Pickup theory question

[David King]

Different pickups will behave differently in FunkyKikuchiyo's guitar model. They might all be easily categorized by output on the lathe test bed but I bet the order changes as soon as you start strapping them to real guitars. I like the 80-20 (30-30) aluminum now because it's going to sound a little more like a piece of wood, but one that anyone anywhere can order up and replicate exactly. To that end I'd skip the dampening fill because that's going to be much harder to control from one person's test to the next.

[ScottA]

A humbucker reacts differently to the entire core equation because there will be slugs, screws and Keepers that dominate the inductance factor. Additionally, and most importantly the permeability of the bar magnet has a large affect on the coupling of the two coils (humbucker)and less of an effect on the inductance factor, which can be observed by calculating the mutual inductance. All this can be derived by measuring inductance.

The effect of the bar magnets permeability can be seen in spectrographs as affecting the bass region to a great extent (supporting David Schwabs observations), which I assume to be from coupling.

That's really interesting, and perhaps the most viable explanation I've seen for the effect of magnet materials in humbuckers.

Analogously, I've measured very significant and reproducible differences in frequency response (and consequently very little - if any - effect on AC electrical measurements) with pole pieces 1) daisy-chained together and grounded, 2) daisy chained but not grounded, and 3) isolated, with the resonant frequency going up in that order (highest when isolated).

This would imply that an (the primary?) effect of magnet materials in humbuckers could be the effect on the efficiency of the mutual inductive coupling of the pole pieces, probably both within and between coils.

This also has implications on the blade connection question that was being bandied about.

[Rick Turner]

Funky, I'm looking to minimize "noise" and come up with a way of looking at pickups...literally looking at their output...as free of such wildly variable and uncontrollable influences as neck and body resonances and coil resonances. I want a reproducible and replicable test bed that allows for a wide variety of tests in which only one factor is changed at a time. I want to be able to see different aspects of what we seem to hear isolated from interfering factors.

I am not satisfied by seeing frequency response plots of pickup coils. I recognize that the superimposition of coil filer characteristics is an important element of guitar pickup tone, but it's not the only thing going on, and clearly even among this group of pickup mavens, there is no agreement on what's going on. Some folks think it's output level differences making the tonal differences; some think it's the inductance changes. I'm trying to level the playing field to see if it's something else.

So the first thing you do is eliminate the variable of input into the pickup. For that you need a reproducible pluck or inductively coupled input. I'm going for pluck of a given strength with the pickups being a given distance from the string. That will give a true and accurate pickup sensitivity rating.

If you want to test for tonal differences among magnets, the second thing you need to do is to have the signal going to your test gear be at the same level no matter what magnets you choose to use. So with a known pluck strength you can then set a reference output level on a pre-preamplifier...one with a lot of headroom and variable gain. Now you're testing for tone independent of output voltage because that is the same every time no matter what the magnet is.

Etc., etc. You can then change one factor at a time whether it's string length, pickup position, pickup aperture, pluck position, afterlength of the string behind the "nut" and bridge, the effect of break angle on string motion, ad infinitum. It's straightforward, albeit time consuming, basic research. And if anyone has really done this, they're not publishing yet.

And there's nothing inherent here that prevents one from testing standard high impedance pickups in a rig like this. I simply think that in order to understand this magnet/coil/string thing, we have to get rid of the electrical filter factor being in the audio band. It's masking what we're hearing too much. I think we need to hear the magnet and string without hearing the effect of the coil. I've experienced this, and it's pretty interesting.

[Mike Sulzer]

The effect of the bar magnets permeability can be seen in spectrographs as affecting the bass region to a great extent (supporting David Schwabs observations), which I assume to be from coupling.

CRU

That is very interesting. Where can one see these spectrographs?

[FunkyKikuchiyo]

Rick - if I understand you correctly, the point of the lathe beds was to be a giant inertia block then, right? My initial suggestion was simply to use something other than a string, so you don't need to worry about pulling it to tension or using those things as an inertia block, and since a pickup wouldn't need to be under the strings, you could mount it in a simpler fashion. I see now that you're going in a different direction. Thanks for the explanation... certainly was better than just insulting me!

I do like your idea of starting with nothing and slowly adding in elements to observe the change in sound. I didn't realize you had ambitions beyond measuring power.

How would you have a known pluck strength? A robotic arm would obviously work, but I don't think any human could reproduce the exact same pluck again and again, or be communicated to another player empirically. Heck, I don't think I have any two guitar picks that sound the same, except maybe the delrin ones straight out of the package.

All that said, I have to disagree with the real scientific value of this. It won't work on all pickups... I make quantum pickups, so the tone will change when you're measuring the output. Sorry guys.

[Joe Gwinn]

How would you have a known pluck strength? A robotic arm would obviously work, but I don't think any human could reproduce the exact same pluck again and again, or be communicated to another player empirically.

Actually, this problem was solved by the physicists a century ago: Pull the string a measured distance to one side using a long cotton thread. To start the vibration, ignite the thread with a match.

[FunkyKikuchiyo]

Brilliant! I'm SO going to relearn Stairway to Heaven that way.

[Rick Turner]

Funky, if you'd go back and read what I've written, you'd see that the inertia block aspect of this is really rather secondary. No, I don't want a particularly resonant test bed, nor to I care too much about sustain. I want something rather neutral, and a bit of sustain should be good for making the kinds of measurements I want. I also want incredible versatility. It's not that much harder to make a universal string/pickup test bed than to just make something for pickups, and features can be added one by one. For instance, the after-length and break angle test fixture doesn't have to be part of the initial pickup test rig.

I'm mostly concerned with pluck repeatability, establishing a few constant distances of pickup top to string, and having total control of the location along the string for both the pickup and the pluck drive. I also want to put piezo pickups at each end optimized for pressure mode vibrations and put the two pickups at 90 degrees to one another to study string phase as it relates to energy into each end of the string and as it relates to magnetic pickups.

This rig can also become a string testing fixture. With the modularity, it could work for piano strings and using hammers, harpsichord strings and use a traditional jack, or even for bowed strings and for that it could have either a mechanical bow driver or a hurdy-gurdy wheel. All that stuff can be designed in the future and easily added if the basic test bed is versatile enough. The 80/20 aluminum beam approach allows for any and all of that and makes changing out fixtures and locations very simple.

Re. the breaking thread...actually doing it with magnet wire is more repeatable. We did that in some of the recordings we did for Mama Bear. Magnet wire is incredibly consistent and will break at a given tension every time.

[JGundry]

If you talk directly to a magnetics house in the USA they will usually ask you for your magnetic circuit so they can determine the energy point of the circuit. They will custom make a magnet for your circuit if you want. I never did this and frankly don't know how it would translate to what we hear in pickups, but my guess is that many of these questions are easily answered by a magnetics engineer that has made magnets for guitar pickups using the energy point as a reference.

Personally I think inductance and coercive force are the two most important factors with magnets when it comes to guitar pickups. In the most dead simple terms the coercive force determines the feel of the pickup in terms of attack and dynamics and focus. And the inductance is a better indicator of the linear tonal response of the pickup. As an example A2 has a low coercive force and has much the same attack and feel as unoriented A5 which also has a low coercive force. But the differences in inductance between the two give the unoriented A5 magnets more mids.

Much of the tone of a picked guitar string is in the first transient moment after it is picked. I think a magnet with a higher coercive force gives a much more focused and restricted quality to the notes transition from being plucked to free vibration. Lower coercive force gives a looser tone and feel in this time frame.

[Rick Turner]

Here's a little bit of homework for those of you who wind humbuckers.

Wind up a couple of pickups with about 1,000 turns of #38 wire. You'll have a low impedance pickup with a self resonance way out of the way. Now try switching magnets from Al 2 to Al 5 to ceramic to whatever else you want to try. Attempt to adjust the gain of your test amp so each magnet change is volume compensated. Adjust the pickup so it's the same distance from the strings for each change.

Now tell me if you hear a difference due to the magnet differences.

That is, in essence, what I want to do with the test rig, but it will be easier to closely monitor, and it will be more convenient to change any variable, and it will be hooked up to a computer with FFT analysis that can read the string at each end and compare that to the mag pickup. I'll also be listening. What a concept...listening!

BTW, we shouldn't really talk about the inductance of a magnet; the magnet affects the inductance of the coil. It's a secondary effect, and I'm hoping to see/hear past that effect.

You have to know the sonic effect of each individual part of a pickup system to really understand how to design them beyond by guess and by gosh...which, indeed, is how some of the iconic pickups were designed. They were serendipitous accidents from a tone point of view. You want one of the greatest accidents in pickup history? The Tiesco Del Rey six porthole with foil with six almost adjustable polepieces single coil. Looks like it should be a humbucker. It's not. It's the cheapest piece of shit you can do given that you've got a punch press on site. It doesn't even use a bobbin for the coil, and the ceramic magnet is absolutely Jurassic. It sounds like one of God's pickups, and in series RWRP with one of my old double horseshoe pickups, it's just beyond heavenly.

[Mike Sulzer]

BTW, we shouldn't really talk about the inductance of a magnet; the magnet affects the inductance of the coil. It's a secondary effect, and I'm hoping to see/hear past that effect.
.

That effect is very small, as I showed above.

Do you listen to these pickups through guitar speakers?

Let's look at the relative potential relative effect of the pickup/cable resonance and the guitar speaker. In the pickup it is a well-damped resonance that increases the response in the upper mid-range and lower highs, and then rolls off at 12 db per octave.

The guitar speaker, on the other hand, has an extremely complicated response throughout its range, and falls at many db per octave at about 5 KHz, depending on a number of factors.

The pickup response is relatively simple. The speaker is a huge mess and provides most of the coloration of the tone.

[Rick Turner]

Mike, when I'm doing this kind of experimentation, I listen through a variety of systems. I have a JBL D-120 loaded '64 Fender Princeton Reverb for my "guitar" rig, an SWR California Blonde for my "good enough to be called a standard" "acoustic" amp, and I have and have had good...very good...stereo systems with custom made tube monoblock preamps...no tone controls, a variety of power amps including modded Allen Organ 60 watt tube monoblocks, a Hafler 220, and a McIntosh 275. My speakers include DynaAudio, Nelson Reed (now at my son's), JBL L-166, and Synthesis, and I have a set of some of the best headphones in the world, Joe Grado PS-1s. And now I also have two Bose L-1 line array systems.

So, yes, I can listen using the equivalent of an audio microscope, and I've got a good old classic Fender guitar amp too.

I learned early on in my recording studio days that it's helpful to listen on a variety of loudspeakers to kind of "hear between the lines" as it were.

My favorite monitoring loudspeakers tend to be 3 way systems with the widest midrange possible to avoid having crossovers interfering with the all-critical vocal range. The Nelson Reeds and the DynaAudio systems are particularly transparent through the mids.

For ultra critical listening, the Joe Grado headphones are incredible. There were only about 1,600 of these made, and they're considered to be among the best dynamic headphones ever built. I've compared them to Stax Lambdas, and they hold their own against really good electrostatics.

And I was a involved in loudspeaker cabinet and array design many years ago as part of the team building the Grateful Dead's Wall of Sound.

So, yes, I know how to listen to hear details...

[Steve Conner]

(impressive kit list...)

So, yes, I know how to listen to hear details...

That doesn't actually follow, it's like saying that everyone who owns a Ferrari is a great driver. You could own all of that gear and yet have serious noise-induced hearing loss from spending too long around the Grateful Dead and their Wall Of Sound. (I'm not necessarily implying that this is Rick's actual position )

Conversely, you could have nothing but a $600 pair of mini monitors (Mackie, Alesis or whatever), a well acoustically treated listening room, and you wouldn't miss much.

I agree with the basic point though: if you only ever listen to guitars through guitar amps, you're missing information (everything above roughly 5kHz) that could help your designs. My latest guitar amp project ended up with an Eminence Beta 8, which is supposed to be a PA driver, but it was sitting in a box just asking to be used for something. I think I convinced myself that it is more "revealing" of the differences between guitars and pickups, or whatever.

[Mark Hammer]

My favorite monitoring loudspeakers tend to be 3 way systems with the widest midrange possible to avoid having crossovers interfering with the all-critical vocal range. The Nelson Reeds and the DynaAudio systems are particularly transparent through the mids.

That's a fascinating comment, and one I've not heard before. But the more I think about it, it makes an enormous amount of sense. "Coherence" is an important aspect of what makes music reproduction pleasing and listenable. Do you feel this makes 3-ways, in some sense, preferable to 2-ways that might insert the frequency split midway in this critical range?

[Rick Turner]

OK, Steve, so the gear does not make the listener, but it sure can help.

On the pro listener side of my career, I've put in a fair amount of time behind mixing consoles both in recording studios and mixing live sound. I'm good at it; and musicians and audiences like my mixes. I like to listen to good gear, and I don't like to push the gear to the max. I've heard some of the best gear in the world, and I hear details well, and all that has helped me listen critically to less expensive gear, too.