X2Ns

[flyneye]

This is a bit like paying attention to whether a bulldozer has chromed bolts on the treads or not.
Extra highK pickups exist because of musicians who demanded distortion for stack amps (which are harder to distort than a low wattage amp) which were only invented because the P.A. had not been yet. Over time these musicians became unaware of dynamics which had disappeared by this time leaving the guitar in the role of ORGAN which the various metal styles still cling to. Doesn't have much to do with guitar playing or dynamics, might as well juice up a hammond. O.K. that wasn't supposed to sound so "all inclusive" and mean every player, but you get my point, no?
Sorry, an industry that promotes misinformation to sell goods no one at his heart really wants gets me worked up like politics.

[Mike Sulzer]

Mike -

Thanks for this summation of that version of a sidewinder.

The single-coil-like coherence of the transient & dynamic responses is one of the main reasons I'm using this format.

Regarding the hum canceling efficiency (relative to other designs) I'll do some tests regarding coil orientation relative to hum source point, direction & polarization and see what comes through.

I will be very interested in seeing your results.

[David Schwab]

Extra highK pickups exist because of musicians who demanded distortion for stack amps (which are harder to distort than a low wattage amp) which were only invented because the P.A. had not been yet.

Ummmm.... no. PAs have been around since like the 40s? How old are you? You never saw concerts, or pictures or videos of concerts from back in the late 60's and 70's with big PAs? Never saw the movie Woodstock? Even the Beatles had a PA at Shea Stadium. Not much of a PA, but they had one.

Here's the Greatful Dead's rig in 1974:



The first high output pickup on the market was the DiMarzio Super Distortion. That came out in 1972. A lot of people used them back then, and still do. One of the most popular pickups is the Duncan JB. Seymour made the first one for Jeff Beck. I certainly think Beck knows about dynamics. Don't you?

The reason loud pickups exists is because they overdrive amps easier. People want and use them. This is not so much of an issue now, but back in the 70's it was. Also overwound pickups just push the amp differently than vintage wound pickups. And look at all the people who use boosters and Tube Screamers and their clones.

Personally I don't like pickups like the X2N or Tone Zone, because they don't clean up well. Even the JB is too dark for me. But I do like hotter than PAF pickups in the bridge position. But if you want dynamics, turn down your volume or pick softer. Has nothing to do with the pickups.

[fieldwrangler]

I will be very interested in seeing your results.

So, I was starting to set up a somewhat controlled test and was soon faced with the specter of yet another issue I don't feel confident about.

And so, I put the question to y'all: What do you think would make a good standard "hum source" (electrostatic/buzz source will come along later) that wouldn't threaten to demagnetize the collection??

So far, a wall wart has stepped forward to volunteer for the position, but I'm sure someone has a better idea.

[David Schwab]

What do you think would make a good standard "hum source" (electrostatic/buzz source will come along later) that wouldn't threaten to demagnetize the collection??

If your computer has a CRT monitor, then you are sitting in front of it! That's how I test hum reduction and shielding.

[Joe Gwinn]

So, I was starting to set up a somewhat controlled test and was soon faced with the specter of yet another issue I don't feel confident about.

And so, I put the question to y'all: What do you think would make a good standard "hum source" (electrostatic/buzz source will come along later) that wouldn't threaten to demagnetize the collection??

So far, a wall wart has stepped forward to volunteer for the position, but I'm sure someone has a better idea.

The best approach will be a Helmholtz coil, Helmholtz coil - Wikipedia, which will yield a fairly uniform field. Humbuckers reject uniform fields (those from distant sources) in favor of those from nearby and thus nonuniform fields (like the vibrating strings, and of course that nearby wall wart.

In practice, you will use a coil of order a meter in diameter, so the center field will be quite uniform. Aside from being big enough, the details are not especially critical.

It will be basically impossible for any reasonable Helmholtz Coil of such a size to demagnetize anything. At least not without a power supply capable of causing the coil to explode.

[Mike Sulzer]

So, I was starting to set up a somewhat controlled test and was soon faced with the specter of yet another issue I don't feel confident about.

And so, I put the question to y'all: What do you think would make a good standard "hum source" (electrostatic/buzz source will come along later) that wouldn't threaten to demagnetize the collection??

So far, a wall wart has stepped forward to volunteer for the position, but I'm sure someone has a better idea.

A solenoid has a constant field inside away from the ends and so it can be small. Find a small round plastic tub about six inches in diameter. (Mine had chorine tablets in it.) Wind the outside with wire, neatly, single layer. #22 plastic insulated hook wire is fine. You can use smaller magnet wire if you like, but you will get tired of winding it on. This does not take much; so do it the easy way. If you run out of wire before you do the whole length, it will work, but the field will be more constant if you make it longer.

Put a block in the tub for the pickup to sit on. It should sit as near to the center as possible.

This is really easy to drive; I use the headphone jack on my mac driven from suitable software. The coil will have a resistance of an ohm or two; put 100 or 200 ohms in series with it so the headphone driver will be happy. You can easily get a couple hundred mV from a single coil placed at the center with its coil axis aligned with the axis of the solenoid. Turn the pickup on its side so the axes are perpendicular and you can decrease that by about 25 db.

You can do any frequency you want, but remember that the sensitivity of the pickup increases linearly with frequency; so you will have to adjust the drive level.

So this should do what you want.

[fieldwrangler]

I must say you fellas are presenting some interesting options.

I won't actually be at my workshop for a coupla weeks (and there are a few pressing projects over there stacking up) but I do feel that this line of inquiry is valuable enough to the common knowledge base that I'm definitely going to pursue it.

[Joe Gwinn]

For the same length and diameter, a Helmholz coil will yield the more uniform field, compared to a solenoid. I mentioned a diameter of one meter on the assumption that you were intending to test pickups installed in a guitar.

If not, you can make a far smaller coil. But it needs to be large enough that you can rotate the pickup in all directions withoiut getting too close to the actual coil or the ends of the coil. The uniform field is a volume in the center. . It's easiest if the coil is large enough that you can put a hand inside.

But beware electrostatic pickup if the pickup isn't shielded. One can also shield the hum-generating coils, but be sure not to make the shield into a shorted turn.

A very common and cheap source of plastic for winding coils is plumbing supply houses and hardware stores that carry PVC drain pipe and fittings. Four inch diameter is very common, and used in houses, but larger diameters are available. By the way, the coils need not be round.

[Mike Sulzer]

Yes it is, but the solenoid is pretty good for this application. For example, a coil 13 cm in diameter and 13 cm long has a variation of about 2% from the center point to a point .5 cm away along the axis. (This uses the results of a problem at the end of chapter 5 of Jackson.)

Using this device gives the following useful results:

1. Reversing the phase of the windings on a Japanese paf type clone changes the hum sensitivity by about 25 db.

2. Removing the slugs from a humbucker coil decreases the hum sensitivity by about 5 db. This is to be compared to the result of using a small coil over one of the slugs, and then removing the slug. This gave 7 db, indicating that using cores increases the signal to hum ratio by a bit as well as increasing the overall sensitivity. An even smaller coil that better mimics the field line spread from a string might show even more difference.

I do the measurement at 100 Hz rather than 60 Hz and use an FFT analyzer for the measurement. This eliminates the problem of environmental hum pickup.

I think four inches is too small to work with conveniently. Since six inch pipe is not readily available at my local hardware stores, I used the small bucket, which I already had. When you are done, you throw any associated stuff in the bucket, put the lid back on and store it for next time.

[Joe Gwinn]

Yes it is, but the solenoid is pretty good for this application. For example, a coil 13 cm in diameter and 13 cm long has a variation of about 2% from the center point to a point 0.5 cm away along the axis. (This uses the results of a problem at the end of chapter 5 of Jackson.)

Testing a perfectly-balanced non-concentric two-coil humbucker in a field with 2% nonuniformity will yield about 2% mismatch of flux and thus output voltage between the two coils of the humbucker. This implies 20 Log10[(1.02V-V)/(1.02V+V)]= -40 dBc of leakage, compared to the sum of signals from the coils. This implies that the measurement floor is -40 dBc, which is more than adequate for measuring humbuckers with 20 or 30 dB of rejection.

For a 13 by 13 cm solenoid, the sweet volume is a rough sphere about a centimeter in diameter. Pickups are larger than 1 cm, and the deviation from uniformity probably increases sharply away from the 1 cm sweet volume. So a larger coil may be useful.

Using this device gives the following useful results:

1. Reversing the phase of the windings on a Japanese paf type clone changes the hum sensitivity by about 25 db.

2. Removing the slugs from a humbucker coil decreases the hum sensitivity by about 5 db. This is to be compared to the result of using a small coil over one of the slugs, and then removing the slug. This gave 7 db, indicating that using cores increases the signal to hum ratio by a bit as well as increasing the overall sensitivity. An even smaller coil that better mimics the field line spread from a string might show even more difference.

I don't quite understand this paragraph (#2). It is true that adding a slug will increase both hum and music pickup, although the music pickup increase will be smaller at higher frequencies due to eddy currents.

I do the measurement at 100 Hz rather than 60 Hz and use an FFT analyzer for the measurement. This eliminates the problem of environmental hum pickup.

If one has the equipment, this is certainly the right way to do it. A lock-in amplifier is even better.

If one does not have the equipment, a better approach is to use a fairly strong test signal, and simply swamp stray hum signals. The test is to measure the output voltage from one or the other coil of the pickup with the drive generator turned off and then turned on, and compute the following: 20*Log10[Voff/Von]= measurement floor in decibels. You will not be able to measure pickups with hum rejection better than this floor.

I think four inches is too small to work with conveniently. Since six inch pipe is not readily available at my local hardware stores, I used the small bucket, which I already had. When you are done, you throw any associated stuff in the bucket, put the lid back on and store it for next time.

It's a nice approach.

Given the above, I think I would wind a Helmholtz coil on the outside of a 5-gallon bucket, costing a few dollars from Home Depot. The diameter is 10" and the windable length is 11". The buckets are made of high density polyethylene, so most glues will not stick. What will stick is hot-melt glue. Or, simply drill holes in the bucket and tie the windings to the bucket.

As I mentioned earlier, the coil need not be round, and one can use a square or octagonal frame or box made of wood.

[Mike Sulzer]

Testing a perfectly-balanced non-concentric two-coil humbucker in a field with 2% nonuniformity will yield about 2% mismatch of flux and thus output voltage between the two coils of the humbucker. This implies 20 Log10[(1.02V-V)/(1.02V+V)]= -40 dBc of leakage, compared to the sum of signals from the coils. This implies that the measurement floor is -40 dBc, which is more than adequate for measuring humbuckers with 20 or 30 dB of rejection.

With the humbucker axis parallel to the coil axis and exactly centered the non-uniformity does not matter by symmetry. Since you can locate the pickup to much better than a centimeter and get it reasonably level, the floor is quite a bit better than 40 dbc. But I do agree that a Helmholtz pair would be much better for serious measurements. But what I have was so easy! I have not come up with a really simple way of winding two compact multilayer coils on the outside of a bucket.

For a 13 by 13 cm solenoid, the sweet volume is a rough sphere about a centimeter in diameter. Pickups are larger than 1 cm, and the deviation from uniformity probably increases sharply away from the 1 cm sweet volume. So a larger coil may be useful.

Yes it would, especially for what Bob wants to do because he has to put a sidewinder (some of which are quite large) in two positions, one of which has the screws (or whatever) perpendicular to the field. This would be a lot easier with a large uniform volume.

I don't quite understand this paragraph (#2). It is true that adding a slug will increase both hum and music pickup, although the music pickup increase will be smaller at higher frequencies due to eddy currents.

The purpose of this measurement is to examine a fairly subtle effect of a pole piece. The field lines from a hum source are close to straight and parallel. The pole piece amplifies the effect. But the effect on a field with lines that spreaad should be greater because the pole piece can greatly reduce the spread. This gets more flux through the windings furthest from the strings. I found a 2 db increase, but the coil was too large to have the same spread as the field of a string, and so it could be more.

If one has the equipment, this is certainly the right way to do it. A lock-in amplifier is even better.

All it takes is a pc with sound card and some software, maybe free. Pretty much everyone should be able to do this. I do not think that the really narrow bandwidth you can get with a lock in is necessary.

Given the above, I think I would wind a Helmholtz coil on the outside of a 5-gallon bucket, costing a few dollars from Home Depot. The diameter is 10" and the windable length is 11". The buckets are made of high density polyethylene, so most glues will not stick. What will stick is hot-melt glue. Or, simply drill holes in the bucket and tie the windings to the bucket.

As I mentioned earlier, the coil need not be round, and one can use a square or octagonal frame or box made of wood.

I like your big bucket approach. You have plenty of windable length since the coils are separated by the radius.

[Joe Gwinn]

With the humbucker axis parallel to the coil axis and exactly centered the non-uniformity does not matter by symmetry. Since you can locate the pickup to much better than a centimeter and get it reasonably level, the floor is quite a bit better than 40 dbc. But I do agree that a Helmholtz pair would be much better for serious measurements.

While in theory symmetry will help, in practice I bet one will be hard pressed to reliably find that center point, and to be sure that one had found it. A far larger sweet volume is needed.

For the same construction effort, a Helmholtz coil yields better uniformity, so the second generation of drive coils might as well be Helmholtz coils.

But what I have was so easy! I have not come up with a really simple way of winding two compact multilayer coils on the outside of a bucket.

While the coils need to be reasonably compact, they don't have to be especially precise. The easiest way to make the coils would be to wind them on a piece of plywood with a ring of headless nails sticking up. One winds the coil on the nails, then stabilizes the coil by lacing it with flat nylon lacing twine (or by wrapping the coil with cloth tape), removes the coil from the form, installs in on the bucket, and then ties the coil to the bucket with more lacing twine threaded through holes in the bucket and wrapped around the coil.

One can use ordinary braided nylon fishline as lacing twine.

There is a cloth tape specifically intended for stabilizing coils wound on a nail board, which is often done to make motor windings.

One can stabilize the coil by soaking it with shellac and allowing it to dry for a week.

Yes it would, especially for what Bob wants to do because he has to put a sidewinder (some of which are quite large) in two positions, one of which has the screws (or whatever) perpendicular to the field. This would be a lot easier with a large uniform volume.

Yes.

The purpose of this measurement is to examine a fairly subtle effect of a pole piece. The field lines from a hum source are close to straight and parallel. The pole piece amplifies the effect. But the effect on a field with lines that spread should be greater because the pole piece can greatly reduce the spread. This gets more flux through the windings furthest from the strings. I found a 2 db increase, but the coil was too large to have the same spread as the field of a string, and so it could be more.

You've put your finger on it - this effect is too subtle to really measure reliably with the present apparatus.

All it takes is a pc with sound card and some software, maybe free. Pretty much everyone should be able to do this. I do not think that the really narrow bandwidth you can get with a lock in is necessary.

Not everyone will have the sound card et al, though I suppose that musicians are more likely to have such than the general population.

As for the lock-in amplifier, I suggested it because I suspect that you have access to such a beast. Even an unloved old unit from the back of the equipment room would work for this application, and will be far more precise that the FFT approach.

I like your big bucket approach. You have plenty of windable length since the coils are separated by the radius.

Yes. A Helmholtz coil 10" in diameter has two windings 5" apart.


There is a whole literature on extensions to the Helmholtz coil using multiple windings of various turns counts and radii, but most of these schemes are hard to build. The classic limiting case of these schemes is to wind a solenoid on the surface of a sphere (not a cylinder). The field within the sphere (often called a fluxball) is quite uniform, and not just in the central volume. Much detail may be found in US Patent Application 20100019764, available via "www.pat2pdf.org".

[Mike Sulzer]

I am afraid I have to disagree with a lot of what you have written. First, one can locate the center of something 13 cm in size to considerably better than a one cm. I do not know what makes you doubt this.

Second, your estimate of the small size of the "sweet spot" on this solenoid is incorrect. This solenoid is approximately as long as its diameter, not half that as with the Helmholtz coil pair. It is true that the H. coil is still better, but not by that much. The field of the H. coil is down about six percent at the location of the coils. The solenoid, with the geometry I have used, is about ten percent down at the same distance (not at the end of the coil). This is not that big a difference. It is also true that the H coil field is flatter near the center, in the sense that more derivatives are zero. This is just not that important in this application.

I think the method you suggest for making the coils is good for when you need quite a bit of wire and it needs to be pretty big to carry significant current. In this application, we need only about 50 turns on each coil, and the wire can be small. I will make a Helmholtz pair next with #30 magnet wire on a somewhat larger bucket. I plan to define the locations of the coils with a number of layers of tape, and then just wind them on the bucket.

No, the effect is not too subtle to measure with this setup. I would prefer a more accurate one, which is the reason for a new bucket, but I think I did measure it.

An old lock in amp is better than an FFT machine? No way. The limitation in the process is in the analog multiplication. Look what happens in a digital receiver where you use digital baseband mixing. The specs are far better because the digital multipiclation is superior. In any case, this is just not a very demanding application. You just need to sample with a lot of bits so that you preserve the dynamic range and then filter to reject the hum and keep the signal. The FFT does that just fine.

Did Bateman use lock in amplifiers in his excellent work on capacitor distortion? No, he did it digitally, and for that very demanding application, he made sure he had really good digital hardware and software.

[Joe Gwinn]

I am afraid I have to disagree with a lot of what you have written. First, one can locate the center of something 13 cm in size to considerably better than a one cm. I do not know what makes you doubt this.

Practicality. While one can find the center, why bother? Use a big enough sweet volume, and you don't need to be that careful.

Second, your estimate of the small size of the "sweet spot" on this solenoid is incorrect.

I used your estimate.

This solenoid is approximately as long as its diameter, not half that as with the Helmholtz coil pair. It is true that the H. coil is still better, but not by that much. The field of the H. coil is down about six percent at the location of the coils. The solenoid, with the geometry I have used, is about ten percent down at the same distance (not at the end of the coil). This is not that big a difference. It is also true that the H coil field is flatter near the center, in the sense that more derivatives are zero. This is just not that important in this application.

Ten percent? That's 20*Log10[0.1/2.1]= -26.4 dBc, which will interfere with clean measurements of pickups with hum rejections of 20 to 30 dBc..

I think the method you suggest for making the coils is good for when you need quite a bit of wire and it needs to be pretty big to carry significant current. In this application, we need only about 50 turns on each coil, and the wire can be small. I will make a Helmholtz pair next with #30 magnet wire on a somewhat larger bucket. I plan to define the locations of the coils with a number of layers of tape, and then just wind them on the bucket.

OK. That works too.

No, the effect is not too subtle to measure with this setup. I would prefer a more accurate one, which is the reason for a new bucket, but I think I did measure it.

You were seeing a 2 dB effect at most.

An old lock in amp is better than an FFT machine? No way. The limitation in the process is in the analog multiplication. Look what happens in a digital receiver where you use digital baseband mixing. The specs are far better because the digital multiplication is superior. In any case, this is just not a very demanding application. You just need to sample with a lot of bits so that you preserve the dynamic range and then filter to reject the hum and keep the signal. The FFT does that just fine.

Umm. Lock-in amps can pull signals 100 dB down; FFTs cannot do this, because there is too much spectral leakage. A FFT plus a FIR or IIR prefilter can dig that deep. Lock-in amps are far simpler. But suit yourself.

Did Bateman use lock in amplifiers in his excellent work on capacitor distortion? No, he did it digitally, and for that very demanding application, he made sure he had really good digital hardware and software.

Yes, but what's the point? Bateman was not trying to pull a signal out of massive interference.