender' between two different coupling caps (to determine how much low end enters a subsequent distorting gain stage); you'd do parallel caps into lugs 1 and 3 of series pot, with output from lug 2.

Going into it, I assumed having the wiper fully in either direction would result in 'fully' the tone of its respective cap (i.e. it would act more like a switch than a blender), with the rationale that if one extreme is zero ohms, the other may as well be infinite regardless of its actual rating (for this purpose, anyway). In practice, the extremes still result in some level of compromise, with larger value pots getting closer to the expectation.

I imagine the explanation for this is that (in the real world) with the wiper in the extreme, the pot still has some resistance on its 'lesser' side. Thus the ratio of lesser to greater resistances approaches zero the higher the pot's rating. Sound right?

If so, are there drawbacks to just using larger pots? I suppose inefficiency in power loss over the net greater resistance? In which case, I suppose it's better to stick with a smaller pot but use more extreme versions of whatever other component (caps in the above example)?

(Apologies if this has come up before, but not in the first several pages of search results)

[FiveseveN]

Welcome to the forum!
What matters is the source impedance and the load impedance. Which I guess covers a lot of things, if not most things.

the ratio of lesser to greater resistances approaches zero the higher the pot's rating. Sound right?

That ratio already approaches zero for any practical pot. Again, you have to take into account the source impedance: even if the pot is "zero", the cap isn't, and it probably has a bunch of stuff "behind" it as well.

[PRR]

> What governs the pot size to select as a variable voltage divider when the specific resistance doesn't matter

You said it yourself. It doesn't matter.

But in a real world, consider the source impedances and the LOAD impedance. A 10K load on a 1Meg pot will have a violent dip in the center; 100K load on 10K pot not so much.

Draw the whole circuit. Pots work in context of other circuit impedances.

[Sifunkle]

Welcome to the forum!

Thanks! I'd always thought electronics was magic, but then I licked a radioactive 9V battery and have been lurking for the last couple of months, learning to control my new powers.

As you've both suggested, I need to remind myself I'm not some DC pleb where only resistance matters (although I still don't fully understand impedance; kinda know about calculating it, but the concepts behind impedance matching source/load, etc haven't sunk in however often I read rules of thumb - I'm missing the 'why' rather than the 'what').

With the capacitor blending example, on the source side of things I suppose the smaller value cap has greater capacitative reactance, so maybe you'd describe the pot as favouring the tone of the larger cap rather than the 50/50 split I expected? In which case, could I calculate an appropriate size resistor to add in series with the larger cap/parallel with the smaller cap that would even the impedance of those parallel branches, thus get the 50/50 sweep?

I have no idea how the post-pot load would affect this situation...

(That reminds me of another roadblock I always encounter when considering the output of a prospective effect - how do you know what load/impedance to plan for when someone might plug the pedal into any number of things? I suppose output buffers exist to help with this sort of thing, although I'm not quite across them yet. So I'll want to bias a BJT: everything I've read says 'Start by picking a collector resistance to suit your load's impedance'... falter at the first hurdle.)

Appreciate the responses 

[Matthew Sanford]

If you're trying to blend two different low pass filters using 2 caps but one pot as the resistor, you're going to change the cut off frequency for both when changing the pots setting. Not sure if that's your intent, but if so look at the BMP tone stack - it's an lpf and hpf blender. The pot is solely to blend.

[Mark Hammer]

Confusing.  Physical "size" or resistance "value"? 

[merlinb]

As an illustrative example, using the pot as a 'blender' between two different coupling caps (to determine how much low end enters a subsequent distorting gain stage); you'd do parallel caps into lugs 1 and 3 of series pot, with output from lug 2.

You'd want the pot resistance to be significantly larger than the reactance of either cap at the corner frequency(s), e.g. a few times larger than the load resistance.