For EE's: diff btw 'magnetic' noise vs 'electric field' noise?! (pickups, shielding)

[Usable Thought]

It doesn't seem to me to be a question of either/or, mains hum is both (i.e. has both electromagnetic and electrostatic components).

This is where things seem to get interesting.

To start with - and this is a very minor point - when you say "electromagnetic" that is not the same thing as saying "magnetic." It may seem that "electrostatic" and "electromagnetic" are paired terms meaning opposite things, but unfortunately the language isn't so neat as that. Electromagnetic refers to anything to do with electromagnetism, which includes both electric and magnetic phenomena. Electrostatic, on the other hand, has to do with the behavior of charge that is not moving. That may seem nitpicking but I do think we need to get our terminology right - or closer to right (I find it difficult).

Beyond that, when you say that it's not "either or" but "both" - well, that sounds like common sense and as if it should be true; but common sense is inadequate for describing this stuff. For example what kinds of forces we experience can depend, in part, on how near or far we are from a phenomenon such as a noisy transformer or power line.

With AC power, the changing current causes magnetic & electric fields to arise. A transformer makes use of this, as we know. It happens that when you're near enough to EM sources, both the electric and magnetic fields can behave almost as if they were separate & thus have powerful effects on their own. For example with a transformer, when you're close to it, apparently the magnetic fields can be more dominant than the electric fields. However once you get far enough away, at a certain point the fields start behaving as EM radiation - which is very different. These different regions are called the "near and far fields." I only just heard about this topic a few days ago & don't pretend to understand it; but in some branches of technology, e.g. radio, it is important. So radio buffs, for example, might know a lot about it because it affects the behavior of broadcast antennas - an antenna can have a near field and a far field and different things happen in these different fields. And near & far EMI is also important in shielding circuits that are highly sensitive to both. See for example this highly technical article about shielding such circuits.

To what extent the near vs. far fields really matter with guitar pickups & wiring, I don't know. Obviously it's not discussed much - but if the near vs. far distinction for EM noise is important for other kinds of technology, there's a good chance it can be extrapolated to electric guitar technology as well.

For example, take the electrostatic ("electric") fields that are discussed in the article I linked to about mains hum: the capacitive coupling discussed in that article would be considered a "near field" phenomenon. Again, this is as opposed to the far field where pretty much all noise is in the form of EM radiation, whether light, radio, or what have you. Thus when we talk about RFI interference from a cell phone or radio getting into a guitar or amp circuit, that is happening in the far field.

Effective EM noise solutions for guitars & pickups have already been found - there are already many products available. So there is not a lot of incentive to really delve into the detail I'm interested in. The reason I got curious is that there is so often disagreement on forums like this about even very simple issues involving shielding and pickups. And people get very vehement but also very confused. I'm in the "confused" camp myself.

So I am just curious enough to want to know a little bit more - and in scientific terms, too, to the extent that I can understand them.

[dai h.]

I'm not much of a science person, but (at least on the surface) it would seem that "electromagnetic" relies on electricity to be magnetic, so still "magnetic" but conditional on the electricity flow.

re: near & far fields, not sure what that is but distance is one of the basic techniques to deal with noise (i.e. move farther away to get away from noise). (Other basic techniques, relative orientation(common example: standing in front of an amp with noisy single coils and moving the guitar around to find the orientation with the least noise) and shielding.)

As far as the points of confusion, perhaps you could list the ones you have found to be clearer?

[Mick Bailey]

Interestingly, the paper on EMI shielding makes no mention of electrostatic effects.

[Usable Thought]

Interestingly, the paper on EMI shielding makes no mention of electrostatic effects.

Are you sure? And would that really be the interesting part? Perhaps I used the word "electrostatics" incorrectly in bringing it up; or perhaps it was used incorrectly in the mains hum article - I'm not the one to say. But there seems to be a lot of common ground shared by the hum article & the shielding article; and that is what I would suggest is interesting.

For example, here is a bit from the shielding article I noticed that is familiar to me from other books & articles I've looked at re noise, including some I can't whose titles I can't find right now:

The EMI coupling paths are actually very few in terms of basic number. Three very general paths are by:
1. Interference due to conduction (common-impedance)
2. Interference due to capacitive or inductive coupling (near-field interference)
3. Electromagnetic radiation (far-field interference)

E.g. some of that shows up in material on grounding and EMI by Duff - for example this: Designing Electronic Systems for EMC: Grounding for the Control of EMI.

Given that the article on mains hum refers to capacitive coupling as electrostatic in nature rather than magnetic; and that the article on shielding from EMI also talks about capacitive coupling, I think that regardless of who said "electrostatics" and who didn't, they are talking about the exact same phenomenon - which involves the propagation of noise through electric fields rather than magnetic fields.

And when the shielding article says this . . .

2. Interference due to capacitive or inductive coupling (near-field interference)

. . . the author is clearly drawing the distinction between electrical (capacitive) and inductive (magnetic) coupling . . . both of which are near-field. So yeah, that's the sort of thing the hum guy was talking about too - similar distinctions being drawn.

I'll get back to my reading & tinkering now. Some of my reading involves Michael Faraday. Now he was the man.

[Usable Thought]

More for them as care about "electrostatics" - possibly I did misdefine it, although you will find similar definitions in many places. For example Wikipedia starts off by saying

Electrostatics is a branch of physics that deals with the phenomena and properties of stationary or slow-moving electric charges.

which is more or less what I said. But William J. Beatty, an EE who likes to present electricity from a physics point of view, vehemently disagrees; and although he's a little crazy I trust him more than I trust Wikipedia on something like this. He has a wonderful rant on electrostatics from a page on "high voltage misconceptions"; here is part of what he says:

Electrostatics is similar to Hydrostatics: it is the study of electric forces and the Electric Charge which creates those forces. It is the study of imbalanced charges in matter, and of voltage and electric fields. Notice that I didn't say anything about "charges at rest." This is intentional, since electrostatic forces don't go away when the charges start flowing in a current. And while electric voltage falls under the heading "Electrostatics," electric voltage is intimately involved with flowing charges. (Analogy: water pressure is intimately involved with water flow.)

Beatty's entire rant, indeed his entire page on misconceptions, is worth reading. And it makes me think the author of the hum article was correct in using the term as he did. But I'll leave it at that.

At this point I'm off this thread. I may read more on my own about such topics & perhaps do some experimenting with guitar pickups. If I find out anything cool I'll post about it; if I don't I won't.

[nickb]

<snip>

Does this mean that if you've got a single coil pickup, and it's only mains hum that you want to ward off, you can get by with some system of shielding and grounding? I can think of quite a few objections:

• A single coil pickup is far more sensitive to magnetic fields than a single 15cm loop of wire, thus his test was not adequate for our purposes.
• Grounding by itself won't work. Right??
• A complete Faraday cage is impossible for a guitar, and a partial cage won't work. So shielding the pickup is out.
• There's no point to only blocking mains hum! Because you'll always have amp transformers nearby contributing magnetic noise.

It would be interesting to do some independent and fairly rigorous tests involving pickups. All you need to start is a pickup - no wiring harness - and a way to isolate it from all but the noise sources you choose to introduce. For example, get the single coil pickup & plug it into a transformerless chipamp run by battery. Take it onto a lawn or other area free of nearby noise sources. Now use a long extension cord to introduce whatever noise source you like. One thing I'd be curious about - does shielding & grounding the wiring harness alone (or basically not having a wiring harness) work by itself for mains hum? After all then the pickup would be grounded too. Lots of "No's!" would be raised to this idea I'm sure. But you could test & find out - w/ no transformers nearby.

Call me cynical, but it has to be pointed that the article was essentially a sales pitch for their Faraday cages so the conclusion that such the noise was from the E field could almost be predicted form just that information alone. Also, bio-electric info is low <50Hz frequency so he would not me interested in the audio band.

A couple more points. In the near field, the ratio of electric to magnetic field is determined by the radiator. In the far field it's the electrical properties of free space that determine the ratio. Therefore the nature of the radiator is very important to us.

Thinking of the amplifier with a traditional linear power supply. The power supply is the biggest source of noise due to the high varying currents. You have the power transformer throwing off primarily magnetic fields. This magnetic field, due the the non-linearity of the core, is very distorted and so contains lots of harmonics. The rectifier circuits too carry even more distorted currents and will radiate both magnetic and electric fields, the ratio I would expect to depend on the design and layout of the circuit. Then you have all the mains wiring. The wires run mostly in pairs that are quite close so, at any significant distance away, the magnetic fields will almost cancel. On the other hand, the wires look rather like a big antenna so I'd expect that the electric field is the dominant effect. The point is the both fields exist and the frequencies will cover the full audio range.

I think it is generally accepted, or perhaps more accurately, I generally accept that 'hum' is low frequency i.e. mains or 2X mains frequency. This is mainly from the transformer field. 'Buzz' is all the higher frequency stuff, the kind of noise that comes from the spikey currents in the rectifiers. Perhaps we can just agree this as a definition, for now.

Hold your single coil PUP anywhere near the amp and you'll hear hum and it's coming from the power transformer magnetic field. You don't need to be that close, a few feet is enough. Move the guitar around and you'll find a place where it lessens. This is magnetic coupling into the PUPs. They are very directional. The field around the transformer is mainly power line frequency so you hear hum

Move well away from any transformer, don't touch the strings and you will likely hear buzz. Your body is one plate of a capacitor that the electric field acts on. Touch the strings and the buzz will be reduced. The current is now flowing through your body to ground and so no is longer picked up by imperfectly shielded wiring. Why HF buzz and not hum? The small capacitance forms a high pass filter so you attenuate all the LF.

Coming to your points.

Yes, a single coil is more sensitive to magnetic fields than electric(-dynamic). Note, not electrostatic fields. We don't care about them. A static field will not induce noise. So as you say his model was not a good one for this situation.

Grounding. Certainly grounding helps. Just as touching strings helps (grounding yourself) so does grounding metal parts in the guitar. The ungrounded parts radiate the E field to the wiring.

Faraday cages, for the E field, only need to be complete in respect of the frequency you are trying to attenuate. Holes still attenuate the noise. The rule of thumb is that the hole should be no bigger than 1/10th of the wavelength. Taking 20KHz, 1/10th of the wavelength is 1.5km! The problem is the circuits are very sensitive. Also, the buzz is small compared to the string signal but very annoying. Therefore you need a lot of attenuation. I don't see why a thin piece of non-ferromagnetic material cannot be placed over the the pickup - it would affect the magnetics hardly at all. The cable shield continues the cage out of the body. The biggest thing would be to ensure the scratch plates make good contact all the way around with the body shielding but you can get flexible conductive gaskets for that. Therefore, I think you should be able to do a pretty decent job of shielding it. A complicating factor might be the non-linearities of the amplifier can make the harder to screen much higher frequencies that are far out of the audio band audible. A well designed amp should be immune to such effects.

"There's no point to only blocking mains hum.". I thought that was the point! Perhaps you meant: There's no point in only blocking E fields.

One last point is that a notch filter could be designed into amplifiers to null out the power line frequency. As far as I know this has never been been done in guitar amps. I guess PUP hum, as opposed to buzz, is not seen as a problem. A comb filter could be used to suppress the power line and all it's harmonics to eliminate buzz. With a high enough Q it needn't have too much effect on the audio, but I can already feel the purist's hair rising on the back of their necks

To be honest, I wasn't aware it was that big a problem. I don't see a crowd of angry guitarist's waving their fists in exasperation.


I guess it just goes to show that Faraday was right, we should have had DC mains after all.

[Fluoroscope 5000]

I'm not sure how you can improve the SNR of electric guitar circuitry beyond the standard methods we've all gone over. Shielding SC PUP's can reduce hum, but can compromise the sound. Orienting more of the coil wire closer to the strings and optimizing the magnetic circuit will make a difference. For that matter, thinner wire is more efficient, as more of the total coil can be closer to the strongest magnetized portion of the strings. Never tried a Tele with the bridge cover plate on. I don't imagine a cover that thin would reduce EM hum much at all, but may significantly reduce the upper-harmonics. I think RFI protection in PUP & control cavities is dubious – perhaps more precautionary. Either way, I'd include this article in your research:
Fender Telecaster® Electric Guitar Central -- No. 1 in the World

[Usable Thought]

To be honest, I wasn't aware it was that big a problem. I don't see a crowd of angry guitarist's waving their fists in exasperation.

It's not a big problem! I thought I made that pretty clear. But I admit that confuses the usual mindset brought to a forum, where generally posts are about problems to be solved, or if not that then silly things.

It's just I'm one of those folks who likes to play around with topics he knows nothing about & learn a little something. I do have a small practical interest in shielding my guitars; and lately ran into a real puzzler where someone who knows what he's doing had shielded his guitar, gotten noiseless pickups, and was still getting noise, which seemingly contradicts the usual wisdom.

But mostly I just like learning weird stuff & if I can recreating simple experiments - for example, I have a very impractical wish someday to recreate what Coloumb was doing, but probably not what Cavendish was doing, thank you very much. On another forum I have someone who tells me I should go back to engineering school to learn about electronics because I'll never pick up the math otherwise. But I'm not sure I have the stamina, time, or money to do that. Maybe in another life.

[Fluoroscope 5000]

Not sure if you got my last post, but it's worth checking. Someone in another forum discovered that an ungrounded metal pickup bracket caused some hum with an HB. The same might possibly happen with ungrounded shielding, depending on proximity to the pickup and/or leads or lugs. It could also short if it comes loose and touches a pickup lug. Unlikely, but a reason I prefer to avoid shielding PUP cavities, but for Al tape underneath to reduce any potential rheostat buzz. Wiring the input jack backwards will cause noise. Sometimes it's just a bad joint or broken wire. Probably best to use new wire and check each piece before soldering. I mistakenly bought some solid core wire – not good for guitar work.

You could test the affect of Cu tape by taking noise measurements of a spare PUP before and after encasing it in Cu tape. Maybe wrap it around a cup without pulling the backing off. I did it with both Al pie tins and tape. The main difference was that with both Al shielding options, harmonics in the ~6kHz range went up, while the third harmonic went down. Here are the results (dots above the lines are the averaged levels):
https://drive.google.com/open?id=0Bx...lRTLTQwam5FS2s

[Usable Thought]

You could test the affect of Cu tape by taking noise measurements of a spare PUP before and after encasing it in Cu tape. Maybe wrap it around a cup without pulling the backing off. I did it with both Al pie tins and tape. The main difference was that with both Al shielding options, harmonics in the ~6kHz range went up, while the third harmonic went down. Here are the results (dots above the lines are the averaged levels): https://drive.google.com/open?id=0Bx...lRTLTQwam5FS2s

Interesting tests, thanks. Curious if it sounds better or worse when shielded? Or just different?

Also, were the spectrograms done with a mic at the speaker, or via signal tracer on a preamp node? Or possibly right from the pickup? Although I don't know if the voltage from the pickup is high enough for that? I use a USB sound card with a signal tracer for that kind of thing. I might try this kind of test myself - I have a cheap single coil coming my way in the mail shortly.

As for the puzzling noise issue that I mentioned, this is someone quite far away from me so I don't have the means to investigate myself; but I trust him when he says he took great care with the shielding; he is quite knowledgeable. Of course it is always possible that a mistake can be made. The other possibility is that some environments may be too noisy even for good shielding - a practice studio right by an electric train line, for example.

[Fluoroscope 5000]

I connected the pickup right to the Hi-Z input of my Presonus Firebox with the input level at nominal: 12 o'clock. ~125k resistance might roll off that 5~6kHz noise a bit, but it's ~10dB below the hum as is. I'd be curious to see if Cu tape affects the hum harmonics at all. If not, Al tape under the PUP and the pickguard might be worth it. I don't think it's needed on the PUP cavity sides, and may roll off the highs a tad if applied there. The Al tape is thin enough that it won't really affect the sound under the pickguard, but does soften the upper-mids at 0.015~0.03”. You could do before and after recordings with a few different layers of Al tape, masking taped to the pickguard around your new SC pickup. I just use the GFS Al pickguard shields. My Wilde Microcoils don't really need it, because the bandwidth Q is very broad (~1.5 vs ~3 for Fender type SC's), and the fundamental harmonics are strong, and can be adjusted via the screw poles. One GFS Al guard sounds great, giving each pickup a touch of a quackiness. It depends on the guitar, strings & pick, and the amount of aggressive bite you want. I use Ni/Fe alloy strings, which sound “sweeter” than most Steel alloys.

[Mick Bailey]

Has anyone got an aluminium bodied guitar - Talbo, Valeno, or one of the milled-from-solid Teles? Interesting to know how this affects hum and tonality. From the recordings I've heard (PIL and Pete Haycock) they don't seem to suffer from any adverse tonal effects, but I don't know how the noise level is affected.

Also, from a magnetic perspective, how do the Trussart steel guitars fare?