Humbuckers, Single Coils and playing dynamics

[ruger9]

In Julian Lage's own words:

He talks about the pickups (and how they built the guitar around these) starting at @0:35 and elaborating more about how it can "handle" information in different registers and helping with his "orchestral" approach starting @3:20 – how the instrument is able to handle his blend of melodic lines and comping.
I know it's an advertising ;-) But he elaborately tells what he wanted from the instrument and how they achieved that. An instrument that is able to translate his dynamic playing style.

I can certainly attest to the fact that some guitars/pickups are better at it than others. So it's not just the player, of course.

[KirkP]

…
First of all, the phenomena to which I refer are not all linear. For example, back EMF is neither constant nor linearly related to signal strength. In fact, it is zero when the string is first struck - so the coil starts to put out the full driving voltage as the tone begins. But as the string vibrates, the back emf grows and opposes the signal. This then reduces both the voltage across the coil and the amount of current it delivers. So the initial attack will be unopposed. But before it can reach its peak, it will start to be attenuated. Even 1 dB at the pickup would be quite noticeable at the speaker, as it would be amplified by the gain of the amplifier. This could certainly cause some true compression by limiting peak voltagw of the signal. As pickups may generate more or less back emf, depending on their design, some pickups will display more attenuation of peak voltage on the attack than others. It doesn't take much to achieve audibility.

The effect of capacitance on frequency-spcific signal passage is also not linear. A pure, simple first order low pass filter (which is what a simple capacitor is) has a transfer function and a phase effect like this:

Attachment 90402

Couple that nonlinearity with the other factors in an RC filter (e.g. inductance, resistance and capacitance in the wiring, inductance in the load etc) and you have a complex function that's definitely not linear. Those phase shifts can combine with the signal to remove an audible part of the spectrum and cause volume drops. This is also compression.

I don’t think you are using the correct definition of linearity. Linearity doesn’t mean having a flat frequency response. Linearity means that if you increase the amplitude of a signal or combine two or more signals in the input, the output’s response will be exactly proportionate.

One of many tutorials online:
The Main Difference between Linear and Nonlinear Circuit

None of the components of a magnetic pickup are non-linear within the range of inputs from a vibrating string. It’s true that magnets can become non-linear if they approach saturation, but I don’t think a vibrating string can cause that. Therefore, any non-linearity of of the response must be due to something before or after the pickup circuit in the signal chain.

The usual downstream non-linearities are distortion and compression pedals, tube saturation, and overdriven speakers. The only upstream non-linearity is the vibration of the string itself. Someone already mentioned that the response of the string will be non-linear if it vibrates against the frets, but I think the compliance (spring action) of a string is also non-linear, causing a non-linear response to picking. Wood has a non-linear response to vibration, so the body of a guitar will contribute to the non-linear response of the string. That’s why even non-amplified guitars are often described as having compression.

When pickup designers describe a passive magnetic pickup as having “compression” I don’t put much weight on it. A high output pickup can increase compression downstream in the signal chain, but it’s not the pickup itself that is non-linear. I think the pickup design may also affect how we perceive non-linearities in the string vibration. For example, a narrow pickup (e.g. Strat) responds to small length of the string while a wide pickup (e.g., traditional humbucker) responds to more of the string. That’s part of the reason narrow pickups tend to be brighter. I have a hunch that the the non-linearities in string vibration affect low frequencies more than high frequencies. If that’s true a brighter pickup may be perceived as less compressed, and therefore having more “dynamics” (a word that seems to have many interpretations to musicians).

I don’t think the OP was looking for a technical essay, but since things went that direction I couldn’t resist. And I do think understanding how each element of the signal chain contributes can help us achieve the sound we are after.

[burchyk]

Don't know if this is just a perception, but I do feel that single coils are less compressed than P90 which are less compressed than humbuckers.. Is it an illusion due to inherently different tonal balance? Maybe..

There is more to it than the pickup type - amp headroom, string gauge, action and pickup height, pick thickness and material, picking angle and position, and ultimately right hand "weight". All of these affect dynamics more than the pickup type and most can be adjusted on the fly. So just choose the pickup type you like the base sound of and learn to control the rest.

[RJVB]

Trying to understand here...

A humbucker is a dual-coil system designed so both coils pick up the same noise which is then cancelled out when merging the outputs because in counterphase? I don't really see why the "other" coil can capture the string vibration signal in counterphase while the noise is captured in the same phase. More importantly: why would this change the sound character as compared to how the individual coils would sound? Is that because you can't make a humbucker by associating two individually usable single-coils, or is that because of the physical separation of the coils?

If the latter, that begs the question if you could get the same noise-canceling effect by using concentric windings (one CW, the other CCW for instance)?

[rpjazzguitar]

Trying to understand here...

A humbucker is a dual-coil system designed so both coils pick up the same noise which is then cancelled out when merging the outputs because in counterphase? I don't really see why the "other" coil can capture the string vibration signal in counterphase while the noise is captured in the same phase. More importantly: why would this change the sound character as compared to how the individual coils would sound? Is that because you can't make a humbucker by associating two individually usable single-coils, or is that because of the physical separation of the coils?
If the latter, that begs the question if you could get the same noise-canceling effect by using concentric windings (one CW, the other CCW for instance)?

One solution for a 3 single coil Strat humming is to use neck and middle together. If they're wired properly (out of phase?), the hum disappears. The tone changes.

As far as the rest of it, the engineering is over my head. I understand enough of it to believe that the systems are nonlinear. Thinking it over, the only time I've noticed non-linearity (apart from overdriven signals) is that it seems like plucking more gently on the highest notes of the instrument might be louder (or maybe nearly as loud) than plucking hard. I've always assumed that was because the string was fretting out if I picked too hard; that is, something mechanical, not electrical. I've never noticed this on lower notes. I think some guitars are worse than others, and, very likely, action height has a lot to do with it.

[burchyk]

(somebody who properly knows electronics please correct)

A humbucker requires two coils that are in opposite phase, but also have the opposite magnetic polarity. That way the signal generated by the string is opposite on the coils which then flip the direction and effectively sum it. Hum has the same polarity on both coils (not dependent on the magnet polarity) and thus gets negated by the opposite coil winding.

Concentric winding does indeed work and it's commonly called a "stack". Since there is no reverse magnetic polarity it negates the string signal except the "difference" between the coil that is closer to the string and the one that is further.

One could also do the RWRP "split" like a P-Bass or some G&L guitar pickups.

[nevershouldhavesoldit]

I don’t think the OP was looking for a technical essay, but since things went that direction I couldn’t resist. And I do think understanding how each element of the signal chain contributes can help us achieve the sound we are after.

I greatly appeciate your thoughtful response and your analysis. And I suspect you're right that the OP didn't have a technical explanation in mind. But he did ask "why?" -

humbuckers seem to compress more, and a large humbucker more then mini humbuckers – which is sometimes welcome, other times not so much....What do you think? Do single coils translate playing dynamics better then humbuckers in general? Which do you prefer and why?

and he got a few answers from which to choose, ranging from "pickups don't compress at all" to the opposite extreme.

I don’t think you are using the correct definition of linearity.

FWIW, linear simply means "in a straight line". Mathematically, a linear function is any function that can be expressed in the form y = mx + b. In this thread, most of us are talking about output voltage as a function of picking force. Let's assume we're talking about a perfectly aligned pickup (completely homogeneous magnetic field around an ideal string lined up perfectly to the axis of the magnet magnetic field etc). Let the voltage be y and force be x. There's always some noise from a pickup, so b is where the output voltage intercepts the y axis with no picking force applied. And m is the slope of the line, which reflects losses in the system that prevent a perfect 1:1 relationship between force and voltage. This function will be perfectly linear no matter how much loss there is, as long as the loss is a consistent percentage of the input. It'll be a straight line that rises from the y intercept at less than 45 degrees.

If you look at the magnetic fields around the strings from different pickups, you'll see that they're all different in size, shape, flux density etc. The position of the string over a polepiece (or two, or a blade, etc) changes the effect of its vibration on output, as does the angle of the string's axis relative to the magnetic axes. The baseplate on a Tele bridge pickup widens the field dramatically along its surface, but it the tapers to a narrower peak over the strings than it would without the baseplate. So even a minor increase in string or pickup height can push a portion of the vibrating envelope out of the field at its widest point when struck hard. This will compress the output because the horizontal string vibration is not all within the field and therefore doesn't generate full output despite high picking force.

There's no end to the factors at play here. But the bottom line is that many of them render pickup output as a function of picking force mathematically nonlinear. Even the ratio of horizontal to vertical vibration of a string (which depends in large part on how you pick) and where along the string you strike it affect both the amplitude and the frequency spectrum of the output signal. Some of these are linear when measured and some are not. Some are probably expressible as quadratic equations and some require truly complex math well beyond my understanding or interest.

I think there's ample audible, mathematical and physical support for the statement that some pickups on some guitars played by some players have a progressive reduction in the rate of rise of output voltage for a given increase in picking force as that force increases toward maximum. I've tried to provide support for this point of view. I love learning more than I knew yesterday, and I thank everybody for the time and effort they put into the discussion. I'm still learning, so I look forward to more input.

To the OP, I think the answer is that some humbuckers compress more than some single coils, and vice versa. Some of each display no audible compression and a few give the impression that external compression is being applied somewhere because their attack doesn't get much louder beyond a certain point if picked harder, but they sustain longer. The Duncan Quarter Pounder in my Kubicki Express just keeps putting out more and more the harder I pick. With 1/4" diameter Alnico 5 rods, it's probably magnetizing my watch when I play it. On the other hand, my old Danelectro Convertible (Alnico 6 bar magnet, 42 gauge wire lipstick pickup) had a limited range of volume available from picking style and strength - it was best for slide, with its volume pot all the way up and my Morley volume pedal controlling my sound level.

[RJVB]

One solution for a 3 single coil Strat humming is to use neck and middle together. If they're wired properly (out of phase?), the hum disappears. The tone changes.

It's clear why the tone changes here: the 2 PUs don't have the same tone in isolation so adding them should give some kind of average.

However:
If you add them together but in counterphase you will surpress all common sounds and end up with something that is a representation of the sound difference between the neck and middle PUs. Probably not what you want (also why mic preamps have a phase switch function). I will presume that the guitar is set up such that you can select any combination of PUs and NOT get this cancellation effect on the string sound. It beats me why that would cancel noise though; there's no guarantee that the 2 PUs are subjected to the same interference or (if they are) that the captured noise signals are going to be in counterphase.

Concentric winding does indeed work and it's commonly called a "stack".

"Does work" as in cancels are attenuates hum? How about the sound character compared to a single coil?

(Aren't the coils perpendicular to the strings, meaning the "distance to string" of 2 concentric coils isn't always greater for the one or the other when the string vibrates?)

[sgosnell]

Don't know if this is just a perception, but I do feel that single coils are less compressed than P90 which are less compressed than humbuckers..

To be pedantic, P90s are single-coil pickups.

[KirkP]

(somebody who properly knows electronics please correct)

A humbucker requires two coils that are in opposite phase, but also have the opposite magnetic polarity. That way the signal generated by the string is opposite on the coils which then flip the direction and effectively sum it. Hum has the same polarity on both coils (not dependent on the magnet polarity) and thus gets negated by the opposite coil winding. ….

That’s basically right. I think an engineer would say opposite polarity but opposite phase is what most people call it. Traditionally the two pickups are connected in series and wired such that the signal from string vibration is in phase while externally induced noise is out of phase. Since the pickups are wired in series the voltage is doubled, so it’s higher output than a single coil. The impedance is also doubled, which gives a different frequency response from a single coil (e.g., different resonance frequency, different high frequency roll-off, different interaction with volume and tone pots). The two pickups are also sampling the string at different locations, so they aren’t sensing identical signals, which also changes the voice compared to a single coil. If you connected the two humbucking coils in parallel you’d no longer be doubling the voltage and impedance, so I think it would sound a bit more like a single coil. (I hope someone will correct me if I’m wrong.)

I think humbuckers are most effective at cancelling noise induced by strong magnetic fields, especially the low frequency hum produced by power transformers. They obviously can only cancel noise induced on the pickup coils. A humbucker will have no effect on noise that enters via the guitar’s wiring, which is usually higher frequency buzzes from dimmers, LEDs, poorly filtered wall warts, etc. Most of those devices didn’t exist in the early days of guitar, so they didn’t worry as much about shielding and grounding, but we sure need to now! I think modern power transformers are better designed to limit radiation than they used to be. So these days, I think pickups should be selected based on how they sound, not the humbucking feature. Then follow best practice in grounding & shielding.

[RJVB]

Traditionally the two pickups are connected in series and [...] which also changes the voice compared to a single coil. If you connected the two humbucking coils in parallel you’d no longer be doubling the voltage and impedance, so I think it would sound a bit more like a single coil. (I hope someone will correct me if I’m wrong.)

I'd be surprised if the output from the 1st coil isn't "coloured" by its passage through the 2nd beyond the increase of impedance.

From what I can see the wiring on my archtop uses shielded cables, but the connections are simply out in the open. That always surprised me, until I learned that the PU itself is probably more likely to capture noise...

Maybe it's just what you get when you try to play guitar on a theremin

[KirkP]

I'd be surprised if the output from the 1st coil isn't "coloured" by its passage through the 2nd beyond the increase of impedance.

From what I can see the wiring on my archtop uses shielded cables, but the connections are simply out in the open. That always surprised me, until I learned that the PU itself is probably more likely to capture noise...

Maybe it's just what you get when you try to play guitar on a theremin

Since the two coils are so close together they are undoubtedly magnetically coupled to each other, so I agree there is more to it than impedance.

I don’t think it’s a big deal if there are a few unshielded connectors, as long as the cables and pickups are properly shielded and grounded. We are trying to keep the wiring from acting as a receive antenna, and antennas need to have some length to be excited—a significant fraction of a wavelength, which is pretty long at audio frequencies.

I’ve got an electromagnetics background but I won’t claim expertise in shielding and grounding. I’m working on hunches informed by some reading plus knowledge of EM basics. But analyzing shielding and grounding is tricky and my intuition on the topic could fail me. I think I’m right about this one though.

[John A.]

I think humbuckers are most effective at cancelling noise induced by strong magnetic fields, especially the low frequency hum produced by power transformers. They obviously can only cancel noise induced on the pickup coils. A humbucker will have no effect on noise that enters via the guitar’s wiring, which is usually higher frequency buzzes from dimmers, LEDs, poorly filtered wall warts, etc. Most of those devices didn’t exist in the early days of guitar, so they didn’t worry as much about shielding and grounding, but we sure need to now! I think modern power transformers are better designed to limit radiation than they used to be. So these days, I think pickups should be selected based on how they sound, not the humbucking feature. Then follow best practice in grounding & shielding.

IME, humbuckers are effective against dimmers and, LED's etc., though not always completely so because of issues with shielding of other wiring. I say this based on playing humbucker- and SC-equipped guitars in spaces where there's lots of buzz from the lighting. Humbuckers are much quieter. My experience is that there are environments where SC pickups are unusable, so I don't agree with idea that this feature doesn't matter anymore. There were lots of EMI emitting things around by the time electric guitars came into wide use (e.g., flourescent and neon lights, electric fans, TV sets). Humbuckers met a need then, and still do. I don't know the EE explanation for this, but I do know empirically that humbuckers are definitely not obsolete.

[JCat]

Here's a perspective on playing dynamics.
A fresh outlook for the electric guitar player:

[KirkP]

IME, humbuckers are effective against dimmers and, LED's etc., though not always completely so because of issues with shielding of other wiring. I say this based on playing humbucker- and SC-equipped guitars in spaces where there's lots of buzz from the lighting. Humbuckers are much quieter. My experience is that there are environments where SC pickups are unusable, so I don't agree with idea that this feature doesn't matter anymore. There were lots of EMI emitting things around by the time electric guitars came into wide use (e.g., flourescent and neon lights, electric fans, TV sets). Humbuckers met a need then, and still do. I don't know the EE explanation for this, but I do know empirically that humbuckers are definitely not obsolete.

My theory can’t trump your experience. ;-) I’ll note that typical Humbuckers (at least those with metal covers) are better shielded than Tele & Strat type single coils pickups and have shielded cables. But you’re right that any noise picked up by the coils rather than the wiring should be cancelled by a humbucker.

[KirkP]

Thursday I saw Julian Lage with his trio in Dortmund. Great concert.

From a gear perspective it was stunning to hear him perform the whole dynamic range from whisper quiet to really loud WITHOUT ever touching the controls on his guitar – a Gretsch-shaped Collings with "Ellisonic" single coils. I contribute most of this to his virtuosic and very dynamic playing. But I imagine that also the pickups provide the basis for translating his huge dynamic range through the ampification.
Personally I go back and forth between the single coils in my tele and ES 330 (P90s so very different construction) and the humbuckers in my Ibanez GB. I imagine that I can hear a difference in the translation of the playing dynamics – the humbuckers seem to compress more, and a large humbucker more then mini humbuckers – which is sometimes welcome, other times not so much.

What do you think? Do single coils translate playing dynamics better then humbuckers in general?
Which do you prefer and why?

Your hunch that the width of the coil has an effect on perceived dynamics might be right, if by “dynamics” you mean being able to hear the effects of small variations in how you pick or fret. The fundamental tone is an overall motion of the entire string. Harmonics are vibrations of sections of the string. A wider coil may sense both the positive and negative lobes of higher order harmonics, so may have less of an ability to accurately reproduce them. A lot of our playing technique (e.g., picking style) is to control the higher order harmonics, so perhaps a wide coil can mush some of that out. Rolling off the treble control has a similar effect, but I suspect the effect of a wide coil is more complex than that.

[John A.]

My theory can’t trump your experience. ;-) I’ll note that typical Humbuckers (at least those with metal covers) are better shielded than Tele & Strat type single coils pickups and have shielded cables. But you’re right that any noise picked up by the coils rather than the wiring should be cancelled by a humbucker.

FWIW, my strat has a shielding on the pickguard and the cavity is painted with conductive paint, and my humbucker guitars are a semi-hollow and a fully hollow archtop, neither of which has that kind of shielding. I don't know exactly what difference these things make, though.