Ibanez HFS pickups - I'm curious

[Mark Hammer]

I was flipping through the current Ibanez catalog and came across their JTK line of guitars. The JTK guitars use a rather unique pickup called the "HFS". From the side view, it appears to be a sort of P-90/Lace-like pickup, but with a cancelling dummy coil wound around the sensing coil. Judging from the catalog, the entire dual-concentric-coil package comes in around the P-90 footprint size.

It's an intriguing idea. Does anyone here have any experience with concentric cancelling coils?

[Mike Sulzer]

That is an interesting idea. The first thing to notice is that it shares one of the disadvantages of the old type of stacked humbucker: the canceling coil picks up some of the signal and cancels it as well. The reason that it does not cancel it all is:
1. the flux from the vibrating string that passes through the main coil has to return to the string. Some of it passes through or inside the canceling coil, reducing the signal in that coil;

2. The canceling coil has larger area than the main coil, and so it needs fewer turns in order to pickup the hum with equal strength, and so it also picks up less signal.

So the inductance of the canceling coil is less than that of the main coil, but I think it is still significant, and so one should expect some loss of highs.

I suppose you could think of this as a dummy coil that is smaller than one would like, and mounted in a non-optimum location.

[frankfalbo]

Originally Posted by Mike Sulzer

I suppose you could think of this as a dummy coil that is smaller than one would like, and mounted in a non-optimum location.

Bingo

[David Schwab]

There was at least one patent for a coaxial humbucker similar to this from 1983. I think their might be more.

4372186 (see attached screen grab). Looks just like the Ibanez design.

I had this idea many years ago, but never tried it. I suppose you can try using large diameter wire for the dummy coil, like the Ilitch system, but you still run into the phase cancelation problem Mike mentioned.

Bill Lawrence had two; 3711619, and 5376754, but in those the outer coils was for tonal adjustment, and not for hum cancelation.

[Mark Hammer]

I think it is fair to note that the ad copy for the guitar in question conveys the impression that it aims for a 60's oddball guitar vibe. In other words, it is not trying to sound like SRV or Billy Gibbons, so all conventional wisdom about the tonal impact of such coils is tossed out. I'm not suggesting that is somehow "less". Rather, the normal thinking of "If I do this, I'm gonna screw up X" is not hemming in design, and whatever the tonal outcome is is treated as musically valid and idiosyncratic rather than "wrong".

That patent diagram David shows sure enough looks like it, though.

[David Schwab]

You hit the nail on the head Mark.

I've had a few guitars with stacked pickups, like the old Duncan stacks. They have to over wind them to make up for the reduced low end, but they sound just like single coils. I have some lawrence L-250s and a dual rail Tele pickup in another guitar. They sound great too. Maybe not exactly like a particular Strat pickup, but, really, who cares?

I think too much emphasis is placed on getting exactly the same tone as SRV or Billy Gibbons or whom ever, and they sure weren't thinking about using a particular pickup to sound like someone else! I love their playing and tone, but I don't want to sound like them.

I miss the 70's where people were happy to try new things. Too many people lack originality these days! Music has become a uniform that you rent.

If you can't get an interesting tone from any guitar with any pickup, there's a problem! And it ain't the guitar.

[the great waldo]

Originally Posted by David Schwab

You hit the nail on the head Mark.

I've had a few guitars with stacked pickups, like the old Duncan stacks. They have to over wind them to make up for the reduced low end, but they sound just like single coils. I have some lawrence L-250s and a dual rail Tele pickup in another guitar. They sound great too. Maybe not exactly like a particular Strat pickup, but, really, who cares?

I think too much emphasis is placed on getting exactly the same tone as SRV or Billy Gibbons or whom ever, and they sure weren't thinking about using a particular pickup to sound like someone else! I love their playing and tone, but I don't want to sound like them.

I miss the 70's where people were happy to try new things. Too many people lack originality these days! Music has become a uniform that you rent.

If you can't get an interesting tone from any guitar with any pickup, there's a problem! And it ain't the guitar.

HOW TRUE

Cheers

ANDREW

[automan]

Originally Posted by David Schwab

Music has become a uniform that you rent.

Forgive me, but I am going to steal that and use it in real life

[Mark Hammer]

Originally Posted by David Schwab

Music has become a uniform that you rent.

If you can't get an interesting tone from any guitar with any pickup, there's a problem! And it ain't the guitar.

I will overlook the "nail" pun and say that the boldfaced line is quotable. That's a good'un. Copyright it.

[David Schwab]

I can't take full credit!

I stole it from this XTC song:

Travels in Nihilon

You've learnt no lessons
all that time so cheaply spent
there's no youth culture
only masks they let you rent

Travels, travels in Nihilon
we've seen, no Jesus come and gone

Fashion, their vampire
drapes itself across your back
as you fall from style
it waits rebirth on its rack

Building your whimsy
hypnotising you to need
dance goes full circle
one step ahead of your greed

You've learnt no lessons
all those years to get it right
flashes of promise
burn out faster than strobe light

Sums up the idea that if you only follow fashion trends, you will be left with nothing but wasted time when the new trend replaces the old.

I get bored real fast with music and other things. I'm always looking for something I have never heard before.

[Possum]

Its an old idea and that patent has been posted here going back five years or more. Yes I tried it a long time ago but the results were very muddy. I suspect they are using some kind of shielding material in between the two coils. Also note that the pickup in the patent probably never saw the production line, I've never seen a pickup like that and note that it says it produces a hi-fi type tone. I imagine it may have been larger than a regular humbucker as well. Just goes to show there really aren't any new ideas, just different iterations of stuff thats been done before....

I would think they are probably using two different guages of wire too, my attempt was with just 42 and the dummy coil part wasn't enough turns to kill the hum.....would be interesting to get hold of one of those pickups and take it apart.....

[Possum]

I wonder if the "metallic frame for magnetic field" is actually magnetic and acting as two outside poles, now THAT would be unique. Its probably something like Mu metal though, I bet. Cool idea....

[David Schwab]

Originally Posted by Possum

I wonder if the "metallic frame for magnetic field" is actually magnetic and acting as two outside poles, now THAT would be unique. Its probably something like Mu metal though, I bet. Cool idea....

Yeah, it is. A lot of the stacked pickups are like that.

Here's a Dimarzio. The shield is touching the bottom of the magnets and leading the field back up to the strings.

The bottom coil only has metal slugs and is would with less turns of heavier wire.

[Mike Sulzer]

Originally Posted by Possum

I wonder if the "metallic frame for magnetic field" is actually magnetic and acting as two outside poles, now THAT would be unique. Its probably something like Mu metal though, I bet. Cool idea....

If it is something like Mu metal, it does tend to act as a partial return path for the the changing flux from strings passing through the core of the inner coil and back to the string, thus reducing the cancellation in the outer coil. It is an incomplete shield, and I would not expect much more than a 6 db effect based on a few experiments I have done.

[Possum]

I'm not quite understanding why heavier guage wire with fewer turns on the dummy coil would be total cancelling, maybe its not? Personally I'm not a fan of fake single coils and don't really even like matched winds on anything, it sounds sterile and too many good frequencies are cancelled. Good for high distortion though, eh?

[Possum]

Would less turns of heavier wire pick up more 60 cycle signal to match the top coil? Meaning thicker wire will induce more noise?

[Mike Sulzer]

Originally Posted by Possum

I'm not quite understanding why heavier guage wire with fewer turns on the dummy coil would be total cancelling, maybe its not? Personally I'm not a fan of fake single coils and don't really even like matched winds on anything, it sounds sterile and too many good frequencies are cancelled. Good for high distortion though, eh?

Well, it is not really perfect for a couple of reasons (but neither are humbuckers).

1. The hum flux changes as a function of location; so the small coil and the larger dummy do not see exactly the same hum signal. But it is very close.

2. The capacitances of the main and dummy coils are different. This makes no difference to the lower hum harmonics, but might make some difference at the highest hum harmonics that kind of sound like a buzzing.

[Mike Sulzer]

Originally Posted by Possum

Would less turns of heavier wire pick up more 60 cycle signal to match the top coil? Meaning thicker wire will induce more noise?

The wire thickness does not affect the voltage induced around a loop to any significant degree. The reason for using heavier gauge wire is to keep the resistance as low as possible so that the series combination of the two coils looks as much as possible like the main coil by itself. At least that is the most important reason.

[Joe Gwinn]

Originally Posted by Mike Sulzer

If it is something like Mu metal, it does tend to act as a partial return path for the the changing flux from strings passing through the core of the inner coil and back to the string, thus reducing the cancellation in the outer coil. It is an incomplete shield, and I would not expect much more than a 6 db effect based on a few experiments I have done.

Someone (Kinman?) patented a stacked humbucker with a mild steel plate at least 0.100" thick between the coils, and I would guess that they get far more than 6db hum reduction. There may also have been thinner steel plates coming up from the middle plate to the top (nearest to the strings).

[Mike Sulzer]

Originally Posted by Joe Gwinn

Someone (Kinman?) patented a stacked humbucker with a mild steel plate at least 0.100" thick between the coils, and I would guess that they get far more than 6db hum reduction. There may also have been thinner steel plates coming up from the middle plate to the top (nearest to the strings).

I think the purpose of that is help prevent signal from getting to the bottom coil by guiding the time-varying flux from the string and coming out the bottom of the top coil away from the bottom coil. I do not know how many db that buys.

The situation I described is for approx. .1" thick steel pieces placed between the coils in a six coil pickup.

If the plates were very large, or approaching infinity, the simple skin depth argument would apply. For example, 5.18A (pp 221-222) in the latest edition of Jackson derives the skin depth for a semi-infinite conductor with a magnetic field parallel to its surface.

For small pieces, you have to solve the differential equation or make a measurement.

[David Schwab]

Originally Posted by Joe Gwinn

Someone (Kinman?) patented a stacked humbucker with a mild steel plate at least 0.100" thick between the coils, and I would guess that they get far more than 6db hum reduction. There may also have been thinner steel plates coming up from the middle plate to the top (nearest to the strings).

Yes, it was Kinman (5,668,520), and then DiMarzio made their own version and got that patented as well. See the photo I posted above.

Kinman had a shield on the top and bottom coil. DiMarzio only does the top coil. Kinman also has a version with a laminated steel bottom bobbin. Duncan came out with a new stack that combines the features of both of those, and got that patented as well.

Here's the Kinman Strat pickup.

[Joe Gwinn]

Originally Posted by Mike Sulzer

I think the purpose of that is help prevent signal from getting to the bottom coil by guiding the time-varying flux from the string and coming out the bottom of the top coil away from the bottom coil. I do not know how many db that buys.

That's my understanding too. I think Kinman says as much in his patent (5,668,520 according to David). But I bet Kinman (and competitors) would not bother for 6 dB, which would be almost inaudible, so the design must work far better than that.

The term "shield" may be a marketing misnomer, as the effect may be due more flux diversion than to shielding.

The situation I described is for approx. 0.1" thick steel pieces placed between the coils in a six coil pickup.

If the plates were very large, or approaching infinity, the simple skin depth argument would apply. For example, 5.18A (pp 221-222) in the latest edition of Jackson derives the skin depth for a semi-infinite conductor with a magnetic field parallel to its surface.

If the magnetic field is parallel to the surface, the shield has already won, even if by default. It's the perpendicular component that matters.

For small pieces, you have to solve the differential equation or make a measurement.

In practice, for such complex shapes, one must use finite-element numerical methods. For quasi magnetostatic systems such as pickups, the standard approach is the Boundary Element Method (BEM): Boundary element method - Wikipedia, the free encyclopedia.

Developing and validating a 3D program to do this well is a major effort, almost a career, which is why the commercial programs that do this are so expensive. But building a more limited solver is not impossible, although it is a lot of work. FEMM is based on the BEM, but is restricted to 2D to make the programming effort practical for one person to do.

I think that using FEMM to analyze a central cross-section of a stacked rail humbucker design would be instructive. The test would be to move the simulated string slightly and note the ratio of the flux changes above and below the shield plate. Beyond that, it's probably easier to make a test article and make measurements.

[David Schwab]

Here's a listing of all the stacked patents of interest:

Kinman

5668520
5834999
6103966
7022909
7189916

DiMarzio:

4442749
5811710
5908998 (this is for added slugs to the bottom coil)

Beller (Duncan):

7166793

Non "shielded" stacks:

Duncan:

4524667

Stich (Bill L)

5789691
20060156911 (Fender scn)

[Mike Sulzer]

We are talking about two situations here, the parallel and the perpendicular. The former applies to the original drawing that Mark presented, which is why I brought up my six coil pickups. The perpendicular case applies to the "shielded" stacked bucker, yes, but that is not what we were discussing at first. As I said, I do not know how many db that shield gives you.

Would they bother for 6 db? I bet they would. All they are trying to do is reduce the level of the signal getting to the bottom pickup. 6 db means that whatever field strength is getting through is reduced by a factor of about 4. When the subtraction is done, this would mean that the resulting signal level would be no less than 75% of what the top coil can provide alone. Since it does not all get down there even without the shield, the 6 db additional from the shield means that one would keep nearly of what the top could can provide. Losing a few percent would not be a problem, so there is no need to make a really good shield as one would like to do for hum reduction.

As we have discussed before, FEMM does not have the accuracy for detecting the very small changes of the field when the string moves in static field. However, there is no need to do it that way. As before, one simply magnetizes the string in the model and looks for the ratio of the fields on the two sides of the shield.

[Joe Gwinn]

Originally Posted by Mike Sulzer

We are talking about two situations here, the parallel and the perpendicular. The former applies to the original drawing that Mark presented, which is why I brought up my six coil pickups. The perpendicular case applies to the "shielded" stacked bucker, yes, but that is not what we were discussing at first. As I said, I do not know how many db that shield gives you.

Would they bother for 6 db? I bet they would. All they are trying to do is reduce the level of the signal getting to the bottom pickup. 6 db means that whatever field strength is getting through is reduced by a factor of about 4. When the subtraction is done, this would mean that the resulting signal level would be no less than 75% of what the top coil can provide alone. Since it does not all get down there even without the shield, the 6 db additional from the shield means that one would keep nearly of what the top could can provide. Losing a few percent would not be a problem, so there is no need to make a really good shield as one would like to do for hum reduction.

The problem is that minimum discernible difference in sound level is around 3 dB. This is a matter of the physiology of the human ear, not electronics.

Human Hearing: Amplitude Sensitivity Part 1 -- Reviews and News from Audioholics

As we have discussed before, FEMM does not have the accuracy for detecting the very small changes of the field when the string moves in static field. However, there is no need to do it that way. As before, one simply magnetizes the string in the model and looks for the ratio of the fields on the two sides of the shield.

I don't recall the discussion any more, but magnetizing the string may work well enough. Even if it doesn't exactly model the effect of the motion of an unmagnetized string in a magnetic field, it may be instructive.

The basic claim is of equivalency, that these two approaches will yield the same answer. I'll have to think about this.

[Mike Sulzer]

Originally Posted by Joe Gwinn

The problem is that minimum discernible difference in sound level is around 3 dB. This is a matter of the physiology of the human ear, not electronics.

When you reduce the level of signal getting to the bottom coil by six db, you assure that the loss of level resulting from the subtraction is less than 2db below the output level of the top coil alone. The 6 db change is not significant by itself.

Originally Posted by Joe Gwinn

I don't recall the discussion any more, but magnetizing the string may work well enough. Even if it doesn't exactly model the effect of the motion of an unmagnetized string in a magnetic field, it may be instructive.

The basic claim is of equivalency, that these two approaches will yield the same answer. I'll have to think about this.

The result of putting a string made of magnetic material in a magnetic field is a magnetized string as long as the field is present. The field from that magnetization is what matters. (It is a small perturbation of the static field; so linearity applies.)

[Joe Gwinn]

Originally Posted by Mike Sulzer

When you reduce the level of signal getting to the bottom coil by six db, you assure that the loss of level resulting from the subtraction is less than 2db below the output level of the top coil alone. The 6 db change is not significant by itself.

Hmm. I guess we need to clearly differentiate between hum reduction and music reduction. It sounded like you were saying that the hum reduction would be only 6 db, which would not be worth the trouble. Reducing music level by 6 db is also bad, but not nearly as bad as not canceling the hum.

The result of putting a string made of magnetic material in a magnetic field is a magnetized string as long as the field is present. The field from that magnetization is what matters. (It is a small perturbation of the static field; so linearity applies.)

I was thinking along these lines as well. But have not come to a conclusion.