|
Post by JohnH on Jul 11, 2017 4:22:30 GMT -5
Thanks. That seems like its quite insightful
|
|
|
Post by ms on Jul 11, 2017 6:27:16 GMT -5
This supports a growing suspicion in the back of my mind, that the humbucking design does have a distinct character of its own, whereas you could consider a single coil design as actually having less character or colouration, at least if it is constructed and loaded in a certain way. I'm trying to lean toward a design that offers maximum flexibility and so that character of a humbucker (which I think we are attributing to phase cancellations here) is a liability. The same character is a benefit not a liability if you are actually seeking a classic sound character that already exists - e.g. a PAF sound. It just irks me because I'm also a stickler for noise reduction, which bring a level of complexity to the table when single coils are required. Again, from my point of view, and with my jaded ears, the difference is not as much as I expected. It could explain why my Firebird pickups don't sound quite the same as singles in spite of many similarities that they have in common - magnetic poles, no steel, low losses resulting in very high loaded Q. Yes. A test set up to allow switching between both coils of a humbucker and just one of them, with circuitry to keep the resonant frequency and Q as much the same as possible, shows just that. The biggest difference is on the low E string, and is what this plot (http://guitarnuts2.proboards.com/attachment/download/1108) implies, a loss of definition in the higher harmonics. The effect gets less noticeable moving to the higher strings, becoming unimportant. That is, the hole in response moves from a very sensitive region of human hearing within the range of the guitar system, to a less sensitive region out the range of the guitar electronics.
|
|
|
Post by ms on Jul 11, 2017 6:32:21 GMT -5
I'm trying to lean toward a design that offers maximum flexibility and so that character of a humbucker (which I think we are attributing to phase cancellations here) is a liability. The same character is a benefit not a liability if you are actually seeking a classic sound character that already exists - e.g. a PAF sound. It just irks me because I'm also a stickler for noise reduction, which bring a level of complexity to the table when single coils are required. If you want humbucking and minimal comb filtering, I think the solution exists in tiny humbuckers, such as Bardens. I'll be the first to admit I wouldn't mistake a Barden for Fender set, but that's a mystery unto itself. Same for Lace Sensors; if it's not comb filtering that make them sound different, then what is it? The stronger magnetic pull of the AlNiCo pole pieces? I think the solution exists in single coil sized hum buckers using multiple coils. It is just my opinion, but I think the pickups described in the HiQ discussion solve essentially all the problems, and with some further development, will do the job completely.
|
|
|
Post by reTrEaD on Jul 11, 2017 10:56:28 GMT -5
reTrEaD mentioned on page 1 the fact that a note will have different harmonics depending on where along the fret board it's played That's close to what I meant. There is likely to be a small difference in the overall harmonic content depending on the length and diameter of the string. But I was referring to the relationship of the pickup sensing window to the vibrating section of the string. For instance: if the pickup sensing window is centered at 1/4 of the vibrating length, it will be directly under a node of fourth harmonic. So we would expect only a tiny fraction of whatever fourth harmonic that is present on the string to be present in the electrical signal the pickup produces. (None at all if our sensing window is infinitely narrow.) Likewise, the pickup window is sitting directly under the antinode of the second harmonic. So we would expect to see all of the second harmonic content present on the string represented in our electrical signal. In fact a greater amplitude relative to the fundamental than is actually present. Since we're sampling the fundamental between it's antinode (midpoint of the vibrating length) and a node (terminus of the string), we have a less than all of the fundamental that's actually present.
|
|
|
Post by antigua on Jul 11, 2017 13:14:03 GMT -5
|
|
|
Post by ms on Jul 11, 2017 16:00:10 GMT -5
This supports a growing suspicion in the back of my mind, that the humbucking design does have a distinct character of its own, whereas you could consider a single coil design as actually having less character or colouration, at least if it is constructed and loaded in a certain way. I'm trying to lean toward a design that offers maximum flexibility and so that character of a humbucker (which I think we are attributing to phase cancellations here) is a liability. The same character is a benefit not a liability if you are actually seeking a classic sound character that already exists - e.g. a PAF sound. It just irks me because I'm also a stickler for noise reduction, which bring a level of complexity to the table when single coils are required. Again, from my point of view, and with my jaded ears, the difference is not as much as I expected. It could explain why my Firebird pickups don't sound quite the same as singles in spite of many similarities that they have in common - magnetic poles, no steel, low losses resulting in very high loaded Q. Yes. A test set up to allow switching between both coils of a humbucker and just one of them, with circuitry to keep the resonant frequency and Q as much the same as possible, shows just that. The biggest difference is on the low E string, and is what this plot (http://guitarnuts2.proboards.com/attachment/download/1108) implies, a loss of definition in the higher harmonics. The effect gets less noticeable moving to the higher strings, becoming unimportant. That is, the hole in response moves from a very sensitive region of human hearing within the range of the guitar system, to a less sensitive region out the range of the guitar electronics. Here is a link to a discussion of this: music-electronics-forum.com/t26893/
|
|
|
Post by antigua on Jul 11, 2017 16:32:16 GMT -5
Based on these sound samples, I think I'd describe the series humbucker versus split as being a combination of two things, 1) attenuation due to the lower resonant peak, and 2) notch filtering that occurs northward of 2.5kHz. Doing one thing or the other get's you half way there, but doing both gets you nearly all the way. In general, that notch filtering come across as a "quack", where a Strat with it's pickups far apart, or the bridge and and neck pickups used together demonstrate extreme degrees of notch-like filtering as a result of comb filtering. A humbucker, with its coils fairly close together, is more mildly effected, and at higher frequencies, so if the in-between positions are full on flavor, you could say a humbucker is "lightly seasoned" by contrast. The Tillman caclutor more or less agrees with the observations here, the www.till.com/articles/PickupResponseDemo/, as the frequency gets higher, the notches are more finely spaced, and when you include two pickups into the demonstration, moving those pickup farther apart results in the notches moving to lower frequencies. Notch filtering through other means, such as splitting a humbucker with a capacitor (resulting in more than one resonance), also results in quack-like tones. If the humbucker were hotter, such as a Seymour Duncan JB, it's likely that the resonant peak would be so low that nearly all of the comb filtering would be above the resonant cut off, and so you're only going to get that peculiar voicing if the PAF is "vintage wound", which might explain some of the preferences guitarists have, in that regard.
|
|
|
Post by stratotarts on Jul 11, 2017 17:57:51 GMT -5
If you want humbucking and minimal comb filtering, I think the solution exists in tiny humbuckers, such as Bardens. I'll be the first to admit I wouldn't mistake a Barden for Fender set, but that's a mystery unto itself. Same for Lace Sensors; if it's not comb filtering that make them sound different, then what is it? The stronger magnetic pull of the AlNiCo pole pieces? I think the solution exists in single coil sized hum buckers using multiple coils. It is just my opinion, but I think the pickups described in the HiQ discussion solve essentially all the problems, and with some further development, will do the job completely. I was very impressed with that. It's just that I'm worried about manufacturability. It has a lot of parts that have to go together. Regarding single-coil sized humbuckers, I've had and liked them, but I wonder whether they need to be close to the string to work properly. I like pickups that you can back away from the string. There are a lot of trade offs to consider.
|
|
|
Post by antigua on Jul 12, 2017 1:22:56 GMT -5
I think the solution exists in single coil sized hum buckers using multiple coils. It is just my opinion, but I think the pickups described in the HiQ discussion solve essentially all the problems, and with some further development, will do the job completely. I was very impressed with that. It's just that I'm worried about manufacturability. It has a lot of parts that have to go together. Regarding single-coil sized humbuckers, I've had and liked them, but I wonder whether they need to be close to the string to work properly. I like pickups that you can back away from the string. There are a lot of trade offs to consider. It would be interesting to find that the only thing stopping a single coil sized humbucker or a Lace Sensor from sounding "authentic" is a lack of string pull. If that's the case, neodymiums might make the difference, causing pleasant harmonic disruptions, short of causing "Stratitus", but being strong enough to induce a subtle chorus-like effect, especially when playing chords, and you have harmonic beating happening at the same time. I have measured Tele Bardens and Strat Lace Sensors, and found that both have loaded Q factors that are "single coil-like", +5dB loaded, around +10dB unloaded, so I think that can be ruled out as a key difference. My money is on magnetic pull. We might be able to test this theory with blind listening tests, using neo buttons to alter the flux.
|
|
|
Post by antigua on Jul 12, 2017 3:34:05 GMT -5
Here is another test of the same format as previously, this time I removed a coil of a humbucker, it happened to be the one closer to the neck: For this test the humbucker is in parallel mode the entire time in order to get an extended harmonic response. In the sound file there are four strums: 1) parallel, both coils in place 2) coil removed but in in circuit, harmonics altered to match the first strum 3) removed coil, with simple attenuation at 5kHz 4) coil removed, unchanged The amplitude of the lone coil is also normalized with the full parallel humbucker, because with one coil removed, the voltage production dropped by about half. www.echoesofmars.com/misc/neck_hb_G_chord_coil_moved_aside_composite.wavBased on the plot above, when the coil was removed (4th sample), the response reaches a lot higher, because even though the removed coil was still in circuit, the inductance dropped and the resonance increased. The 3rd sample is a simple attenuation above 5kHz, so it's closes to what would happen if the inductance had remained the same. The second sample, the harmonically altered sample, shows how close the two become (both coils in place, versus one removed) if the harmonics are modified individually. In terms of overall tone they all sound very alike, but then the first sample, the intact humbucker, has a richness to the initial strum it that the single coil doesn't have. Altering the harmonic levels (#2) didn't entirely endow it with that richness of the intact pickup (#1). With one coil removed, the sounds seem thinner, more sterile, for a lack of technical terms. The richness seems to only be apparent right at the very beginning, within the first second. The remaining four seconds of each sample sound fairly alike, so I suspect the cause of the "richness" is transient related. Another difference of note is that having one of the coils removed reduced the string pull somewhat, so it's possible that the rich transient sound is related to magnetic string pull. There seems to be some comb filtering at 3.4kHz and 4.6kHz, and in a couple instanced the amplitude of the harmonic appeared to be higher with the parallel coils, so it doesn't seem to be strictly subtractive. In the previous tests it seemed that the modified harmonics did more to make one sample sound like the other, and I think the key difference in this case is that the remaining coil is the one that is closer to the bridge, where as in the series/split test, it split to the coil that is closer to the neck. What this tells me is that the comb filtering effects that make the tonal difference here are not caused by dual coil comb filtering, but rather are caused by the more simple fact that the sum of harmonics is coming from closer to the bridge of the guitar. It's hard to say what it could be, since the transient strum is different then the decay in several ways, such as the fact that the strings displace the most, and In other words, why does a humbucker neck sound different than a single coil neck? Because the humbucker extends in the direction of the bridge. Most people think of single coil neck tones as a Tele neck or a Strat neck. What do they have in common? They put the pickup right up against the neck. I suspect that if you modified Teles and Strats such that the neck pickup was moved one third of an inch towards the bridge, people would think the humbucker and hot single coil sound sound a lot more alike. The same is probably true of a HSS versus an SSS, where the SSS has a single coil bridge that is overwound to match the inductance of a humbucker; even though it's placed at an angle, the side of the pickup that reads the treble strings is more towards the bridge. Placing a humbucker in the bridge picks up more harmonic content further away from the bridge. Of course, this doesn't explain it 100%, because the intact humbucker had that rich transient that disappeared once a coil was removed from under the strings. I'll try to think of a test setup that can make this "rich" transient appear and disappear with as few changes as possible, but it might be difficult, not knowing what the "richness" really is. Maybe modifying the amount of string pull will yield something. I can say anecdotally that when I set a pickup closer to the string, it also has a "richer" sound, though I'm not positive it's this "richer" sound, but they do have one thing in common; more string pull = more of this richness.
|
|
|
Post by wgen on Jul 12, 2017 8:13:59 GMT -5
Here is another test of the same format as previously, this time I removed a coil of a humbucker, it happened to be the one closer to the neck: For this test the humbucker is in parallel mode the entire time in order to get an extended harmonic response. In the sound file there are four strums: 1) parallel, both coils in place 2) coil removed but in in circuit, harmonics altered to match the first strum 3) removed coil, with simple attenuation at 5kHz 4) coil removed, unchanged The amplitude of the lone coil is also normalized with the full parallel humbucker, because with one coil removed, the voltage production dropped by about half. www.echoesofmars.com/misc/neck_hb_G_chord_coil_moved_aside_composite.wavBased on the plot above, when the coil was removed (4th sample), the response reaches a lot higher, because even though the removed coil was still in circuit, the inductance dropped and the resonance increased. The 3rd sample is a simple attenuation above 5kHz, so it's closes to what would happen if the inductance had remained the same. The second sample, the harmonically altered sample, shows how close the two become (both coils in place, versus one removed) if the harmonics are modified individually. In terms of overall tone they all sound very alike, but then the first sample, the intact humbucker, has a richness to the initial strum it that the single coil doesn't have. Altering the harmonic levels (#2) didn't entirely endow it with that richness of the intact pickup (#1). With one coil removed, the sounds seem thinner, more sterile, for a lack of technical terms. The richness seems to only be apparent right at the very beginning, within the first second. The remaining four seconds of each sample sound fairly alike, so I suspect the cause of the "richness" is transient related. Another difference of note is that having one of the coils removed reduced the string pull somewhat, so it's possible that the rich transient sound is related to magnetic string pull. There seems to be some comb filtering at 3.4kHz and 4.6kHz, and in a couple instanced the amplitude of the harmonic appeared to be higher with the parallel coils, so it doesn't seem to be strictly subtractive. In the previous tests it seemed that the modified harmonics did more to make one sample sound like the other, and I think the key difference in this case is that the remaining coil is the one that is closer to the bridge, where as in the series/split test, it split to the coil that is closer to the neck. What this tells me is that the comb filtering effects that make the tonal difference here are not caused by dual coil comb filtering, but rather are caused by the more simple fact that the sum of harmonics is coming from closer to the bridge of the guitar. It's hard to say what it could be, since the transient strum is different then the decay in several ways, such as the fact that the strings displace the most, and In other words, why does a humbucker neck sound different than a single coil neck? Because the humbucker extends in the direction of the bridge. Most people think of single coil neck tones as a Tele neck or a Strat neck. What do they have in common? They put the pickup right up against the neck. I suspect that if you modified Teles and Strats such that the neck pickup was moved one third of an inch towards the bridge, people would think the humbucker and hot single coil sound sound a lot more alike. The same is probably true of a HSS versus an SSS, where the SSS has a single coil bridge that is overwound to match the inductance of a humbucker; even though it's placed at an angle, the side of the pickup that reads the treble strings is more towards the bridge. Placing a humbucker in the bridge picks up more harmonic content further away from the bridge. Of course, this doesn't explain it 100%, because the intact humbucker had that rich transient that disappeared once a coil was removed from under the strings. I'll try to think of a test setup that can make this "rich" transient appear and disappear with as few changes as possible, but it might be difficult, not knowing what the "richness" really is. Maybe modifying the amount of string pull will yield something. I can say anecdotally that when I set a pickup closer to the string, it also has a "richer" sound, though I'm not positive it's this "richer" sound, but they do have one thing in common; more string pull = more of this richness. Very interesting, thank you very much for the efforts...It would be very interesting understanding what that richness is. I'm also playing with a guitar that I modified with a overwound humbucker (at the bridge in my case) wired in parallel though, so that's always nice and bright. I had a single coil there before, and it was a s----y Duncan Hot ssl-3, overwound and low resonant peak I'd bet. No matter what the sound now seem to have a little richer quality than when I had the single coil. Before it had a more "twangy" response, a dull twang but still that kind of more "vintage vibe". Now it lost much of that twang, even if the sound is way more clear than before. Could it be that a more extended string pull across the strings of a humbucker is kind of similar to when you change strings for thicker gauges, so that the movement of the string is more controlled and less of the transient is allowed? I really don't know how to check frequencies, but I experimented a lot with change of string gauges in the past and I found that if you change 9 for 11-12 string gauges for example, you're going to hear the difference: less twang and more apparent low mids, which could simply be more lower order harmonics...but it's just a shot in the dark
|
|
|
Post by antigua on Jul 12, 2017 13:12:39 GMT -5
Very interesting, thank you very much for the efforts...It would be very interesting understanding what that richness is. I'm also playing with a guitar that I modified with a overwound humbucker (at the bridge in my case) wired in parallel though, so that's always nice and bright. I had a single coil there before, and it was a s----y Duncan Hot ssl-3, overwound and low resonant peak I'd bet. No matter what the sound now seem to have a little richer quality than when I had the single coil. Before it had a more "twangy" response, a dull twang but still that kind of more "vintage vibe". Now it lost much of that twang, even if the sound is way more clear than before. Could it be that a more extended string pull across the strings of a humbucker is kind of similar to when you change strings for thicker gauges, so that the movement of the string is more controlled and less of the transient is allowed? I really don't know how to check frequencies, but I experimented a lot with change of string gauges in the past and I found that if you change 9 for 11-12 string gauges for example, you're going to hear the difference: less twang and more apparent low mids, which could simply be more lower order harmonics...but it's just a shot in the dark I can test the magnetic pull issue by doing the test again by leaving the coil there, then just splitting to the bridge-side coil, then doing the harmonic modifications again, and seeing how similar they sound that way. The purpose of keeping the coil in circuit was to retain the same RLC values, but it appears that didn't happen anyway, the inductance dropped by some amount. The SSL-3 appears to be a super hot overwound Strat pickup, 15k ohms, probably using 44 AWG, so I think the SSL-3 is liable to "destroy" a lot harmonic information, I'd guess from around 3kHz and up. I don't believe stronger string pull is similar to thicker, stiffer strings. I can see why you'd suspect that, but what I found it actually does is dampen harmonics, the ones whose anti-nodes are most centered above the pickup. It's similar to pinch harmonics, but you're not deadening the harmonics completely like you do with your finger, you're just sucking some amount of energy away from them, as though you were damping it with a feature, which leads to more complicated intermodulations, which I think accounts for the "mild chorus" effect. I'm not sure that the "richness" quality is related to that, hopefully more testing till help flesh it out more. This is a reason why damping would disproportionately effect the transient; the string is at peak displacement immediately after excitation, and comes closes to the pickup's magnetic field at that time, and as it decays, is suspended at a further distance from the magnetic field. I should try damping the strings with a feather too, maybe it would make something similar happen.
|
|
|
Post by ms on Jul 12, 2017 20:21:45 GMT -5
I think the solution exists in single coil sized hum buckers using multiple coils. It is just my opinion, but I think the pickups described in the HiQ discussion solve essentially all the problems, and with some further development, will do the job completely. I was very impressed with that. It's just that I'm worried about manufacturability. It has a lot of parts that have to go together. Regarding single-coil sized humbuckers, I've had and liked them, but I wonder whether they need to be close to the string to work properly. I like pickups that you can back away from the string. There are a lot of trade offs to consider. They are a bit difficult to build, but perhaps parts that make it simpler could be found. I do not see any reason why the same partial completion of the magnetic circuit would not work with narrow hum bucker coils, allowing reasonably high output in an SC size. The two coils would go close to the center, and each would "complete" with ferrite going up the corresponding side (away from the center line). You would need to use the thin ferrite plates in order to leave as much space as possible for the coils. For the same reason, a ferrite blade would be a good core, allowing more wire than the standard size cylindrical pole pieces. I find it difficult to get as high a wire density in a long coil as in a cylindrical one, but it can be done and is necessary in order to get the high output. But I do like the six coil pickups where the polarity changes from coil to coil; the hum rejection is so good. Next I will try this for a neck pickup where both the polarity and the location along the string change from coil to coil.
|
|
|
Post by antigua on Jul 15, 2017 12:55:28 GMT -5
Here's an observation I just noticed while setting up a test, it somewhat relates to this experiment, and one I don't think I've seen made before: if you split a humbcuker, the second "coil" is still partially active, not the electrically, but magnetically. You can witness this by carefully tapping on the pole pieces of the secondary coil of a split humbucker. You will hear that the primary very loud, and active, but that the secondary "out of circuit" coil is still producing a voltage with you tap the tops of it's pole pieces. Even though the coil itself is out of circuit, the permeable cores are still in the "magnetic circuit", and the strings above those pole pieces are still being magnetized and producing a time varying flux, a signal, that reached the active primary. Since the string and pole pieces of the unused coil are still producing almost the exact same magnetic field as they would if the secondary were in circuit, the contribution the secondary's magnetic field makes to the primary coil is the same as it would be if both coils were in circuit. The only real difference is that a much smaller proportion of the secondary's magnetic field is reaching the primary than would have reached its own. This same principle of producing magnetism on one side and "reading" it on the other is somewhat at work in a HiloTron design, the difference there being that the magnet is much stronger and more directly aligned.
In summary, splitting a humbucker is not producing a true single coil, it's just reducing the output amplitude of one of the two halves of the pickup by a great amount.
The significance of this is, the RLC values change, and the voltage produced drops, but arguably, none of those difference change the "voicing" of the pickup, they merely change the resonance, where as the reception of magnetic change from the secondary coil does change the harmonic balance (a non-uniform attenuation that is specific to each note and string), and that does change the "voicing" of the pickup.
Another take-away from this; I really love the split sound of "Hot Rails" type single coil sized humbuckers, because the lone coils have a high enough inductance to match that of a stock Strat single coil, but now it's apparent that there is a second reason it sounds convincing: assuming that second coil is still contributing flux change though the primary, the harmonic information is still very close to that of a narrow sized Fender single coil pickup. Therefore, a split "hot" single coil sized humbucker sounds more like a true single coil for the fact that it has close enough RLC values, as well as geometric similarity to a true single coil, in any wiring mode that it's in. OTOH, I have both Duckbuckers and Cool Rails in guitars, and I'd be lying if I said they sound like true singles, so I might have to analyze those soon to see if the RLC values are to blame. The steel pole pieces are liable to create higher eddy currents, so it's possible that Seymour Duncan set a low inductance to offset the effects of the low Q.
|
|
|
Post by antigua on Jul 15, 2017 13:36:43 GMT -5
I just recorded a sample, humbucker split with the unused coil removed, and split with the unused coil in place, and it looks like the unused coil increased the output of the split humbucker by at least 5dB. It's an easily heard difference. Some part of this difference is due to inductance, but I just checked the L value of the split coil with the unused neighbor in place and removed, and it only change the inductance by a small amount: 2.030H with the coils side by side, and 2.015H with the unused coil moved away. Given a difference only 15 mH, it appears that the big amplitude difference owes to the higher concentration of flux through the primary coil, as a result of the secondary coil's pole pieces extending the magnetic circuit of the primary. The more complete magnetic circuit will increase the amplitude for the harmonics over the primary coil, as well as introduce harmonics over the second coil, that would have been missing completely otherwise, with no pole pieces to magnetize that segment of string. Knowing that a combination of magnets and permeable agents placed beside a primary can have such a great effect, might lend itself to product ideas. For example, maybe you could fatten up the tone of a Strat bridge by placing some sort of ferrite bar beside it, on top of the pick guard. It could be painted to match the color of the pick guard.
|
|
|
Post by ms on Jul 15, 2017 14:01:08 GMT -5
I just recorded a sample, humbucker split with the unused coil removed, and split with the unused coil in place, and it looks like the unused coil increased the output of the split humbucker by at least 5dB. It's an easily heard difference. If tapping the magnets of the unconnected coil makes a weaker sound than tapping the magnets of the connected coil, I would expect the signal from the string to be similarly weaker. A 5 db increase means that the signal from the unconnected side is almost as strong. I wonder why?
|
|
frankfalbo
Meter Reader 1st Class
Posts: 74
Likes: 1
|
Post by frankfalbo on Jul 15, 2017 16:27:45 GMT -5
There are several things making the tapping sound audible through the active coil. First, the screwdriver or other item takes the shock and that is within the field of the active coil. So e people make the mistake of tap testing for the active coil with the screwdriver actually going across the active coil toward the shorted one. If, instead, you tap the high/low E poles, with the screwdriver extending out and away from the coil you're testing, you'll notice it's a lot quieter.
In addition, the magnetic circuit is coupled underneath, through the keeper bar, through the magnet, etc. Sticking the screwdriver to the inactive pole shocks the entire circuit. Your result would be different in a Stag Mag type of pickup where the two fields are not mechanically linked.
Next, the pole of the inactive coil is reverse polarity. So insofar as it IS contributing to the audio generated by the active coil, that component is phase inverted. I would look to the elongated aperture and coil saturation for the amplitude change. We've also had discussions about the return path, and how it relates to the magnetism at the string. By widening the field you're also betting more productivity from the outer turns of the coil. (Toward the center of the humbucker)
|
|
|
Post by antigua on Jul 15, 2017 16:46:04 GMT -5
I just recorded a sample, humbucker split with the unused coil removed, and split with the unused coil in place, and it looks like the unused coil increased the output of the split humbucker by at least 5dB. It's an easily heard difference. If tapping the magnets of the unconnected coil makes a weaker sound than tapping the magnets of the connected coil, I would expect the signal from the string to be similarly weaker. A 5 db increase means that the signal from the unconnected side is almost as strong. I wonder why? I'm claiming that's a 5dB increase over having the secondary coil move to the side, away from the the primary. It means the magnetic circuit of the secondary has increased the total flux through the primary, to such an extent that a +5dB voltage increase results. Having the secondary coil in place increases the total flux through the primary in two different ways: 1) replace some portion of the primary magnetic fields "air" return path with the permeable steel poles of the secondary, and 2) introduce flux change of the secondary coil's field into the coil of the first, by way of the secondary's return path. So even though the secondary coil is no in circuit, it's core and magnetized strings are still working with the primary in the "magnetic circuit" context. I haven't come across anyone who seems to know more about magnetic circuits than you, so if you disagree, I'd love to get your take on it.
|
|
|
Post by antigua on Jul 15, 2017 16:55:43 GMT -5
There are several things making the tapping sound audible through the active coil. First, the screwdriver or other item takes the shock and that is within the field of the active coil. So e people make the mistake of tap testing for the active coil with the screwdriver actually going across the active coil toward the shorted one. If, instead, you tap the high/low E poles, with the screwdriver extending out and away from the coil you're testing, you'll notice it's a lot quieter. You're correct that you're shocking, or displacing flux, within the field of the active coil, but by that same virtue, so too is the string above it, hence the idea that the secondary coil is still producing some signal in the primary. The fact that the pole piece of the unused coil is still so "noisy" is the easiest most obvious way to demonstrate that this principle is at work. As for the screwdriver test, you're correct that it seems to be a lot louder when the screwdriver is above both coils, but the abrasion sounds are still rather loud, even when the metal piece is not situated above the primary. Besides that, the screw driver is permeable, so it's also enhancing the flux density by completing more of the magnetic circuit between the two coils/cores, so this result isn't too surprising. In addition, the magnetic circuit is coupled underneath, through the keeper bar, through the magnet, etc. Sticking the screwdriver to the inactive pole shocks the entire circuit. Your result would be different in a Stag Mag type of pickup where the two fields are not mechanically linked. You bring up a good point with the Stag Mag that I was thinking about, but omitted for some brevity; since AlNiCo poles are lower permeability than steel, those two cores will have a lot less benefit from being side by side. Therefore, splitting a Stag Mag should lead to a more convincing single coil result, all other things being equal. It also leads a Stag Mag to be qualitatively different from a PAF in terms of the magnetic relation to the strings, even if they were to match the RLC values. Next, the pole of the inactive coil is reverse polarity. So insofar as it IS contributing to the audio generated by the active coil, that component is phase inverted. I would look to the elongated aperture and coil saturation for the amplitude change. We've also had discussions about the return path, and how it relates to the magnetism at the string. By widening the field you're also betting more productivity from the outer turns of the coil. (Toward the center of the humbucker) It's not phase inverted, because the return path is opposite polarity. Each coil is in its neighbors return path.
|
|
|
Post by ms on Jul 16, 2017 7:32:26 GMT -5
test with this hum bucker:
two screw coils 4.8K per coil 1.8 H per coil one coil connected to preamp and scope other coil open
small driver coil used over individual pole pieces, 300 Hz rest signal
The response was measured over various of the pole pieces in the connected coil and the unconnected coil. The ratio of the responses, connected to unconnected, was about 14db. This seems consistent with the interpretation that the field passing through the excited pole piece in the unconnected coil loops down and back through the connected coil.
I would think that this would be close to the ratio that should be measured when using a string.
|
|
|
Post by antigua on Jul 16, 2017 13:25:40 GMT -5
test with this hum bucker: two screw coils 4.8K per coil 1.8 H per coil one coil connected to preamp and scope other coil open small driver coil used over individual pole pieces, 300 Hz rest signal The response was measured over various of the pole pieces in the connected coil and the unconnected coil. The ratio of the responses, connected to unconnected, was about 14db. This seems consistent with the interpretation that the field passing through the excited pole piece in the unconnected coil loops down and back through the connected coil. I would think that this would be close to the ratio that should be measured when using a string. Is the 14dB in contrast to the 5dB I see in the audio recording? That's a much bigger difference than what it appears I get from a practical test, far more than would result from inconsistent strumming between the two audio samples.
|
|
|
Post by ms on Jul 16, 2017 14:50:35 GMT -5
test with this hum bucker: two screw coils 4.8K per coil 1.8 H per coil one coil connected to preamp and scope other coil open small driver coil used over individual pole pieces, 300 Hz rest signal The response was measured over various of the pole pieces in the connected coil and the unconnected coil. The ratio of the responses, connected to unconnected, was about 14db. This seems consistent with the interpretation that the field passing through the excited pole piece in the unconnected coil loops down and back through the connected coil. I would think that this would be close to the ratio that should be measured when using a string. Is the 14dB in contrast to the 5dB I see in the audio recording? That's a much bigger difference than what it appears I get from a practical test, far more than would result from inconsistent strumming between the two audio samples. I think something is wrong. For messing around, perhaps an easier way to disable the unconnected coil would be to take the screw poles out.
|
|
|
Post by antigua on Jul 17, 2017 12:59:50 GMT -5
Is the 14dB in contrast to the 5dB I see in the audio recording? That's a much bigger difference than what it appears I get from a practical test, far more than would result from inconsistent strumming between the two audio samples. I think something is wrong. For messing around, perhaps an easier way to disable the unconnected coil would be to take the screw poles out. It's actually the slug coil I'm removing, which, I believe, is the coil that people usually disable when splitting. It would be more work to popup the slugs out. One thing I can try, though, is removing the screws from the primary. It will decrease the inductance, but it would be an effective test because it would demonstrate how much voltage / signal is produced by the return path of the disabled coil. One "trick" I've seen out there for how to get a humbucker to sound like a single coil, is to remove the screws of one of the coils, and continue to run it in series. Based on these observations, it seems that this is a pretty good idea. Having the screws removed removes one coil from the magnetic circuit, and it also reduces the inductance, therefore raising the resonant peak. The Q factor will still be lower, but you probably have 500k pots in the guitar, which helps on that end. Of course, the drawback is that you can't put the screws back in with the flip of a switch, but it reinforces an feeling I've had for a while that pickup mods might need to consider mechanical solutions, rather than electrical ones.
|
|
|
Post by antigua on Jul 19, 2017 4:23:50 GMT -5
I've conducted this test again, this time with an exciter coil driving a constant amount of string vibration in the A string. The difference is smaller than I observed with strumming, though a single steady state string is qualitatively a bit different from strumming a chord. Below is an a wave file with alternating states of having the second disconnected coil in place, or removed. The louder portions are where the second coil was in place. I pause recording, change it, then resume, and repeat, to make sure that the result was consistent. So the presence of the second coil does increase the signal, it's just a matter of degree. The stats window for this "coil removed" portion shows average RMS -26.9dB: This portion with the coil in place shows average RMS -25.49dB So the difference here is only 1.4dB. When I was strumming, the difference seemed audibly obvious, and it the recording showed a greater than 1.4dB diifference overall, so I'll do some more experimenting to see why there might be a less dramatic difference with the single steady string. Here is a pic of the setup:
|
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jul 19, 2017 7:50:28 GMT -5
Just a question,
since there are low latency polyphonic guitar synths out there that work with the stock regular pups, I guess it would make sense to talk about : "Making one pickup sound like any instrument with modern software Synth technology"
|
|
|
Post by antigua on Jul 19, 2017 15:15:26 GMT -5
Just a question, since there are low latency polyphonic guitar synths out there that work with the stock regular pups, I guess it would make sense to talk about : "Making one pickup sound like any instrument with modern software Synth technology" The idea here would be to modify the signal, subtracting some some harmonics out and adding others back in. A synth, otoh, wholly replaces the signal with a synthetic timbre. AFAIK, the synth has to monitor each string individually, because as can be seen from the spectrogram, it would be hard for a signal processor to tell the difference between a fundemental of a higher note, versus the harmonic of a lower one. Are you playing the higher E, or is it sensing a harmonic of the low E? You can only tell by evaluating the strings individually. The same would apply here, though. You've have to tweak the harmonics on a per string basis. To some extent you, can manipulate the timbre manually, based on where you pluck the strings. If you engage the neck pickup and pick really close to the bridge of the guitar, you excite particular harmonics to where it almost sounds like you're using the bridge pickup, and that does constitute a change in timbre.
|
|
|
Post by JohnH on Jul 19, 2017 17:21:55 GMT -5
A few thoughts and observations:
Due to my 'CRD' problem (Compulsive Rewiring Dementia'), I have a number of guitars with humbuckers where I can split coils, change phase and do series to parallel switching. In these scenarios, split coil refers either to bypassing or disconnecting one coil electrically, but of course its still in place physically and magnetically.
The split coils sound 'single coil like', they have some of the edge of a true single, much more than parallel wiring of coils which provides more of a brighter humbucker tone. All ars uncovered.
The splits are a workable option if one has to play 'Sultans' on an LP. But a good true single has much more of that edge.
So the factors identified that make that difference include eddy current damping due to extra metal parts and magnetic coupling of the bypassed coil. I think those are definate.
Other effects are; Added electrical damping if coil split is done by shunting one coil. True in theory, my hunch is its not very significant TBC even though weve been talking about that effect here for more than a decade.
Magnetic pull - my hunch is this is not so significant
Coil position - at the neck, which coil you split makes very little difference to tone. I found this on two guitars with symetrical humbuckers (each coil with same winds, same poles). At the bridge, it does make a difference due to a significant % difference from the bridge.
Pickup tilt - hard to be sure. I think its a subtle effect but not enough to change a single-coil tone to one that isnt. I had it built into GuitarFreak but didnt see much interest graphically so I took it out - which simplified the analysis a lot.
I had a tought about improving single-coil tone from humbuckers. Obviously in practice, a single coil switch needs to just be electrical. I wonder how some of tha half out of phase variants would compare? ie, the unwanted positive contribution from a bypassed coil due to magnetic coupling could be negated by feeding a small amount of its electical signal out of phase?
|
|
|
Post by antigua on Jul 19, 2017 18:18:44 GMT -5
A few thoughts and observations: Due to my 'CRD' problem (Compulsive Rewiring Dementia'), I have a number of guitars with humbuckers where I can split coils, change phase and do series to parallel switching. In these scenarios, split coil refers either to bypassing or disconnecting one coil electrically, but of course its still in place physically and magnetically. The split coils sound 'single coil like', they have some of the edge of a true single, much more than parallel wiring of coils which provides more of a brighter humbucker tone. All ars uncovered. The splits are a workable option if one has to play 'Sultans' on an LP. But a good true single has much more of that edge. So the factors identified that make that difference include eddy current damping due to extra metal parts and magnetic coupling of the bypassed coil. I think those are definate. Other effects are; Added electrical damping if coil split is done by shunting one coil. True in theory, my hunch is its not very significant TBC even though weve been talking about that effect here for more than a decade. Magnetic pull - my hunch is this is not so significant Coil position - at the neck, which coil you split makes very little difference to tone. I found this on two guitars with symetrical humbuckers (each coil with same winds, same poles). At the bridge, it does make a difference due to a significant % difference from the bridge. Pickup tilt - hard to be sure. I think its a subtle effect but not enough to change a single-coil tone to one that isnt. I had it built into GuitarFreak but didnt see much interest graphically so I took it out - which simplified the analysis a lot. I had a tought about improving single-coil tone from humbuckers. Obviously in practice, a single coil switch needs to just be electrical. I wonder how some of tha half out of phase variants would compare? ie, the unwanted positive contribution from a bypassed coil due to magnetic coupling could be negated by feeding a small amount of its electical signal out of phase? You didn't mention the one thing that likely trumps all those factors combined: the lower inductances of the individual coils. On the DiMarzio PAF Pro I measured only 1.8 henries per coil, and loaded resonant peaks in the high 4kHz range guitarnuts2.proboards.com/thread/7737/dimarzio-super-distortion-analysis-review All the other stuff is peanuts in comparison to that. I think if you split a coil and end up with ~2.5H inductance, you get something a lot closer to a true single coil, and at that point the disappointment stems from the position of that coil not being identical to the positioning of pickups in a Strat or Tele. I think the idea that the inactive coil is still active by virtue of return path is similar to those other things you listed; possibly more for the sake of argument and something that has to be taken into account. Although, this influence has two things going for it 1) the varied harmonic content of the second coil, it's not a linear attenuation, and 2) the effect appears to have the potential to eclipse 1dB, the magic threshold of perceptibly. I'll admit it it's probably not soiling good single coil tone, I think that's mainly due to inductance and positioning.
|
|
Deleted
Deleted Member
Posts: 0
Likes:
|
Post by Deleted on Jul 20, 2017 1:43:59 GMT -5
Just a question, since there are low latency polyphonic guitar synths out there that work with the stock regular pups, I guess it would make sense to talk about : "Making one pickup sound like any instrument with modern software Synth technology" The idea here would be to modify the signal, subtracting some some harmonics out and adding others back in. A synth, otoh, wholly replaces the signal with a synthetic timbre. AFAIK, the synth has to monitor each string individually, because as can be seen from the spectrogram, it would be hard for a signal processor to tell the difference between a fundemental of a higher note, versus the harmonic of a lower one. Are you playing the higher E, or is it sensing a harmonic of the low E? You can only tell by evaluating the strings individually. The same would apply here, though. You've have to tweak the harmonics on a per string basis. To some extent you, can manipulate the timbre manually, based on where you pluck the strings. If you engage the neck pickup and pick really close to the bridge of the guitar, you excite particular harmonics to where it almost sounds like you're using the bridge pickup, and that does constitute a change in timbre. I think they have done it : www.boss.info/us/products/sy-300/ .
|
|
|
Post by antigua on Jul 20, 2017 13:42:05 GMT -5
The idea here would be to modify the signal, subtracting some some harmonics out and adding others back in. A synth, otoh, wholly replaces the signal with a synthetic timbre. AFAIK, the synth has to monitor each string individually, because as can be seen from the spectrogram, it would be hard for a signal processor to tell the difference between a fundemental of a higher note, versus the harmonic of a lower one. Are you playing the higher E, or is it sensing a harmonic of the low E? You can only tell by evaluating the strings individually. The same would apply here, though. You've have to tweak the harmonics on a per string basis. To some extent you, can manipulate the timbre manually, based on where you pluck the strings. If you engage the neck pickup and pick really close to the bridge of the guitar, you excite particular harmonics to where it almost sounds like you're using the bridge pickup, and that does constitute a change in timbre. I think they have done it : www.boss.info/us/products/sy-300/ . I see that they offer some aggressive filtering, but what I don't see id adaptive harmonic alteration, the difference between EQ and timbre. For example, if it detects a frequency of 82Hz, the 3rd harmonic is at 330Hz, and you want the bridge pickup to sound more like the neck, you have to remove that 330Hz component specifically, since the neck pickup sits over the anti-node of the 3rd harmonic. The exact harmonic changes that would be required for each string and fret position would be different, but a pickup with six outputs that's fed into a digital processor could be made to do it.
|
|