A Better Tone Pot

[frets]

[ssstonelover]

Hi Frets,

Have you tried this yet? How would you describe the thinking behind it and the results you achieved?

[frets]

Hi ssstonelover😸😸😸,

The thinking behind it is to provide a standard 22nF tone. When pulled up, you have on one side a tone very similar to Q-Filter Tone, a sweet and mellow tone that is somewhat akin to a single coil, the other side is a midrange scoop. I feel it provides 3 uniques tones beyond a typical 3-Way cap Varitone. It is quite unique I think in terms of tonal diversity.

[ssstonelover]

Thanks Frets,

I've never tried a Q-filter, but I'm now reading the the literature and seeing comments very similar to your words. It's neat that you added two additional tonal options on the same switch. You've made it a "Swiss Army knife" of tone control. LOL.
I have a rail humbucker guitar this might work on.

[gckelloch]

Mar 15, 2022 16:05:27 GMT -5 frets said:

Hi ssstonelover😸😸😸,

The thinking behind it is to provide a standard 22nF tone. When pulled up, you have on one side a tone very similar to Q-Filter Tone, a sweet and mellow tone that is somewhat akin to a single coil, the other side is a midrange scoop. I feel it provides 3 uniques tones beyond a typical 3-Way cap Varitone. It is quite unique I think in terms of tonal diversity.

Do you know the cap value determines where the midrange starts to dip? The design of the inductor is also very important for the right result. It will be more of a high-end shelf above mid dip if the inductor resistance is fairly high. That's why the Wilde Q-filter has very high compression on a very small coil to bring up the inductance of a very low resistance coil.

I like your idea, but I'm not sure the point of the 220k resistor. I don't think it will do any more than no resistor at all. That side will have more bass than the other, but there will be a midrange dip on both sides. The side with the 18nF cap will start to dip at a slightly higher freq than the 33nF cap, determined by the pickup inductance.

I'd try a 22~47nF cap on one side for the standard midrange dip, but use a much lower cap value in the 3~7nF range on the other side for a dip that starts above 1kHz in most cases. I have that on one guitar and it's very useful for sweetening without lower-mid loss. I'd also use the recomended 5k or 10k resistor for some bass reduction at lower knob settings, or no resistor if you want to retain full bass ad have a strong midrange peak when the knob is down.

[Deleted]

To me both sides will be working as one

220K(33nF)+Pot Value X
250K or 500K-Pot Value X+15K(18nF)

Both going to the same point of the 1H5 inductor(transformer)

[frets]

No, It’s perfect😸

The critical factor of any tonifier is how it sounds and how useable it is. I have tested many combinations and this configuration provides the best differentiation to my ears for tones that are useable with Humbuckers or single coils.

[sumgai]

ang actually has it correct, although he might not have known it - the resistor in parallel with each cap is also in series with the inductor.

So let's look at the circuit for a moment, without the cap. Now we see a straightforward bass-cut circuit, don't we. (Simple in design, but more costly to implement than the more-standard cap-and-pot we usually use.)  Re-inserting the cap, we now have a high-cut in series with a low-cut, don't we. And this whole thing is acting as a bypass, right? So at some point, the highs are rejected from passing to ground (the inductor), and the lows are rejected from passing to ground (the cap), and what actually does go to ground are the mids.  With the mids removed (they went to ground), all that's left are the highs and lows.  We guitarists most often call this a mid-scoop, though the electronics industry knows it better as a notch filter.

At cap values of 18nf and 33nf, we should expect to see knee values (the point where the frequency starts dropping) approximately one octave apart.  (18 x 2 = 36)  This of course is at the extreme end of rotation. When "blended" at some point along the track, the overall effect should be "interesting".  Which is where we came in, with ang pointing out that the two sub-circuits are actually in parallel, albeit with a large dose of resistance between the two.  All we're actually doing is varying the amount of pickup signal that gets to each sub-circuit.  frets states unequivocally that the final results suit her just fine, and I'll presume that her customers are also happy.

I must caution you though.... all of this depends on the inductance value of each pickup.  A so-called 'hot' pickup may have an inductance value two or three times higher than a bone stock (OEM) pickup, and that may well throw off the values stated by frets.  Not by much, but enough to make the final results less pleasing than expected.  Expect to do some serious experimenting.

HTH



sumgai

[gckelloch]

Experimentation is good. I use GuitarFreak to determine values for a given inductance range, and then adjust by ear. Incidentally, the low peak would actually shift down ~1/2 an octave when doubling the cap value. 4x is an octave. You'll notice a slight lower-mid change from 33nF to 18nF. Nothing wrong with it, but there really is no perfect, right or wrong for these things. There is still 220k of R in front of the inductor when the pot is turned all the way towards it. No appreciable amount of lows would pass to it. You'd need a substantially lower R value for it to make a difference in the lows.

[MattB]

I had a go at modelling this circuit in LTSpice.

Here's the basic circuit, with a 500k standard tone contol and 500k volume, both at 10.

And here's the circuit with the tone pot pulled up:

As gckelloch noted, the resistance of the inductor has a fairly large effect, so I made three plots with different resistances.

The first shows an ideal inductor with no series resistance, the second has 2k of added resistance in series with the inductor, and the third has 5k.

These plots show the output from 0 to 10 on a 500k log pot, with run 1 at 0 and run 11 at 10 on the knob.

0k:

2k:


5k:

Mar 16, 2022 13:21:43 GMT -5 gckelloch said:

There is still 220k of R in front of the inductor when the pot is turned all the way towards it. No appreciable amount of lows would pass to it. You'd need a substantially lower R value for it to make a difference in the lows.

Using a smaller resistor cuts low end, but it also reduces the notch depth. This plot shows the same circuit as above, with R3 changed from 220k to 22k, and no additional resistance in series with the inductor:



The low end is cut by around 2.3dB, and the notch depth is reduced by around 9dB. As you said, there's no right or wrong answer here, but the bigger resistor gives much more contrast between the sounds at opposite ends of the knob.

[MattB]

This graph shows the same circuit as above, but with a 500k linear pot instaed of a log pot.

The overlapping traces are hard to see, so here's a closeup:

The top line is the grey trace, at 5 on the knob. Next are the pink and dark green traces overlapped, at 4 and 6 on the knob. The next pair of lines is light and dark blue, at 3 and 7, then red and brown at 2 and 8 and finally blue and purple at 1 and 9 on the knob. 

So with equal resistances the two sides sound almost identical over most of the range.

A log pot seems like a much better choice. All the duplicated sounds are squeezed into a narrow band between about 7.5 and 9.5 on the knob, and it gives more precise control over one side of the circuit, which is almost as good as having it over both sides.

[frets]

Thanks guys for doing all the work testing this circuit.  I think Sumgai's comments are very relevant to the circuit.  I think it sounds great; but, I have a tendency to use hot pickups.  Anyway, the idea of 3 tone choices to me is worth the experimentation.