Science, Scientism, and Understanding Pickups

[Mike Sulzer]

It is convenient to divide the process of understanding pickups into two parts:
1. the waveform that it produces,
2. the perceived sound of the waveform when it is processed in an appropriate manner.

The first is in principle entirely explainable by science, although in practice you would not expect it to be possible to describe the waveform sufficiently accurately through measurements to explain everything that can be heard. The second is only partially understandable through science, although many measured properties do have a perceived equivalent. For example, we know roughly what increasing the high frequency content of a signal sounds like.

The first part is explainable through science because electricity and magnetism is very well understand. The relevant laws were discovered in the nineteenth century, and since that time they have been integrated into physics in a way that any problem with the accuracy of the E&M laws (Maxwell's equations) would result in many inconsistencies. For example, Einstein discovered the laws of special relativity by looking for consistency with Maxwell's equations. Special relativity has been verified countless times, even through the operation of fairly normal devices. For example, the klystrons for our 430 MHz radar at Arecibo have a beam of electrons that is relativistic, meaning that the electrons move at a significant fraction of the speed of light, requiring Einstein rather Newton in order to describe the physics. These klystrons could not work as they do if there is a problem with the consistent laws of special relativity and E&M.

Part two is not well understood through science because the operation of the brain is not well understood. It might never be; even whether it can be understood is not known.

Scientism as it is used here is an attempt to use science to explain what it cannot or does not. For example, consider the idea that the law of magnetic induction and associated laws describing electrical circuits are sufficient to explain the waveform produced by any pickup. This is within the realm of science because these laws are fully understood. Therefore it is not scientism. On the other hand, it is scientism to say that every detail of the perceived sound of a pickup is understood through science. For example, it is reasonable to say that if you increase the high frequencies in a pickup, you expect it to sound brighter. It is not reasonable to say "This new pickup is brighter than the old one; therefore it has more high frequencies." You had better measure it. There could be other reasons why a pickup could be perceived as brighter.

The more subtle the properties of a perceived sound, the less likely it is that you can explain it entirely by science. But that is no reason not to use science where you can. It is certainly wrong to say that because you use science, you do not listen.

[Rick Turner]

The issue is not to believe only in the current questions that science asks.

In 1970 those of us with ears could hear that "high fidelity" power amps sounded different from one another, and that at that time, high quality tube amps, like the McIntosh 75s and 350s that we used in the Grateful Dead's sound system sounded better than the main transistor competitor...the Crown DC 300 which had, according to the then current science, much better performance. Then suddenly Walt Jung came up with a way to measure a previously unknown factor, Transient Intermodulation Distortion. Bingo...starting with that, there was a new measurement that much more closely tracked what golden ears had been hearing while the scientific solid state audio engineers had been poo-pooing them. Now we use a more "modern" spec...slew rate...which seems to track TIM distortion, and fast slew rate is good...very good...though we still don't know all the reasons why many ears will prefer the "more musical" sound of a high quality tube amp.

By the way, I'm not talking Marshalls and HiWatts and Mesa Boogie overdrive tube characteristics; I'm talking high end hi-fi tube gear vs. high end hi-fi solid state.

So the issue is not science vs. anything else; it's putting science into the service of understanding real human perception which is best tested in double or even triple blind setups.

There are people with absolutely amazing hearing. Eric Johnson is legendary...hearing the difference between brands of batteries in effects; being able to locate a partially blown speaker out of a whole array of them. Laurence Juber is like that. When we first did listening tests on our D-TAR Mama Bear digital acoustic guitar preamp, he was able to detect an algorithm of a cedar topped koa back and side built guitar...playing his own rosewood and spruce Martin. He said "I'm hearing cedar and koa...", and we all went running for the list of digital guitar models, and sure enough, it was a cedar/koa Ed Claxton guitar.

So when science can show us really what we hear, I'm all for it. But attempting to squeeze what we hear into current textbook explanations of Maxwell's equations and not being open to anything beyond doesn't make it in my book. Pickups operate in a three dimensional dynamic system, and until you can understand all the points in that matrix, you'll not get away with simplistic explanations of pickup tone.

BTW, "brighter" can mean a sharper upper midrange spike with less extended highs...

See my "high end capacitor" thread.

[Mike Sulzer]

Exactly, nobody with ears considered the DC300A a high fidelity amp next to a good tube amp. And the better the speakers, the more obvious the differences. I remember TIM as Marshal Leach's thing. He was then, and still is now, a professor at Georgia Tech. There were a couple of Finnish guys involved, too. It was obvious from the start that TIM was the same mechanism as slew rate limiting, something that I thought Leach failed to point out. Slew rate limiting was well known as a limiting factor in the performance of op amps at the time, which is why I was so appalled when some idiots put 741s in their audio consoles. A 741 cannot produce a sine wave at 20 KHz at anywhere near rail-to-rail level.

In any case, TIM is not what made solid state amps sound so bad. I am not sure that it ever has been sorted out properly, but it must relate to the fact that transistor distortion is ugly sounding and full of higher harmonics. But if you make the distortion incredibly low, as it is in modern amps, you do not hear it.

Yeah, I can understand still liking the sound of tubes amps, especially for guitar, even below the overload level. But for audio, not so much. SS is much cleaner. I even retired my VT52 amp some years ago. I designed a lot of amps back then, but this was the best. All had solid state regulated power supplies, since a solid supply is very important. And this one used a good op amp as the low level stage. Even in the late 70s, a good op amp at low level was cleaner than a low level triode. It took quite a few years for that to propagate up to the higher powers.

I think you do not quite understand what I am saying about Maxwell's equations. They are correct; it is our ability to understand what they imply in specific situations that is limited. That is what makes the simultaneous application of science and listening so interesting in audio design.

[David Schwab]

Mike, I had asked a question in the other thread, and I wasn't being facetious. We were talking about aperture and coil shape. My question was how you explain scientifically why a tall thin coil like a Strat sounds different from a wide squat coil, like a Jazzmaster. Lets assume for this that they have the same number of turns on the coils. I can see that magnet length is different, but what about the coil shape? If aperture on a single coil is restricted to the pole, then why does the wider coil sound more mellow? Why is the tall coil brighter? This is the same effect on tone that a wider pole vs. a narrow pole has.

[David Schwab]

They are correct; it is our ability to understand what they imply in specific situations that is limited.

That was my point. We often utilize things that we don't fully understand.

[Mark Hammer]

My thinking about a lot of this was transformed when I read some short pieces about Thomas Edison's early wax cylinder recordings. Apparently, he made some coin by going to country fairs and charging people money to record their voices onto his was cylinders. News reporters at the time described the recordings as being difficult to distinguish from the actual voice, they were THAT lifelike. (You can stop rolling your eyes now)

Of course, now we would laugh at such comparisons, just as we would giggle at the "Is it Maxell or is it Ella Fitzgerald?" comparisons in an era of 24-bit digital recording.

So what happened? Whenever there is a technological innovation, people are impressed at the very fact of it, and tend not to notice its imperfections or limitations. Over time, they begin to notice them, though. With even more time, they begin to be able to describe those limitations in a more systematc fashion, and eventually that leads to being able to measure them (a technological innovation in itself). Once you can measure them, you can begin to understand what produces them. And once you begin to understand how to predict those limitations, you can become able to control or eliminate them. Voila! Technology takes another big leap forward.

And of course, as David rightly notes, sometimes someone does something different without truly understanding how it works or why it works, and over time we start to comprehend the hows and whys, leading to yet another serendipitous advance.

[David Schwab]

Mark, that's a good observation. Also some people become accustomed to older technology, like vinyl gramophone records. When I first heard a CD in person I thought it was kind of brittle sounding. (and some of the early ones were). But after I got my own CD player, I cue up an XTC album on the same track on vinyl and the CD and A/B'd them as they played. I was shocked at how mushy sounding the vinyl was! What sounded warm and punchy before was reduced to soft and mushy. Especially in the drums and bass. The high end sounded smeared too.

And then of course you start to perceive the surface noise and flutter. When my turn table's stylus broke I sold it and put all my vinyl in boxes in the basement! I miss a few albums that aren't on CD, but in general I can't listen to that format anymore.

And now a couple of decades later, and all the music we listen to in our home and car are via iTunes and iPods.... They still sound better than vinyl!

[frankfalbo]

Which is the one that makes Tom Cruise into an idiot?

[Mike Sulzer]

Mike, I had asked a question in the other thread, and I wasn't being facetious. We were talking about aperture and coil shape. My question was how you explain scientifically why a tall thin coil like a Strat sounds different from a wide squat coil, like a Jazzmaster. Lets assume for this that they have the same number of turns on the coils. I can see that magnet length is different, but what about the coil shape? If aperture on a single coil is restricted to the pole, then why does the wider coil sound more mellow? Why is the tall coil brighter? This is the same effect on tone that a wider pole vs. a narrow pole has.

Sorry, I thought I had answered, but it might have gotten lost in that shuffle. The coil inductances are different. One would somehow have to take this effect on the frequency responses out in order to look for more subtle effects. No doubt there are some, but are they really well-described as a difference in aperture? It seems unlikely considering how the field is concentrated near the poles.

[David Schwab]

Which is the one that makes Tom Cruise into an idiot?

Any on them! I can't stand that guy. He went to high school the next town over from here (Glen Ridge NJ). I've seen the house he used to live in to.

But then the DeLeo brothers from STP lived there, so they make up for it.

[David Schwab]

Sorry, I thought I had answered, but it might have gotten lost in that shuffle. The coil inductances are different. One would somehow have to take this effect on the frequency responses out in order to look for more subtle effects. No doubt there are some, but are they really well-described as a difference in aperture? It seems unlikely considering how the field is concentrated near the poles.

So, if you made two pickups with the same rod magnets, and same wire gauge and number of turns, but one was shaped like a Stat, and the other was like a Jazzmaster, would the inductance read different between them?

And how would the difference in induction explain the difference in tone? It's not like one if overwound and sounds darker. Jazzmaster have plenty of top end, but just sound different.

And lastly, I'd like to know why the inductance is different because of the coil shape.

[Mike Sulzer]

So, if you made two pickups with the same rod magnets, and same wire gauge and number of turns, but one was shaped like a Stat, and the other was like a Jazzmaster, would the inductance read different between them?

Yes, but it is not so easy to say how much. The difference should be significant.

And how would the difference in induction explain the difference in tone? It's not like one if overwound and sounds darker. Jazzmaster have plenty of top end, but just sound different.

Higher inductance gives a lower resonant frequency. The large flat coil would also be expected to require more wire, leading to a higher series resistance and thus a flatter peak. Both could affect the tone significantly.

And lastly, I'd like to know why the inductance is different because of the coil shape.

For an air core coil, it is a fairly simple problem. The differences are related to the different flux linkages between turns which occupy nearly the same space and ones that are further separated, as well as the areas of the turns. The flat coil has its turns more nearly occupying the same space and the average area of a turn is greater.

So the tall thin coil is expected to have less inductance than the flat one, but the cores can make this a more complicated problem, altering the difference since they can help connect the flux from one turn to another, and increase the effective area of a turn. But the effect of the rod cores used in pickups is a lot less than closed cores such as toroids or the common rectangular types. So one needs to make measurements. I would expect the flatter coil to have a somewhat higher inductance and a higher resistance leading to a somewhat lower frequency and less peaked resonance. This is the kind of difference that should be audible to someone familiar with electric guitar sounds.

[Rick Turner]

Mike, this gets back to one of the things I was hoping you'd read and understand from some of my previous posts.

I've made a fairly complete study of classic pickup forms wound to low Z specs (typically 600 Ohms DCR) using 7/44 Litz wire. I wind these so the physical mass of the coil is very similar to original specs. This essentially takes coil impedance, capacitance, inductance and all that way out of the audio pass band. The pickups sound different from one another with certain elements of classic signature tone. The Tele still sounds Tele-ish. The Strat still sounds Strat-ish. The humbucker still sounds humbucker-ish. There are effects of magnetic field shape and coil space that are not explained by inductance. This is the 3D dynamic response of the interaction between field, strings, and coil space.

I sent a J bass up to my pals at Bass Player Magazine just for them to try out. They were amazed. The pickups made funky strings sound really shitty, but with good strings, one of the guys said he could hear his fingerprints...

I did these experiments about 20 years ago, and don't have the pickups on hand. May have to do this again when I retire...if I ever do...

I'd encourage anyone to try winding low Z coils on your standard bobbins. Go down to 36 gauge wire or some such, and wind 'til the coil looks the right size. Forget about output; this isn't about that. Just listen for the tone when the frequency response is really opened up and the resonant peak isn't there for you to hear. You'll start hearing your pickups in a new way, I guarantee.

[Mike Sulzer]

This is the 3D dynamic response of the interaction between field, strings, and coil space.

You have not explained what this could mean.

I can think of a couple of reasons why humbuckers and single coils retain some of their characteristics even when wound for low impedance. First is the sampling function. The humbucker does it in two places and this establishes a high frequency filtering function for each string. It does not change with impedance established by the winding. The second is eddy currents. The effect is very different for steel slugs and pole screws as opposed to alnico magnets.

[David Schwab]

Yes, but it is not so easy to say how much. The difference should be significant.

I'd like to see real numbers... I don't own an inductance meter, but if anyone here wants to make a tests.. please do.

Higher inductance gives a lower resonant frequency. The large flat coil would also be expected to require more wire, leading to a higher series resistance and thus a flatter peak. Both could affect the tone significantly.
For an air core coil, it is a fairly simple problem. The differences are related to the different flux linkages between turns which occupy nearly the same space and ones that are further separated, as well as the areas of the turns. The flat coil has its turns more nearly occupying the same space and the average area of a turn is greater.

Remember that the test was specified as both coils be wound the same. It seems Jazzmasters had more turns than Strats. The Jazzmaster pickup has 8500 turns. Jaguar pickups also have 8550 turns. They are very different sounding, but then Jaguars also had that steel comb on them. I see some Strat pickups listed as 8,000 turns. maybe someone who knows more about Fender pickups will chime in.

I'll have to do some tests.

So the tall thin coil is expected to have less inductance than the flat one, but the cores can make this a more complicated problem, altering the difference since they can help connect the flux from one turn to another, and increase the effective area of a turn. But the effect of the rod cores used in pickups is a lot less than closed cores such as toroids or the common rectangular types. So one needs to make measurements. I would expect the flatter coil to have a somewhat higher inductance and a higher resistance leading to a somewhat lower frequency and less peaked resonance. This is the kind of difference that should be audible to someone familiar with electric guitar sounds.

Unfortunately the available pickups of various shapes are also not wound the same. So is the difference in tone from more windings, shorter magnets, or coil shape, or a combination of the above?