|
Post by ms on Jan 4, 2017 11:41:48 GMT -5
Ken, We have to have some ferromagnetic material in play to guide the field and increase the inductance since an air coil pickup would be too large to fit any of the standard packages and it would be a bear to get the magnetic field guided to the strings. Ferromagnetic material does not have to be conductive. There are ferrites that can do the job with any permeability that you want. If you want to get permanent field to the strings in an air core pickup, just use Neo magnets inside the coil under the strings. Neo's permeability is about 2% higher; so it is effectively the same. If you want such an air core pickup for some reason, use very fine wire.
|
|
|
Post by ms on Jan 4, 2017 11:30:51 GMT -5
I actually think that the tone does change as you fret higher, and becomes a purer tone with more emphasis on the fundamental. I can think of three reasons: 1. The main reason would be because as the vibrating length gets shorter due to fretting, and the pickup is at a fixed distance from the bridge, it is becoming at a greater % of the vibrating part of the string and so getting relatively more fundamental. 2. The same effect is happening to the picking distance which, if you keep picking in one place, is getting nearer to the centre of the vibrating string as you fret higher. 3. There's possibly a third effect too due to bending stiffness of the strings which rends to limit the highest frequencies since the string resists bending so tightly. These frequencies represent a lower order of harmonics being curtailed if you fret higher. Actually, I got entangled in my point. I realize that the tone changes as you go up. What I meant is that it seems to do it fairly proportionally, without huge peaks and valleys. I'm wondering why those aren't clearly audible if the harmonic nodes are positioned over the pickup aperture. The audible effect is a smooth, linear change of tone in proportion to the increase in fret. As a rough analogy, the human vocal tract modulates the harmonics in different ways which sound like vowel expressions. That is what I'd expect to hear from filtering different harmonics. Kind of like a vocorder. On a related note, one would expect the effect to be stronger with a single pickup because the nodes overlap and "fill each other in". But I've never noticed any difference like that between single and both pickups engaged. I think the sound of neck and bridge pickups together is mostly explained by a significant reduction in certain harmonics that have opposite polarity over the two pickups and are not too different in amplitude.
|
|
|
Post by ms on Jan 4, 2017 6:32:24 GMT -5
Based on ms' comments and my thoughts above, I think the 120hz Inductance measurement may be able to be used as another very positive fixed point, by using it to match impedance between model and real pickup at that frequency. Yes, exactly. That is how I use it, and I have not seen any pickups for which it is not true. (But always on the look out for exceptions and improvements!) In the low frequency range eddy current effects should fall off with the square of frequency, and thus go away fast as the frequency goes down ( MutualInductanceLoading.pdf (65.82 KB)).
|
|
|
Post by ms on Jan 3, 2017 14:59:57 GMT -5
Suppose you have a Seymour Duncan Quarter Pound at one extreme, and at the other you have a Strat pickup with little tiny Neodymium pole pieces. All other things being equal, will there be a difference in harmonic content between the two pickups? One reason this is an important(ish) question is because you have pickups like the Quarter Pound, and it's worth knowing exactly what their value proposition is, and you also have Strat pickups with various degrees of bevel around the pole pieces. I just got some pickups from China that feature a dramatic bevel, while some Strat pickups have none at all. What would happen if the AlNiCo pole piece was sharpened to a pencil point? The tests that I have done show that having two sampling regions with humbucker coil spacing, has a significant filtering effect on the wound strings, less noticeable on the plain strings. These higher harmonics are what one might call picking transients since they die out very quickly. You can both measure and hear the harmonic differences resulting from activating the second coil with the wound strings, most effect on the E. I doubt that the different narrower sampling widths that you mention when using different single coil pickups matter, but I have not done that test.
|
|
|
Post by ms on Jan 3, 2017 14:41:10 GMT -5
Here are my notes about modelling pickups with multiple components, exploring how well they match or don't match test results and how more complex models can be optimised to get better matches. It uses covered and uncovered tests on a '57 Classic as a basis for modelling and comparison (thanks antigua and stratotarts) It's a bit too long for a post, se its in the form of a pdf. I hope it will be of interest to some. Hey, thanks. I was just going over some of your stuff looking for this information over the weekend, and now you put it in one place! I think it is very useful to be able to express the pickup response in terms of a few simple ideal components. But I think it is also important to keep the "original three" all unchanged as you add more because they have such direct physical interpretation with measurement or simple calculation. For example, the coil inductance is the inductance measured at a low frequency. It includes effective of the permeability of the cores, etc., but it does not include the effect of eddy currents since they are significant only at higher frequencies, and thus should be represented by additional components. Eddy currents are a mutual inductance effect, and thus L2 and R2 should have physical significance as analogues of the leakage inductance and secondary resistance in an imperfectly coupled transformer, but if this were a perfect analogy, the effect of R2 should go to zero as the frequency does since the mutual coupling vanishes. But it does not, and thus I wonder if there is a more physically compelling way to do this. But I cannot think of what it might be! Again, thanks for writing it up.
|
|
|
Post by ms on Jan 3, 2017 14:05:15 GMT -5
Those calculations assume an open string. Play the chromatic scale ascending on a string. Does each note sound "different"? I think not very much. Yet according to this theory, completely different harmonic nodes are present over the fixed pickup. There is an obvious discrepancy between theory and practice here. The relative level of harmonics is "filtered" by the pickup location and sampling width (or widths, in the case of a hum bucker pickup , which has two). The sound of the string does shift gradually as you run up the scale because the effective pickup location changes. But other things change, too, such as the ratio of length to stiffness. But there is no doubt that the difference in sound between the bridge and neck pickups changes as you fret higher, especially at the highest frets, because the changes in effective pickup location are different.
|
|
|
Post by ms on Dec 28, 2016 15:25:57 GMT -5
In this post I would like to discuss the waveform used in the measurement outlined in diagram B in a post above. The idea is to measure the impedance at many frequencies, and for simplicity, to use a single waveform that contains all the frequencies. (But this single waveform is repeated many times to overcome the effects of random noise.) Since the measurement involves taking a ratio of voltage to current at each frequency, it is not necessary that the waveform contain equal amounts of power at all frequencies: the ratio is independent of the actual value since both the voltage and current are affected by the same factor. However, it is not good if it contains too little power at some frequencies because the measurements at such frequencies have a a poorer signal to noise ratio than you would like. A short pulse, that is with width something like one over the bandwidth we want to cover, contains all frequencies in that range. However, we cannot repeat the pulse too often, something like one over the frequency resolution we want, and so a short pulse makes very poor use of the average power capability of the amplifier that creates the waveform. Thus the SNR would be poor, and we would have to repeat the waveform many many times to get good results. Such a waveform can be said to have "low duty cycle", and we would like to use a waveform with high duty cycle. One way to get high duty cycle is to use a waveform that has only two values, the positive and negative maxima that the amplifier can produce, or at least near to those values. But how do we get the bandwidth? That comes from some appropriate time pattern of changes between + and -. There is no pattern that gives perfectly distributed frequency values. How do I know? This is a very well researched problem because radars are a very important tool. Suppose you have a radar that emits rf pulses. You often want a short pulse to give good range resolution, that is, to allow to close targets to be easily seen. However, you also want to transmit high power to get good sensitivity. It turns out that radar transmitters usually can achieve their full average power output only by transmitting a pulse longer than you want. That is, the peak power is not really as high as you would like. Can you make a long pulse act like a short one? Yes, under some useful conditions. You need to make the long pulse have a spectrum that is wider like a short pulse. You do this by modulating the pulse, often with the two value waveform that gives high duty cycle. Then you have to do some special processing in the receiver, but that is another matter. So how do you find such waveforms? You can search for them. One way is to do a computer search over billions of waveforms until you get one that is good enough. Or you can do a literature search. That is, you look for a paper where someone else shows what good codes they have found. That is the easier way. The top plot in the attachment shows the code of the length I wanted found by literature search. It is in arbitrary time units. The next plot down shows the spectrum in arbitrary frequency units. The bottom two plots show a randomly selected code. The spectrum dips much deeper, and so it is not nearly as good.
|
|
|
Post by ms on Dec 23, 2016 13:41:36 GMT -5
Neo magnets are cheap and available in many many sizes I did a bit of googling. Seems like the sizes would be reasonable to design around (1/8" and 1/4 diameter) but I couldn't find the Goldilocks size (3/16" diameter like a Strat magnet or HB slug). www.kjmagnetics.com/products.asp?cat=1
|
|
|
Post by ms on Dec 23, 2016 5:54:00 GMT -5
My original thoughts were regarding ceramic permanent magnets in the shape of slugs. Since they aren't electrically conductive, the eddy current issue is mitigated. And the low permeability suggests that the inductance would be lower than that of steel and perhaps lower than AlNiCo? But the idea of highly permeable ferrite as pole-pieces with external magnets sounds interesting as well. I think that would be really great, but the question is, where to find them? Or any magnets in the right size, for that matter. The strat magnets fit the holes, but they're way too long. I'd like to ditch the bar magnet because I think it would be great to make pickups flatter. Neo magnets are cheap and available in many many sizes, suitable for putting on the backs of individual pole pieces (or even on top, but be careful, even the plated ones corrode if exposed to guitar players). They are very small for the field strength. Ferrite beads are available shaped and sized like pole pieces.
|
|
|
Post by ms on Dec 20, 2016 6:53:24 GMT -5
When computing the response of, for example, a pickup from the impedance, you have to correctly locate the source in the circuit in order to get the correct response. This is not so difficult with a pickup, but I always get a bit confused when considering eddy currents as well as the more obvious components. Speaking of eddy currents and how to understand them, here is the beginnings of a theory, something I started a few years ago, but have not done much wth since. MutualInductanceLoading.pdf (65.82 KB)
|
|
|
Post by ms on Dec 19, 2016 14:24:51 GMT -5
Since the ceramic is not even a metal, it's not conductive or permeable, not to any significant extent, if at all. They have to use steel in order to deliver the flux alignment of the ceramic magnet to the strings on the opposite side of the pickup. They are a very special ceramic, a so-called ferrite. Although it is true that ceramic (ferrite) permanent magnets tend to have low permeability, the soft magnetic versions can have very high permeability. They also can be very lossy at higher frequencies (used to suppress rf pickup) or not. In the audio range ferrites in general are quite low loss, and permeabilities can be as high as several thousand.
|
|
|
Post by ms on Dec 19, 2016 14:16:43 GMT -5
Bear with me. Just some noob musings to follow . . . I get the impression that steel is the real culprit in ceramic pickups. I have to wonder if ceramic magnets in the same shape as the AlNiCo slugs might not be so offensive in terms of shifting the resonant peak to a lower frequency. I would imagine it's much easier to machine steel into a slug and slap a magnet or two on the backside of the flatwork. But it doesn't seem as if it would be all that difficult to bake ceramic magnets in the same shape as the slugs. Yet we don't see that. Any thoughts on this? Sure. In the case of humbuckers, there are many reasons. I believe the main reason for using steel is to reduce the number of magnets to only one. That reduces material and assembly cost. Ferrite and Ceramic magnet material is brittle, it can't readily be made into screws. The number of players that actually adjust the filister screws is miniscule, yet there they are. It's my impression that the reason they are used instead of slugs is mainly historical and aesthetic. IIRC, Seth Lover's prototype PAF had no screws, and they were added to make it look more like a P-90. In the case of the Fidelitron, they help to secure the coils. It would be possible to either use magnets directly, as with Firebirds or Strat/Tele, or to replace steel with Ferrite, which has high permeability, while retaining the single under-slung magnet. Ferrite pole pieces would definitely be feasible, but are not currently available in the correct dimensions. A pickup made with those would have to have them in both coils, which would lead to a flat, shiny, screwless top. That would be a hard sell to consumers who don't have any understanding of poles, and may not even realize that there are a second, hidden set of poles under the cover. Any such pickup would appeal to a smaller market, since many players have become accustomed to the tonal response of the steel based designs, and would have to follow a learning curve in order to benefit from the extended range of an improved design, while still being able to reproduce the tones that they already have. You can find so called ferrite beads very close in size to a standard pole piece. They have a small hole down the length, but that does not matter.
|
|
|
Post by ms on Dec 19, 2016 12:27:41 GMT -5
Bear with me. Just some noob musings to follow . . . I get the impression that steel is the real culprit in ceramic pickups. I have to wonder if ceramic magnets in the same shape as the AlNiCo slugs might not be so offensive in terms of shifting the resonant peak to a lower frequency. I would imagine it's much easier to machine steel into a slug and slap a magnet or two on the backside of the flatwork. But it doesn't seem as if it would be all that difficult to bake ceramic magnets in the same shape as the slugs. Yet we don't see that. Any thoughts on this? Sure. In the case of humbuckers, there are many reasons. I believe the main reason for using steel is to reduce the number of magnets to only one. That reduces material and assembly cost. Ferrite and Ceramic magnet material is brittle, it can't readily be made into screws. The number of players that actually adjust the filister screws is miniscule, yet there they are. It's my impression that the reason they are used instead of slugs is mainly historical and aesthetic. IIRC, Seth Lover's prototype PAF had no screws, and they were added to make it look more like a P-90. In the case of the Fidelitron, they help to secure the coils. It would be possible to either use magnets directly, as with Firebirds or Strat/Tele, or to replace steel with Ferrite, which has high permeability, while retaining the single under-slung magnet. Ferrite pole pieces would definitely be feasible, but are not currently available in the correct dimensions. A pickup made with those would have to have them in both coils, which would lead to a flat, shiny, screwless top. That would be a hard sell to consumers who don't have any understanding of poles, and may not even realize that there are a second, hidden set of poles under the cover. Any such pickup would appeal to a smaller market, since many players have become accustomed to the tonal response of the steel based designs, and would have to follow a learning curve in order to benefit from the extended range of an improved design, while still being able to reproduce the tones that they already have. So called Ferrite beads with very close the pole piece dimensions are available from places such as Amidon. You have to look around and see who has the right sizes at the moment. The small hole down the middle does not matter. I have made both singe coil and humbucker pickups using them. They are lower loss than alnico, and there are several different permeabilities available, with different losses. Remember, effective permeability never gets high with short open poles.
|
|
|
Post by ms on Dec 19, 2016 12:09:11 GMT -5
Do you think the low field readings you got are the result of the "de-charge" mentioned the material you quoted?
|
|
|
Post by ms on Dec 17, 2016 11:38:49 GMT -5
Curious. In "B" you measure the voltage across the voltage source. Don't you already know the value, since you defined it and it's not affected by the load because the load is in parallel with it? 1. It is not a perfect voltage source (zero output impedance) just because I drew it that way for simplicity. It must have some impedance. 2. Even if it had zero output impedance, that does not mean that you know the frequency response or amplitude perfectly. I use a digital waveform, a so called code, that by its very nature does not have equal response at all frequencies. Since this response can be modified by the D/A and the analog amplifier, it is better to measure what you actually have, than to try to predict it. This way the ratio is very close to correct at all frequencies of interest
|
|
|
Post by ms on Dec 17, 2016 6:31:27 GMT -5
There are some similarities, but the intent and the results are very different. In the setup you show using a 1M resistor (an improvement over the sys comp diagram using a 56K resistor; is that a typo?), you are just driving a pickup from a high impedance and measuring the voltage across it to get the resonant frequency. The Q is limited by the resistor and the scope input impedance. Diagram B drives a pickup in series with a small resistor with a voltage source. This voltage is measured; also the voltage across the resistor is measured. The loading on the small resistor is either negligible or easily corrected. The voltage across the pickup is found by subtracting the two samples. It can be affected by the small stray capacitance across the pickup, but you could look at this part of the measurement since you always have some with the pickup in the instrument also. This voltage also could be affected by cross coupling in the channels in the measuring instrument, but that is small. The current through the pickup is found from Ohm's law using the voltage across it. This voltage is not huge, but you are using a measurement device intended to keep noise low, gain flat, and have good cross talk rejection. So the idea of the processing is to divide the current through the pickup into the voltage across it resulting in the impedance. All frequencies are obtained at once by using an appropriate waveform, and results are averaged long enough to get quiet measurements.
|
|
|
Post by ms on Dec 16, 2016 19:21:56 GMT -5
Consider two diagrams for measuring pickup impedance by dividing the voltage across the pickup by the current through it. Diagram A appears simpler, but it has the disadvantage that the finite impedance of the voltmeter is across the pickup, a high impedance device over some frequency range. It also has the disadvantage that a computer controlled floating input high impedance voltmeter and a computer controlled ammeter are expensive. Diagram B can be realized with inexpensive hardware. The resistor can have a low value and so loading by the voltmeter is not as much of an issue. The voltage across the pickup is found by subtraction of the lower sample from the upper, and so there is no loading, while the current through the pickup is the lower sample divided by the resistor. So it is preferred for at least two practical reasons.
|
|
|
Post by ms on Dec 16, 2016 17:52:02 GMT -5
Thanks for detailing exactly how that's done. Did you use Matlab? One thing that has held me up is the cost and licensing for Matlab. Even supposing it's $150, which isn't insignificant, just having to email them to get a quote, and figure out the license, and worrying about the fact that the license is term limited, is kind of stressful when you're capacity is as a hobbyist. I love free things more for the fact that they're convenient, even more than the lack of cost. I can see how the scripting approach would simplify automation, and permit the +6dB slope in the AC source, and all sorts of other things that would go beyond the capabilities native LTSpice, and SPICE in general. I think the real sell with LTSpice, though, is the ability to drag and drop components, and not even have to be aware of how they components and their values mathematically relate, or even a scripting language, aside from some a handful SPICE directives. That, on top of the fact that it's free, greatly reduces the barriers to entry. So if we were ranking the options by of "ease of use" versus power, I'd go GuitarFreak, then LTSPice, then interactive Python. No Matlab in there at all; it is all Python, and it is all free. Matplotlib is a Python package; it is a plotting package that works very much like the plotting in Matlab. I think different people prefer different approaches. That is good; checking by different means is essential. My main point here is that programming it from scratch is a simpler now the it was in the past. If one wants even more simplicity, the six lines of code I used could be put a single routine with the list of component values as arguments. That would not be best for me, but I think this approach could be made as simple as the others if anyone wanted to.
|
|
|
Post by ms on Dec 16, 2016 13:06:36 GMT -5
An alternative to using spice: Programming languages are quite powerful and easy to use now. For example Python can be used in an interactive mode, like Matlab, for example, and arrays, single numbers, real and complex, all can be mixed in expressions. This makes it really easy. Once in interactive Python, the six commands below are all it takes to get the attached plot. The first line brings in the file pupc, which has the code shown at the end. The second line gives the little project a name, (ap stands for Antigua's pickup) and defines the start frequency, the end frequency, and the number of frequency points. No need to worry about the frequency after that. Then we need the serial and shunt impedances to put into the method for finding the response of a voltage divider. the values come from Antigua's plot just below "Getting Started". fZL finds the impedance pf an inductor vs. frequency given its inductance; likewise for fZC for a cap. vdiv finds the response using its input as the series and shunt impedances. Then it is plotted. In [1]: import pupc In [2]: ap = pupc.pupc(20., 20.e3, 500) In [3]: Zser = ap.fZL(2.2) + 5800. In [4]: Zshu = ap.par(ap.fZC(120.e-12 + 500.e-12), 250.e3, 250.e3, 1.e6) In [5]: res = ap.vdiv(Zser, Zshu) In [6]: ap.plres(res, 'Antigua\'s pickup') This is the code in pupc that is imported above: from matplotlib.pyplot import * import numpy as np import time # Starting and ending frequency and number of frequencies class pupc(): def __init__(self, fs, fe, fnum): self.f = np.linspace(fs, fe, num = fnum) # Method to return impedance of a capacitor def fZC(self, C): return -1.j/(2.*np.pi*self.f*C) # Method to return impedance of an inductor def fZL(self, L): return 1.j*2*np.pi*self.f*L # Method to parallel 2 or more impedances def par(self, Zf, *Zs): admit = 1./Zf for Z in Zs: admit += 1./Z return 1./admit # Method to find the response of a voltage divider def vdiv(self, Zser, Zshu): return Zshu/(Zser + Zshu) # Method to make a plot of the results of vdiv def plres(self, res, str): semilogx(self.f, 20.*np.log10(np.absolute(res))) grid() xlabel('Frequency (Hz)') ylabel('Response (0 = unity gain)') title('{0:s} {1:s}'.format(str, time.ctime()))
|
|
|
Post by ms on Dec 14, 2016 13:34:53 GMT -5
What about the amp? I have a ProCo MusicMan 10 foot cable that measures 380pF. When I plug that into the Cube 80x, I see 1,900pF. A tube amp would be about 50 pf, with the Miller effect amplifying the grid plate capacitance of the 12AX7. A solid state could be anything, including near zero. I cannot see why anyone would make it over 1000 pf. Measuring with an input resistor in parallel should be possible on the par setting of the Extech.
|
|
|
Post by ms on Dec 13, 2016 16:18:14 GMT -5
Well, I have a somewhat different take on the tone issue. If you want to play clean with the bright sound usually identified with clean electric guitar (OK, not most jazz), and want to play without a pre-amp in the guitar and the correct cap before the amp to set the resonance right, then you must play with the volume on 10 if you want to get it exactly right. (Even a treble bleed does not sound right wth the volume reduced.) And you must select a cable with the right capacitance. It does not require measuring the cap of each of the cables in that pile on the floor; it just requires selecting the one that sounds right.
if you want to play with a lot of distortion, then you probably want to cut treble, and there are a lot ways to do that, and some of them include not even noticing that you did it.
|
|
|
Post by ms on Dec 13, 2016 8:05:58 GMT -5
A wire near the edge of a flat conductor is very much like two wires since the charge in the flat conductor will mostly be along the edge near the wire. So your result shows the consistency of your measurements since you got almost the same result as two wires.
The method of images (https://en.wikipedia.org/wiki/Method_of_image_charges) is useful for predicting and understanding the effect of a shield. Of particular interest is the statement "Because electric fields satisfy the superposition principle, a conducting plane below multiple point charges can be replaced by the mirror images of each of the charges individually, with no other modifications necessary."
|
|
|
Post by ms on Dec 13, 2016 7:28:56 GMT -5
Why the emphasis on low capacitance? What you need is the right capacitance to get the tone you want, and for most guitarists, this is not as simple as "as bright as possible." Anyway, making the resonance too high can make the sound less bright. But I do think you might want to use cable with a lower capacitance per foot when you need to play with a longer cable.
|
|
|
Post by ms on Dec 12, 2016 8:00:18 GMT -5
I agree that I do not have the necessary symmetry for high frequency gain difference cancelation due to cables and input capacitance since the output impedance of the line out on the 2X2 is 110 ohms. I need to add about 900 ohms. (For some reason I though it was 1K until I read the manual again.) However, I do not think it matters. With 1K as the highest impedance and the small capacitance from short cables, it does not seem to be an issue. I do not think this is a device loading issue. The voltage across the 1K resistor measures the current through the pickup very accurately, and the voltage across the series combination is used to derive the voltage across the pickup by subtracting the other sample from it. The 1K resistor is very slightly loaded, but you can compensate for this if you want. There is an over all gain matching issue, of course. This was easier with the Apogee Duet than the 2X2. Of course yo do not have to make the channel gains the same; you can introduce a compensating factor in the software. But you still have to get the gain right. Okay, then. For a lower output impedance you could use the headphone jack, if it matters. To deal with gain, I suggest some kind of calibration procedure. Edit - for example, you would insert some test load (resistor?) into the fixture, and run a calibration routine. It would adjust levels and save a gain constant in a file. The test routine would read the calibration constant stored in the file, and incorporate it in the calculations. No, I do not see any need for lower output impedance; I was just thinking of matching it to the 1K of the resistor, although this is not really necessary. One calibration procedure is to measure the dc resistance of the pickup independently, and then make fine adjustments on one of the channel gain controls until you match it. Then the other parameters should be correct as well. But it is more convenient to have a piece of equipment quantized accurate gain adjustment, well matched between channels.
|
|
|
Post by ms on Dec 12, 2016 7:19:22 GMT -5
"The magnetic polarity of metal comb is "north", while the polarity of the rubber magnet strips (as well as the metal casing) the on the edge is "south", which makes for a very short magnetic return path, since the flux hops directly from the "north" over to the "south"."
I think you cannot say very much about this without knowing the permeability of the rubber magnet material, at least in regards to the return path for the flux from the vibrating string.
|
|
|
Post by ms on Dec 11, 2016 18:10:27 GMT -5
I am creating this thread to continue the sub-discussion at other thread and give it its own subject, because I believe it will continue to be "on the radar". It concerns member ms's offsite posts Measuring pickup impedance with an A/D recording interface and Software for performing pickup analysis with a recording interface. There are good reasons to pursue this approach. Mainly, by capturing a complete black box representation of the pickups behaviour by recording complex impedance at all frequencies, it may be possible to produce accurate bode plots for an endless variety of load situations, without any curve fitting or analysis (as might be required to interpret data recorded only as a bode plot). It would still be possible to perform such curve fitting or other analyses, but it becomes optional. Member ms has suggested a simple test circuit that allows this to be done without any specialized amplifier circuits (such as the V5 integrator). There was some discussion about the effects of cable capacitance. I think that ms says it does not cause any problems of loading of the device under test (pickup). I do have my doubts about that, but I may be wrong. In any case, I have two suggestions to address that. One is to use the high impedance portion of the V5 as a buffer, omitting the integrator. Another is to introduce a compensating leg in the drive circuit. The idea is to allow the effects of capacitance in the two input channels to cancel, like this: Honestly, I am not sure about this, it's just an idea, and I have no way to test it right now. I agree that I do not have the necessary symmetry for high frequency gain difference cancelation due to cables and input capacitance since the output impedance of the line out on the 2X2 is 110 ohms. I need to add about 900 ohms. (For some reason I though it was 1K until I read the manual again.) However, I do not think it matters. With 1K as the highest impedance and the small capacitance from short cables, it does not seem to be an issue. I do not think this is a device loading issue. The voltage across the 1K resistor measures the current through the pickup very accurately, and the voltage across the series combination is used to derive the voltage across the pickup by subtracting the other sample from it. The 1K resistor is very slightly loaded, but you can compensate for this if you want. There is an over all gain matching issue, of course. This was easier with the Apogee Duet than the 2X2. Of course yo do not have to make the channel gains the same; you can introduce a compensating factor in the software. But you still have to get the gain right.
|
|
|
Post by ms on Dec 11, 2016 9:53:34 GMT -5
We've only talked about eddy currents in the cores, does anyone know if hysterisis losses are at work here also? Would hysterisis losses be higher for alnico poles than steel? Histerisis is not an issue because the currents are too small to move the medium along the curve
|
|
|
Post by ms on Dec 11, 2016 7:24:11 GMT -5
This is all very interesting stuff. we have had quite a number of discussions I this group about what and how to measure to capture pickup response. I had thought about the impedance measurements before, and they can be done either as real and imaginary parts, or magnitude and phase angle.
But I don't think that they capture a pickup response alone, in the way that the bode plats can do. At each frequency, in addition to two impedance parameters (real + imag, or mag + phase), we also need an output voltage. I think those three values then do capture the response, and can be feed into an analysis or simulation of downstream components.
Mike, my interest in all this is to model pickup response by spreadsheets, to build the resulting pickup models into larger models that include other electrical and non electrical aspects (see GuitarFreak ).
Inside the spreadsheet, I use exactly the kind of impedance data that you measured directly. But I've been deriving this data from curve fitting a 6 part model of fixed components to match the sets of bode pots from Antigua and stratotarts. In my mind, the reason for having to discuss how things like Inductance changes with frequency is because the truth is more complex than can be captured with just a single fixed value. My method, which is a couple of steps better but is still always an approximation, uses two fixed inductances plus resistors and cap to lock onto the main response characteristics. I have not yet found any of the response tests that have been posted here that can not be captured quite well across a range of frequencies and loading conditions, including pickups with significant eddy effects. I agree eddy effects are often best thought of as being derived from poorly coupled 'transformers'. My models imply that the effects of these transformers can be reflected back into the primary, which I understand is a recognised approach in transformer theory (at this point, im beyond the theory that I can remember)
J I agree that a model with a few theoretically perfect components (that is, no frequency dependence in these components) is the way to go. It is an interesting question, what you need besides impedance measurements. I like to think of pickup response as composed of two parts, the law of induction part, and the circuit part. The impedance measurement is aimed at the circuit part. The other part consists of at least the 6 db/octave increase, but is there more? I think the answer to this is part of the eddy current issue. The response to eddy currents can be divided into two parts. First, is the effect on the impedance. Measurements seem to show that this is the dominant one. But there is another part. If the vibrating string excites a current in, for example, a cover, then this current causes a changing magnetic flux through the pickup coil that subtracts from the effect of the string. This effect is at least partly accounted for in an impedance measurement since such a measurement also excites current in the cover, but I doubt that it is the full effect. Thus I think that there is a small effect that the impedance measurement does not capture, but I do not really know. This seems like a good thing to investigate from both theory and measurement directions.
|
|
|
Post by ms on Dec 11, 2016 7:01:09 GMT -5
Edit for an afterthought - you should test the idea of independence from cable/ADC capacitance by comparing results when adding a cap in parallel with the 1k load resistor. It's fine to theorize that it's okay, but it's a really easy practical test that would put to rest any concerns about that. My apogee Duet died and I bought a temporary replacement for a bit over $100, a Tascom US 2x2. The cheap pots make adjustments difficult, and trying to get the channel gains the same is really hard. Also there appears to be something funny about the digitation approaching 20 KHz. I use the preamp in both channels because flaws tend to be greatly reduced when the technique involves taking a ratio as long as the two channels are very similar. The test I use to convince myself that the high frequency impedance measurements are OK is to measure the impedance of a 330K resistor. In addition to the real part, you get a small imaginary part that ramps up from zero. Some of this is associated with the resistor itself, some with the measuring device. At 20 KHz, this imaginary part affects the magnitude of the impedance by less than 1% (the real part is a bit noisier than it should be, however), and the phase is a bit less than four degrees. This might sound impossibly good, but it is the result of the correcting power of a ratio in the measurement process. Errors in the frequency response that are the same in the two channels are divided out (unless so severe that they destroy the SNR). Thus the effect the capacitance of the one foot mass produced cables that I use is nearly all gone, and so on.
|
|
|
Post by ms on Dec 10, 2016 14:15:32 GMT -5
The output of the SH-1n typical guitar circuit is -13dB at the self resonance frequency of 7650Hz. Any differences at that frequency are for all intents and purposes, inaudible. So I don't see the importance. It might be an overly simplistic model, but the effect of eddy current losses in all the experiments I have done, have been on the Q and never the frequency, except by extremely small numbers. I have performed many experiments where the components that produce eddies have been removed and compared - with almost no change in resonant frequency. The dominant eddy current loss is from the cores (if steel). Did you really replace the cores with something with the same permeability, but without the conductivity? That is not such an easy thing to do. How accurate is that self resonance measurement? How much capacitance does your set up have?
|
|