Hi! Long post with impedance/parallel humbuckers/q's

[pauk]

Hi! I've browsed this forum for a while (whenever I'm tinkering with guitar wiring) so first, big thanks to everyone for the interesting discussions, ideas and data. This idea particularly was inspired by micguy on the talkbass forum, cooltone and antigua here on GN2.

micguy's post 1; post 2;

cooltone's post here;

antigua's entire catalog of pickup analyses etc



That being said, I have a bunch of questions about a bunch of different things, so I'll get started.


Background info/guitar data:

I recently picked up an Ibanez RGRT421 (purchased new online in 2022), HH configuration with 5way blade selector switch, 1 vol 1 tone. Selector switching arrangement uses a specific Ibanez 5way giving these positions:

1 - Bridge only ;

2 - bridge/neck both split (humcancelling);

3 - bridge/neck;

4 - neck only (wired parallel);

5 - neck only.

I like these selections generally, but I don't like the volume difference between the standard (series) positions 1,3,5 and the split or parallel positions 2,4. In addition, the bridge pickup is quite a bit brighter, so every time I switch between neck and bridge, I have to give the tone knob a turn to get the sound where I want. The stock volume pot is 500k linear, tone is 500k audio. A quick volume-and-tone-maxed multimeter read on the other end of the guitar cable shows 7.36kohm resistance at neck pickup standard series mode, 1.87kohm neck in parallel, 3.95kohm both humbuckers, 1.98kohms both split, 8.31kohm bridge standard series mode. The pickups are the stock Ibanez Quantum set with ceramic magnets. Both coils of each pickup have adjustable hex screw pole pieces, smallish, not huge Invader-sized poles. The cable I'm using has 0.46nf (460pf?) capacitance. I tend not to roll the tone down very far, just to "take the edge off" - probably staying in the 7-10 range on the knob.




Goals:
I want to go to 2 vol, 2 tone, 1 3way blade switch, 1 push pull. Due to space constraints in the cavity, I'm looking at adding 1 new hole between the default volume and tone locations. Neck volume with push pull, Bridge volume, dual concentric neck tone/bridge tone. I have been reading up on wiring humbuckers in parallel to reduce impedance and inductance, then using capacitors to bring the resonant peak frequency back down to expected ranges. Both pickups would be wired parallel internally. I would use one pole of the push/pull to switch between 2 capacitors in parallel with the neck pickup, one for a darker "series humbucker", lower frequency resonance peak and one for a more sparkly, higher frequency resonance peak somewhere in the vicinity of where a single coil's peak would be. I would use the other pole of the push/pull to do the same idea with the bridge pickup. I'm going to select cap values independently for each pickup. I don't mind that the overall "bright" sound won't be "this is a single coil", it just needs to land in the ballpark. The net desired result: 2 vol, 2 tone, 3way selector. Pop the push/pull up for a brighter tone. Stays hum-cancelling in all positions, and no volume jumps or drops due to coils being switched in and out of series mode. I don't really care that I'm losing the "2 pickups split" setting, and that the overall output volume will be the lower, non-series dB level. The wiring for the whole scheme shouldn't be too daunting... but I still do have a bunch of things to figure out, like component values.



Here are my questions (I am hoping to be corrected if I'm wrong on any assumptions, whether it's in question or statement form.)



1) Due to the drastically lowered impedance of running parallel-only (I won't have any series humbucking modes, or even single coil modes in this scenario), would it be reasonable to run 250k pots or lower all-around? micguy on talkbass seems to make the case in the 2nd post of his that I linked to, that due to the lowered impedance he is able to run a 100k or 150k volume pot and 100k tone pot and that there is still less high-end loss even with the drastically lower pot values. He's referring specifically to a p-bass, which typically has 1 vol, 1 tone (250k/250k), and a typical p-bass pickup apparently is wound a little heavier - like 10k-11k dc resistance - versus these Ibanez pickups coming in around 7.5k/8.5k. His description also makes it sound like a treble bleed may be unnecessary in this scheme.

I use 1800pF as a starting figure, but.....the value that replicates your series tone when wired in parallel depends in part on what capacitance your cable has. The parasitic capacitance cancels itself out when you consider the pickup's impedance dropping by a factor of 4, so adding 3x the capacitance of your favorite cable makes the load capacitance 4 times bigger, tuning things to the same frequency. The pots you use are also a consideration - I tend to use a lower value (100K or 150K) volume control, and either 100K or 250K tone pots. As the pickup is 4x lower in impedance, this actually means you have less loading than when the pickups are wired in series with 250K pots - your tone control will "go to 11" (have more range on the "up" side") even with 100K pots. Having load capacitors on the pickup side of the volume control and a lower value volume control means that your volume control will act more like a pure volume control - not a combo of volume and tone.

2) Will the frequency response of a humbucker wired parallel, with very meticulously selected capacitor(s) in parallel, match that of a humbucker wired in series? Or in other words, can the frequency response curve that is changed due to this reduction in inductance be directly substituted back to what it was with capacitance?



3) I just want to check my understanding on tone pots and volume pots - Each pickup (unloaded) has a resonant peak frequency, peak frequency width or Q, and peak height dB. The tone pot for a large portion of its travel dampens the peak height, then establishes a new, lower resonant peak due to the cap. The volume is a voltage divider, but as you turn down the knob, the rising series resistance between the pickup and everything following in the circuit raises the impedance and acts with the cable and amp input capacitance to behave as a low-pass filter. Is this a decent summary of why the tone typically becomes muddy with a non-treble bleed volume control?



4) If that understanding of a tone pot is correct: In conjunction with JohnH's data showing that the cap value doesn't matter until you're pretty far down on the tone, and since I basically don't use it that low anyway, is there anything that necessitates having a cap at all? How would the function be different with a small resistor in series instead of a cap? It looks to me that it should be essentially the same while in the "tone cap value is irrelevant" region, but would the area where the cap comes into play and below be drastically different? Would it just be a "muffle" or "mud" knob from that point on, with no new resonance peak to be established? If that's the case, then obviously using a cap is a no brainer.



5) The volume pot value has a strong effect on the amplitude of the resonance peak. Under typical H/H guitar convention, you use 500k vol/tone, or perhaps 250k vol, no tone, so that the pickup sees a net 250k load when the pot is at 10. The higher value is needed for humbuckers than single coils so the sound isn't too muddy, and the compromise is that this further raises the impedance of the whole circuit, exacerbating high-end loss as you roll down the volume. Hence the need for a treble bleed. Considering that a series humbucker is approximately double the impedance of a single coil, and a parallel humbucker is about half that of a single coil, does it follow logically that series humbucker uses 500k, single coil uses 250k, parallel humbucker would use 125k (or 100/150)?


Thanks for any and all input!
---Paul

[newey]

pauk-

Hello and Welcome to G-Nutz2!

I'll try to deal with those questions I can, not in order however:

Jun 24, 2023 19:33:16 GMT -5 pauk said:

I just want to check my understanding on tone pots and volume pots - Each pickup (unloaded) has a resonant peak frequency, peak frequency width or Q, and peak height dB. The tone pot for a large portion of its travel dampens the peak height, then establishes a new, lower resonant peak due to the cap. The volume is a voltage divider, but as you turn down the knob, the rising series resistance between the pickup and everything following in the circuit raises the impedance and acts with the cable and amp input capacitance to behave as a low-pass filter. Is this a decent summary of why the tone typically becomes muddy with a non-treble bleed volume control?

That's a pretty good explanation.

4) If that understanding of a tone pot is correct: In conjunction with JohnH's data showing that the cap value doesn't matter until you're pretty far down on the tone, and since I basically don't use it that low anyway, is there anything that necessitates having a cap at all? How would the function be different with a small resistor in series instead of a cap? It looks to me that it should be essentially the same while in the "tone cap value is irrelevant" region, but would the area where the cap comes into play and below be drastically different? Would it just be a "muffle" or "mud" knob from that point on, with no new resonance peak to be established? If that's the case, then obviously using a cap is a no brainer.

A cap selectively cuts a range of frequencies; a resistor cuts them all, across the board. Substitute a resistor and it's not really a tone control anymore, it's like a volume pot that starts out muddy and with less output and gets ever moreso as you turn it down. So, when you say "Would it just be a "muffle" or "mud" knob from that point on, with no new resonance peak to be established? If that's the case, then obviously using a cap is a no brainer.", you are correct.

5) The volume pot value has a strong effect on the amplitude of the resonance peak. Under typical H/H guitar convention, you use 500k vol/tone, or perhaps 250k vol, no tone, so that the pickup sees a net 250k load when the pot is at 10. The higher value is needed for humbuckers than single coils so the sound isn't too muddy, and the compromise is that this further raises the impedance of the whole circuit, exacerbating high-end loss as you roll down the volume. Hence the need for a treble bleed. Considering that a series humbucker is approximately double the impedance of a single coil, and a parallel humbucker is about half that of a single coil, does it follow logically that series humbucker uses 500k, single coil uses 250k, parallel humbucker would use 125k (or 100/150)?

Treble bleeds are useful for both single coil pickups as well as HBs, for the same reason- as you roll down the volume, you not only cut the output overall, but you cut the highs to a certain extent. The treble bleed circuit is a compromise way around this, regardless of the pot values. If your goal here is to have the somewhat brighter sound of parallel HBs, I would think you'd want to stick with 500K pots, but it's really a matter of taste. You already said you turn the tone down a little; you might just have to nudge it a bit more.

I would use one pole of the push/pull to switch between 2 capacitors in parallel with the neck pickup, one for a darker "series humbucker", lower frequency resonance peak and one for a more sparkly, higher frequency resonance peak somewhere in the vicinity of where a single coil's peak would be. I would use the other pole of the push/pull to do the same idea with the bridge pickup.

Putting a cap in parallel will cut highs. Different cap values will do more or less. But if you want brighter, one cap would need to be in series, not parallel.

2) Will the frequency response of a humbucker wired parallel, with very meticulously selected capacitor(s) in parallel, match that of a humbucker wired in series? Or in other words, can the frequency response curve that is changed due to this reduction in inductance be directly substituted back to what it was with capacitance?

Others can probably answer that better than I can, but my off the cuff response is "no". A pickup's frequency response is a product of a number of different factors which interact in complex ways. Capacitors are a sort of "brute force" compromise way to work around some things, but you'll probably not really match the frequency response.

As you said, testing will be needed to select what cap values work best for you. I would not worry about achieving a particular curve on a graph- work out what sounds good to you.

[gckelloch]

I took the liberty of experimenting with the parallel wired PAF Pro pickup option in the GuitarFreak 6 database file. The 1.46H value should average what your pickups will be in parallel. The GF 6 database compiled by John H shows the actual electrical response of different pickups.

The Fe/Steel poles in PAF, P90, etc pickups and your Ibanez models create eddy currents in the coils that roll off the high end above ~1kHz. So, higher resonant peaks will be lower in level. The DCR readings you took from your guitar cable are ~10% low due to the pot value load. Your neck HB should actually be ~8kHz, which is generally in the ~4.5H range with Fe poles. You should get a 1~2dB peak at ~6kHz on it (in parallel) with your cable and 2 x 250k pots, and closer to ~5kHz for the bridge HB.

I'd recommend a 2~2.5kHz peak with one cap on the p/p, and a standard ~600Hz peak with the other cap. A 3~4nF cap should get you the first peak. That would be up ~5dB when the tone knob is down, which is fine in that freq range, and you'll get some very nice useful tones along the entire sweep. A standard 47nF cap should work for the lower freq peak, but you might want to wrap a 22nF over it if it's not dark enough. 250k pots should be fine. You might try making one pot a lower value, or just keep the tone knob down a few notches if it's too bright.

[JohnH]

Great work gckelloch !

GuitarFreak has three pickup models based on tests by Antigua where he made specific tests on full HB, single coil and parallel wiring of humbuckers. These show how the real results differ a bit from the simple theory by which values divide or multiply x2 or x4 etc.

Its a good way to explore what is and isn't significant. Final choices are best made by listening and choosing by ear though.

[pauk]

Thanks for the welcome, newey! I had a more fleshed-out post typed up but it got banished into the ether by my 14-month old. So here's version 2.0.

Jun 25, 2023 6:46:46 GMT -5 newey said:

pauk-

Hello and Welcome to G-Nutz2!

I'll try to deal with those questions I can, not in order however:

4) If that understanding of a tone pot is correct: In conjunction with JohnH's data showing that the cap value doesn't matter until you're pretty far down on the tone, and since I basically don't use it that low anyway, is there anything that necessitates having a cap at all? How would the function be different with a small resistor in series instead of a cap? It looks to me that it should be essentially the same while in the "tone cap value is irrelevant" region, but would the area where the cap comes into play and below be drastically different? Would it just be a "muffle" or "mud" knob from that point on, with no new resonance peak to be established? If that's the case, then obviously using a cap is a no brainer.

A cap selectively cuts a range of frequencies; a resistor cuts them all, across the board. Substitute a resistor and it's not really a tone control anymore, it's like a volume pot that starts out muddy and with less output and gets ever moreso as you turn it down. So, when you say "Would it just be a "muffle" or "mud" knob from that point on, with no new resonance peak to be established? If that's the case, then obviously using a cap is a no brainer.", you are correct.

I think I probably took that to an illogical extreme, "for any normal tone cap value, there won't be an audible difference until pot resistance is very low" is probably more the gist of that.

Jun 26, 2023 12:17:07 GMT -5 gckelloch said:

I took the liberty of experimenting with the parallel wired PAF Pro pickup option in the GuitarFreak 6 database file. The 1.46H value should average what your pickups will be in parallel. The GF 6 database compiled by John H shows the actual electrical response of different pickups.

The Fe/Steel poles in PAF, P90, etc pickups and your Ibanez models create eddy currents in the coils that roll off the high end above ~1kHz. So, higher resonant peaks will be lower in level. The DCR readings you took from your guitar cable are ~10% low due to the pot value load. Your neck HB should actually be ~8kHz, which is generally in the ~4.5H range with Fe poles. You should get a 1~2dB peak at ~6kHz on it (in parallel) with your cable and 2 x 250k pots, and closer to ~5kHz for the bridge HB.

I'd recommend a 2~2.5kHz peak with one cap on the p/p, and a standard ~600Hz peak with the other cap. A 3~4nF cap should get you the first peak. That would be up ~5dB when the tone knob is down, which is fine in that freq range, and you'll get some very nice useful tones along the entire sweep. A standard 47nF cap should work for the lower freq peak, but you might want to wrap a 22nF over it if it's not dark enough. 250k pots should be fine. You might try making one pot a lower value, or just keep the tone knob down a few notches if it's too bright.

Thanks for the values, gckelloch. It looks like I may be on the right trail, I was expecting the parallel mode wiring to have a higher peak frequency than a single, so 6khz vs 4khz(ish?) makes sense to me. I need to get Excel and run GuitarFreak. I have no idea what to expect my peak amplitude to be, that will be pretty important. My plan with the push/pull is to bring the cap in, in parallel to the pickup, totally a separate thing from the tone control (Down position would be a larger cap, to hopefully land the peak freq around 2-2.5khz, Up position a smaller cap to land it more around 4khz). If you already got that from what I said, my bad! I was thinking along the lines of this post by JohnH:

Feb 19, 2014 2:55:47 GMT -5 JohnH said:

Small Tone Caps

Carrying on from above, with the tone pot at minimum (or not there at all), varying the tone cap creates a moveable peak in the response that can either slightly darken the natural pickup sound, or bring the resonance right down into the mids.

Here is a range of small tone caps, with 250k Volume pot and SC pickup:



The right-most trace has no added cap and shows the natural pickup output, as the tone pot is 'no-load' THe left trace represents 10nF, which matches the lowest of the values tested in the previous post

I've tried a few extra caps like this, and find in practice that about 2.2nF makes nice difference, and smaller than that, there's not as much audible change that you might expect from these plots. So it's worth a try but important to experiment with the real thing.

As a side note, now that you mentioned switching the tone caps, I could probably switch the tone caps at the same time, that's an interesting idea. I don't want to add any more knobs or switches, but I might not have to.
I still need to audition and try out the caps and pots etc, but I wanted to get the knowledgeable take from GN2 before spending the time and money to start acquiring/trying out parts. I'm going to punch up a visual for the scheme I have in mind.

[JohnH]

Heres some GuitarFreak plots, FWIW:

Reference tones are represented by the upper grey line , being an SD59 series wired humbucker, with 500k pots and a 0.5nF cable capacitance.
And also the dashed dark blue, being a Fender Fat50 with two 250k pots, and the 0.5nF cable

Pale blue is a parallel-wired SD59, with two 250k pots at full, and 1.5nF cap added (so cable plus an extra 1nF) - looks pretty bright!

Then, green curve, same as above but turn the tone down to about 6 (or add a resistor across the pickup of about 100k), still with 1.5nF added capacitance. The green curve is a close tonal shape match for the dashed curve

Red curve is the parallel SD59, with 2.2nF total added capacitance, 250k pots and the tone dialed down to about 5.4, or a 68 k resister across the pickup. Red is a close tonal match for the full humbucker.








Here they are again, normalised for volume:

[pauk]

Jun 27, 2023 0:55:48 GMT -5 JohnH said:

Heres some GuitarFreak plots, FWIW:

Reference tones are represented by the upper grey line , being an SD59 series wired humbucker, with 500k pots and a 0.5nF cable capacitance.
And also the dashed dark blue, being a Fender Fat50 with two 250k pots, and the 0.5nF cable

Pale blue is a parallel-wired SD59, with two 250k pots at full, and 1.5nF cap added (so cable plus an extra 1nF) - looks pretty bright!

Then, green curve, same as above but turn the tone down to about 6 (or add a resistor across the pickup of about 100k), still with 1.5nF added capacitance. The green curve is a close tonal shape match for the dashed curve

Red curve is the parallel SD59, with 2.2nF total added capacitance, 250k pots and the tone dialed down to about 5.4, or a 68 k resister across the pickup. Red is a close tonal match for the full humbucker.








Here they are again, normalised for volume:

Nice, thanks for the plots JohnH. I was worried that the peak amplitude might not be high enough, but it looks like there might be more there than I need! Sounds like I can park the 250k tone pots somewhere in the middle and be right where I want. I'm looking forward to getting some components and giving it a whirl.

I also worked up a wiring diagram, I'll try and post that up for review maybe tonight or tomorrow.

[gckelloch]

It would make more difference going down to a 4-4.5kHz peak if you started with a lower C cable for the no cap setting. Something in the 200pF range works well for a lot of things. 10~12' of this would be good: store.haveinc.com/p-62046-sommer-cable-300-0091-sc-spirit-llx-low-loss-instrument-cable-per-foot.aspx

It's excellent cable, and the HiCon noise-free connectors are great. All German made. That's what I use.

You can check resonant peaks with an SPL graph plug-in by running white noise on a mobile phone through an earbud pressed to the pickup. That's what I do.

[pauk]

Jun 27, 2023 18:39:37 GMT -5 gckelloch said:

You can check resonant peaks with an SPL graph plug-in by running white noise on a mobile phone through an earbud pressed to the pickup. That's what I do.

That sounds pretty useful. I'm definitely going to want to capture that kind of data before I start taking it apart.

Here's the diagram I made. It should be functionally accurate, I just represented the push/pull as a separate dpdt switch and the tone pots as individuals instead of dual concentric because it was getting really messy. None of the values are finalized.