Mike Sulzer

Joe's right that ground loops are really not an issue in guitars. I have tried just the experiment he mentioned. However, it is possible that they could be with really high electric field interference. The basic idea of ground loops is this: metal parts in the guitar, pot podies, shields, etc., have small current flowing in them due to electric fields things using elecrical power. This is 60 Hz and harmonics. (In fact, shielding works because such currents are excited in the shield.) Small voltages develop across the metal parts, and these voltage are a potential source of hum and buzz. Careless wiring with multiple grounds can cause these voltages to appear in series with the pickup signal and so be potentially audible. Single point grounding is intended to prevent this from happening. It does not matter whether the amplifying device is in the guitar or the amplifier; what matters is how big the voltages are and how much they are eventually amplified. However, this is much more of a problem in amplfiers because they have a power supply: bigger currents flowing in the chassis making bigger voltages than in a guitar which is not so close to a power supply. In any case, ideally, the single point ground of the amp should include the guitar shields. That is, there should be a ground path to the guitar connected to the amp chassis (using the shield of the cable). The slider of the volume pot and the "low" side should both go on wires (inside the shield) to the input stage of the amp. In the real world, a single point ground inside the guitar just does not make that much difference, bit it costs little to do the wiring right just to be sure, and to establish good wiring habits. A ground loop is not possible in a passive guitar. Don't worry about it. It's an internet myth that comes up whenever someone grounds something twice. A ground loop is more than multiple grounds. FWIW, I never ground pots in a tele, other than the output jack ground connection to the volume control. if so, then how do they add noise? A ground loop (according to its real electrical definition) requires that two ground points are operating at some difference in potential (voltage), AND that there is some signal voltage (noise) developed across those two ground points, AND that there is some active device (tube/transistor) that can detect and amplify that signal voltage developed between the ground points. All three conditions have to exist. Despite the name, the definition of "ground loop" simply doesn't apply to multiple paths to ground from the same point in the circuit. There is absolutely no way for multiple paths to ground from the same point in the circuit to cause noise. You can do an experiment to prove it. Take your guitar (a tele works best), plug it in. Now take a wire (a test lead with alligator clips is easier). Connect one end to the bridge (which is grounded) and the other end to the control plate (which is also grounded). You have just intentionally created an internet wisdom ground loop. There will be no difference in noise regardless if the wire is connected or disconnected between the two grounds.

I have built a guitar with a bar between the bridge and tailpiece. This bar pulls down on the strings, and so you can adjust the break angle of the strings over the bridge. A very shallow angle sounds bad, but once the angle gets to some reasonable size, there is no further improvement. To me, this indicates that although string through is a really good way to get sustain, there is no reason why a tail piece cannot be just as good as long as it achieves a good angle over the bridge. I do have my doubts about some combination bridge/tailpiece designs. Such a thing needs to be very strong and stiff.

On the raising the heater voltage: Joe is right about the effect being different from just estalishing a solid reference. However, it is best to use the 100 ohm resistors and also connect their "center point" to an elevated voltage. I have measured both effects separately and together. In the real world I would agree that you do not really need to do both, unless you have extremely low hum pickups.

Power supply: You mean the point labeled 35V? Yes that is to bias the heaters up. It is well worth doing because it reduces the hum significantly. It matters most for the first stage in the preamp.

The cable capacitance is bigger than the pickup capacitance, so presumably different length cables have more of an effect than scatter winding. Also, hearing a difference and hearing a difference that really matters are two very different things. OK, I agree about the tension. About scatter wounding: The supposed effect is that a coil that is wound so that every next turn of wire is placed just next to the one before will have a different capacitance than a coil that is scatter wound. The difference in capacitance will of cause change the way a guitar circuit behave. So I was a little too hard on the scatter wound phenomenon. Of cause there will be players that hear the difference. There are even players that hear the difference (in blind tests) between the battery brands in their pedals! But Mr Joe Average will not hear the difference. Scatter winding maybe make up for at most (very personal opinion) like 5% of the sound of the pickup. But scatter wound is mostly used as one big marketing hype that doesn’t come close to reflecting the impact it has on the sound. But what do I know. As I hand guide (very few guys do actual hand WINDING) the wire on my pickups, they are all scatter wound so I cannot make real comparisons.

But it is already shaped. Easiest thing to do is start over. (If this is a two piece body you could resplit it lengthwise, then find a really big saw slice the resulting pieces lengthwise the other way to make top and back pieces. Then it is a matter of making a middle out of a different kind of wood and regluing. This would not really be worth doing, but if you did, you would have a story you could tell for the rest of your life!)

Yea, and the problem looks worst at the nut end. So replace the nut. If I wanted to find out if i would be happy with a new nut, I would tape off near the nut and fill the #1 e slot with a very hard epoxy mixed with silica power to prevent it from flowing. Then I would carefully cut a new slot displaced a bit. This would be a temporary solution that would neither last nor sound so great., but you could see if the guitar would play better. Do not try this unless you know what you are doing.

Everything that goes to the case of the pot would move to the sleeve of the jack. The wire from the pot to the tip of the jack would be removed, and the black wire from the pickup would be moved to the tip of jack. If I said all that right, then there is nothing left on the pot and you can toss it out.

This is a pretty standard long tail phase splitter using a pair of triodes connectetd as a kind of differential amplifier. Signal goes in the U2/1, and feedback goes into U2/2. The two outputs drive the power tubes with opposite polarity. Most people would think of this as part of the power amp.

Thanks, I will check out the meter. For analysis, I use Signalscope and Signalsuite on a mac. Not free, but inexpensive. The guitar volume pot can cause a big change in the frequency response. The cable capcitance is part of the pickup resonant circuit, so as you turn the volume control down, the resonant frequency changes and the level of the high frequencies reaching the other end of the cable changes due to the series resistance. The gain should be changed with the first gain pot on the amplifier; the gain stages before this control need to be linear, that is, not overloaded. You can use a speaker, or just wrap some wire around a steel core. I prefer to use white noise as a signal source and use the scope software in its spectrometer mode, averaging enough times to get a good response.

My experiments have been different from yours. I use a small Neodymium magnet on each pole piece. It is easy to vary the field strength by varying the size of the magnet (, although I have never measured the field as you have done*). I do not hear any significant difference in the sound of the pickup when the magnet size is changed until the onset of string pull is reached. You always need to listen at the same volume since a louder sound is clearer and more present even if otherwise no different. It is possible that I have missed differences because of not really listening carefully enough. That is, perhaps through too much reliance on the theory: increasing the degree of magetization of the string should change the volume, but nothing else as long as the vibration of the string is not affected by the magnets and the pole pieces are not saturated by strong er magnets. *How do you measure the field?

When you chopped up that Gibson alnico magnet, you observed exactly what the measured properties and theory of magnets predict: alnico magnets get weak quickly as you make them short in the direction of the field (from pole to pole). Ceramic magnets do no lose strength as fast. That is, while a long Alnico magnet might be stronger than a ceramic of the same shape, a short one will not be. This has to do with the B-H, or hysteresis loops, but we probably do not want to get into that here. I have to disagree about the relatiionship between magnetic field strength and high frequencies. I think that there is no relationship at all unless the field strength is quite strong. This at the onset of "string pull" or "stratitis", and I think the sound is better described as harsh rather than just just more high frequencies. It is, after, a non-linear affect. I disagree. It is not the material that the magnets are made of that produce the difference in sound. It is the magnetic strength. The stronger magnets the more treble. Most Gibson style pups use ALNICO5 (A5). The 5 indicates the magnetic strength. A higher number means stronger magnets. A2 up to A8 have been used in pickup making. Ceramics are stronger than A5 , similar to A6 and way weaker than A8. An A8 humbucker will have considerable more treble than an A5 AND a ceramic magnet. The boosted treble is the thing that fools some people to think that ceramic magnets are harsher than ALNICO. That is only true in comparison with A5 (OK, definitely the most used ALNICO).

Check the wiring carefully, too. Coud wires be broken or reversed at the output jack or pots?

Remember also that you should not expect too much from shielding. It takes care of the electric fields, but magnetic fields must be canceled by humbucker, dummy coil, etc. But the kind of hum that is reduced when you touch the strings is electric, and so you should be able to reduce it with shielding. Then you will turn up the gain some more and find out what is left.

"Sound vibrations travelling 5X's faster in one direction than another? I'm not literate on that topic, but I'd be interested for the statement to sight references so I can investigate further." The easiest way to understand this is to find a piece of spruce, or whatever, cut for the top of acoustic guitar. Note that it bends much more easily across the grain than along the grain. The degree of stiffnes and the density enter into the equation for the speed of sound in the medium in a given direction direction. Stiffer, that is, harder to bend, means a higher sound speed. I am a bit concerned about using the grain cross ways. It would depend on the type of body you are making. The critical area is the neck joint and hole for the neck pickup. Some designs do not have a lot of extra material in that area and weakening it might not be good.

"Hum is not present at high frequencies." Joe, most interference is not the fundamental, but harmonics, which extend up into the KHz range. Magnetic fields containing such harmonics can be produced when electrical currents are chopped by light dimmers or from circuitry in CRT monitors, etc. "Wow. Where do you get this stuff from? It is so far from reality it is not even worth disputing!!" AM radios do not use ferrite core antennas? They sure do. I am not the only one to have done the test that described above that shows how the hum cancelation descreases as you remove the screws.

The pole pieces in a humbucker have two purposes: 1. Magnetize the strings. (The magnet magnetizes the poles pieces; the pole pieces magnetize the strings. 2. Amplify external magnetic fields (from the vibrating strings, or external interference fields). It is the second purpose that is in question here. Nearly all magnetic sensors use a ferromagnetic core to amplify the field being sensed. For example, nearly all AM radios sense the magnetic part of radio wave using a coil with a ferrite core. A ferrite that functions at frequencies up to about 1.6 MHz, or higher, is used. Without the ferrite core, it would not work so well, and one would probably need to use a long wire to sense the electric field from the wave. The short steel cores used in a guitar pickup do not have a huge amplifyig effect, but it is there. This is easy to demonstate. Connect a humbucker to an amplifier. Make sure that it is well-shielded against electrical fields. Now remove the screws one at a time. If there are interfering magnetic fields present, the hum increases some as each screw is removed.

Joe, you obviously have never tried this. Do you think the original PAFs from the 50s were "computer wound"? Nobody had a computer outside of a (large) lab then. The capacitance between the windings reduces the output at high frequencies. If two coils in series have different capacitances the output voltages of hum at the higher frequencies are different and will not cancel so well. A minor but real effect. What is the sarcasm for? There is nothing bad about being wrong, but you are making it hard for you to ever admit it. Greg, the longer thinner screw matches with the thicker shorter slug pretty well, but I think you are right about this being the limit. Humbuckers with coils with identical poles can be better.

I agree you should try shielding. In an unshielded guitar, touching grounded strings can make you the shield (but an imperfect one) and thus reduce the hum.

Hand wound pickups can cancel hum very well. It is only necessary to use a counter to get the number of turns the same on the two coils. Fralin sells a model called the Unbucker with more wound on the screw coil. Is this what you have? If so, you cannot expect it to cancel hum as well. (Although it is possbile to design pole pieces with different sensitivities in order to compensate for the turns, but I do not think the Unbucker does.) It would be posible to wind and connect a pair of Unbuckers for good cancelation with both on, but this is the first I have have heard that it might actually work that way. Very interesting information - thanks! I was not aware of problems such as these in hand-wound pickups. I guess that explains why copper etc. didn't really help with the noise. Do you have any idea how off the winds would have to be to begin picking up noise in this fashion? I might touch base with Fralin to see if this is a common occurrence. thanks, roah

Since EMG pickups use lower value pots than passive pickups, shielding the cavity is less important. Nothing wrong with doing so; the shielding cannot hurt, and might help a little in some really bad electrical environment.

Look at AX84.com

Have you tried taking one of those pickups you like and holding it over the strings of an electric to see if it gives the sound you want? If not, you might be hoping for a sound you cannot get with a solid body electric.

When a shield blocks an electric field from affecting you circuit, currents must flow into and out of the shield. The idea of star grounding is to prevent these currents from getting mixed up with the signal currents and effectively puttting a small hum signal in series with the guitar signal. Star grounding is less important in a guiitar than an amplifier where you have much larger currents from power supplies, etc., but it is still the proper way to do it. For star grounding to be really effective, the separate shielding connecton should continue all the way to the amplifier rather than coming together with the signal at the jack, but it is hard to fight tradition.