avoiding capacitance from unused portion of coil with a tap

[stratotarts]

I was thinking about Antigua's observation that the unused portion of a tapped coil adds a huge capacitance when the tap is selected. One solution is to bring four wire out and disconnect the unused portion. I was thinking - what about winding them in the same direction, with one end of each coil made common and connected to ground, then the coils selected with the traditional two way selector switch? Would it be different enough from Gibson's patent on bifilar coils to be considered outside the scope of the patent?

Oh, wait - that does add extra wire. Never mind.

[antigua]

Hmm, I'm not picturing this. You say extra wire is a downside, but with 44 or 45 AWG I think you can as exotic as you want, in terms of wire length. 

It's been known for decades that tapped coils underwhelm. Knowing the industry, it wouldn't surprise me is this capacitance issue has gone unknown the whole time, and that with a remedy in hand, a true dual function single coil can be a reality, for any pickup company that is looking to make some cash. Can you patent disconnecting the two sides of the coil, or is that considered "obvious"? I'm surprised the bifilar method would be patented, that seems like an idiotic patent. If someone patents the idea, does the fact that we're talking about it here, on the 29th of March, 2017 invalidate any patent of such with a future date?

[stratotarts]

The way I started thinking, was why not ground the tap and then use the other two connections to select between the two winds. The capacitance of the other coil is out of the circuit that way, but since they are wound in the same direction, the result is that one coil will be out of phase. So then I thought, okay, there is no way to solve that other than reversing the connection of the out of phase coil. But then you can't put them in series, so you need separate coils.

We see a lot of aspects of pickup patents that have some of these characteristics:

1) Have dubious technical merit - no proof of actual performance or effective operation (essentially cosmetics masquerading as technology)
or
2) Subsequently not applied by the patent holder
or
3) Don't meet the criteria of being non-obvious
or
4) Do work in certain configurations but not for the stated reasons

It's impossible to know the actual level of sincerity behind the applications. We have seen that most designers in this industry are very self confident but very few have the actual engineering background that it would take to translate an idea into a sound patent. Perhaps in some cases, the applicants have been coached to inflate or confabulate claims in order to gain acceptance. But I suspect that the other factor, is an attempt to consolidate territory - to effectively block competition by raising the barrier just high enough to make it not worthwhile to risk an infringement with similar designs that actually do work, or utilize the method in some way that makes it actually effective. Look towards the larger companies that have a long term outlook and can actually afford it, for this.

I am not a lawyer, so I'm not sure if a patent could be challenged on the basis of disuse.

[stratotarts]

The four wire method has another advantage - there are actually six potentially useful connections:
1. Series in phase
2. Series (one coil reversed)
3. top coil
4. bottom coil
5. Parallel in phase
6. Parallel (one coil reversed)

But some of these would likely be underwhelming.

[antigua]

Mar 30, 2017 8:43:22 GMT -5 stratotarts said:

The four wire method has another advantage - there are actually six potentially useful connections:
1. Series in phase
2. Series (one coil reversed)
3. top coil
4. bottom coil
5. Parallel in phase
6. Parallel (one coil reversed)

But some of these would likely be underwhelming.

The way they make tapped single coils is they wind a regular coil X turns from the first eyelet, then tie it off on a second eyelet, then wind it another Y turns and terminate at the third eyelet, so the first coil is inside and the second coil is outside, and the three wire setup is a thing of convenience. The four wire setup would require a fourth eyelet, and a little more work, which is probably why this issue has gone unnoticed forever. 

[col]

Mar 29, 2017 14:45:55 GMT -5 antigua said:

Hmm, I'm not picturing this. You say extra wire is a downside, but with 44 or 45 AWG I think you can as exotic as you want, in terms of wire length. 

It's been known for decades that tapped coils underwhelm. Knowing the industry, it wouldn't surprise me is this capacitance issue has gone unknown the whole time, and that with a remedy in hand, a true dual function single coil can be a reality, for any pickup company that is looking to make some cash. Can you patent disconnecting the two sides of the coil, or is that considered "obvious"? I'm surprised the bifilar method would be patented, that seems like an idiotic patent. If someone patents the idea, does the fact that we're talking about it here, on the 29th of March, 2017 invalidate any patent of such with a future date?

Yes. IANAL, but everything I have read about patents supports that. It is one of the required conditions for a patent to be considered valid.

[col]

By the way, I've been working on a scheme which does exactly this (disconnects unused coils): a scheme for splitting picks and combining coils.

What about whole pickups 'hanging from hot'? I assume that they too are a source for undesirable capacitance - yes?

I also assume that coils/pickups hanging from ground are not a problem - yes?

One other thing: are shunted coils/pickups a source of capacitance? I'm, again, guessing not.

[antigua]

Mar 30, 2017 16:42:02 GMT -5 col said:

By the way, I've been working on a scheme which does exactly this (disconnects unused coils): a scheme for splitting picks and combining coils.

What about whole pickups 'hanging from hot'? I assume that they too are a source for undesirable capacitance - yes?

I also assume that coils/pickups hanging from ground are not a problem - yes?

One other thing: are shunted coils/pickups a source of capacitance? I'm, again, guessing not.

You say "splitting", are you referring to a humbucker or a single coil with a tap? This issue is mostly specific to a single coil, where the unused coil is closely hugging the "in use" coil. Split humbuckers have similar issues. 

All of those configurations will capacitively couple. You get capacitance whenever you have two conductors on the same circuit with a charge difference between them, so it doesn't matter which end of the unused coil is connected, because as long as any end if connected, the far end of that disconnected coil will have a charge difference with its neighboring coil, which is "further away" in the circuit, separated by resistance and inductance, creating a charge difference. The end you choose to disconnect merely decides where in the circuit the capacitor happens to be. 

Shunting both ends of the tap... that's interesting. I had wired mine with a SPDT on the hot side of the pickup. What would happen is you'd get a combination of capacitive and inductive coupling, because the unused coil would have continuity. I'd expect that, due to proximity, the inductive coupling will be rather profound. I can't believe I didn't think to try that. I'll test this later. Ideally though, you don't want any relation between the used and unused coil. The only way to get there is disconnecting both ends. 

As a trivial matter, even a fully disconnected coil is still "active", because the open spool of wire is still "closed" by way of it's own inter-winding capacitive coupling, and it's leaches from the "in use" coil by way of inductive coupling. It's a very tiny effect, but it can be seen with split humbuckers when the second coil is disconnected at both ends. It's virtually trivial, and just to say that an open spool of wire is not the same as "air". 

[antigua]

I was excited to try shunting the unused portion of coil in a tapped pickup, so I did this right when I got home, and the result: crazy eddy current losses. 



Interestingly, with the inner coil active and the outer shunted, their is a resonance that is not seen when the reverse is true; outer active, inner shunted. In either case, the parasitic relationship induced by the continuity of the shunt destroys the response of the pickup. So we've learned another new thing; tapped single coils are picky about how you wire them. 

So what does this wiring sound like in a guitar? Obviously weak, probably dark. 

I haven't tested full disconnection, because it would be very difficult to mod the SSL-4 to allow for it, without destroying the pickup. A little hair of 43AWG peaks out from under the tape, and that's all the acces I have to the tap leads. Removing the tape can rip the coil, so that's risky. Maybe one of these days I'll chance it, but there's not much question as to what effect would result, anyway.  

[col]

Mar 30, 2017 17:01:24 GMT -5 antigua said:

Mar 30, 2017 16:42:02 GMT -5 col said:

By the way, I've been working on a scheme which does exactly this (disconnects unused coils): a scheme for splitting picks and combining coils.

What about whole pickups 'hanging from hot'? I assume that they too are a source for undesirable capacitance - yes?

I also assume that coils/pickups hanging from ground are not a problem - yes?

One other thing: are shunted coils/pickups a source of capacitance? I'm, again, guessing not.

You say "splitting", are you referring to a humbucker or a single coil with a tap? This issue is mostly specific to a single coil, where the unused coil is closely hugging the "in use" coil. Split humbuckers have similar issues. 

All of those configurations will capacitively couple. You get capacitance whenever you have two conductors on the same circuit with a charge difference between them, so it doesn't matter which end of the unused coil is connected, because as long as any end if connected, the far end of that disconnected coil will have a charge difference with its neighboring coil, which is "further away" in the circuit, separated by resistance and inductance, creating a charge difference. The end you choose to disconnect merely decides where in the circuit the capacitor happens to be. 

Shunting both ends of the tap... that's interesting. I had wired mine with a SPDT on the hot side of the pickup. What would happen is you'd get a combination of capacitive and inductive coupling, because the unused coil would have continuity. I'd expect that, due to proximity, the inductive coupling will be rather profound. I can't believe I didn't think to try that. I'll test this later. Ideally though, you don't want any relation between the used and unused coil. The only way to get there is disconnecting both ends. 

As a trivial matter, even a fully disconnected coil is still "active", because the open spool of wire is still "closed" by way of it's own inter-winding capacitive coupling, and it's leaches from the "in use" coil by way of inductive coupling. It's a very tiny effect, but it can be seen with split humbuckers when the second coil is disconnected at both ends. It's virtually trivial, and just to say that an open spool of wire is not the same as "air". 

Yes, I meant a splitting a humbucker.

Just to be clear, do you mean that capacitance is still an issue with an 'out of circuit' coil, with one end connected to ground? If so, I am surprised by that. And, is the same true for a coil/pickup shunted at ground?

[col]

Mar 30, 2017 19:55:17 GMT -5 antigua said:

I was excited to try shunting the unused portion of coil in a tapped pickup, so I did this right when I got home, and the result: crazy eddy current losses. 



Interestingly, with the inner coil active and the outer shunted, their is a resonance that is not seen when the reverse is true; outer active, inner shunted. In either case, the parasitic relationship induced by the continuity of the shunt destroys the response of the pickup. So we've learned another new thing; tapped single coils are picky about how you wire them. 

So what does this wiring sound like in a guitar? Obviously weak, probably dark. 

I haven't tested full disconnection, because it would be very difficult to mod the SSL-4 to allow for it, without destroying the pickup. A little hair of 43AWG peaks out from under the tape, and that's all the acces I have to the tap leads. Removing the tape can rip the coil, so that's risky. Maybe one of these days I'll chance it, but there's not much question as to what effect would result, anyway.  

I am very surprised by those results. This matter has been discussed here in the past (at length). Although there were no firm conclusions, (as I recall) the prevailing opinion was that a shunted coil/pickup would have little effect and certainly less than a coil/pickup hanging from hot (one end connected to hot). If your graphs are correct, then this would be a very noticeable effect and could be detected easily by just listening. But past discussion suggests otherwise. I expect that others more involved with those past discussions will wish to comment more about this.

[antigua]

Mar 30, 2017 22:13:03 GMT -5 col said:

Just to be clear, do you mean that capacitance is still an issue with an 'out of circuit' coil, with one end connected to ground? If so, I am surprised by that. And, is the same true for a coil/pickup shunted at ground?

Right, one connected end is still "in circuit". Here's another way to think about it, all of the shielding in the guitar capacitively couples with the positive circuitry in the guitar, even if the shielding is only connected in one spot. When it comes to capacitance, "air" is effectively a conductor, but instead of carrying current, it's trading electrostatic force, which results in the movement of current on either side of the capacitor. The unused portion of coil is essentially the same as shielding. It probably even serves as a layer of shielding if the outer coil is connected by one end on the ground side of the circuit, but the capacitance effect happens either way.

[col]

Mar 30, 2017 22:25:33 GMT -5 antigua said:

Mar 30, 2017 22:13:03 GMT -5 col said:

Just to be clear, do you mean that capacitance is still an issue with an 'out of circuit' coil, with one end connected to ground? If so, I am surprised by that. And, is the same true for a coil/pickup shunted at ground?

Right, one connected end is still "in circuit". Here's another way to think about it, all of the shielding in the guitar capacitively couples with the positive circuitry in the guitar, even if the shielding is only connected in one spot. When it comes to capacitance, "air" is effectively a conductor, but instead of carrying current, it's trading electrostatic force, which results in the movement of current on either side of the capacitor. The unused portion of coil is essentially the same as shielding. It probably even serves as a layer of shielding if the outer coil is connected by one end on the ground side of the circuit, but the capacitance effect happens either way.

I think you have just blown up a lot of Gnut 'theory' and 'best practice' advice with the above. It will be interesting to see what some of the other members have to say.

Thank you for the explanation, antigua.

[antigua]

Mar 30, 2017 22:22:59 GMT -5 col said:

Mar 30, 2017 19:55:17 GMT -5 antigua said:

I was excited to try shunting the unused portion of coil in a tapped pickup, so I did this right when I got home, and the result: crazy eddy current losses. 



Interestingly, with the inner coil active and the outer shunted, their is a resonance that is not seen when the reverse is true; outer active, inner shunted. In either case, the parasitic relationship induced by the continuity of the shunt destroys the response of the pickup. So we've learned another new thing; tapped single coils are picky about how you wire them. 

So what does this wiring sound like in a guitar? Obviously weak, probably dark. 

I haven't tested full disconnection, because it would be very difficult to mod the SSL-4 to allow for it, without destroying the pickup. A little hair of 43AWG peaks out from under the tape, and that's all the acces I have to the tap leads. Removing the tape can rip the coil, so that's risky. Maybe one of these days I'll chance it, but there's not much question as to what effect would result, anyway.  

I am very surprised by those results. This matter has been discussed here in the past (at length). Although there were no firm conclusions, (as I recall) the prevailing opinion was that a shunted coil/pickup would have little effect and certainly less than a coil/pickup hanging from hot (one end connected to hot). If your graphs are correct, then this would be a very noticeable effect and could be detected easily by just listening. But past discussion suggests otherwise. I expect that others more involved with those past discussions will wish to comment more about this.

A lot has been learned recently through practical experimentation. With dropping prices of oscilloscopes and powerful software, it's easier than ever to actually test effects that people have only pontificated about in the past.

So the eddy current issue is not so surprising if you read a report Ken Willmott (our own stratotarts) made on eddy current effects kenwillmott.com/blog/wp-content/uploads/2016/11/Pickup_Cover_Geometry.pdf . Once the unused coil is permitted continuity by the shunt, it becomes free to inductively couple with the primary coil. Current flows in the unused coil, producing a magnetic field that opposes the primary magnetic field, due to Lenz's law. This effect increases with frequency, due to Farday's law, which is why you see the voltage drop with increased frequency. 

[antigua]

Mar 30, 2017 22:29:57 GMT -5 col said:

Mar 30, 2017 22:25:33 GMT -5 antigua said:

Right, one connected end is still "in circuit". Here's another way to think about it, all of the shielding in the guitar capacitively couples with the positive circuitry in the guitar, even if the shielding is only connected in one spot. When it comes to capacitance, "air" is effectively a conductor, but instead of carrying current, it's trading electrostatic force, which results in the movement of current on either side of the capacitor. The unused portion of coil is essentially the same as shielding. It probably even serves as a layer of shielding if the outer coil is connected by one end on the ground side of the circuit, but the capacitance effect happens either way.

I think you have just blown up a lot of Gnut 'theory' and 'best practice' advice with the above. It will be interesting to see what some of the other members have to say.

Thank you for the explanation, antigua.

IIRC, that's only with respect to humbucker splitting, I'm not aware that they have an opinion on tapping single coils.

[antigua]

The thing about a humbucker, as opposed to a tapped single coil, is the unused coil is a little further away from the primary. All the same applies, but to a lesser degree, because there is far less inductive and capacitive coupling. 

Here is a Seymour Duncan Pearly Gates bridge tested with various splitting combinations, with a standard load of 470pF and 200k ohms:


(5dB divisions)




(1dB divisions)





You can see that there is a difference, but it's very small, about 2dB difference at resonance at the most, and this is an uncovered humbucker. The resonance frequency doesn't really change at all.

Technically, shunted does provide the strongest resonance, but it's such a small difference that I think it's best to wire the split however is most convenient. I don't think you'll hear any difference.