|
Post by ms on Mar 17, 2017 22:24:44 GMT -5
Then why doesn't every guitar amp and/or pedal begin with a 5K brick wall filter? Or 10K even? If those frequencies are meaningless why are they allowed? If you ignore 5-25kHz your conclusions suffer. If you promulgate it you're guilty of the same misleading propaganda the likes of which you've accused the marketing charlatans and profiteers. As for the return path, that meter isn't going to tell you anything about it. If something is to small to matter, why would you go to the trouble of filtering it out? Consider 15KHz. To get it down to 3KHz to have to intermod it with 12KHz or 18KHz. Both terms of the product are small in either case, and so the resulting product is doubly small. I think these products are insignificant.
|
|
|
Post by ms on Mar 17, 2017 20:58:00 GMT -5
Two reasons, 1) many, maybe most guitar speakers have a response curve that drops off around 5kHz 2) not many pickups have a resonant peak that allows for much transfer of sound above 5kHz. Filter'trons are the only ones I know of. Most Strat type pickups attenuate between 3.5kHz and 4.5kHz, and PAFs between 2kHz and 3kHz, as a result of inductance values that are typical to those models. That's where you lose me. Pickups are won and lost between 5k-25k. Between 5 and 15 especially. Treating it like it's marginalized because it's 3/6/12/24/48dB is ignorant of the gain chain. Even frequencies above 20kHz will alias down into the audible frequency range under gain. By "aliasing" do you mean the production of sum and difference products, that is, intermodulation distortion? Small signals produce small intermod products while the larger signals below 5KHz make larger products which are more important. Also, such high frequency string harmonics are small and die out very quickly because they are highly damped.
|
|
|
Post by ms on Mar 17, 2017 18:53:41 GMT -5
I am not saying that the strength of the magnetic field causes more bass; I just mean the sort of "wowing" sound. But, I believe I observe the wowing sound even with PAF style pickups, which have very weak magnetic pull, showing a flux at the pole piece of no more than 300 gauss, compared to a Strat with A2 of ~700G , or A5 poles with 1050G. I strongly suspect the effect is due to geometry, and not damping factors. True, but you are pulling in two places rather than just one.
|
|
|
Post by ms on Mar 17, 2017 16:16:45 GMT -5
I don't think this rules out tonal consequences of proximity, but it does appear that it takes any credit away from comb filtering. Why would you reach that conclusion? Also it sounds like you only use the term comb filtering as it relates to the string window, from the string itself. The other comb filtering opportunities are in the return path, and other shifted relationships between the flux and the coil, within the coil. DB and frequency response charts aren't granular enough to show that. Also FYI it might be more effective with the other pickups removed. They're providing baseline magnetism that is sort of a magnetic noise floor in your experiment. That could be affecting your numbers at the increased distances. Remove the pickups, degauss the strings, repeat. Or to establish the other baseline just remove the pole from your test pickup and check the dB to approximate the minimum contribution from those existing pickups' magnets. . What do you mean by "comb filtering opportunities are in the return path"? I do not know of any such effect.
|
|
|
Post by ms on Mar 17, 2017 16:14:13 GMT -5
Thanks for this insightful thread. My strat has Texas Specials for N and M, so I watch out for discussions about them. They have a anecdotal rep for being somewhat 'harsh'. I don't find this myself, but the usual forum advice is to lower them to smooth out the tone. This could be considered consistent with Tillman's theories of 'sensing width' combined with the offset tests above, showing a more concentrated sensing at higher/closer pickup positions. Here is a calculated envelope plot. It shows a Tx Sp (6.2k) neck pickup, plucking string 1 only, with sensing width varied from 6mm to 25mm. It is Tillman, modulated by harmonic smplitudes of string vibration,fed through the electrical filter of the pickup, strat wiring and cable. It shows a few db of difference at higher frequencies, but is it enough to explain the effect or is there more? I think lowering the pickup makes the sound less harsh because you are pulling on the strings less. The window is too narrow in either position to affect the sound very much.
|
|
|
Post by ms on Mar 17, 2017 16:10:57 GMT -5
You are describing the consequences of too strong a B field when you get too close to the string. In still stronger form this is the classic "stratitis". One reason you here more bass when you move the pickup closer to the string is that the volume goes up (Fletcher Munson). If you want to prove a real effect, you have to measure carefully and turn the volume down on the amp to compensate. I know what you mean by string pull causing Stratitus, but I think I hear more bass even with weaker pickups that have weak magnetic pull. At this point it's just something I believe I've observed, I haven't proven it with an FFT plot (though with the ebow on hand, it's not too hard of a test to perform) so I'm not able to press the point much further. I don't think it's Fletcher Munson related, because the effect seems to remain, even if the amp's volume is adjusted after the fact. This is another tedious type of test to perform, but I'll try to get to it also. I am not saying that the strength of the magnetic field causes more bass; I just mean the sort of "wowing" sound.
|
|
|
Post by ms on Mar 17, 2017 12:32:23 GMT -5
Yeah that's the same thing I thought after reading your test and having tried various "pickup width" apertures in the Tillman demo...there should be a much higher aperture to achieve a real difference I guess..! Not to say that in practice I won't be able to go lower than a certain height with pickups, trying to achieve a greater distance from the strings...so it seems to me that, for most applications, what other players hear as "a world of difference" when they talk about different pickups heights, they're mostly referring to the effects of a stronger signal of a pickup closer to the strings versus a weaker signal, which obviously may have an influence on the rest of their signal chain (overdrives clipping more or less, tubes in the amplifier compressing more or less, and so on..) I don't think this rules out tonal consequences of proximity, but it does appear that it takes any credit away from comb filtering. To my ear, there is a lot of tonal difference with pickup height. It seems that I hear 1) more bass and 2) more tonal fluctuation with time, when a pickup is closer to the strings. Maybe the reason there would be more bass when the pickup is closer is due to the higher permeability of the lower wound strings, so closer pickup = even stronger magnetized string. Maybe I should repeat this test on a lower wound string before I move on to other pole piece types. The other thing, when a pickup is closer to the strings, it sounds like the tone "evolves" more, as if a wah wah pedal was being rocked very slowly. If you look at a guitar string when you pluck it, you can see its movement pattern change with time as it decays, because when you pluck it, you pluck it in a particular direction with a lot of force, and the string rebounds from that initial vector force as it vibrates. The test above shows that when the string an pole piece and coil are all close, it has a narrower aperture. My suspicion is that the narrow aperture causes that evolving rebound movement to have a more pronounced effect on the tone, and when the aperture is wider, it has a less profound effect. I can't prove that, though. I just see a correlation at this point. Here's a video someone made in January demonstrating pickups heights. Towards the middle of the video, he lets the strings ring out longer, and I think you can hear more of that "slowly rocked wah" tone when the pickup is closer to the strings. You are describing the consequences of too strong a B field when you get too close to the string. In still stronger form this is the classic "stratitis". One reason you here more bass when you move the pickup closer to the string is that the volume goes up (Fletcher Munson). If you want to prove a real effect, you have to measure carefully and turn the volume down on the amp to compensate.
|
|
|
Post by ms on Mar 17, 2017 7:40:59 GMT -5
Since this is such an active topic I couldn't help but do this sooner than later. Here are the results of my experiment. Very nice! One way to describe the variation of the window with distance up from the pickup would be this: As you start from far away, the window is broad and narrows as you get closer. But as you get very close the narrowing slows down. This is because the magnetic source now appears extended rather than like a single dipole. The length and permeability of the pole piece have an effect on the shape of the field, and so they are expected to play some rolel in where the window stops getting narrower. Your measurement nicely captures both effects I described, magnetization of the string and induction in the coil. I am looking at the second effect now since I like to be able to see the two effects separately if possible..
|
|
|
Post by ms on Mar 16, 2017 16:59:08 GMT -5
I understand that string bending is imperfect because it's receives an assist form the neighboring pole piece. hence the need for a test with a single pole piece.. The variable you're missing is that it's not just an assist. It's a null. Let's pretend wends referring to flat poles, no stagger. The assist is vertical by focusing the field, but depending on how close the pickup is to the strings, you're coming in and out of a null. With the pickup far from the strings you have a smeared convolution. Come closer and you'll have 6 focused, concentrated zones with convolution in between, some overlap. Closer still and you'll find the null/dead zone. None of which you'll see during this first rudimentary look. Compound that by the comb filtering associated with one lone pole, mid-coil, generating non-standard audio into the coil in association with its return path. My point is, you won't have a clean number on the amplitude delta across the lateral axis because it's height dependent and altered by neighboring poles. Drive it with a tine like a Fender Rhodes if you want to get a 3 dimensional dB plot. I do not understand the null you mention, nor the comb filtering associated with one lone pole.
|
|
|
Post by ms on Mar 16, 2017 15:10:00 GMT -5
Antigua is preparing to measure treble-to-bass window, possibly as it relates to drop off when bending between poles, by using only one pole piece in a coil. You don't see the problem with that? One pole, with its own unaltered return path and productive coil body on either side of it? I believe he is not interested in string bending itself so much but rather he is using this technique as a way of measuring the sensitivity versus location of one pole piece. Thus I think he is doing what he wants. He will see string magnetization fall off, and he will see loss of flux from the vibrating string through the pole piece. It is true that he will see some flux from through the coil without the pole piece, but we have no perfect way of measuring the fall off along a string. This seems useful to me, but maybe he can give you a better idea what he is doing.
|
|
|
Post by ms on Mar 16, 2017 13:51:16 GMT -5
One pole is meaningless and will have a phase altering return path within the coil along the treble-to-bass axis. When neighboring, like-oriented poles are flanking it (for the center 4 strings) or it is against the coil edge with only one neighboring pole, the like orientation imparts rejection at the top, completely changing the shape of the field both at the string, as well as the return path. Beveled pole pieces splay the field as well, but while that affects the return path along the string axis, it exacerbates the rejection and focusing betweeen the poles. Depending on how close the pickup is, beveled magnets may have an increased, or a decreased window delta. Also with regard to -10dB, a measley 2:1 compression Or clipping ratio knocks that down to -5dB. Rarely is guitar played through a totally linear signal path, sometimes clipping under 90dB of gain or more. This doesn't change the physical string window, but it does compress the information present, tantamount to lowering the Q. There is a reason why we take this slowly and go through one step at a time. Does anything you just wrote cast doubt on the measurement that I made showing approximately where the string magnetization in the direction along the string is 3 and 10 db down? I do not understand what you wrote, and I do science for a living mostly, electro magnetic type of stuff, and I started this off by pointing out that there are two sorts of approximate answers, width of the pickup, or with of the string, the first a result of "expert" opinion, the second indicated by theoretical analysis. Does what you are saying affect the validity of the evidence so far that it is the width of the pole piece that is important? "One pole is meaningless and will have a phase altering return path..." There is no phase to alter; what do you mean?
|
|
|
Post by ms on Mar 16, 2017 12:07:24 GMT -5
The field diverges like a dipole further from the pole piece and so the window widens with the diverging field lines as yo get far away. It is only the component of flux pointing through a loop that contributes. "Sideways" pointing flux through the loop has no such component. Not "sideways", but at any angle other than perfectly perpendicular to the loop. Flux of course never points perfectly straight, at least not outside of a coil, so much, if not most, of the returning flux from the string will have some degree of angle to it. At a distance, the angles would likely be more extreme, but that off-axis flux would still constitute change through the loop, all the same. The main difference would be that it's less dense, because of the added distance, and because the loop is smaller when it's angled by virtue of perspective; it becomes increasingly oval, until you are 90 degrees off axis with the loop, at which point the loop looks like a straight line, and so there is no loop from that angle. Also I think we need a "system" at some point to clarify things; we're talking about "aperture widths" because it's easy, but we all know we're talking about a gradient which has no defined boundary. The rate of drop off probably has to be expressed mathematically, but I'm not a mathematician. You mentioned that -10dB means half as loud, so maybe we can define aperture width as, for a given distance between pickup and string, the distance from center, where the voltage generated is -10dBV down, from center. Another complication to keep in mind is that this boundary, however defined, it probably different depending on string gauge, because the permeable mass differs, and the boundary also likely changes as the string displaces, though that should "average out" as the string should be equal parts near and far. I'm still planning to set up a different kind of test this weekend. It just takes more free time to get this sort of thing set up. I'll tell you what I'm planning in case you see a problem; an ebow would excite a string, probably the G or D string. I'd take a single coil pickup, and mount only a single AlNiCo 5 pole piece in the pickup, I'm thinking in one of the two center-most slots, and then I'd map out dBV for measured x/y offset between the single pole piece and the excited string. Guitar strings are all small compared to a pole piece diameter. I do not see why the string gauge would matter much for the aperture size. I say use 3db as you do for any filter. As the measurement have shown, it falls of quickly after that. These measurements respond to both intensity and angle, and so you can see how fast the combination of both factors works.
|
|
|
Post by ms on Mar 16, 2017 5:41:50 GMT -5
Mmm, I don't know. It sounds like you're saying the pickup has a more "needle" like precision at a great distance. That just seems counter intuitive. More often than not, reception becomes more diffused at a distance, unless you have a focusing agent, like a pin hole camera or a parabolic dish. Keep in mind that the flux orientation never has to be perfectly on axis, even side ways flux constitutes change through the loop. The field diverges like a dipole further from the pole piece and so the window widens with the diverging field lines as yo get far away. It is only the component of flux pointing through a loop that contributes. "Sideways" pointing flux through the loop has no such component.
|
|
|
Post by ms on Mar 15, 2017 20:55:43 GMT -5
I would expect the window to get smaller as the distance between the string and the pole piece decreases. The field diverges as it gets further away from the pole. But that is not the whole story. If you have a pole piece of diameter D, and you are looking significantly closer to it than D, I would expect the window to change slowly, but as you get further away it should change more quickly. All magnetic sources look like dipoles if you get far enough away, but up close it is different. Consider the field of a solenoid (http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/solenoid.html). Inside the solenoid the field is uniform, but as the lines come out an end, they diverge, but right outside they are nearly straight, especially near the center of an end. (The field strength is proportional to the density of lines.)
|
|
|
Post by ms on Mar 15, 2017 8:10:57 GMT -5
That seems very interesting and clear for the single coil. I've also though Tillman had it too wide, and so I was using coil width instead of his value greater than that. But I see it should be reduced again. On the HB's, it would be interesting to see if the response dips significantly between poles and then how it falls off outside the poles The humbucker will be different. For one thing, the pickup in this guitar currently under measurement has long pole pieces that tend to give a tight field. A humbucker has shorter pole pieces, and thus a broader pattern of magnetization. What we are looking at with this test so far is the response of a pickup coil to the string magnetization, that is, how it changes with distance along the string. That is, the sensing pickup coil is right there close to the magnetization. In actual pickup operation, the magnetization out there along the string is sensed by a pickup could that is not there, but back at the center. Thus it sees a weaker signal, and so the dbs down would increase some more. With this pickup under test, the difference might not be so great since it is already quite narrow in the magnetization, but with a humbucker, you might have to look at both effects together. It might be useful to use a driven sensing coil moved along the string to get this second factor and then take the product with the first factor. The hum bucker does have a deep drop in the middle: the field is weak and horizontal there.
|
|
|
Post by ms on Mar 15, 2017 7:59:54 GMT -5
That's an interesting approach. My plan was to sort of approximate what happens when you bend a guitar string such that it's between two pole pieces. I would mount a pickup over the string, but offset so that only one pole piece is over one moving string. I'd use an eboy to excite the string, then slide the pickup perpendicular to the string in measured increments and measure the change in dBV. One reason I expect the drop off to not be abrupt is because performing string bends between pole pieces into the "dead space" is not all that bad. Some guitarists are bothered by it, so they use rail style pickups, but I've personally hardly ever noticed a volume drop as a result of string bending. The other open question is whether or not compression effectively widens the reach my amplifying that content that is further away from the pole piece. I don't know enough about audio signal compression to intuit an answer. Yes , unless you are playing very clean, you will probably not notice that the gain falls off between poles since the distorted level does not fall all that much. Also remember that it takes about a 10 db drop to sound half as loud.
|
|
|
Post by ms on Mar 14, 2017 8:55:29 GMT -5
This is follow up on some discussion begun here: guitarnuts2.proboards.com/thread/7905/seymour-duncan-little-analysis-review. These are not easy measurements to make, but at the very least one should be able to decide this: is the length of string sampled closer to the diameter of the pole piece, or to the width of the pickup? Analysis of the physics of the pickup says that "pole piece" is the right answer. What do measurements indicate? I will start with a very simple measurement; discussion and further measurements will be required. Start with a single coil sized pickup in the guitar. This pickup uses ferrite poles pieces about .2 inches in diameter, and each pole piece has a 3/16 inch diameter, 1/32 inch thick neo magnet on top. This is not a single coil, but rather six individual coils, but that does not matter because the measurement is made with a single coil pickup placed above the strings over the guitar pickup such that it can be slid along the strings to sample the field produced by the vibrating string magnetized by the magnet in the guitar pickup. The measurement pickup is air core for this first measurement (wound on a single core plastic bobbin). Three measurements are made by carefully plucking the string, one centered over the guitar pickup, and then slid along the strings by 3/32 and 3/16 inches. The measurements are converted to db referred to the centered measurement, and the measurements are 0 db, -3.1 db, and -18.4 db. Thus 3/32 inches is the half width measurement or 3/16 inches the full width of the sampling length at approximately the 3 db points. The fall off is very rapid after that and so this defines the effective sampling length. Thus this measurement says that the sampling length, based on sampling the magnetization of the string by the magnet in the guitar pickup is a pole piece width rather than a pickup width. I do not see any ambiguity in this conclusion, but I do see that additional refining measurements would be useful. Next I will repeat the measurement with a hum bucker in a guitar, same sampling coil, and then repeat both with a core in the sampling coil.
|
|
|
Post by ms on Mar 13, 2017 12:15:54 GMT -5
No, I do not disagree, as implied by my previous message. By the way, another indication that small displacements matter is found in this material: Kirk T. McDonald www.physics.princeton.edu/~mcdonald/examples/guitar.pdf. He showed that there is a frequency doubled component in the pickup output from vibration across the pole piece. This shows that the field varies even for small displacement; it does not apply completely to change in sensitivity along the direction of the string, but it is a pretty good indication since the magnetization from a pole is pretty symmetrical around the axis pole piece. This work shows that there is some change in sensitivity even for the small distance of the string vibration, and so the fall off should get large quickly as the distance is increased. I fixed the wording up, thanks for calling attention to that. I don't deny that there's a change in sensitivity over the pole piece, at not point is there a perfect or ideal homogeneous flux field, at least not with a round pole piece, a blade is a different story. I'm not sure that speaks to a particular distance / amplitude ratio though. I'd like to do a practical test, but they're a pain to set up. This is hard to measure with great accuracy. I am going to start by using an unmagnetized pickup over a pickup in the guitar. This can be slid along the strings and the output measured. I am sure you can see some problems with this, so you have to start by making some measurements that tell you how big the problems are, such as measuring the change in signal from the pickup in the guitar when the unmagnetized pick up is placed over it. I will just have one pole piece in the unmagnetized pickup; the other holes can be used as sights to help determine the displacement along the string.
|
|
|
Post by ms on Mar 13, 2017 5:07:43 GMT -5
This is what you are trying to show: "The first effect is that a wider pickup will see more harmonics, because it physically spans more nodes." Showing that the fourth harmonic is no longer nulled when you sample it off center does not show that. Maybe it was a poor choice of word to say "more nodes", but do you disagree that a wider neck humbucker would be seeing more of that 4th harmonic displacement? No, I do not disagree, as implied by my previous message. By the way, another indication that small displacements matter is found in this material: Kirk T. McDonald www.physics.princeton.edu/~mcdonald/examples/guitar.pdf. He showed that there is a frequency doubled component in the pickup output from vibration across the pole piece. This shows that the field varies even for small displacement; it does not apply completely to change in sensitivity along the direction of the string, but it is a pretty good indication since the magnetization from a pole is pretty symmetrical around the axis pole piece. This work shows that there is some change in sensitivity even for the small distance of the string vibration, and so the fall off should get large quickly as the distance is increased.
|
|
|
Post by ms on Mar 12, 2017 20:44:19 GMT -5
I'm not following; the blue colored diagram I posted above shows the neck pickup intersecting the anti node of the 4th harmonic. A thin Strat pickup will land dead center on that 4th anti node, but a wider humbucker will sample some portion of the 4th harmonic, since one of its coils will be off center of the anti-node. In a similar respect, a bridge humbucker extends away from the bridge, picking up content that is otherwise out of reach of a single coil bridge pickup. Leo Fender slanted his single coil bridge pickup so that the bass half of the pickup would see more fundamental movement, and generate more bass amplitude. To my ear, it's apparent hear that a neck humbucker sounds different than a single coil sized pickup, even if the frequency response is otherwise the same, so if doesn't explain it, I feel that something else must. This is what you are trying to show: "The first effect is that a wider pickup will see more harmonics, because it physically spans more nodes." Showing that the fourth harmonic is no longer nulled when you sample it off center does not show that.
|
|
|
Post by ms on Mar 12, 2017 17:33:39 GMT -5
Even Tillman acknowledged that the approximation is difficult to define www.till.com/articles/PickupResponse/index.html "This isn't completely accurate, the pickup sensitivity will be greater in the middle than at the ends, but this makes a fine first approximation.", and by going with the width of the pickup, I've even chose narrower windows than he had: "Figure 10 shows the response of a neck pickup 1.0 inches wide on the low E of a Stratocaster. Figure 11 is the same thing, but the pickup is 2.5 inches wide."In the Tilrman article you referred to he makes this statement: "Pickups do not sense the string at a single point source, but rather over an area due to the width of the magnetic field. This sensing area is called the "aperture" of the pickup and is about an inch wide on a thin single coil pickup and about 2.5 inches wide on a wider pickup such as the Gibson hum bucker." This is wrong. Tillman has never taken the trouble to understand how this works in theory or even to make decent simple measurements. I think it is appalling that this statement has remained on the web for so long when it is so clearly wrong. The one inch sampling width implies the filtering out of picking transients that are present in spectral measurement made right after the string is picked. Harmonics are present that imply that it is the width of a pole piece that is the limiting factor.
|
|
|
Post by ms on Mar 12, 2017 15:41:17 GMT -5
Just for my own satisfaction, I held a connected single coil over a Stratocaster, and approached the string as it vibrated, and I could hear quite a lot of volume before the pickup was even above the string at all, so I think I might have actually underestimated when I went with the width of the pickups themselves, but I still think 2.5 inches is probably too generous for a humbucker. Accurately presenting harmonic response isn't really the goal though; just looking for some kind of difference, given consistent methodology, is, and it appears that, for a given methodology, there is not much change in the neck pickup, but there does appear to be a bigger difference with respect to the bridge single coil versus humbucker width. the plucked open note lands over an anti-node of the 4th harmonic. If that pickup is made to be wider, it won't be completely blind to the fourth harmonic. A neck humbucker extends towards the bridge, relative to a single coil, so looking at the diagram, you can see that it would pickup a little more of everything shown, aside from the fundamental, as that movement decreases closer to the bridge. When I try your test I get very different results. The level half a pole piece width from the edge of the pole piece compared to the level at the center is 6 db down. You normally defines the width of a filter as where it is three db down, but I do not have enough precision to make that measurement. A whole pole piece width away is more than 10 db down, and so it is a rapidly falling function . This is not an easy measurement; careful plucking is required, and there are significant errors I am sure, but the width of the pickup is not a good indicator of the width of the string it samples. I do not buy your other argument at all. Making the sampling region wider in such a way that the center of of the sampling region changes will reduce the amplitude of harmonics other than the one you are concentrating on, especially higher ones. The harmonics we are concerned here are the really high ones, on the order of the separation between the two coils of a hum bucker, or even higher, the width of a pole piece.
|
|
|
Post by ms on Mar 12, 2017 8:06:46 GMT -5
Question 2, how does the smaller size effect the tone? The difference in width has two effects, but both ultimately result in different proportions of harmonic content being heard from the guitar. The first effect is that a wider pickup will see more harmonics, because it physically spans more nodes. For this reason, a wider pickup tends to sound more harmonically rich. The second effect is comb filtering, which is where the physical width of a harmonic oscillation is roughly divisible by the width of the pickup, so that for each "up swing" of the harmonic, there is an equal corresponding "down swing" of that same harmonic, causing a net cancellation of the harmonic. J Donald Tillman created a "calculator" to quickly plot out the expected response of a pickup based on width an position www.till.com/articles/PickupResponseDemo/index.html. Going by the rough widths and position of the Little 59' and a full sized '59 neck we get.. A pickups does not sample a region of the string equal to its width. The sampling is confined to an area over the pole piece. This is the product of two factors. First, the the pole piece magnetizes the string with a signifiant component of B field pointing through the coil only over and close to over it. Second, the pole piece directs the changing flux from the vibrating string through the coil, and it does it best with the flux originating right over the pole piece. Since this is the product of two factors, the overall function falls off quickly as the distance from the pole piece increases along the string. A pole piece is small enough so that its size is not a factor in decreasing audible harmonics for any string. However, a hum bucker has two sampling areas for each string, one for each coil. These add coherently, and so there is cancelation when an "up" and a "down" are over the two pole pieces. In general, there is a function that relates the degree of reduction to the harmonic number, and so it is not just a matter of canceling or not. The function is a comb filter much like the function for two pickups spaced very closely (which is what it is, after all), with complete attenuation beginning when the wavelength of the harmonic approaches the pole piece width. Whether this makes a difference is a function of which string. That is, harmonics of the same number for the high E string are four times higher in frequency than for the low E string. For a standard hum bucker, it works out that this is audible for the three wound string (barely for D), but very little for the three plain strings. For these strings, the effects are pretty much outside of the 5KHz window. With the close spacing of the single coil sized hum bucker, perhaps effects could be heard on the low E string, but is very small for other strings. Also this: "The first effect is that a wider pickup will see more harmonics, because it physically spans more nodes." makes no sense to me even if a pickup did sample uniformly across its width. A wider uniform sampling window starts losing response at lower harmonic numbers than a narrow one, meaning that the wider response function results in fewer harmonics..
|
|
|
Post by ms on Mar 8, 2017 12:41:00 GMT -5
These are a little dark for my tastes. They do well for distorted bass, but if you're into slap bass, they don't snap like vintage output pickups. I think it makes more sense to go active than to go for QP's, but if you're not a demanding tone connoisseur and don't want to deal with dead batteries, it's a viable option. In another discussion, we heard the desire to put things learned in these measurements and discussions to use to make better pickups. In particular, there was the tradeoff in pole piece materials between Alnico (low loss, but also low inductance) and steel (high loss and high inductance). Then there are ferrite rods (usually known as beads, www.amidoncorp.com/fb-73-1801 ***) which give high inductance and very low loss. They can be used to make higher Q pickup coils, and, although this can be useful for certain tradeoffs in guitar pickups, the application to a bass pickup seems obvious. With the right design, you should be able to get significantly more "snap" with a passive system. Something I will try when I get some time. ***This link does not work when you click on it. But if you copy/paste it (I need to start outside the link to get it to copy and then delete the extra characters after pasting), it does.
|
|
|
Post by ms on Mar 5, 2017 6:44:34 GMT -5
To cancel hum from magnetic fields you need to add two induced villages of opposite polarity. I do not see how what you are proposing does this.. Pafs cancel hum because you have two coils connected with opposite polarity. The signals through the two coils add because they also have opposite polarity from opposite string magnetization. Suppose you take the HiloTron, and instead of putting just one coil off to the side of the bar magnet, you put a second reverse wired coil on the other side. The polarities through the strings above the two coils will be opposite polarity, and from this you achieve RW/RP. It's essentially similar to a PAF, where the two coils and their pole pieces flank a bar magnet laying on it's side. This arrangement could be used to create a flat humbucker, one that is at least as flat as a Hilo'tron. With ceramic or neodymium, you can probably also get away with using a thinner bar magnet in between the coils. Right, I see. But why omit the pole pieces? They help guide the flux from the vibrating string through the coil.
|
|
|
Post by ms on Mar 4, 2017 7:08:58 GMT -5
It would be humbucking, since the returns path are opposite polarities, which is the basis for the magnetic circuit of Filter'trons and PAFs. This whole idea of placing magnets beside coils instead of below them seems under explored. With magnets that are stronger than AlNiCo, the 1/4" bar would be replaced with something much thinner. To cancel hum from magnetic fields you need to add two induced villages of opposite polarity. I do not see how what you are proposing does this.. Pafs cancel hum because you have two coils connected with opposite polarity. The signals through the two coils add because they also have opposite polarity from opposite string magnetization.
|
|
|
Post by ms on Mar 4, 2017 7:01:18 GMT -5
I was watching an interesting video on the Youtube channel of Reverb.com where I discovered some great tones, in my opinion, by these Supro reissue guitars... I found that one of the most iconic pickups which were used in these guitars were these Vista Tone single coils...on this page I found that they seemed somewhat similar to the TV-HT reviewed here design-wise; www.premierguitar.com/articles/21134-cult-coils-lesser-known-vintage-pickups?page=6I now have a doubt regarding the harmonics sensing of this type of pickups. The screws pole pieces are reading the strings, but what about the magnets that are close to the coil? Don't they sample the region of the strings above them? From the review of Antigua he seems to consider the smaller area above the screws, but not the one above the magnets. In this article there's this info, "According to inventor Ralph Keller’s 1952 patent, “An object of this invention is to provide a pickup device which establishes a magnetic field extending for a substantial distance along each string, with the magnetic lines of force lying substantially parallel to the strings for the major portion of said distance.” The wide magnetic field spans the width of the housing, interacting with approximately two inches of the strings’ length." ...which lets me think that the magnets should have a role here, perhaps? Thank you very much! It is the changing magnetic flux pointing through the coil from the vibrating string that induces the voltage in the coil. It is hard to see how magnetizing the string far from the col along the strings contributes much to that.
|
|
|
Post by ms on Feb 27, 2017 7:22:42 GMT -5
Nice job! I remember that modeling with FEMM can be used to show similar results, but I forget if there were any careful numerical comparisons. What you can see with FEMM is how much flux "escapes" out the sides of the pole piece rather than continuing straight on through. I appreciate your irony. I used to think that Fender pickups did an OK job of getting the flux to the lower part of the coil until I realized that the low permeability of alnico is important.
I find it remarkable that there are so few pickups using ferrite rods to replace the steel in a magnet backed pickup, or ferrite rod magnets by themselves. Then you can high high permeability and low loss. In fact the Q might be a bit too high, but that is easily fixed with a resistor. (But I have been told that guitarists do not want to see any external components in their replacement pickups!) But the result can be a higher output single coil pickup or a humbucker with a higher resonance and no steel sound.
|
|
|
Post by ms on Feb 23, 2017 17:47:26 GMT -5
I was reading a document called Self-Capacitance of Inductors that somewhat covers multi-layer coils and conductive core coils. Most of the math is over my head, but he seems to disagree about, or at least not preclude, capacitance between turn-to-turn windings in the same layer. i think turn to turn no the same layer is out. A reference in the document that Antigua quotes from is convincing. If the field pointing between two turns is very small, there is no reason for charge to collect and thus not much capacitance.
|
|
|
Post by ms on Feb 22, 2017 13:27:01 GMT -5
Straight wire? Equations 3 and 4 in your reference contain the diameter of the coil, not the length of the wire. I do not understand that relationship. I misunderstood this line "Hence, when considering the physical processes in a single-layer coil at frequencies close to the frequency of its self-resonance it is necessary to abandon the model of lumped elements as baseless and to consider the inductor as a transmission line." but after reading again they just mean that the capacitance doesn't compound, but increases in proportion to wire length. Any way, this is for single layer coils; very different for a guitar pickup. Right, but as I said, the nugget of information is that the capacitive coupling occurs perpendicular only, no matter what kind of coil is being discussed, so that creative methods of insulating windings coil make use of this fact to most efficiently shield the wire. As it says, shielding turns within the same layer is ineffective, but shielding turns of different layers would be effective. It strikes a blow to scatter winding as a means of reducing capacitance, because how much perpendicular space are you adding by simply crisscrossing the wire a little bit more? It would be a lot of work, but if a CNC coil winding machine could stop every so often to somehow create a coating or layer, then resume winding, you would have lots of perpendicular spacing, and relatively little parallel spacing. You'd block the vector of the electric field that causes the capacitance, and ignore the direction in which doesn't lead to capacitance. I have thought it is possible that scatter winding lowers capacitance because there are no well defined layers, but I have to admit that I know less about coil capacitance than I thought I did!
|
|