A better single coil sound from a humbucker?

[megi]

Hi folks, apologies for not posting very much, I do still look on the forum a lot and appreciate the expertise available - plus I am still very happy with two of John H's circuits I have on my strat and tele.

Having found this "partial phasing" circuit for a humbucker on Entwistle's website, I was just wondering if anyone has tried it or has any views. It is described as tending to give a better single coil sound from a humbucker, compared to a conventional coil split.

www.entwistlepickups.com/assets/Humbucker%20wiring.pdf

Thank you!

[newey]

megi-

Welcome back!

What Entwhistle Pickups is calling "partial phase" we call half out-of-phase (HOOP). Although I don't know that anyone here has done this between the two coils of a HB, usually it's done between neck and bridge pickups, as is described in the thread I linked above.

The two coils of a HB, wired out-of-phase, will ordinarily sound very thin. Since the coils are so close, and virtually identical, the cancellation of frequencies is almost total, so you get a very thin, low output sound. Using a cap on one of the two coils changes the frequency response of that coil, making it less of a "match" with the other, and therefore resulting in fewer frequencies being cancelled. This can make for a more useful OOP sound.

I don't know about their claims that it sounds more like a single coil, but then, I've never tried it, either. It's still out of phase, so I would think it would still sound OOP as opposed to more like a SC.

[long813]

From 4real, by the sounds of it, he did the HOOP on the neck pickup? (Go to easter egg part)

Maybe he'll come along and chime in some more.

Jan 31, 2011 4:59:45 GMT -5 4real said:

I played around with this idea a bit today in my HSS strat...

I have it in a push pull on the master tone control and it reverses the phase on the neck pickup...



I tried a few caps but indeed found the .01uF to get the best results...but different tastes and pickups might be worth playing around with.

It does change some interactions in surprising ways...in some combinations the effect is subtle, perhaps more apparent in notes above the 12th fret or on the open bass string frequencies...while in series mode it can be a really dramatic but useful sound...

The interesting thing is an EASTER EGG in this...not only does it tame down the thin and weak sound but still fairly funky...but it also changes the sound of the pickup itself alone while switched. On my neck pickup it turns the warm sound to a brighter airier sound...so not just a 'phase switch' but also changes the character of the pickup itself BONUS!

This mod is well worth trying especially if you usually find a full phase a little extreme...

[sumgai]

megi,

Yes, it's been rather longish since you last graced us with your presence... but we'll let bygones be bygones, so long as you don't let it happen again.


The real problem here is that the two coils are so close together in terms of position along the string. In essence, they both receive very nearly the exact same node/anti-node information from the vibrating string, thus they will tend to either reinforce or cancel the same frequencies. It's probably of interest from an experimental standpoint to try this, but I think the partial reversing of phase will be quite drastic, and the resulting tone will be more 'novel' than anything else. Usefulness will be, of course, up to you.

HTH




sumgai

[newey]

long-

]From 4real, by the sounds of it, he did the HOOP on the neck pickup? (Go to easter egg part)

4real has the switch putting the neck pickup "Half-out-of -phase" with the other pickups, not with itself. From his diagram, the neck looks to be a single coil anyway, as it only has 2 wires coming from it.

The "easter egg" part is, as he says, when the neck pickup is on by itself, and thus not OOP with anything. Flipping the switch brings the cap into the circuit, changing the sound of the neck pickup by itself. This is as opposed to a regular phase switch, which has no effect on the sound when the pickup in question is not paired with another.

[megi]

Thank you for your thoughts chaps - I think it may well be an interesting sound, but possibly only Entwistle's idea of "better single coil tone". Perhaps I will have to experiment at least one time. The reason I'm asking is that I'm planning to modify a two humbucker guitar, with a standard 3 way toggle and single volume and tone control. I'm going for a couple of push-pull pots, and probably this circuit to give series-parallel switching for each pickup www.guitarelectronics.com/product/WD2HH3T11_04/Guitar-Wiring-Diagram-2-Humbuckers3-Way-Toggle-Switch1-Volume1Tone003.html - unadventurous I know! but is this a well designed circuit can I ask, or would you change anything?

Anyhow, I was just contemplating the Entwistle option insead of the series-parallel switching, but I think I'll leave that for now. I'll try to be a bit more active on the forum sumgai, I do feel a bit out of my depth with some of the guitar electronics discussions, although I find it all very interesting to read.

[newey]

but is this a well designed circuit can I ask, or would you change anything?

The circuit is fine, no worries if that's what you want to do. I like the "parallel HB" sound (depending on the HB in question).

Perhaps I will have to experiment at least one time.

A gutsy, Nutzy™ thing to do. You could dedicated one of the push/pulls to that wiring on one of the HBs, and, if it's not to your liking, then rewire it as per the diagram for series/parallel.

Doing so on the bridge pickup would probably give you less cancellation (and thus a more "worthwhile" sound, depending on what one likes), since the slight difference in coil position along the string is accentuated closer to the end of the string length.

I do feel a bit out of my depth with some of the guitar electronics discussions

We all learn from each other. I owe a lot to many of the long-time members who, just about 5 years ago, put up with my dumb questions.

At some point, by participating in discussions, I learned how to locate the business end of a soldering iron. But, there are still many topics where my ignorance is on full display.

So, don't feel intimidated, a lot of us didn't come into this with any more knowledge or background than you have.

[4real]

Hi Nutzers...

It is right that my strat has a 'HOoP switch' in the tone control push pull and operating on the neck pickup...so that has an effect only in combination with other pups. It is a 'milder' version of an OoP switch like I use on my tele, not so extreme and suits this guitar.

The "Easter Egg" is theat the 'spin-select' thing will wind down one coil of the HB...so you can have 1.5 coils. The push pull selects which coil is would down, standard is the bridge most coil, pulling the switch winds down the neck most coil of the HB...all the way it is a 'single coil' and noise reduction is diminished as it gets closer to the full split. Not that it's noisy, the whole guitar is extremely quiet.

The 'twist' was that on this second split, so leaving the bridge most coil of the HB has a samll cap in there, this lets some of the bass through from the other coil so giving this setting an interesting effect and alittle more 'body' and effect in the travel.

...

This entwistle idea gives you both coils for noise cancelling, and the cap will diminish the effect exactly as a HOoP switch does, but between the two HB coils in this case. An HB out of phase with itslf is very extreme because the coils are so close together and canceeling almost everything...the cap, exactly as in the HOoP will moderate that...it is labeled 'partial phasing' there and as used in the HB SC sound too.

A very similar and fully noise cancelling effect can be done with wiring the HB in parallel as has been noted. I have this on my Khaler LP. It works well for high output HBs and does have a SC like sound. My LP has both splits and parallel in its HBs and I prefer the sound of parallel over the split and it is still completely noise cancelling. It's certainly a good option.

None of these things will sound like a real SC, one thing to consider is that even if you 'split' an HB, even if you made an HB out of two fender SC pup, switching one coil out, does not remover the reverse polarity magnet sitting right next to the other and so not the same magnetically as a SC on it's own.

The main thing about all these strategies is that it sounds good for your purposes. All will be diffeerent to with different pickups and combinations. I tinkered with my tele's wiring after doing teh strat and HOoP thing for instance, putting in the HOoP mod as the phase can be exreme...but was not happy with that at all in that guitar and pickup combination so put it back, I like the OoP sound in the tele though I don't use it often.

There is a 'split' plus the bass of the other coil (and a little noise cancelling) thing, it is effectively what my 'spin-select' thing does but that is variable, you could do a similar thing with a switch. If you included a trim pot inside the gutiar with the cap, perhaps you could trim to taste. It is worth trying a few cap values, but in all these things, all of them are very small values.

[megi]

Cheers again chaps! Maybe I'll try your suggestion and just wire the partial OOP switching in the bridge to try out newey, thanks for running your eye over that wiring diagram too. 4real, much appreciate your detailed replies - I do understand your point about not getting massively close to true single coil sound - I already have a guitar modded with switching to give both coil-split and series-parallel swithching for each pickup, and to me both options give a pretty similar tone, but I do like the idea of keeping everything humbucking for the current project, hence my liking the parallel wiring option. So I do think that is the strongest contender for me - if it just gives me a brighter but less powerful sound I will be happy. The pickups are not high output either in this case, just normal PAF types, 7 and 8K, with alnico 2 magnets. I use the guitar for clean jazz stuff, mainly with the neck p'up - so largely going for that warm woody jazz tone, but sometimes it would be nice to be able to switch to a lighter, less full tone.

[4real]

Cool...that reminds me...

In my last 'jazz box procject' I installed a hidden series parallel switch after being so impressed with the S/P in my LP with an original PAF in it and works well and quiet for exactly this kind of thing...

guitarnuts2.proboards.com/index.cgi?board=wiring&action=display&thread=5970



Here's me playing it...you can just see the hidden S/P and 'tone' switch in teh sound hole there...

Not quite as 'impressive' on the PAF but it certainly takes the 'heat' out of the HB...not a single coil sound as such (unlike the hotter pup in the LP which sounded very much like the split but without noise) but still a nice tone and of course quiet. Perhaps teh active buffer had some bearing on that as well...

Glad to be of service if I have...pete

[megi]

You have indeed helped 4real, cheers! Taking the heat out of the humbucker - that actually puts it very well, thats kind of what I'm after - so the tones I have available will remain in the "useful for jazz" category (I don't want, or need it to sound like a strat!), but it would still be nice to have an lighter alternative to the big, fat, warm jazz sound I normally go for (which the guitar does very well, used with a Polytone Minibrute amp). I do remember your project thread for that guitar of yours - and a very nice bit of work you did there!

This is the guitar I'm modding btw, for no reason other than an excuse to show it off ha ha, a Korean built Shine SIL-510 semi-hollow/chambered guitar, currently available in the UK for a ridiculous discounted price of £99 plus postage, but I've found it to be a really excellent axe nontheless.



The pickups are not the originals, I put a set of alnico 2 PAF types in, bought from Axesrus in the UK, but I think made to their specs by Artec.

[JohnH]

that is a nice looking guitar, and a perfect base for trying mods.

My favorite 'improved single coil' sound is an in-phase series wiring of the coils, with one coil not completely cut out but bypassed with a resistor or cap, so that some remains. On my PIAF bridge humbucker. a 0.047 cap across one coil does this.

I read that PRS do something similar for coil cuts on there guitars, but I don't know exactly what they do.

On this half out of phase circuit, we might learn something by modeling it, so off I go to do that......

J

BTW..was doing Spell check, and accidentally clicked when I shouldn't.

This post is now about 'getting a better single-coil sound from a Hamburger'.

It's all about that top-end sizzle...!

[sumgai]

John,

Dec 14, 2012 14:41:05 GMT -5 JohnH said:

BTW..was doing Spell check, and accidentally clicked when I shouldn't.

This post is now about 'getting a better single-coil sound from a Hamburger'.

It's all about that top-end sizzle...!

See if you can get Runewalker to send you down some of his Texas BBQ sauce - that'll fix it right proper!

[JohnH]

yes indeed!

But here's this HooP circuit:





It represents a PAF humbucker, split into two out of phase coils, connected in parallel via the 2nF cap.

The two signal sources are out of phase 180 degrees, so this represents the extreme positioning of two coils right on top of each other, which is close to the truth in a single Hb.

The graphs show the '2nF', swept in 0.5nF increments from 0 (red trace) to 3nF (peak furthest to the left).

So, it does do something - and seems to move and enhance the response peak in the high end.

cheers
John

[megi]

John - I'm very grateful for what you've done there! I'm afraid my lack of knowledge re frequency response graphs etc. means I'm still not quite sure what the modelling results actually mean in terms of the sound produced - how would the graphs compare with that for a normally wired PAF humbucker, or a true single coil for that matter? And in your opinion, is there much mileage in using this instead of a coil-split or series-parallel switch? Does it work with hamburgers also? ;D Apologies for all the questions, it's really interesting though, cheers!

[geo]

Is there a format you can export those graphs in? I wonder if it can't be too hard to hack together a little sine wave generator to play notes/chords EQ'd based on those graphs.

Would that be useful? It probably wouldn't sound anything like a guitar, but I suppose it would tell you what the circuit itself sounds like. (Compared to an entirely flat frequency distribution, at least.) If that program can accurately model different pickups you might really have something.

I can try to hack something together over winter break if you think it'd be useful, but if it probably wouldn't be of much use I'd rather not invest the time.

[JohnH]

All right! Very interesting questions and ideas from megi and geo.

Megi

These kind of analyses try to pick up the action of a guitar circuit as an electrical filter. Ie, what is fed in is a pure linear frequency response at the signal generator, and we try to learn something about how the circuit will affect it. So what it can’t do is represent the true response of the strings interacting with the guitar body, and the effects of tonality dependent on the position of the pickup and whether it is part of a humbucker or a single coil etc. But, they can be very useful none the less for comparing two traces, to see how they differ and how they are the same, and to test the effects of different component values.

So that being said, the key features of the various traces are the linearity of the response at lower frequencies, followed by the height, width and frequency of the first peak, after which it drops off at higher frequency. That low frequency linearity tends to be a given unless there are special tone control circuits, so it is the peak that is of most significance. It depends on the pickup (mainly the inductance), plus the pots and the capacitance of the cable.

So, looking at those traces above for this unusual HooP circuit, I think the most interesting one there is the second from right which has a peak at about 3kHz, and its about 12db up above the general level. Thats a pretty high peak and it will sound – good or bad? We wont know unless its tested. Maybe and even smaller cap than the 0.5nF of that trace will be better.

But getting to your question, a similar analysis for the same humbucker coils, connected normally in series, with the same assumptions about the cable, would give a peak of about 4.5 db, at about 3kHz. A typical true single coil has a calculated peak of about 6db at 4kHz

Note, for the analysis above, I used a 500k tone pot and a 250k volume pot – a compromise just for the example, smaller pot resistances for tone and volume suppress the peaks.

If you want to play, the GuitarFreak spreadsheet in the reference section will model these things easily, with buttons and sliders too.

None of these calcs tell you how it sounds however, but can be a good guide to how to change a sound.

Geo

Yes..i think there is definitely some interest in what you suggest. A table of values can be produced, I think from that 5Spice modelling program, but certianly from my spreadsheet noted above. So that’s no problem.

The first version of what you describe would be to feed in some notes and harmonics, for single notes and chords with some arbitrary relationship between the amplitudes of each harmonic. It would need the ability to play around with those proportions until it sounded somewhat like a guitar note. Then, feeding this signal mix into two different calculated response traces, you could begin to experience whether say, a 3db reduction in the peak response height, is significant, or not..etc..so useful right there.

The next iteration would be to try to add some more reality to it, so, we would record some actual guitar signals, through a known electrical setup (eg a standard guitar circuit), then replay that through a calculated filter that is the difference in amplitudes of two traces, to respresent some new circuit that we want to test. That way, the final result is based on a real signal, adapted to test new tweaks that we want to investigate.

So yes, I think it’s a very good idea!
John

[ashcatlt]

I have recently switched to LTSpice. It is not quite as User friendly, but much more powerful. It will take a .wav file as input and/or generate a .wav output so that yes, you can hear some approximation of what the circuit will do.

Problem with John's little deal there is that the polarity reversal is flat across the frequency spectrum - as he says it's like the two coils take up exactly the same physical space. Even in an SC sized HB this will not be the case. Higher frequencies will have a greater chance of being different phase between the two coils, and therefore less likely to show the total cancellation which that graph shows.

I don't know for sure how to model that action. Perhaps an all-pass filter on one of the coils? That wouldn't be exactly the same either, but might be closer?

[megi]

Dec 15, 2012 16:39:20 GMT -5 geo said:

Is there a format you can export those graphs in? I wonder if it can't be too hard to hack together a little sine wave generator to play notes/chords EQ'd based on those graphs.

Would that be useful? It probably wouldn't sound anything like a guitar, but I suppose it would tell you what the circuit itself sounds like. (Compared to an entirely flat frequency distribution, at least.) If that program can accurately model different pickups you might really have something.

I can try to hack something together over winter break if you think it'd be useful, but if it probably wouldn't be of much use I'd rather not invest the time.

Thanks for the offer geo, I would have to say please don't go to such trouble on my account - although I find this interesting, I pretty sure I'm going with the standard series-parallel switching circuit for my guitar now, having thought about things. I think this will better suit my intended use of the guitar for jazz playing, plus keep all the settings hum-cancelling. Trouble is I constantly use this guitar to gig with, so am having to wait for a suitable window in which to do the wiring work.

Many thanks to everyone else on the forum as well, especially John H, this is a great place!

[geo]

megi:
It looks like ashcatlt is telling us there's already functionality for that in LTSpice.

[ashcatlt]

Dec 20, 2012 13:29:22 GMT -5 geo said:

megi:
It looks like ashcatlt is telling us there's already functionality for that in LTSpice.

Yes I am. I've not yet tried using a .wav as an input, but I have made it spit out the .wav output from a circuit fed by its own "virtual" sine wave generator. Can't tell you exactly how close that was to what the circuit would actually sound like, but it was pretty convincing and very cool! Rendering is pretty slow.

[JohnH]

Interesting!

What I want then, are two simple clips, one from a standard Strat, the other from an Lp. Each to be recorded as a DI signal with no amp and 'de- EQ'ed' with the inverse of the frequency response calculated by Spice for those ghitars. Then we would have reference signals that can be fed back in as wav files into ltspice, modelling whatever new circuit we want to try, and hear a close approximation of how it will sound. The proviso would be that it would only be true for the same pickup arrangements as the reference recordings. Ie if you want to try a new tone control with N+B, youd need N+B in tbe recording

J

[sumgai]

John, et al,

While having a 'real world' sample of a certain guitar is nice, it is limited to being representative of that guitar only. Characterizing it as typical of the majority of guitars in that class is a major shortcoming, IMO.

Better would be to meld the input of LTSpice with the output of this applet:

www.till.com/articles/PickupResponseDemo/index.html

Long time members here will recognize that link instantly, so for the rest of you..... It's a Java applet that charts the harmonic output of a pickup located at a given position along the string. The neat part is, you can add additional pickups and plot the harmonic interaction between them - that's priceless when it comes to visualizing the effect of how two or more pickups will act in concert. There's even a button to reverse the polarity of each pup!

(Note: My browser configuration blocks Java applets, so I can't see this app in action. Hopefully it still active for the rest of you.)

Seems to me that this would be the best way to see what's happening in near-real-time, as standard Spice programs don't allow for setting where along the string the pickup is located. Which, IMO, greatly curtails the usefullness of said emulations.

Doing such "melding" would be, more or less, the Cat's Meow in terms of a Spice program made specifically for guitarists.   But this is, after all, only a suggestion.





sumgai

[ashcatlt]

Dec 20, 2012 16:05:14 GMT -5 JohnH said:

Interesting!

What I want then, are two simple clips, one from a standard Strat, the other from an Lp. Each to be recorded as a DI signal with no amp and 'de- EQ'ed' with the inverse of the frequency response calculated by Spice for those ghitars. Then we would have reference signals that can be fed back in as wav files into ltspice, modelling whatever new circuit we want to try, and hear a close approximation of how it will sound. The proviso would be that it would only be true for the same pickup arrangements as the reference recordings. Ie if you want to try a new tone control with N+B, youd need N+B in tbe recording

J

Of course for that one would need exact values of all the Ls and Cs and Rs in the circuit. Then I have absolutely no clue how to design the upside down and backwards circuit to "de-EQ" the thing. Might be easier to use a paragraphic EQ to create an upside down version of the response chart.

[JohnH]

Dec 20, 2012 20:17:56 GMT -5 ashcatlt said:

Dec 20, 2012 16:05:14 GMT -5 JohnH said:

Interesting!

What I want then, are two simple clips, one from a standard Strat, the other from an Lp. Each to be recorded as a DI signal with no amp and 'de- EQ'ed' with the inverse of the frequency response calculated by Spice for those ghitars. Then we would have reference signals that can be fed back in as wav files into ltspice, modelling whatever new circuit we want to try, and hear a close approximation of how it will sound. The proviso would be that it would only be true for the same pickup arrangements as the reference recordings. Ie if you want to try a new tone control with N+B, youd need N+B in tbe recording

J

Of course for that one would need exact values of all the Ls and Cs and Rs in the circuit. Then I have absolutely no clue how to design the upside down and backwards circuit to "de-EQ" the thing. Might be easier to use a paragraphic EQ to create an upside down version of the response chart.

My thought was to use an EQ curve entered into audio software to do the de-eq. Audacity will do this so im sure others do too. That gives the input wav file.

Assumptions about values only need to be best guess and consistent to give a useful result because the system is self correcting. Ie when you then feed this new wav file back through spice with the same assumed circuif the original recording should be returned. But now if any new tweaks to the model are made, their effects will be audible.

Sumgai- I'll check out that link, but a system based on a real recording would pick up more subtleties of dynamics and string attack than a fabricated signal.

[geo]

Sumgai- I'll check out that link, but a system based on a real recording would pick up more subtleties of dynamics and string attack than a fabricated signal.

I don't think anyone's ever gonna debate you on that.

[sumgai]

geo,

You're correct, I won't debate John's point about reality. But I do say that a wave file is nothing more than a 'frozen snapshots in time', and equally as undesirable, it represents only one guitar, not a subset of the entire class of guitars.

Why is that bad? Because it's too close to the mark. A modeling program, such as a Spice, should be a step removed from what we're calling reality so that it can give us the details we're looking for, and not muddy the waters with inconsequential data. I liken this to schematics versus wiring diagrams - the latter clutter up the design process in an attempt to inform us both what we're doing, and how to do it. That detracts from the 'what we're doing' aspect, which a schematic lets us do without worries about where to place components, etc.

Ditto here - a wave file tells us too much about all the little stuff that is of much lesser concern than what we're after, which is the harmonic content, the interaction between pickups, and how to achieve what we, as individual users, consider the optimal tone.

Now, that's not to say that outputting a wave file is bad, just the opposite. Therein, the exact pickup characteristics would be desirable, but then again, no one can pick the string exactly the same way every time.... the player changes, just a minute fraction, every time he/she plays. Over a lengthy period, an astute listener can easily discern a changing/changed picking style, no matter that the musician thinks he/she's still doing it the same way as always.   That alone subtracts from the overall usefulness of starting with a wave file.

I guess it's a matter of personal taste, and how one's brain is wired. Me, I prefer (and like) the abstract style, it lets me imagine stray musings that only KIIMH could appreciate. Others prefer to start from concrete data, and wander away from that starting point in search of The Ultimate Mojo^®. To each their own, as they say.

HTH





sumgai

[newey]

I guess it's a matter of personal taste, and how one's brain is wired. Me, I prefer (and like) the abstract style, it lets me imagine stray musings that only KIIMH could appreciate.

OK, way too obvious . . .

I dunno which is more scary, sg channeling KIIMH, or KIIMH channeling sg . . .

[sumgai]

Well, they do say that Great Channels Run Through The Same Mind.......

[gumbo]

Dec 22, 2012 18:08:33 GMT -5 sumgai said:

Well, they do say that Great Channels Run Through The Same Mind.......

...and that Channels are the result of pushing the mud to the sides???