Pickup theory question

[Steve Conner]

OK, I'd have been surprised if you hadn't tried the pickup through another amp. But here's a thought.

What's the first thing we plug a pickup into? That's right, a 12AX7 running off 250V with a 100k plate resistor and 1.5k cathode resistor. That represents the first stage of most classic amps. With a gain of 50, and maybe 50V RMS of headroom, it's not quite clipping with Strat pickups, and getting mashed with hot humbuckers.

Maybe the tonal changes you hear with magnet strength are caused by the first stage getting overdriven more or less. Because 95% of tube amps share the same first stage circuit, and 95% of modelling amps run simulations of it, we all hear the same tonal changes, and we all agree on them. (Not. )

But if you ever used a completely clean amp that didn't distort the slightest bit, then the different magnets would all sound the same, just louder or quieter.

I present this as a hypothesis for debate. Thoughts? Any big cheeses like Rick Turner have an opinion?

[Joe Gwinn]

Embrace the Heaviside.

Groan...

[Mike Sulzer]

The strength of the magnetic field is reduced a desired amount to achieve pleasing tonal quality...

What a wonderful statement. Did it never occur to you that you need to understand under what conditions, if any, that might be true and why?

Where is you sense of skepticism? This is a field with more manure that the original Barnum and Bailey circus, including the elephant tent.

And I do not care what other amps you have played the pickup through or how many stories you have about how this must all be true. You posted distorted clips with the expectation that these would support your argument. You cannot then say "but, but, I do have real evidence".

[peskywinnets]

A great thread.

There's some info/analogies I will take away to use elsewhere (& imply I came up with them)....anyway, don't let me stand in your way....

fight fight fight FIGHT!

"There's only one rule about the pickup makers' forum - you don't talk about the pickup makers' forum"

[Mike Sulzer]

No.

I was really interested in how each magnet sounded. In some instances the large neos made the tone very muddy. That was very surprising. A smaller neo cleaned up the tone. But then I rewound the pickup with a lot less turns.

This is an interesting observation. Can a strong neo change the tone of a pickup, or do we have to look at the whole system in order to see what is happening?

A neo is essentially a nearly perfect source of magnetic field; it is not affected significantly by other magnetic material. [That is, its permeability is low, only about 2% larger than a vacuum.] It is also cheap and you can get it in a wide variety of sizes., and so it is the ideal material to use in new designs.

So how does it make a pickup muddy sounding? I do not think that there is any way it can, directly. I think what is happening is that a few hundred Gauss is plenty to cause string pull, subtle or not depending upon various factors, such as the height of the string. The magnet is affecting the vibration of the string, not affecting the tone of the pickup directly.

But this is not something that neo does; it is something that a strong field does; neo and other types of magnets can do this. It is more useful to classify according to cause rather than the type of material.

[Rick Turner]

In changing the magnet material from Alnico to ceramic or samarium cobalt or neo, you're also changing the resistance of the magnet and thus the eddy current performance; and of course, you're changing the inductance.

I still believe there are things going on in the 3 dimensional space that is the interface among strings, coils, and magnets that affect tone independent of output. One way to see this would be to simply put a gain stage between pickup and test amp or FFT device, and adjust the to a constant standard output. Then look at FFTs for harmonic content and phase response of the pickup(s) with different magnets, but all at a constant distance from the string(s). The other issue is to standardize the string pluck which is pretty easy using a loop of magnet wire and pulling 'til the wire breaks. Of course you have to put the wire loop at exactly the same spot on the string every time. That also allows very controlled string phase/vector analysis, at least of the initial attack.

My hunch is that when all other aspects are equal, you will find tonal differences among different magnet materials.

It would also be interesting to do this with low impedance pickups whose resonance is well above 30 kHz. Just wind some low output humbuckers and swap magnets and test. Adjust gains for some sort of standard, and then look at the FFTs. Then you'd get the whole inductance/capacitance bullshit out of the way and you'd be looking at/hearing only the magnetic interface between string and coil.

This gets back to my Litz wire pickup experiments where the coils were basically audibly transparent...no frequency response issues, and yet I heard definitely identifiable familiar differences among all the classic cores I tried winding. I came to describe what I hear as the "magnetic tone" of the pickup, and it consists of how the coil(s) interface with the field(s) and the strings in a 3 D dynamic system. Believe me, a low impedance Tele neck pickup does not sound the same as a low impedance Strat pickup, and you can hear the effect of the steel plate on a Tele bridge pickup even if the frequency response is out to bats' ears places. So it's not just inductance and coil resonance of high Z pickups that establishes signature tone.

[Mike Sulzer]

In changing the magnet material from Alnico to ceramic or samarium cobalt or neo, you're also changing the resistance of the magnet and thus the eddy current performance; and of course, you're changing the inductance.

These effects exist but they are not very big. For eddy currents you have the base plate; changing the magnet cannot affect things by too much. For the inductance, you have steel cores in the coil; a lower permeability material outside the coil does not have a large effect.

The measurements just made show about a 4% increase in the height of the resonant peak when the alnico magnet is removed, and about a 3% decrease in the frequency of the peak.

[Mike Sulzer]

Also, I hate to say it, but if you suddenly take the magnet away, there is some magnetism left in the string and the pickup "model" becomes solely and totally a moving magnet pickup. No more variable reluctance, no "disturbing the static field".

There really is no distinction between these two cases. A field can be thought of as the sum of two or more other fields, and this is a fundamental concept. The moving magnet interpretation is just as valid with the static field present as without it.

On the other hand, the variable reluctance view is just as valid without the static field as with it. The VR concept sets up a magnetic force analogous to a voltage which operates in a circuit with reluctance analogous to electrical resistance. The string magnetization provides the necessary magnetic force in this view of the process, and the static field can be considered the source of this magnetization, but not an explicit part of the magnetic circuit. The motion of the string varies the geometry of the circuit, resulting in the voltage in the coil.

"Disturb the static field" is intuitively appealing but remember, there is no way to disturb the field (at these low frequencies) except by magnetizing the string, and then it is correct and simplest to consider just the resulting field.

If you think of disturbing the field as moving the flux lines, this also should work just as well if you consider the total field, or just the field due to the string magnetization. In the former case you have a strong field, that is, a lot of closely spaced flux lines that move a small amount when the string vibrates. If you consider just the field from the string magnetization, then you have a much weaker field, that is, widely spaced field lines. Then when the string vibrates, these field lines move a lot. The result is the same in either case.

[Rick Turner]

Mike, would you address my concept of testing with the pickup output standardized through a variable gain buffer? This should eliminate the issue of output level and impedance matching and allow one to compare the "tone" of one magnetic circuit with another.

There is enough empirical evidence that such differences exist to warrant much more rigorous test methodologies than I see posted here or elsewhere. And if it turns out that it is a matter of overloading typical guitar amp/preamp inputs, then let us know that.

I still don't think the formula can be reduced to magnetic strength at one particular point in Cartesian space. That ignores so many other aspects of the coil/magnet/string interface that it just simply fails to address what many guitarists seem to actually hear. Once again, I'm all for the science...as long as it isn't constrained into a box which may be convenient for scientists, but ignores the reality experienced by musicians. Simple reductionism is an easy cop-out. The path of science is littered with outmoded theories that didn't hold up under the next level up or down of understanding. Scientists barely have a better track record than magicians, priests, and montebanks over the past 2,000 years, but they're really good at backtracking and writing grant proposals!

[Mike Sulzer]

Mike, would you address my concept of testing with the pickup output standardized through a variable gain buffer? This should eliminate the issue of output level and impedance matching and allow one to compare the "tone" of one magnetic circuit with another.

There is enough empirical evidence that such differences exist to warrant much more rigorous test methodologies than I see posted here or elsewhere. And if it turns out that it is a matter of overloading typical guitar amp/preamp inputs, then let us know that.

I still don't think the formula can be reduced to magnetic strength at one particular point in Cartesian space. That ignores so many other aspects of the coil/magnet/string interface that it just simply fails to address what many guitarists seem to actually hear. Once again, I'm all for the science...as long as it isn't constrained into a box which may be convenient for scientists, but ignores the reality experienced by musicians. Simple reductionism is an easy cop-out. The path of science is littered with outmoded theories that didn't hold up under the next level up or down of understanding. Scientists barely have a better track record than magicians, priests, and montebanks over the past 2,000 years, but they're really good at backtracking and writing grant proposals!

Agreed, we should find the cause, whatever it might be!

Yes, adjusting the levels using a variable gain buffer (as linear as possible) is the way to go. But, to say it again, let's make sure that the pickup gets loaded correctly so that we have the expected sound. This means using the correct input resistance if the buffer is located at the amp, but if it is located at the guitar, one needs to add a capacitor at the input to account for the cable capacitance.

Not so sure about your view of science. Newton's theory of gravity has been displaced in a fundamental sense by Einstein's. But Newton's theory is still just as correct and useful over its range of applicability, that is, the world around us of ordinary sizes, energies, etc. Those other folks you mention are usually just plain wrong.

[Rick Turner]

Mike, the whole point of such an experiment would be to eliminate things like cable capacitance or coil resonance. Get that stuff out of the way completely in order to hear and hopefully be able to see only the changes in tone attributable to different strengths and shapes of the magnetic field. This is why I think using low impedance coils with a resonance in the 30 to 40 kHz or beyond range would be important.

When I did my Litz wire experiments, I also tried to have the new coil come close to matching the physical size of the original coils. I did Tele bridge and neck, Strat, J-Bass, P-Bass, Hagstrom/Guild, and Gibson humbucker pickups using original factory coil forms or bobbins. What I found was that even though the coil resonance was completely out of the picture, each pickup retained a certain degree of it's "signature tone" identity. I attributed that to the fact that each pickup style has a unique 3 dimensional dynamic relationship to the strings based on the varying (from string motion) magnetic field and how it passes through each layer and turn of the coil. If you change the type of magnet, the static density of the field will change at different points in the coil.

[salvarsan]

RE: Rick Turners' mention of eddy current changes from magnet changes.

These effects exist but they are not very big. For eddy currents you have the base plate; changing the magnet cannot affect things by too much.

The famous example is the Barden Blade which uses ceramic magnets (viz., non-conductive). Copper foil electrically connects the blades since the pickups are otherwise too bright sounding.

Electrical and magnetic circuits are isolated without the interconnect, and eddy current effects are a plausible suspect in this circuit.

I am open to other explanations, but all of them must account for the audible difference between the connected-blade and isolated-blade implementations.

The definitive test would involve measuring AC resistance of a Barden T with and without the interconnect.

[Mike Sulzer]

Mike, the whole point of such an experiment would be to eliminate things like cable capacitance or coil resonance. Get that stuff out of the way completely in order to hear and hopefully be able to see only the changes in tone attributable to different strengths and shapes of the magnetic field. This is why I think using low impedance coils with a resonance in the 30 to 40 kHz or beyond range would be important.

When I did my Litz wire experiments, I also tried to have the new coil come close to matching the physical size of the original coils. I did Tele bridge and neck, Strat, J-Bass, P-Bass, Hagstrom/Guild, and Gibson humbucker pickups using original factory coil forms or bobbins. What I found was that even though the coil resonance was completely out of the picture, each pickup retained a certain degree of it's "signature tone" identity. I attributed that to the fact that each pickup style has a unique 3 dimensional dynamic relationship to the strings based on the varying (from string motion) magnetic field and how it passes through each layer and turn of the coil. If you change the type of magnet, the static density of the field will change at different points in the coil.

I see what you mean, but as I see it there are two types of questions to answer:
1. How big can these magnetic effects be?
2. How big are they with standard high impedance pickups.

I think 2. is the one to start with because the first thing to do is to decide if these effects attributed to the field are really something else such as a non-linear preamp stage. To do this we need to duplicate the normal conditions.

[Mike Sulzer]

The famous example is the Barden Blade which uses ceramic magnets (viz., non-conductive). Copper foil electrically connects the blades since the pickups are otherwise too bright sounding.

Electrical and magnetic circuits are isolated without the interconnect, and eddy current effects are a plausible suspect in this circuit.

I am open to other explanations, but all of them must account for the audible difference between the connected-blade and isolated-blade implementations.

The definitive test would involve measuring AC resistance of a Barden T with and without the interconnect.

Eddy currents are extremely important. But in the standard humbucker the magnet sits on the metal base plate, and so I think it has little additional effect. That is what the measurements I made last night show. I am not very familiar with the Barden T pickups. I believe there a coil around each blade? If so, I cannot see why connecting the blades together electrically would affect the brightness because there is no greater potential around both blades to induce larger currents.

[Rick Turner]

Either way, Mike, as long as you have the load on the pickup stay exactly the same and just vary the gain to bring the voltage level after each change of magnet to the same place.

A number of years ago I designed a test rig monochord that would have reference piezo pickups at each end of the string so you could duplicate string plucks and have a true reference that does not depend on all the variables.

I'm really tired of the totally bogus specs, if they can even be called that in our business. This pickup is hot; this one is clean and low output... The microphone and loudspeaker industries have both agreed upon a standardized set of tests, and thus sensitivity and noise specs actually mean something. The pickup industry is totally insane. Nobody agrees upon color codes for wires. Nobody agrees upon standard designations for wind start and end or clockwise vs. counterclockwise winding. Magnetic polarity is whatever one wants at the moment. No wonder it's so hard to talk about this stuff!