PAFs that sound like sh*t

[Possum]

need more...

OK I looked at that page for all of five seconds and it seems aimed not at measuring a pickup as a unit but measuring a piece of steel. I'm more interested in plugging something directly into a pickup as Dr. X says DM does to test its pickups. Measuring samples of steel is kind of pointless because they are not cut to size for use and aren't in a functioning pickup, to me anyway. Besides I don't read circuit diagrams :-)

Dirt is never gotten from the pickup. Pickups are kind of related to amplifiers, they can be hi-fi or more compressed, saggy the same way tube amps are. So called "dirty" pickups are usually wound real hot to push the preamp stage hard, in themselves they don't distort the signal much, they just have a certain frequency response. Humbuckers that have alot of metal in them are more squishy compared to a strat pickup which only has the magnet rods so they are very direct sounding no matter what alnico you use, though the metal alloy of those has also a noticeable effect on eddy currents and tone. Right now my most useful too is the LCR meter, I use inductance and AC resistance side by side to see whats going on.

[Possum]

a bit more...

The reason I'm most interested in setting up some kind of rig to do the BH thing with an assembled pickup is it will be more easy to swap out different steel parts and see what happens to that curve. There are some computer oscilloscope programs out there that would probably be better than buy some old thing with a green tube and a huge grid on it too....

[Steve Conner]

So, RG, want to collaborate on designing some kind of computerised pickup analyser? :P

Part of me thinks it would be interesting, but another part of me calls BS. When people start talking about things that "sure sound different but don't show up on charts" I start to think of experimenter expectancy, and how hard it must be to do a properly controlled blind test on guitar pickups.

At the end of the day, I bet Eric Clapton could have recorded the Beano album with a pickup made of roofing nails, a fridge magnet and wire from a doorbell transformer, and he would still have been on fire, provided the wire was thin enough, there was plenty of it, and the magnet was strong.

I don't even necessarily believe the statement "Some PAFs sound like sh*t": maybe all PAFs sound identical and some other variables were responsible for some people experiencing sh*t tone on PAF-equipped guitars. Which brings me on to my next point...

When I stated my earlier hypothesis, I left out one important thing. A pickup is a multi-input system since there are 6 strings and a guitar body. Each string, and the body vibration, will have its own individual transfer function to the output. So there are 7 transfer functions. (8 if you count direct microphony from airborne sound waves, 9 if you count EMI and hum pickup, etc, but let's call it 7 for now.)

But again I will stick my neck out and say that these transfer functions are of the form I stated, and if two pickups have the same set of transfer functions, a player will be unable to tell them apart in a properly controlled blind test. If this is true, then these functions must describe the mojo of a pickup adequately.

Of course these 7 functions then depend on the adjustment of the pole screws and the height and angle that the pickup was mounted in the body, what the mounting ring is made of, whether the pickup baseplate is touching the guitar body, etc. If someone did tests where different pickups were swapped into the same guitar, were these variables properly controlled?

[Possum]

testing ideas...

Well, sound tests would also have to include the player, Robben Ford and BB King on the same guitar will sound totally different. Seriously though, one of the reasons the LCR meter is useful to me is that it tests at two static frequencies, 100hz and 1000hz. After awhile of using you get to know what the numbers mean, especially in AC resistance.

So what would I as a pickup maker want in a single testing device? Something that gives direct digital numbers at say 8 relevant frequencies. LCR meters according to Joe, that go up to 10khz don't get correct readings at that frequency. If there is a way to get digital numbers up there in that range etc. it sure would be useful. Almost like a frequency response chart really but exact numbers with accurate methods of obtaining them. So if you pulled one pole screw out and swapped a different one of a different alloy it would pick that up. The ability to put this on a chart but the numbers are more important to me. We've gone over this stuff before on the forum and Joe has a method that uses a frequency generator, a resistor connected to the pickup and a multimeter, to detect resonant peak but it never really worked for me except rarely and was real time consuming too. So there's my input....

[DoctorX]

an interesting document...

Here is a link to a paper called "Measurement of the Electromagnetic Properties of Electric Guitar Pickups". No diagram detailing BH curve testing of completed pickups (I'm workiing on that), but the information is interesting.

You may have seen this document already, but in case you haven't:

http://online.physics.uiuc.edu/cours...asurements.pdf

DoctorX

[R.G.]

So, RG, want to collaborate on designing some kind of computerised pickup analyser? :P

Maybe... I'm going to do a pickup test bed where I can isolate influences for testing. But I think we might hold off on the computerized tester until we know what to test.

Part of me thinks it would be interesting, but another part of me calls BS. When people start talking about things that "sure sound different but don't show up on charts" I start to think of experimenter expectancy, and how hard it must be to do a properly controlled blind test on guitar pickups.

You and me both. However, logic tells me that (a) some pickups sound different from others (b) no one is able (or willing) to tell me what it is that makes them sound different (c) nothing I can find in the literature, including patents too fatuous to believe points to what affects what.

Taking all that, I think that no one has done the testing to find out what affects what at the lowest levels. A dark horse is our friend Joe Gwinn. I suspect he's quietly cooking something in the back room. He's certainly capable.

I have seen precious few charts on pickups. Don Tillman, Stephan Moeller, Lemme, a couple of others.Those are all aimed at defining what happens, not teasing it apart to see what makes it tick. Ears hearing what is not on the charts is almost certainly BS at least in part, but given the paucity of charts, I'm giving the ears the benefit of the doubt and saying that we don't have the right or enough charts.

At the end of the day, I bet Eric Clapton could have recorded the Beano album with a pickup made of roofing nails, a fridge magnet and wire from a doorbell transformer, and he would still have been on fire, provided the wire was thin enough, there was plenty of it, and the magnet was strong.

Yes - the human brain and hands are powerful controllers that can make up for huge variances in the plant being controlled. Nonlinear, time varying, multidimensional all get controlled in some cases if you have the right human in the loop.

I don't even necessarily believe the statement "Some PAFs sound like sh*t": maybe all PAFs sound identical and some other variables were responsible for some people experiencing sh*t tone on PAF-equipped guitars.

I absolutely believe that to some people, some PAFs sound bad. It is reasonably well documented that they were made in non-critical commercial way to meet production, so process and material drift were certain. The issue is the definition of "bad". We all have heard tales of Eric Johnson. I know his amp tech, and there are plausible reasons for ... some... of Eric's preferences. It's just that there is no way to say that someone does not find one pickup worse than another which you believe is identical. Audiology research has proved that even if you hear them as identical, someone else can hear them as different.

Which is why I punt on the question. Good? Bad? Like marshmallows with a cherry back note and a hint of parsnips on the finish? OK. No way to measure that. But I CAN measure frequency response, spectrum per string, X-Y sensitivity, and so on and relate that to the voltage that comes out of the pickup. That changes the question to one of isolating effects and determining what to measure. Hence my mania about what affects what. It's a question about isolation of effects. Where do we dig?

I don't think anyone has done that yet. Maybe someone has but hasn't published.

Which brings me on to my next point...When I stated my earlier hypothesis, I left out one important thing. A pickup is a multi-input system since there are 6 strings and a guitar body. Each string, and the body vibration, will have its own individual transfer function to the output. So there are 7 transfer functions. (8 if you count direct microphony from airborne sound waves, 9 if you count EMI and hum pickup, etc, but let's call it 7 for now.)

13. All non-blade pickups have both X and Y responses to string movement. The reluctance varies differently with X and Y string motion, so until proved different, we have to include the possibility of different transfers that way.

But yes, I get your point.

But again I will stick my neck out and say that these transfer functions are of the form I stated, and if two pickups have the same set of transfer functions, a player will be unable to tell them apart in a properly controlled blind test. If this is true, then these functions must describe the mojo of a pickup adequately.

I don't think that's sticking your neck out at all. Only the remaining tribal peoples (N.B. that includes anyone who is not bought into the scientific method; much of the population of North America and Europe, unfortunately) would question that. Not believing it equates to insisting that there is something that CANNOT be measured, but which appears when you pluck strings. Santa Claus, perhaps?

Of course these 7 functions then depend on the adjustment of the pole screws and the height and angle that the pickup was mounted in the body, what the mounting ring is made of, whether the pickup baseplate is touching the guitar body, etc. If someone did tests where different pickups were swapped into the same guitar, were these variables properly controlled?

How about tests where they're put into a rig which cannot feed back microphony, and which can be controlled? Isolate what affects what on the the pickup; materials, construction, process, etc. then add that to the guitar body once you know the pickup in isolation. S'what I've been proposing. Check the sketch of a test bed on my web page.

[JGundry]

R.G.,

One factor that your test bed does not take into account is the effect that the plucked string might have on the pickup response. I think this might particularly be an important variable when using Alnico magnets. The way I understand it is that part of the tonal difference due to Alnico magnets is there is a minute loss of magnetism with peak input levels. The magnetism is quickly regenerated but the input peaks are compressed giving Alnico pickups a swell and compression that ceramic magnets lack.

[R.G.]

One factor that your test bed does not take into account is the effect that the plucked string might have on the pickup response.

That's right, and in some private emails I have already decided that both steady state and plucked responses are needed. I have designed a "standardized plucker" mechanism to do the plucked tests.

I think this might particularly be an important variable when using Alnico magnets. The way I understand it is that part of the tonal difference due to Alnico magnets is there is a minute loss of magnetism with peak input levels. The magnetism is quickly regenerated but the input peaks are compressed giving Alnico pickups a swell and compression that ceramic magnets lack.

That's certainly true in speakers where there is a coil forcing a reversing coercive force onto the magnet. Alnico has a lower coercivity than other permanent magnets like ceramic and rare earth.

I could be wrong on this, but I don't think that the string exerts a demagnetizing coercivity onto the magnet to any significant degree. The strings change the reluctance of the flux path. All the flux that comes out of the magnet still comes out of the magnet. The string causes a tiny amount of that flux to move from one place to another in the 3-space that the coil occupies, and so the coil gets a voltage impressed. If a current can flow, it flows in the direction to oppose the string motion, so there might be some small coercivity effect there, but I think it's so small that it really can't change the magnet flux much. It's very much like Steven Conner's comment on the signal causing the B/H curve traversed to be so small that it's negligible.

I'm not sure what I think causes the rep that alnico has for "compression". It's possible that it really does, but it's also possible that people have extended what they think they hear based on what has been found out about speakers.

Don't know.

An interesting experiment would be to make a test bed with a standardized plucker and a steady-state exciter. The bed would also have an optical pickup to determine what the string motion really is as well, free of magnetic magumbas. Then pluck and drive (no, not shuck and jive ) the strings and measure output versus string motion. If the output quits being linear with string motion, there is a compression going on and it can be hunted down. One should also be able to sub in a ceramic magnet pickup of similar construction and see if the same compression, or lack of it, is present.

So I think I can come up with tests to record whether compression is there or not.

Here's another thought - put a Hall effect device right under the string, on top of the magnet. This will now read the magnet's flux, with great fidelity. You can tell directly whether the flux coming out of the magnet compresses or not.

The test bed is just a way to give similar string motion, but to exclude the other stuff that confuses things in actual guitars. Once we understand, then we can start wading through the weeds on a real guitar.

[JGundry]

There is another factor that I think any test bed should incorporate. Let me start by saying the magnetic testing methods R. G. is talking about are completely beyond my depth right now. I use an LCR meter and a gauss meter but other than my ears that is it. So there is plenty I have to learn from a technical standpoint. But I do have quite a bit of experience doing modal analysis of free violin plates as a hobbyist violin maker. How this relates to pickups is the mechanical resonance of the pickup. I think this is very often overlooked but still a very important part of the equation. The pickup itself will have it's own inherent mechanical resonance. Things like pickup covers, potting, pickup construction, the amount of wire on bobbin, perhaps even the tension of the wire will all have some effect either on the pitch of the mechanical resonance or the efficiency of the resonance. I think potting drives the efficiency way down and makes the resonance pitch much less defined. But if the pickup is so microphonic that you can clearly hear your pick tapping the pickup through your amp then I think it is really a very large part of the sound of the pickup when in a guitar. Some pickups are so microphonic that you can talk into them. One possible method of testing this would be to mount a pickup directly to a plank of wood and drive the wood with an exciter of some sort that would transmit a frequency sweep from a frequency counter. There is a little hammer that is used for the purpose of testing finished musical instruments but I believe it is pretty pricey. I think one could figure an alternate method by using a small speaker mounted directly to the test plank.

[R.G.]

Remind me to buy you a glass of your favorite foaming mental lubricant if we get together. That idea is very much on target.

I'm currently thinking of the test bed as too massive and rigid to let resonances get in the way, but there sure needs to be a resonance test on the pickup itself.

Great idea! I'll get on it!

[Possum]

plucking

the idea of a plucking test seems kind of vague. After all what happens when you pluck a string? Basically you are generating a PULSE. So what should be tested instead of something not very repeatable, would be a frequency pulse with decay. I have a Mac program called FuzzMeasure thats designed to test speakers and speaker enclosures and how speakers react in a room. I believe it uses this method, it hits your coil with a bang. What we should be looking at I think is how a peak pulse when you bang on the strings gets reacted to by the pickup as a unit. Does it react quickly or sag? Making a string plucker is full of faults, how can you quantify how hard the string is being plucked, how new the string is, what kind of string, how the string reacts to temperature etc. etc. etc. I say don't go there. The other things that are important are resonant peak, which no one can agree on in our previous discussions, on how to measure that. The Lemme method is to use a driver coil with very low winds. I've used this method and FuzzMeasure to do frequency response charts and it works well. You get resonant peaks in the 4khz range vaguely if I remember, but using some other method like Duncan uses you get them in the 10khz range. There is a physics course in musical instruments and pickups at some university that shows how they tested pickups with a very complex setup you should try to find, I think someone posted it recently again but before you try to reinvent the wheel it would be worth looking at their work. Another thing I've never seen anyone measure is harmonics, I don't even know how you would measure for that. BAck the to the BH cuver thing I suspect that DM used it for checking pickups by seeing if it had the correct shape and if the shape was out of whack they disqualified the pickup. So more than likely getting into tiny specifics with that method might not be so useful...

backing up to resonant peaks again, the university physics guys were getting very SHARP resonant peaks on strat pickups in the 10khz range if I remember right, the stuff I was getting was very smooth broad peaks which is typical of hand wound pickups, but even with factory stuff I couldn't get the very sharp tall peaks these guys were getting, wonder why....

[R.G.]

the idea of a plucking test seems kind of vague. After all what happens when you pluck a string? Basically you are generating a PULSE.

I think what happens is that you're forcing the string to one side, stretching it like a spring, and then releasing it more-or-less instantly. My idea wasn't all that vague - put a guitar pick in a fixture that could pull the string aside by a measured distance, then move the pick perpendicular to the direction stretched until the pick disengaged from the string, ensuring that the pick then moved out of the string's rebound. Since it depends primarily on the tension and stretchiness of the string, I think it's repeatable per string.

But I'd love to hear other suggestions.

So what should be tested instead of something not very repeatable, would be a frequency pulse with decay. I have a Mac program called FuzzMeasure thats designed to test speakers and speaker enclosures and how speakers react in a room. I believe it uses this method, it hits your coil with a bang.

Impulse response tests were common with speakers at one time for testing frequency response, until they found that you can get the same result with a pseudorandom noise signal, called a Maximal Length Sequence, or MLS. You get the same results as with an impulse, but without some of the difficulties of doing a real impulse test.

However, that test does not generate the instantaneous peak signal that a pickup might do, as in:

What we should be looking at I think is how a peak pulse when you bang on the strings gets reacted to by the pickup as a unit. Does it react quickly or sag? Making a string plucker is full of faults, how can you quantify how hard the string is being plucked, how new the string is, what kind of string, how the string reacts to temperature etc. etc. etc. I say don't go there.

I'd agree it's not going to be perfect, but to some extent, the energy stored in the string as it's pushed to one side is suddenly released into the string when the string is released. String tension and gauge determine the amount of energy released into the string. It's more repeatable than it sounds. Think of it as the string release used in competitive archery. The string+bow stores a tightly controlled amount of energy into a spring by being pulled away from rest, then releases it when the string is released.

May not work, but the advantage of building my own test bed is that if it doesn't work, I'll dismember the body, burn it, and bury it out back...

The other things that are important are resonant peak, which no one can agree on in our previous discussions, on how to measure that.

Yeah, I found that very interesting in reading the archives.

The Lemme method is to use a driver coil with very low winds. I've used this method and FuzzMeasure to do frequency response charts and it works well. You get resonant peaks in the 4khz range vaguely if I remember, but using some other method like Duncan uses you get them in the 10khz range. There is a physics course in musical instruments and pickups at some university that shows how they tested pickups with a very complex setup you should try to find, I think someone posted it recently again but before you try to reinvent the wheel it would be worth looking at their work.

Well you know how I feel about reinventing the wheel. I've pored through all of that I can find. I'm still chasing down one Australian guy's doctoral thesis on guitar pickups, but I've seen the MI physics stuff.

Another thing I've never seen anyone measure is harmonics, I don't even know how you would measure for that.

Fortunately that's easy for me. I just bought a Picoscope, a USB plugin that's an oscilloscope on one end and a program on the other. Plugged into my laptop, I can put the signal into the Pico, and out comes the signal. The software will turn that into a spectrum for me on the spot. That's in the list.

BAck the to the BH cuver thing I suspect that DM used it for checking pickups by seeing if it had the correct shape and if the shape was out of whack they disqualified the pickup. So more than likely getting into tiny specifics with that method might not be so useful...

I bet you're right there. Easy to use it as a go-no-go test.

backing up to resonant peaks again, the university physics guys were getting very SHARP resonant peaks on strat pickups in the 10khz range if I remember right, the stuff I was getting was very smooth broad peaks which is typical of hand wound pickups, but even with factory stuff I couldn't get the very sharp tall peaks these guys were getting, wonder why....

Don't know. I'll be looking. Could be sub-sections of the pickups resonating.

[JGundry]

I don't know if this helps but harpsichords use delrin (guitar pick material) quills to pluck the strings. You might be able to come up with a variation on the mechanism for plucking a guitar string. Each string could have its own quill for one uniform simultaneous pluck of all strings. Here is a video of replacing a delrin Harpsichord quill. http://www.youtube.com/watch?v=14lDkcogzvk

[DoctorX]

How about using an EBow to 'pluck' the string? You could sustain the string vibration longer to take steadier measurements. It might make symmetry between tests easier....maybe.

Any thoughts on this?

DoctorX

[Steve Conner]

Personally, I think that a test bed that mechanically excited a guitar string would be letting the numbers of variables under test get out of control, which would bog the experiment down and risk it never producing a meaningful result.

What's more, I'd argue that some of those variables being tested don't properly belong to the pickup, but to the string and the rest of the test bed.

I'd vote for exciting the pickup with a field from an electric current in a test coil. Maybe passing a heavy current from an audio power amp through a copper wire placed where the guitar string normally is would do it. The geometry of the fields generated by this might be wrong, in which case a different shaped coil might be needed. Maybe you need two tiny coils each the size of a single pole screw, with a delay between them to model the speed of sound in a guitar string, or whatever.