okabass 4/12/2018 6:14 AM
Measuring pickups Capacitance?
It seems that PU's capacitance is not so easy to measure. I think normal method is to define the resonance frequency and inductance and then calculate the capacitance.
It would be nice if you could get the capacitance easier to calculate the resonance frequency and have a feeling how the PU will behave.

I thought that not so sophisticated or brutal method would be to make a tap at half point the coil. Then wind the rest. Then disconnect or cut the coil at half tap, measure the capacitance at the start and end.

So if you are winding 10000 turns, tap it at 5000 turns. Wind the rest, cut the wire at the tap, measure the capacitance, solder the wire back.
Never tried, but if someone has, please tell if it doesn't work.
 
Helmholtz 4/12/2018 7:41 AM
Sorry, your method will not give useful information on the capacitances that determine the PU's resonance(s). The equivalent circuit of your arrangement consists of the series wiring of two coupled windings of a transformer with distributed and reflected capacitances and a rather small interwinding capacitance (<50pF or so) in between. (Please see thread titled "Switchable additional winds"). The result will be more or less the interwinding capacitance. But this does not relate to any PU resonance.

You can measure PU capacitance with an LCR meter that allows measuring at 100KHz. With typical high impedance PUs the accuracy is very good. The absolute measuring error (always negative) caused by the presence of inductance is below 3pF for a PU inductance of 1H and decreases proportionally with higher inductances.

Generally the importance of PU capacitance on the frequency response seems overrated. Its impact on the loaded resonant frequency is much smaller than inductance. A difference of 100pF corresponds to a 3 foot change in length of a typical guitar cable. Thus cables of different lengths can be used to evaluate audible effects of different PU capacitances.
 
okabass 4/12/2018 8:12 AM
HI

Thank you. Really informative answer. I thought that if the measuring the cap. is so easy, it have been used.

Yes the pots, cable etc. impact is very big: several kHz. It is surprisingly big, when you first time notice the difference with lunloaded and loaded PU. Like:
http://i.imgur.com/ryNSHbQ.png
 
bbsailor 4/12/2018 12:38 PM
Quote Originally Posted by okabass View Post
It seems that PU's capacitance is not so easy to measure. I think normal method is to define the resonance frequency and inductance and then calculate the capacitance.
It would be nice if you could get the capacitance easier to calculate the resonance frequency and have a feeling how the PU will behave.

I thought that not so sophisticated or brutal method would be to make a tap at half point the coil. The wind the rest. Then disconnect or cut the coil at half tap, measure the capacitance at the start and end.

So if you are winding 10000 turns, tap it at 5000 turns. Wind the rest, cut the wire at the tap, measure the capacitance, solder the wire back.
Never tried, but if someone has, please tell if it doesn't work.
Pickups sound a certain way due to the unintended consequences of winding enough wire on the pickup to generate enough voltage to drive the input stage of early guitar amplifiers which were derived from tube- based PA amplifiers. These amplifiers typically had a high impedance of about 1M ohm and the consequence of this high impedance design has audible effects. This early design has not evolved much. Les Paul recorded his "Recording" model guitars directly into mic inputs using low impedance pickups without the typical electric guitar sound. Below is the reason for the electric guitar sound.

1. The amp input acts as a bridging input impedance that is about 10 times higher than the source impedance to minimize loading to get the maximum input voltage. At pickup resonance, this bridging impedance affects pickup loading the most as the pickup resonance point is where the pickup impedance is the highest.

2. Using thin magnet wire AWG 42 or 43 allows many thousand turns to be placed on the bobbin to get a high enough output voltage to drive the amplifier required input level. Since this thin wire also has a very thin non conductive coating, there is a very slight capacitance between adjacent strands laying in very close proximity to each other. Multiply this very slight capacitance by the thousands of pickup turns and you begin to get enough capacitance to form a resonant peak in the region of human hearing where the ear is most sensitive. See this link https://sonicscoop.com/wp-content/up...son-curves.png

3. Hand winding pickups puts more air between the winds than machine winding where the wire density is more compact. Thus, hand winding has lass capacitance than machine winding. Typically, pickups have between 80 pf to about 160 pf of capacitance but this is heavily overshadowed by the typical guitar cable capacitance of about 35 pf per foot or about 350 pf for a 10 foot cable. To hear what a guitar sounds like look up "Tillman buffer amplifier" where you can see how to build a JFET in the guitar end of a guitar cable to put a 1Meg to 3 Meg ohm input load right at the guitar end of the cable and also isolate the pickup from the cable capacitance to hear a higher shift in the natural pickup resonance due to the elimination of cable capacitance effects.

4. Guitars have a fundamental frequency range from 82 Hz on the open low E string to 1312 HZ on the 24th fret of the high E string. However, guitar pickups respond to the way a string is strummed as the horizontal movement affects the 2nd harmonic (twice the fundamental frequency) more because when a string passes a pickup pole piece it generates a peak voltage each time is passes the pole piece. Strings typically oscillate in an oblong pattern with both horizontal and vertical movement and depending on which pickup you use, where you strike the strings and how you strike the strings can have an audible impact on the initial string movement and sound. The first 30 up to 50 milliseconds has the most hearing impact on the perceived quality of the sound. Try this simple experiment either by just listening or by listening and observing the pickup output on an oscilloscope. Turn on the neck pickup. Pinch the low E string over the pole piece and pull sideways about .125 inch and then let go. Now, observe and listen the very first few oscillations. Pinch the string again in the same place above the pickup pole piece and pull it up about .125 inch and release (or a little less for low action guitars) and now listen and observe the sound. This vertical pick should sound a little louder as the vertical string motion contributes more to string output level then horizontal string motion. Try it on the other open strings also.

As you can see, pickup capacitance, cable capacitance and other loading (amp input impedance, volume controls and tone controls) can contribute to the perceived sound but there are more variables involved. Playing at a low level where the ear is most sensitive between 3 KHz to 5 KHz and the bass frequency sensitivity is very low is a lot different than when playing in a band where the typical listening level is higher and the threshold of hearing sensitivities becomes more flat across the guitar fundamental and harmonic range up to about 5 KHz where the typical high impedance pickup response tends to quickly fall off after the pickup resonant peak.

I hope this helps?

Joseph J. Rogowski
 
okabass 4/12/2018 1:08 PM
HI

Thank you for your in-depth answer.

I've wind only some dozens PU as hobby, but it is nice to have theory also. My interest is now in 51 7ender Precision bass PU type. It is quite simple but sounds surprisingly good.Try to find a good balance with clarity and lo end. Now It feels that around 10000 turns 42 AWG (7,7 kΩ), 3,3 Henrys is quite close.
 
Antigua 4/12/2018 3:31 PM
Quote Originally Posted by bbsailor View Post
4. Guitars have a fundamental frequency range from 82 Hz on the open low E string to 1312 HZ on the 24th fret of the high E string. However, guitar pickups respond to the way a string is strummed as the horizontal movement affects the 2nd harmonic (twice the fundamental frequency) more because when a string passes a pickup pole piece it generates a peak voltage each time is passes the pole piece. Strings typically oscillate in an oblong pattern with both horizontal and vertical movement and depending on which pickup you use, where you strike the strings and how you strike the strings can have an audible impact on the initial string movement and sound. The first 30 up to 50 milliseconds has the most hearing impact on the perceived quality of the sound. Try this simple experiment either by just listening or by listening and observing the pickup output on an oscilloscope. Turn on the neck pickup. Pinch the low E string over the pole piece and pull sideways about .125 inch and then let go. Now, observe and listen the very first few oscillations. Pinch the string again in the same place above the pickup pole piece and pull it up about .125 inch and release (or a little less for low action guitars) and now listen and observe the sound. This vertical pick should sound a little louder as the vertical string motion contributes more to string output level then horizontal string motion. Try it on the other open strings also.
I came up with a way to demonstrate the horizontal vs. vertical string movement difference. You abut the guitar string with a guitar pick (or something similar), directly over the selected pickup, and then pull and release the string with your fingers, while the string is still against the pick. The presence of the pick forces the string to only move along one axis. Of course, the friction between the pick and the string causes the string to stop moving very quickly, but for the brief transient, you can hear the differences. If the guitar pick is parallel to the face of the guitar, or "side to side", and you release the string, you hear almost nothing. If you re-orient the guitar pick so that it's perpendicular to the face, or "up and down", year hear the transient very clearly. This shows how one axis of string movement produces a clearly audible signal, while movement along the other axis does not.
 
Antigua 4/12/2018 3:38 PM
Quote Originally Posted by Helmholtz View Post

You can measure PU capacitance with an LCR meter that allows measuring at 100KHz. With typical high impedance PUs the accuracy is very good. The absolute measuring error (always negative) caused by the presence of inductance is below 3pF for a PU inductance of 1H and decreases proportionally with higher inductances.
How interesting. The DE-5000 has a 100kHz setting. I've calculated the capacitance of dozens of pickups based on the peak frequency. I'll give C measurements a try with a DE-5000 and see how close it comes to those peak f based measurements.
 
Helmholtz 4/12/2018 4:56 PM
Quote Originally Posted by Antigua View Post
How interesting. The DE-5000 has a 100kHz setting. I've calculated the capacitance of dozens of pickups based on the peak frequency. I'll give C measurements a try with a DE-5000 and see how close it comes to those peak f based measurements.
Just make sure to select parallel mode to measure Cp at 100kHz and serial mode to measure inductance (Ls) at low frequency.

If you find a noticeable deviation from the capacitance calculated via frequency response, the reasons are:

1) Inductance at resonance frequency is lower than the value measured at a lower frequency. This is typical for PUs with high steel cores, where inductance decreases with frequency caused by the magnetic skin effect. An indication is the 1kHz inductance being lower than the 100Hz value.

2)The peak frequency of a low Q parallel resonant circuit is not identical to the theoretical resonant frequency. (I think this is discussed in Terman's book)

In any case the 100kHz measuring gives the most reliable result.
 
Antigua 4/15/2018 10:21 PM
Quote Originally Posted by Helmholtz View Post
Just make sure to select parallel mode to measure Cp at 100kHz and serial mode to measure inductance (Ls) at low frequency.

If you find a noticeable deviation from the capacitance calculated via frequency response, the reasons are:

1) Inductance at resonance frequency is lower than the value measured at a lower frequency. This is typical for PUs with high steel cores, where inductance decreases with frequency caused by the magnetic skin effect. An indication is the 1kHz inductance being lower than the 100Hz value.

2)The peak frequency of a low Q parallel resonant circuit is not identical to the theoretical resonant frequency. (I think this is discussed in Terman's book)

In any case the 100kHz measuring gives the most reliable result.
It works! I don't know how you knew this would work, but you're exactly correct. I measured a Seymour Duncan SSL-1 neck pickup, and got the exact same capacitance value 103pF , reported here a few months ago Seymour Duncan SSL-1, Analysis and Review | Fender Stratocaster Guitar Forum

Here's a pic of the measurement:

[ATTACH=CONFIG]48494[/ATTACH]

Same story with a Seymour Duncan SH-1N or "'59 neck", measured 98pF here Seymour Duncan SH-1N "59" Neck, Analysis & Review | GuitarNutz 2 , showing 99pF with the DE-5000

[ATTACH=CONFIG]48495[/ATTACH]

100kHz was the only test frequency that worked. For the SSL-1, at 10kHz, it shows 4.86pF in PAR, 39pF in SET, both of which are way off, and lower frequncies are increasingly off, showing capacitance values in the nano-farad range. At 100kHz, switching between SER and PAL shows the same value to within 1pF, for bot the SSL-1 and the SH-1N, so while I assume there is a good reason to use PAL mode, SER appears to be very close as well.

Note in the picture that I have the lead wires far apart from each other, and not touching, as when they're brought close together, you instantly see the capacitance rise by a few picofarads.

Since it appears the DE-5000 reliably measures capacitance as well as inductance (only SER mode is suitable for inductance, and the lowest frequency setting of 100Hz is most accurate), and that these value agree with those derived from impedance plots, it's therefore possible to calculate the resonant peak of the pickup using nothing more than the DE-5000.

The resonant peak itself is rarely useful, since in an electric guitar, a high degree of additional capacitance will bring the true resonant frequency down to a much lower value. For the surveying of pickups I've done, I also include resonant peaks with a 470pF capacitor across the pickup, which is intended to represent a guitar cable, and give a more realistic representation of what the pickup will do in situ. I will try putting a 470pF cap across the pickup, then I'll measure the capacitance again, and see how closely the capacitances sum, as well as determine how closely the calculated resonant peak "with load" comes to the peak measured with an oscilloscope. If it turns out the DE-5000 can effectively acquire all these data points, then the only advantage that would remain for bode plot measurements is in determining how much resonant damping occurs due to eddy currents.

Thanks again for tipping me off to the fact that an LCR meter can calculate the capacitance of a pickup at 100kHz, this is very valuable information to have.
 
okabass 4/16/2018 3:23 AM
Thank you for the tip Hemholtz and Antigua. Just ordered DE-5000 LCR meter from eBay.
 
Helmholtz 4/16/2018 6:30 AM
You are welcome.

I knew that it would work before I bought an LCR meter with the 100kHz option some years ago. My considerations were based on understanding of:

- the PU's equivalent circuit
- the measuring principle of LCR meters in series and parallel modes.

The measuring error can be estimated from the formula for the apparent capacitance in parallel resonant circuits, which can be derived from the imaginary part of the admittance (1/Z). It is essential that the measuring frequency is far above the circuits resonance, where the frequency response of the impedance shows a clear -6dB/octave (= purely capacitive) behavior.
A measuring frequency closer to resonance (i.e. 10khz) will give a lower meter reading, as the meter shows apparent capacitance. The latter decreases towards resonance and = 0 exactly at resonance.
 
Antigua 4/16/2018 1:20 PM
Quote Originally Posted by Helmholtz View Post
You are welcome.

I knew that it would work before I bought an LCR meter with the 100kHz option some years ago. My considerations were based on understanding of:

- the PU's equivalent circuit
- the measuring principle of LCR meters in series and parallel modes.

The measuring error can be estimated from the formula for the apparent capacitance in parallel resonant circuits, which can be derived from the imaginary part of the admittance (1/Z). It is essential that the measuring frequency is far above the circuits resonance, where the frequency response of the impedance shows a clear -6dB/octave (= purely capacitive) behavior.
A measuring frequency closer to resonance (i.e. 10khz) will give a lower meter reading, as the meter shows apparent capacitance. The latter decreases towards resonance and = 0 exactly at resonance.
Since most inductors have a much lower inductance, in the mH range, and would have a very high resonant peak, is it somewhat unusual to be able to find the parasitic capacitance of an inductor with an LCR meter, even if it does have a 100kHz test range?
 
Helmholtz 4/16/2018 1:42 PM
I am not sure, if I understand your question.

As noted earlier, the measurement error increases with 1/L and the measuring frequency has to lie in a frequency region, where the impedance drops proportionally with increasing frequency (i.e. -6dB/octave). In practice this means that the measuring frequency (100kHz) should be at least about a factor 10 higher than the resonant frequency, as a rule of thumb.
 
Antigua 4/16/2018 2:07 PM
Quote Originally Posted by Helmholtz View Post
I am not sure, if I understand your question.

As noted earlier, the measurement error increases with 1/L and the measuring frequency has to lie in a frequency region, where the impedance drops proportionally with increasing frequency (i.e. -6dB/octave). In practice this means that the measuring frequency (100kHz) should be at least about a factor 10 higher than the resonant frequency, as a rule of thumb.
So if 100kHz needs to be 10x above the resonant frequency, so this means 10kHz would be the highest resonant frequency for which this would work with accuracy. AFAIK, very few inductors have a resonant frequency that is as low as 10kHz, and that guitar pickups are special in that that have a very high inductance, generally above 2 henries. In a way, that means this is sort of a "trick", in that that this method would not work for most inductors, just those with especially low resonant peaks, and so we've "lucked out", in a sense.

Thanks again for participating here, your insights have really moved the ball forward in my pickup research. Are you involved with Gitec?
 
Helmholtz 4/16/2018 2:40 PM
Quote Originally Posted by Antigua View Post
So if 100kHz needs to be 10x above the resonant frequency, so this means 10kHz would be the highest resonant frequency for which this would work with accuracy. AFAIK, very few inductors have a resonant frequency that is as low as 10kHz, and that guitar pickups are special in that that have a very high inductance, generally above 2 henries. In a way, that means this is sort of a "trick", in that that this method would not work for most inductors, just those with especially low resonant peaks, and so we've "lucked out", in a sense.

Thanks again for participating here, your insights have really moved the ball forward in my pickup research. Are you involved with Gitec?
I would not call it a trick, just applied physics. Fortunately it works fine with typical high impedance PUs. You could use the method for resonant frequencies above 10kHz but would need to calculate the error caused by inductance and compensate. Inductors with even higher resonant frequencies require meters/equipment with higher measuring frequencies.
I know GITEC and had some contact. They don't seem to be interested in exchange/communication with people who refuse to become a member of the club. But I highly appreciate Manfred Zollner's book "Physik der Elektrogitarre".
 
Mike Sulzer 4/21/2018 4:58 AM
 
Mike Sulzer 4/21/2018 5:18 AM
Quote Originally Posted by Helmholtz View Post
Just make sure to select parallel mode to measure Cp at 100kHz and serial mode to measure inductance (Ls) at low frequency.

If you find a noticeable deviation from the capacitance calculated via frequency response, the reasons are:

1) Inductance at resonance frequency is lower than the value measured at a lower frequency. This is typical for PUs with high steel cores, where inductance decreases with frequency caused by the magnetic skin effect. An indication is the 1kHz inductance being lower than the 100Hz value.

2)The peak frequency of a low Q parallel resonant circuit is not identical to the theoretical resonant frequency. (I think this is discussed in Terman's book)

In any case the 100kHz measuring gives the most reliable result.
The lower apparent inductance measured at frequencies above about 100 Hz is mostly due to eddy currents in metal pickup parts. For example, the cores act like the secondary of a poorly coupled transformer. I write "apparent" because the meter is capable of measuring two numbers, the real and imaginary parts (or amplitude and phase) of the impedance, and therefore can be used to model one reactive component (L or C) and one resistance (series or parallel). A pickup is a more complicated circuit, with the eddy current losses involving loss (resistance) as well as inductance. (The Q of some pickups is determined mostly by the eddy current losses rather than wire resistance.) Therefore the measured inductance above 100 Hz is somewhat in error because a correct measurement requires more than two numbers. At a high enough frequency, the the inductive reactance of the eddy current effect should dominate, and therefore the capacitance as measured at 100 KHz should be very close. But i think it is important to evaluate the error with all this in mind.
 
Antigua 4/21/2018 10:52 AM
Regarding capacitance measurements with the DE-5000 at 100kHz, I mentioned some problems in this thread http://music-electronics-forum.com/t46007-3/#post493453 where an SSL-1 measures the same as what could be derived by measuring the inductance and the peak resonant frequency (about 100pF), but a Fender Fat 50 pickup had given a reading that was much too low (about 60pF instead of 120pF). I'm moving this over to this thread since this one is about capacitance and that one is about tapped single coils, which is have issues that go beyond capacitance.

I created extended impedence plots for both the SSL-1 and Fat 50, directly driving the pickup with a function generator, as opposed to using an external inducer coil.

Fat 50
[ATTACH=CONFIG]48571[/ATTACH]

SSL-1
[ATTACH=CONFIG]48572[/ATTACH]

This shows that very near the test frequency of 100kHz, the Fat 50 has some sort of secondary resonance at 98kHz. The SSL-1 has a similar secondary resonance, but it's 153kHz. It appears that the overlap of the DE-5000's test frequency and the secondary resonance prevents the Fat 50 from measuring correctly. Any idea what the source of that secondary inductance and resonance is? Could it be related to the lead wires?
 
Joe Gwinn 4/21/2018 11:38 AM
Quote Originally Posted by Antigua View Post
This shows that very near the test frequency of 100kHz, the Fat 50 has some sort of secondary resonance at 98kHz. The SSL-1 has a similar secondary resonance, but it's 153kHz. It appears that the overlap of the DE-5000's test frequency and the secondary resonance prevents the Fat 50 from measuring correctly. Any idea what the source of that secondary inductance and resonance is? Could it be related to the lead wires?
Without disturbing the lead dress, short the coil and use the DE-5000 to measure the inductance of its own leads. Given that number, how big must the capacitance be to explain the observed resonant frequency? This may be a clue.

Change the lead dress: First time, with leads twisted together. Second time, with leads as far apart as possible.

Does bringing a piece of soft ferrite, steel, copper, stainless steel close to the coil have any effect? And so on.
 
Helmholtz 4/21/2018 12:05 PM
What are all these needle artefacts? I don't see them in my impedance measurements. Also the high and low frequency slopes appear not to be correct. Did you measure as I proposed without the field coil?
The zigzag anomality is the result of an additional series and parallel resonance with higher resonant frequencies. I could show in simulations that such behaviour can be the result of partially shorted windings. Another explanation could be a very sloppy wind, where the winding is not carefully layered but outer turns are used to fill lower spaces thereby causing an uneven distribution of the distributed capacitance. I have not found a way yet to prove this idea wrong or right, as I am not winding.
 
Antigua 4/21/2018 12:19 PM
Quote Originally Posted by Helmholtz View Post
What are all these needle artefacts? I don't see them in my impedance measurements. Also the high and low frequency slopes appear not to be correct. Did you measure as I proposed without the field coil?
The zigzag anomality is the result of an additional series and parallel resonance with higher resonance frequencies. I could show in simulations that such behaviour can be the result of partially shorted windings. Another explanation could be a very sloppy wind, where the winding is not carefully layered but outer turns are used to fill lower spaces thereby causing an uneven distribution of the distributed capacitance. I have not found a way yet to prove this idea wrong or right, as I am not winding.
I don't know what causes the spikes, I assume some sort of extraneous noise, but patters such as resonances are evident regardless of the noise, so it's not preventing me from conducting tests.

Those are both interesting possibilities: 1) a particularly uneven layer distribution causing a non uniform distributed capacitance, or 2) an internal short that would essentially create a small shorted coil within the larger coil. I'm not so sure about #1 because the SSL-1 has this second peak too, and they're known to be machine wound.

#2 seems like an attractive explanation, a short creating a smaller shorted coil within the coil, but that raises more questions, such as how does the short occur, and why would there only happen to be one of them per each tested coil?

I'll test some more single coils and get more data points on the second high freq. resonances.
 
Helmholtz 4/21/2018 1:45 PM
Please do not use the field coil. It seems to distort the impedance frequency response. The field coil coupling only makes sense for plotting the transfer response, but tends to introduce EMI effects and distorts especially high frequency response.
 
Antigua 4/21/2018 1:46 PM
Quote Originally Posted by Helmholtz View Post
Please do not use the field coil. It seems to distort the impedance frequency response. The field coil coupling only makes sense for plotting the transfer response, but tends to introduce EMI effects and distorts especially high frequency response.
These recent plots are direct, with a 1meg resistor, no inducer coil.

I'm gathering more data points now, I just want to keep it all to one post.
 
Helmholtz 4/21/2018 1:58 PM
Here are some of my measurements of strat PUs, some showing anomalies:[ATTACH=CONFIG]48574[/ATTACH]
Please note the straight +/- 6dB/octave slopes below and above resonance. If the anomaly lies at 100kHz, the LCR meter will give a wrong result, as it can only read impedance/admittance at the single 100kHz frequency.
I used a 100k series resistor, as I did not care for open loop Q and 100K is close to the loading with two 250K pots and 1M amplifier input impedance.
 
Antigua 4/21/2018 2:21 PM
Quote Originally Posted by Helmholtz View Post
Here are some of my measurements of strat PUs, some showing anomalities:[ATTACH=CONFIG]48574[/ATTACH]
Please note the straight +/- 6dB/octave slopes below and above resonance. If the anomality lies at 100kHz, the LCR meter will give a wrong result, as it can only read impedance/admittance at the single 100kHz frequency.
Thanks for providing the plots. It's too bad arbitrary test frequencies can't be specified with the affordable meters. I'd think you could test a handful of higher frequencies and achieve good accuracy that way.

I'm trying to determine if it could be the lead wires could be involved, but so far it doesn't seem likely.

I notice your plots show about one prominent anomaly per pickup, all somewhat close together, with similar Q factors, and all above 100kHz.
 
Helmholtz 4/21/2018 2:40 PM
BTW, the one with almost no anomaly (Fralin) sounds by far best to me. It is the only one that has the great brilliance of a good vintage strat PU. (I own a set of original '59 strat PUs for reference.)
 
Antigua 4/21/2018 2:44 PM
Quote Originally Posted by Helmholtz View Post
BTW, the one with almost no anomaly (Fralin) sounds by far best to me. It is the only one that has the great brilliance of a good vintage strat PU. (I own a set of original '59 strat PUs for reference.)
I'm testing a Lollar Blackface neck, so far I'm up to 200kHz with no anomaly. The Fralin and the Lollar and hand guided pickups, where as the Fat 50 and SSL-1 are high volume production pickups, there might be something to that.
 
Antigua 4/21/2018 3:20 PM
Quote Originally Posted by Joe Gwinn View Post
Without disturbing the lead dress, short the coil and use the DE-5000 to measure the inductance of its own leads. Given that number, how big must the capacitance be to explain the observed resonant frequency? This may be a clue.
The anomalies are observed strictly with the Velleman bode plotter, and seem to vary from pickup to pickup, so I think the rig itself is mostly ruled out.

Quote Originally Posted by Joe Gwinn View Post
Change the lead dress: First time, with leads twisted together. Second time, with leads as far apart as possible.

Does bringing a piece of soft ferrite, steel, copper, stainless steel close to the coil have any effect? And so on.
That's a good idea about twisting the lead wires. I gave that a try and it didn't change the frequency at which the anomalous peak occurred.

I tried putting 470pF across the pickup, the resonant peak dropped form ~7kHz down to 3.9kHz, but the anomolous peak only appeared to drop very slightly, from 98kHz down to around 94kHz, with a Q that was lower by about half. Those higher frequency figures are sort of rough estimates, as you can see from the rather low resolution of the plot images. Cap values higher than 470pF seem to drown out the anomalous resonance.

As for placing permeable materials around the pickup, with steel Tele baseplates I've only ever been able to increase the inductance by about 150mH, so I don't think that wouldn't alter the circuit much. If theres value in seeing how inductance changes the anomalous peak, the better trick would probably by to find an steel pole Strat pickup with the anomaly, and then remove the pole pieces. That's easier said than done though, so I'd only do that if there were a hypothesis in place first.
 
Mike Sulzer 4/21/2018 4:32 PM
Quote Originally Posted by Helmholtz View Post
BTW, the one with almost no anomaly (Fralin) sounds by far best to me. It is the only one that has the great brilliance of a good vintage strat PU. (I own a set of original '59 strat PUs for reference.)
The pickup is loaded with about 500 pf and played through a system with about 5KHz bandwidth. It is hard to believe that anomalies at about 100 KHz have any effect on the sound.

It is also hard to believe that old Fender pickups have any particular special qualities. The brilliance of Fender pickups as compared to for example, humbuckers, is the result using Alnico cores that have lower conductance than steel. I suppose you could argue that the Alnico produced then has different eddy current losses than that produced now, but I have my doubts that this is a significant effect.
 
Helmholtz 4/22/2018 6:14 AM
The pickup is loaded with about 500 pf and played through a system with about 5KHz bandwidth. It is hard to believe that anomalies at about 100 KHz have any effect on the sound.
Right, the sound difference described can hardly be explained by the shown frequency responses. But I thought I should mention it nevertheless.

If it wasn't for such hard to understand sound effects, I would not waste my time (and some money) with countless parameter and response measurements, material analyses, simulations, literature researches, listening tests and so on. The standard PU filter response and parameter measurements can get quite boring and frustrating over time, as they often don't vary much and can explain only part of the PU's sound. But I am a physicist and want to find out.
 
Antigua 4/22/2018 12:40 PM
I measured four other machine wound Fender pickups, three from a Mexican Strat with a DCR or 6.9k, and one from Japan with a DCR of 5.4k, the three Mexican pickups showed single secondary peaks of 114kHz, 125kHZ and 150kHz, and the Japan made single coil have a second peak at 150kHz.

I made a LTSpice model that seems to result in a similar plot, it has another resonant coil in series with the lumped capacitance. Maybe this models what is happening, maybe not:

[ATTACH=CONFIG]48582[/ATTACH]

Another possible clue is that this doesn't appear to be something that effects inductors in general, as far as I can tell from Google searches, so aspects that make a pickup like most ordinary inductors can probably be ruled out, leaving qualities that are more specific to Stratocaster pickups.

The second peak frequency is definitely specific to the pickup, if I test the same pickups a second time, they show the same second peak. I've tried fiddling with the lead wires, but moving them around, twisting them, etc. doesn't make any difference.

All four tested pickups have steel pole pieces, and the three Mexican pickups are more or less identical in shape and size and probably turn count. If the pole pieces were a factor, I'd think the frequencies would all be a lot closer.

It's interesting that the second resonance varies from 100kHz to 150kHz from pickup to pickup. These machine wound pickups have coils that are rather flat, so it's likely the traversal is fairly uniform, and in fact the "hand guided" pickups don't seem as likely to have the anomalous peak. So oddly enough, the pickups with a uniform manufacturing method show a randomness in this second peak, where as the hand guided pickups, with randomly laid wire, are possibly more uniform, insofar as they don't have this peak, though it could be the case that's it's just at a much higher frequency and/or the resonance is suppressed to the point of being unobservable.




Quote Originally Posted by Helmholtz View Post
Right, the sound difference described can hardly be explained by the shown frequency responses. But I thought I should mention it nevertheless.
You have to account for cognitive bias in order to draw a conclusion with respect to hearing.
 
Mike Sulzer 4/22/2018 1:32 PM
The capacitance of a coil is a surprising subject. The capacitance of a single layer coil is a transmission line effect, not what you might think. A multilayer coil is complicated, and there is certainly some effect from electrostatic coupling between layers. But it might be with certain winding techniques you effectively get a multiple component circuit perhaps involving more than one transmission line (with capacitive reactance).

Quote Originally Posted by Antigua View Post
I measured four other machine wound Fender pickups, three from a Mexican Strat with a DCR or 6.9k, and one from Japan with a DCR of 5.4k, the three Mexican pickups showed single secondary peaks of 114kHz, 125kHZ and 150kHz, and the Japan made single coil have a second peak at 150kHz.

I made a LTSpice model that seems to result in a similar plot, it has another resonant coil in series with the lumped capacitance. Maybe this models what is happening, maybe not:

[ATTACH=CONFIG]48582[/ATTACH]

Another possible clue is that this doesn't appear to be something that effects inductors in general, as far as I can tell from Google searches, so aspects that make a pickup like most ordinary inductors can probably be ruled out, leaving qualities that are more specific to Stratocaster pickups.

The second peak frequency is definitely specific to the pickup, if I test the same pickups a second time, they show the same second peak. I've tried fiddling with the lead wires, but moving them around, twisting them, etc. doesn't make any difference.

All four tested pickups have steel pole pieces, and the three Mexican pickups are more or less identical in shape and size and probably turn count. If the pole pieces were a factor, I'd think the frequencies would all be a lot closer.

It's interesting that the second resonance varies from 100kHz to 150kHz from pickup to pickup. These machine wound pickups have coils that are rather flat, so it's likely the traversal is fairly uniform, and in fact the "hand guided" pickups don't seem as likely to have the anomalous peak. So oddly enough, the pickups with a uniform manufacturing method show a randomness in this second peak, where as the hand guided pickups, with randomly laid wire, are possibly more uniform, insofar as they don't have this peak, though it could be the case that's it's just at a much higher frequency and/or the resonance is suppressed to the point of being unobservable.






You have to account for cognitive bias in order to draw a conclusion with respect to hearing.
 
Mike Sulzer 4/22/2018 1:50 PM
Quote Originally Posted by Helmholtz View Post
Right, the sound difference described can hardly be explained by the shown frequency responses. But I thought I should mention it nevertheless.

If it wasn't for such hard to understand sound effects, I would not waste my time (and some money) with countless parameter and response measurements, material analyses, simulations, literature researches, listening tests and so on. The standard PU filter response and parameter measurements can get quite boring and frustrating over time, as they often don't vary much and can explain only part of the PU's sound. But I am a physicist and want to find out.
An excellent goal, IMO. My own bias is to look near the resonance for effects that can alter the harmonic content from what one might expect. For example, a pickup with steel cores must have a somewhat non-standard impedance shape around the resonance because of frequency varying eddy current losses (and reactance). I have not yet tried to compare the measured shape to what one expects with no eddy current losses, but it is a project for the future. My favorite way to measure pickup impedance is to put a resistor (say 2K) in series and drive this series combination with a signal with a broad frequency spectrum, sampling across the series combination, and across the resistor. Then with suitable processing, you can get the voltage across the pickup and the current through it. The ratio is the impedance with essentially no loading effects. You can measure hundreds of frequency points at once, giving a very good measurement of the frequency variation.
 
Helmholtz 4/22/2018 2:22 PM
The capacitance of a coil is a surprising subject. The capacitance of a single layer coil is a transmission line effect, not what you might think. A multilayer coil is complicated, and there is certainly some effect from electrostatic coupling between layers. But it might be with certain winding techniques you effectively get a multiple component circuit perhaps involving more than one transmission line (with capacitive reactance).
Thanks, this is exactly the direction/background of my second possible explanation. Now we need a volunteering winder who is willing to produce some extreme samples for measuring and verification purposes. In the meantime I will try to master transformers with varying degrees of coupling in LTSpice (being a beginner still) to simulate differently coupled parts of the PU coil.
 
Antigua 4/22/2018 2:30 PM
Quote Originally Posted by Helmholtz View Post
BTW, the one with almost no anomaly (Fralin) sounds by far best to me. It is the only one that has the great brilliance of a good vintage strat PU. (I own a set of original '59 strat PUs for reference.)
You have the Velleman setup as I do, it would be cool if you could share plots of the 59's you have on hand. I'm also curious to know what the inductance values are.

Someone mentioned the AlNiCo grades varrying in the 50's, I've received some pickups from China that supposedly had AlNiCo 5 pole pieces, but showed a Br value that was about 10% below expected, and yielded a higher Q factor than expected, due to higher resistivity. I don't know what made that Chinese AlNiCo different, be it a composition difference, or a production difference (though I'd think composition on account of the resistivity issue). Maybe the AlNiCo in your 59's has a similar sort of difference. I'd also be curious to know what the flux reading is at the tops of the pole pieces.

Regarding the notion that they sound subjectively better, there is an obvious bias to favor all things vintage, which might owe to nostalgia, or simple scarcity or it's perceived worth to others, or the famous guitarists who promote vintage gear, such Eric Clapton and Keith Richards. This bias impacts nearly every aspect of electric guitar, including caps, pots, hookup wire, body wood, the finish coat, steel hardware, even guitar amps and effects pedals. It seems that guitar manufactures just can't make a great guitar anymore, can they? It's such a pervasive and enduring bias that will power alone can't assure impartiality.
 
Helmholtz 4/22/2018 3:49 PM
You have to account for cognitive bias in order to draw a conclusion with respect to hearing.
I am well aware of possible cognitive bias. My ears and the attached computer are certainly biased by my personal sound preferences. But I have absolutely no desire to fool myself. More than often my listening tests did not confirm my (biased) expectations. Still, what counts in the end is sound and not measurements. I am not trying to persuade anybody to trust my assessments, instead I encourage everyone to do his own listening comparisons/evaluations and not rely on data only. But of course having and understanding measurement data helps to narrow down the variety of candidates.
It would be naive, though, to assume the a PU's total transfer reponse could be completely descibed by a simple linear passive low pass filter composed of lumped elements. A PU is essentialy a non-linear transducer. Sorry for going astray.
 
Helmholtz 4/22/2018 4:01 PM
Someone mentioned the AlNiCo grades varrying in the 50's, I've received some pickups from China that supposedly had AlNiCo 5 pole pieces, but showed a Br value that was about 10% below expected, and yielded a higher Q factor than expected, due to higher resistivity. I don't know what made that Chinese AlNiCo different, be it a composition difference, or a production difference (though I'd think composition on account of the resistivity issue). Maybe the AlNiCo in your 59's has a similar sort of difference. I'd also be curious to know what the flux reading is at the tops of the pole pieces.
Give me some time to look up my notes. Not going to dissect my old strat at the time, so no frequency plots. But I took resistance, inductance, capacitance and B values. Will send PM eventually.
 
Antigua 4/22/2018 4:02 PM
Quote Originally Posted by Helmholtz View Post
It would be naive, though, to assume the a PU's total transfer reponse could be completely descibed by a simple linear passive low pass filter composed of lumped elements.
It's not naive, it's a real possibility. I'd even argue it's the more probable possibility. You're well versed in scientific principles regarding physics, but you're not giving much regard to the necessity of double blind testing.
 
Helmholtz 4/22/2018 5:17 PM
You're well versed in scientific principles regarding physics, but you're not giving much regard to the necessity of double blind testing.
How could you know? I consider this a disrespectful allegation and a personal offense. Of course in the end my pickups have to appeal to me.

I hope you do extensive (double blind) listening tests with trained ears and experienced players- and not only rely on measurements.
 
Antigua 4/22/2018 5:29 PM
Here are extended impedance plots of four pickups I had made at home, which were intentionally wound with low uniformity and varying degrees of tension, including one which is a loose mess. Each is 8,000 turns with a DC resistance right around 6k ohms.

[ATTACH=CONFIG]48587[/ATTACH]


[ATTACH=CONFIG]48588[/ATTACH]

[ATTACH=CONFIG]48589[/ATTACH]

[ATTACH=CONFIG]48590[/ATTACH]

(the big messy pickup)
[ATTACH=CONFIG]48591[/ATTACH]

The earlier theory that a less consistent pickup might exhibit fewer or no anomolous high secondary resonant peaks appears to be squashed, as some of these pickups show several such peaks. The large, bushy Strat pickup which could never fit a cover over the top, has two very prominent secondary resonances, one at 86kHz and another at 153kHz. Two others show a single dominant secondary resonance, with additional peaks that look like ripples. Only one of the four, #3, appears to have no secondary resonance, although there is possibly a subtle knee at 64kHz.

Based on this information, I wondering if there is some sort of event that occurs randomly in the production of a pickup, one that causes multiple, smaller resonant circuits to appear within the coil.

Maybe it has to do with a segment of wire being laid, which comes physically very close to a segment of the coil that is many turns removed. For example, sometimes at the ends of the coil, you get crevices between the coil and the flat work, because the wire doesn't come right up to the edge of the flat work as it traverses back and forth. Then, at some random point, a segment of wire will manage to slip down into the crevice, which would cause that segment of wire that slipped down to be side be side with a portion of the coil that is maybe several hundred turns removed from itself.

While it's true that these higher peaks don't manifest in the audible frequency ranges, I think it's important to figure out what causes them, since it is a feature of the pickup all the same. It might even serve as a clue to indicate how a coil was made, without having to disassemble the coil to look at it directly.
 
Antigua 4/22/2018 5:32 PM
Quote Originally Posted by Helmholtz View Post
How could you know? I consider this a disrespectful allegation and a personal offense. Of course in the end my pickups have to appeal to me.

I hope you do extensive (double blind) listening tests with trained ears and experienced players- and not only rely on measurements.
In the past I had a similar view, that there is something special about pickups, and that my goal is to find it, but some other people pointed out to me that I haven't established with certainty that anything special really exists to find. It might be how I feel about the pickups, some good psychological connotation, that makes me believe there is something special. That's how it remains to this day, I'm not sure that I wasn't imagining the thing I set out to look for.

The easiest way to marry specs with subjective experience would be to, not only have a blind fold and a friend help you conduct an test free of extrinsic factors, but to also have as many specifications as possible about those pickups. Unfortunately, only the DC resistance is readily available, so I'm working towards providing other guitarists more extensive specs for popular pickups, including capacitance measures.
 
Helmholtz 4/23/2018 6:15 AM
Maybe it has to do with a segment of wire being laid, which comes physically very close to a segment of the coil that is many turns removed. For example, sometimes at the ends of the coil, you get crevices between the coil and the flat work, because the wire doesn't come right up to the edge of the flat work as it traverses back and forth. Then, at some random point, a segment of wire will manage to slip down into the crevice, which would cause that segment of wire that slipped down to be side be side with a portion of the coil that is maybe several hundred turns removed from itself.
This is what I ment. I suppose that the effect requires a contiguous portion of the winding with a discontinuity in magnetic coupling and/or distributed capacitance. Magnetic coupling decreases with distance from the center/core. Distributed capacitance changes with the arrangement of the turns.
I tend to think that the effect would show strongly (while at lower frquency) in a bifilar wound coil with the two windings wired in series. But this is just speculation at this point.
 
Helmholtz 4/23/2018 6:29 AM
In the past I had a similar view, that there is something special about pickups, and that my goal is to find it, but some other people pointed out to me that I haven't established with certainty that anything special really exists to find. It might be how I feel about the pickups, some good psychological connotation, that makes me believe there is something special. That's how it remains to this day, I'm not sure that I wasn't imagining the thing I set out to look for.
I don't quite understand what you are talking about. Probably due to my limited command of English. Not sure if I really need to know. Anyway, I prefer technical arguments/discussion.

But it definitely does not sound like an excuse to me.
 
okabass 4/23/2018 7:15 AM
Great info on the thread. Not that witchcraft or caveman level physics, which is so common when you read web pickup forums.
Thanks.
 
Helmholtz 4/23/2018 7:33 AM
My favorite way to measure pickup impedance is to put a resistor (say 2K) in series and drive this series combination with a signal with a broad frequency spectrum, sampling across the series combination, and across the resistor. Then with suitable processing, you can get the voltage across the pickup and the current through it. The ratio is the impedance with essentially no loading effects. You can measure hundreds of frequency points at once, giving a very good measurement of the frequency variation.
This sounds very interesting. But my expertise in signal and system theory is rather limited, so I have no feeling for the power and benefit of such method. What kind of signal do you use? A kind of noise?
But, being a pragmatic, my main question is: How do your results differ from those of standard single frequency point measurements with a current source? Can you show some?
 
Antigua 4/23/2018 8:34 AM
Quote Originally Posted by Mike Sulzer View Post
My favorite way to measure pickup impedance is to put a resistor (say 2K) in series and drive this series combination with a signal with a broad frequency spectrum, sampling across the series combination, and across the resistor. Then with suitable processing, you can get the voltage across the pickup and the current through it. The ratio is the impedance with essentially no loading effects. You can measure hundreds of frequency points at once, giving a very good measurement of the frequency variation.
Quote Originally Posted by Helmholtz View Post
This sounds very interesting. But my expertise in signal and system theory is rather limited, so I have no feeling for the power and benefit of such method. What kind of signal do you use? A kind of noise?
But, being a pragmatic, my main question is: How do your results differ from those of standard single frequency point measurements with a current source? Can you show some?
This sounds like what we've been doing in this thread, using the VElleman bode plot and function generator, similar to what is described here http://www.syscompdesign.com/assets/...ar-pickups.pdf but using a 1meg resistance instead of 56k.

Another option is to use the Velleman's frequency sweep with persistance, which yields the peak like this:

[ATTACH=CONFIG]48592[/ATTACH]

I though there was a way to feed it white noise with the function generator, which would reveal a peak in the FFT view even more quickly, but I'm not seeing the option.



Quote Originally Posted by Helmholtz View Post
This is what I ment. I suppose that the effect requires a contiguous portion of the winding with a discontinuity in magnetic coupling and/or distributed capacitance. Magnetic coupling decreases with distance from the center/core. Distributed capacitance changes with the arrangement of the turns.
I tend to think that the effect would show strongly (while at lower frquency) in a bifilar wound coil with the two windings wired in series. But this is just speculation at this point.
I remember you mentioning this possibility. If this is what is going on, it should be possible to model with LTSpice somehow. I'll work on that more later.
 
bbsailor 4/23/2018 10:22 AM
Quote Originally Posted by Antigua View Post
In the past I had a similar view, that there is something special about pickups, and that my goal is to find it, but some other people pointed out to me that I haven't established with certainty that anything special really exists to find. It might be how I feel about the pickups, some good psychological connotation, that makes me believe there is something special. That's how it remains to this day, I'm not sure that I wasn't imagining the thing I set out to look for.

The easiest way to marry specs with subjective experience would be to, not only have a blind fold and a friend help you conduct an test free of extrinsic factors, but to also have as many specifications as possible about those pickups. Unfortunately, only the DC resistance is readily available, so I'm working towards providing other guitarists more extensive specs for popular pickups, including capacitance measures.
When measuring pickups put a 200K Ohm resistor for single coils or 350K Ohm resistor in parallel for humbucker pickups with about 350pf capacitor (resistor and cap in parallel) across the pickup output to simulate what the pickup looks like loaded by the typical volume pot and the typical coax cable capacitance. These value represent the full load on the pickup when either a 250K Ohm pot or 500K Ohm pot is in parallel to the typical amp input impedance of 1 Meg Ohm. Doing the same measurements with this added load will probably change the upper frequency peaks that you are seeing. The main audio effect will be a slight reduction of the peak resonance near 3KHz to 5KHz due to the fact that the coil impedance is highest at pickup resonance and the pot loading will reduce the peak somewhat. The capacitance loading of the coax cable (350pf simulated load) will also lower the resonant frequency.

Bottom line: Try to do all tests in the same environment that the pickup sees when mounted in the guitar and used in typical situations about 10 ft from the amplifier. This will allow your ears to be in tune better with what you see on the graphs.

Joseph J. Rogowski
 
Helmholtz 4/23/2018 10:51 AM
This sounds like what we've been doing in this thread, using the VElleman bode plot and function generator, similar to what is described here http://www.syscompdesign.com/assets/...ar-pickups.pdf but using a 1meg resistance instead of 56k.
I don't quite agree with some details in this papers. As a consequence the f-responses in figure 5 are wrong. Will elaborate if someone cares.
 
Antigua 4/23/2018 10:59 AM
Quote Originally Posted by Helmholtz View Post
I don't quite agree with some details in this papers. As a consequence the f-responses in figure 5 are wrong. Will elaborate if someone cares.
This PDF is a prominent search result when searching for information on how to measure the response of guitar pickups. Any critique you have would be valuable to anyone who finds both that PDF and this thread.
 
Mike Sulzer 4/23/2018 12:00 PM
Quote Originally Posted by Helmholtz View Post
This sounds very interesting. But my expertise in signal and system theory is rather limited, so I have no feeling for the power and benefit of such method. What kind of signal do you use? A kind of noise?
But, being a pragmatic, my main question is: How do your results differ from those of standard single frequency point measurements with a current source? Can you show some?
The current version uses Golay complementary sequences, that is, a pair of codes that together weight all frequencies equally, or have no side lobes, as a radar person might say. This gives faster more accurate measurements than the noise like signals I have used before. (Since the measurement is a ratio, taken in the frequency domain, equal weighting in the code is not strictly necessary for accuracy, but it does give more uniform signal to noise ratio.)

There is no need for a very high input impedance amplifier to drive the sampler, nor for a good current source. It can be done with a two channel recording interface, cheap these days, something many people already have.

I use an audio package in Python, incorporated into custom software for the signal processing.

Here is an example taken with an earlier version, but it is interesting because it shows two measurements with the same coil, alnico cores and steel.

[ATTACH=CONFIG]48595[/ATTACH]
 
Antigua 4/23/2018 1:07 PM
Quote Originally Posted by bbsailor View Post
When measuring pickups put a 200K Ohm resistor for single coils or 350K Ohm resistor in parallel for humbucker pickups with about 350pf capacitor (resistor and cap in parallel) across the pickup output to simulate what the pickup looks like loaded by the typical volume pot and the typical coax cable capacitance. These value represent the full load on the pickup when either a 250K Ohm pot or 500K Ohm pot is in parallel to the typical amp input impedance of 1 Meg Ohm. Doing the same measurements with this added load will probably change the upper frequency peaks that you are seeing. The main audio effect will be a slight reduction of the peak resonance near 3KHz to 5KHz due to the fact that the coil impedance is highest at pickup resonance and the pot loading will reduce the peak somewhat. The capacitance loading of the coax cable (350pf simulated load) will also lower the resonant frequency.

Bottom line: Try to do all tests in the same environment that the pickup sees when mounted in the guitar and used in typical situations about 10 ft from the amplifier. This will allow your ears to be in tune better with what you see on the graphs.

Joseph J. Rogowski
I use 200k and 470pF as fixed "loaded" values. I use the same values for humbuckers and single coils for the sake of consistency. Myself and someone else had settled on these values, and then I found out later Helmuth Lemme used the exact same values here http://www.planetz.com/wp-content/up..._Technique.pdf , so it seems to be a reasonable in-between standard.
 
Helmholtz 4/23/2018 2:43 PM
I use 200k and 470pF as fixed "loaded" values. I use the same values for humbuckers and single coils for the sake of consistency. Myself and someone else had settled on these values, and then I found out later Helmuth Lemme used the exact same values here http://www.planetz.com/wp-content/up..._Technique.pdf , so it seems to be a reasonable in-between standard.
After extensive testing I have settled with guitar cables having 1000pF to 1200pF, both for my Strats and my Les Pauls. The wiring harness in a typical vintage LP adds 300 to 500pF. A tube amplifier input adds another 150pF typically. I use vintage style PUs.
A realistic load resistance for strats is 100k to 200K(bridge PU) and 200k for LPs. This includes typical amplifier/pedals input resistance.

I don't think any pro player would be comfortable with a stage cable of 10ft or less.
 
Antigua 4/23/2018 2:53 PM
Quote Originally Posted by Helmholtz View Post
As a guitar player I have settled (after extensive testing) with guitar cables having 1000pF to 1200pF, for my Strats as well as my Les Pauls. The wiring harness in a typical LP adds 400+pF.
400pF for the pots and hookups? I measured 70pF per foot for the braided wire used with PAF humbuckers, but 400pF overall seems high.

Fender guitars usually feature non shielded hookups and wiring, so the capacitance there is really low, well under 50pF I'd recon.

Quote Originally Posted by Helmholtz View Post
A tube amplifier input adds another 150pF typically. I use vintage style PUs.
A realistic load resistance for strats is 100K..200K and 200k for LPs. This includes typical amplifier/pedals input resistance.
Since peak freq. will vary from rig to rig, I think it's best to settle on a "center", and then let people shift the frequency up or down mentally, depending on their own rig. So if you know you like 1000pF cables, you know that you will have a peak that is somewhat lower than the standard loaded data points. If the intrinsic L and C are known, then any loaded peak freq. can be solved for, and because inductance factors more prominently, L is is the more valuable metric to have on hand.

Quote Originally Posted by Helmholtz View Post
I don't think any pro player would be comfortable with a stage cable of 10ft or less.
Pros often use wireless units too, which sometimes have selectable capacitance, or a fixed values. I analyzed a Line 6 G10 and found that it imparted about 120pF capacitance http://www.strat-talk.com/threads/th...ay-g10.467237/
 
Helmholtz 4/23/2018 4:08 PM
400pF for the pots and hookups? I measured 70pF per foot for the braided wire used with PAF humbuckers, but 400pF overall seems high.
The neck PU signal in a LP runs through around 3ft of coax wire between PU and jack. My Stew Mac and Allparts coax wires measure over 120pF/foot. But this value strongly increases with ambient humidity in summer months. The values I specified are quite realistic.
The self-capacitance of humbuckers with cloth-insulated coax cable is often dominated by the cable attached .
The capacitance of the guitar cable is the strongest influencer of PU frequency response besides inductivity.
 
Helmholtz 4/23/2018 4:32 PM
The current version uses Golay complementary sequences, that is, a pair of codes that together weight all frequencies equally, or have no side lobes, as a radar person might say. This gives faster more accurate measurements than the noise like signals I have used before. (Since the measurement is a ratio, taken in the frequency domain, equal weighting in the code is not strictly necessary for accuracy, but it does give more uniform signal to noise ratio.)

There is no need for a very high input impedance amplifier to drive the sampler, nor for a good current source. It can be done with a two channel recording interface, cheap these days, something many people already have.

I use an audio package in Python, incorporated into custom software for the signal processing.

Here is an example taken with an earlier version, but it is interesting because it shows two measurements with the same coil, alnico cores and steel.

alandsteelcores.png
Does this alternative method reveal any additional information compared to standard methods? Is there any direct comparison? Of course the effect of a finite source impedance can always be compensated via calculation.
 
Mike Sulzer 4/23/2018 6:04 PM
Quote Originally Posted by Helmholtz View Post
Does this alternative method reveal any additional information compared to standard methods? Is there any direct comparison? Of course the effect of a finite source impedance can always be compensated via calculation.
Well, I suppose there could be no more standard way of measuring impedance than by forming a ratio, as a function of frequency, of the voltage across and the current through the device, using superior inexpensive technology. But let's leave this aside for now. But yes, there is more to do. Consider the capacitance: it is of little importance itself because it is so small compared to other capacitances in the guitar circuit. But it does get in the way because it causes a resonance that makes it hard to see the effects of the metal on the impedance. So the capacitance is found, just for the purpose of removing it from the impedance, by a non-linear least squares fit to a set of samples taken well above the resonance where the capacitance is dominant. The low frequency inductance and the resistance are in the model of the impedance, and the parameters fitted to are the capacitance, and two parameters associated with the coupling to the metal, k, the coupling coefficient, and Rse, the effective resistance of the "secondary" reflected back to the primary. Once the C is found it is taken out of the impedance, and you can see in the plots that the real part increases above the pickup resistance as frequency increases, and the imaginary part decreases below the reactance of the coil inductance.
 
Antigua 4/23/2018 6:32 PM
Quote Originally Posted by Helmholtz View Post
My Stew Mac and Allparts coax wires measure over 120pF/foot. But this value strongly increases with ambient humidity in summer months. The values I specified are quite realistic.
I first read about the cloth shield hookup varying in capacitance by humidity in Manfred Zollner's book Physik der Elektrogitarre, and it's a rather significant claim, because people make hay over the littlest things, but never talk about how their pickups sound different between dry and wet climates. I tried a little experiment where I left shielded cloth hookup wire outside, measured the capacitance, the put it in the oven to dry it out, then measured again, but the capacitance didn't change much. I might give it another try, though, ensuring that it's very damp and then very dry.
 
Antigua 4/24/2018 12:43 AM
I think this model jives somewhat closely with the notion that having the winding fall into the edges puts a capacitance across two distant sections of the coil:

[ATTACH=CONFIG]48611[/ATTACH]

In this model, the series inductance and resistance are broken into three sets, with the middle "RL" set having a capacitance in parallel with it, as would happen if, say, the 3,000th winding somehow managed to come side by side with the 2,500th winding, by falling into a gap at the edge of the coil, or something or that sort. The result of the model is a side by side impedance dip and spike, which isn't 100% like what is seen in the practical plots, which appears to have only an impedance spike, but it has the similar characteristic of a second high frequency resonance. Maybe with a little more refinement the model can duplicate the practical plot completely.
 
freefrog 4/24/2018 1:32 AM
Quote Originally Posted by Helmholtz View Post
The neck PU signal in a LP runs through around 3ft of coax wire between PU and jack. My Stew Mac and Allparts coax wires measure over 120pF/foot. But this value strongly increases with ambient humidity in summer months. The values I specified are quite realistic.
Even humidity aside, my experience agrees with your statements.

I've measured 268pF per meter in normal conditions on the braided shielded that I use.

And there's a serious lenght of cable in a LP... This site recommends to have 5ft at disposal for such a wiring: Six String Supplies ? How to Wire a Les Paul (50s Wiring)

Let's add to it the cables coming from the pickups themselves and the stray capacitance of other components: it forms a highly capacitive inner wiring. Its tonal effect is especially obvious IME when both pickups are selected.

I don't think any pro player would be comfortable with a stage cable of 10ft or less.
BTW, Helmoltz, you have an outstanding brand of cable in Germany: Sommer... Their LLX coax. wire measures 52pF per meter (published value that I've checked with a lab meter). And IME, it's a sturdy cable, whose only relative flaw is its limited flexibility.
Of course, this observation hasn't much interest for you since you use high capacitance cables. But it would be a pity not to share with all potential readers a possibly useful info, while Sommer cables are so unequally known by musicians around the World... :-)
 
Helmholtz 4/24/2018 7:31 AM
BTW, Helmoltz, you have an outstanding brand of cable in Germany: Sommer... Their LLX coax. wire measures 52pF per meter (published value that I've checked with a lab meter). And IME, it's a sturdy cable, whose only relative flaw is its limited flexibility.
Yes, this Sommer cable is fine, as is e.g. Klotz GY 107 ("La Grange"). I use both types with different lenghts. But generally I have no need for extra low specific capacitance, as this forces me to buy and use cables measuring 15m or more in length.
 
Helmholtz 4/24/2018 8:11 AM
Well, I suppose there could be no more standard way of measuring impedance than by forming a ratio, as a function of frequency, of the voltage across and the current through the device, using superior inexpensive technology. But let's leave this aside for now. But yes, there is more to do. Consider the capacitance: it is of little importance itself because it is so small compared to other capacitances in the guitar circuit. But it does get in the way because it causes a resonance that makes it hard to see the effects of the metal on the impedance. So the capacitance is found, just for the purpose of removing it from the impedance, by a non-linear least squares fit to a set of samples taken well above the resonance where the capacitance is dominant. The low frequency inductance and the resistance are in the model of the impedance, and the parameters fitted to are the capacitance, and two parameters associated with the coupling to the metal, k, the coupling coefficient, and Rse, the effective resistance of the "secondary" reflected back to the primary. Once the C is found it is taken out of the impedance, and you can see in the plots that the real part increases above the pickup resistance as frequency increases, and the imaginary part decreases below the reactance of the coil inductance.

Thanks. Seems like a powerful and useful tool. Unfortunately I don't have the time to dig any deeper and make myself familiar with your method at present.
 
Helmholtz 4/24/2018 4:07 PM
Quote Originally Posted by Antigua View Post
This PDF is a prominent search result when searching for information on how to measure the response of guitar pickups. Any critique you have would be valuable to anyone who finds both that PDF and this thread.
I edited my first post and changed it completely as I noticed, that it was not the author's intend to plot the PU's impedance accurately. For the Lissajous method a series resistor of 56k seems fine.

Here are my comments on this http://www.syscompdesign.com/assets/...ar-pickups.pdf paper.

First of all I acknowledge that it gives some useful information regarding measurement methods of PUs' parameters. The description of a magnetic PU as a Variable Reluctance Sensor is perfect. But:

Measuring PU inductance via Lissajous figure
This is an excellent and accurate method to determine the inductance of a parallel resonant circuit like a PU. But it requires the capacitance to be known exactly.

There are two problems associated with measuring the inductance with an LCR meter at a fixed frequency:

1) The meter can only measure apparent inductance. The apparent L of a PU (parallel resonant circuit) increases steadily with increasing frequency below and up to resonance, caused by the effect of capacitance. Apparent L has no practical meaning for PUs and is only a theoretical way to descibe the systematic measuring error of LCR meters. As this error increases with frequency, the value at the lowest measuring frequency is the most meaningful.

2) Eddy current effects in conductive parts (especially in ferromagnetic cores ->magnetic skin effect) reduce the effective L with increasing frequency. Inductors with conductive, ferromagnetic cores do not have a single true inductance. Instead L is a function of frequency. This means that the L value at 100Hz is not per se better or truer than the value at a higher frequency.

What we actually want to know is the (effective) L at or close to the resonant frequency in real life operation. This is where the Lissajous method comes in. Done carefully, it can deliver the correct effective L at the chosen resonant frequency of interest.

As mentioned before, for accurate results the total capacitance Ctot= Cpu+Cadd needs be known exactly. If Ctot is too low by 10%, your calculated L will be too large by 10%.
Cadd can be easily measured with an LCR but also Cpu should be determined beforehand at least approximately.
The method indicated in the article, namely "overpowering" an unknown Cpu by a huge Cadd of several nFs, will give the effective L at a much too low frequency. The result will only be useful for PUs where L does not depend on frequency. But in these cases you may as well use your LCR meter at 100Hz.


And here is the more important part of my comments, dealing with measuring the PU's transfer response:

Measuring PU transfer response requires access to an input port. Inserting a signal voltage source in series with the inductor part as typically done in simulations is not possible in real life. Instead, the well accepted method is to use the PU coil as secondary in a current transformer arrangement. The idea is to inject a current into the PU coil (inductance) via a coupled external coil driven by constant current and measure the resulting voltage across the PU terminals. Mind that driving the external coil directly by a (low impedance) voltage source would load down the PU and change its frequency response.
The induced constant current in the PU coil produces a voltage across its inductance, rising proportionally with frequency and consequently the PU shows a typical bandpass behaviour.
The main requirement for the external primary circuit is that the drive current must stay constant for all frequencies to be measured. This means that not only the self-resonance of the field coil has to lie far above the highest frequency of interest but also that the impedance of the field coil stays negligible compared to the total series resistance (279 Ohms in the example). With the values given in the article the corner frequency for this requirement is around 1.2kHz. Above this frequency the drive current drops with 6dB/octave and distorts the measured frequency response as can be seen in the PU responses of figure 5. The cure is to increase the L/R ratio by a factor of 20 or more.
 
Antigua 4/25/2018 12:54 PM
Quote Originally Posted by Helmholtz View Post
1) The meter can only measure apparent inductance. The apparent L of a PU (parallel resonant circuit) increases steadily with increasing frequency below and up to resonance, caused by the effect of capacitance. Apparent L has no practical meaning for PUs and is only a theoretical way to descibe the systematic measuring error of LCR meters. As this error increases with frequency, the value at the lowest measuring frequency is the most meaningful.

2) Eddy current effects in conductive parts (especially in ferromagnetic cores ->magnetic skin effect) reduce the effective L with increasing frequency. Inductors with conductive, ferromagnetic cores do not have a single true inductance. Instead L is a function of frequency. This means that the L value at 100Hz is not per se better or truer than the value at a higher frequency.

What we actually want to know is the (effective) L at or close to the resonant frequency in real life operation. This is where the Lissajous method comes in. Done carefully, it can deliver the correct effective L at the chosen resonant frequency of interest.

As mentioned before, for accurate results the total capacitance Ctot= Cpu+Cadd needs be known exactly. If Ctot is too low by 10%, your calculated L will be too large by 10%.
Cadd can be easily measured with an LCR but also Cpu should be determined beforehand at least approximately.
The method indicated in the article, namely "overpowering" an unknown Cpu by a huge Cadd of several nFs, will give the effective L at a much too low frequency. The result will only be useful for PUs where L does not depend on frequency. But in these cases you may as well use your LCR meter at 100Hz.
Thanks for the write up. For some reason this forum truncates this URL http://www.syscompdesign.com/assets/...ar-pickups.pdf on your posts.

As for inductance varying with frequency, I've noticed that Fender pickups, with AlNiCo pole pieces and little to no other metal parts, show about the same inductance at 1kHz test freq as they do at 100 or 120 Hz. It's only pickups with steel cores that show incorrect readings. One reason I prefer taking down the loaded and unloaded resonant peaks of pickups is because 1) it's a value that relates more closely to audible performance, and 2) it overcomes errors that might arise from trying to solve for peak freq. from incorrect values for L and C.

Quote Originally Posted by Helmholtz View Post

Measuring PU transfer response requires access to an input port. Inserting a signal voltage source in series with the inductor part as typically done in simulations is not possible in real life. Instead, the well accepted method is to use the PU coil as secondary in a current transformer arrangement. The idea is to inject a current into the PU coil (inductance) via a coupled external coil driven by constant current and measure the resulting voltage across the PU terminals. Mind that driving the external coil directly by a (low impedance) voltage source would load down the PU and change its frequency response.
Can the field coil still load down the pickup even if the coupling factor is very small compared to a traditional transfomer?

Quote Originally Posted by Helmholtz View Post
The induced constant current in the PU coil produces a voltage across its inductance, rising proportionally with frequency and consequently the PU shows a typical bandpass behaviour.
The main requirement for the external primary circuit is that the drive current must stay constant for all frequencies to be measured. This means that not only the self-resonance of the field coil has to lie far above the highest frequency of interest but also that the impedance of the field coil stays negligible compared to the total series resistance (279 Ohms in the example). With the values given in the article the corner frequency for this requirement is around 1.2kHz. Above this frequency the drive current drops with 6dB/octave and distorts the measured frequency response as can be seen in the PU responses of figure 5. The cure is to increase the L/R ratio by a factor of 20 or more.
The PCSU200 shows "output impedance: 50ohm" https://www.velleman.eu/products/view/?id=407512 , so the field coil's impedance would need to be well below 50 ohms, otherwise series resistance must be added?
 
Antigua 4/25/2018 1:00 PM
Also a question, does anyone know these plots typically show a slope that is much lower than 6dB/oct below 200Hz?

[ATTACH=CONFIG]48632[/ATTACH]
 
Mike Sulzer 4/25/2018 1:25 PM
Quote Originally Posted by Antigua View Post
Also a question, does anyone know these plots typically show a slope that is much lower than 6dB/oct below 200Hz?

[ATTACH=CONFIG]48632[/ATTACH]

Without checking all the details of what you are doing, I would guess that it is the effect of the coil resistance.
 
Antigua 4/25/2018 1:32 PM
Quote Originally Posted by Mike Sulzer View Post
Without checking all the details of what you are doing, I would guess that it is the effect of the coil resistance.
This plot is with a 1meg resistor in series with the pickup, then comparing the voltage across the resistor and the pickup, but the same thing happens when using an external inducer coil in a transformer configuration.
 
Mike Sulzer 4/25/2018 1:35 PM
Quote Originally Posted by Helmholtz View Post
And here is the more important part of my comments, dealing with measuring the PU's transfer response:

Measuring PU transfer response requires access to an input port. Inserting a signal voltage source in series with the inductor part as typically done in simulations is not possible in real life. Instead, the well accepted method is to use the PU coil as secondary in a current transformer arrangement. The idea is to inject a current into the PU coil (inductance) via a coupled external coil driven by constant current and measure the resulting voltage across the PU terminals. Mind that driving the external coil directly by a (low impedance) voltage source would load down the PU and change its frequency response.
The induced constant current in the PU coil produces a voltage across its inductance, rising proportionally with frequency and consequently the PU shows a typical bandpass behaviour.
The main requirement for the external primary circuit is that the drive current must stay constant for all frequencies to be measured. This means that not only the self-resonance of the field coil has to lie far above the highest frequency of interest but also that the impedance of the field coil stays negligible compared to the total series resistance (279 Ohms in the example). With the values given in the article the corner frequency for this requirement is around 1.2kHz. Above this frequency the drive current drops with 6dB/octave and distorts the measured frequency response as can be seen in the PU responses of figure 5. The cure is to increase the L/R ratio by a factor of 20 or more.
I use coils with diameter equal to or smaller than a pole piece radius with 3 to 6 turns, driven from an audio amp through an 8 ohm resistor, with a current of about 1 amp. Coupling is very small. Driving a pickup with a pickup size coil, as some otherwise clever people do, seems like asking for trouble.

You can make a response model from the parameters derived from an impedance measurement that works well. I think the only reason for using a driving coil is to include the eddy current loss encountered in passing through an extra thick cover.
 
Mike Sulzer 4/25/2018 1:37 PM
Quote Originally Posted by Antigua View Post
This plot is with a 1meg resistor in series with the pickup, then comparing the voltage across the resistor and the pickup, but the same thing happens when using an external inducer coil in a transformer configuration.
In either case, the coil resistance places a lower limit on the magnitude of the coil impedance.
 
Helmholtz 4/25/2018 2:22 PM
Can the field coil still load down the pickup even if the coupling factor is very small compared to a traditional transfomer?
I have no values for the coupling factor. Generally a coupling factor below 100% introduces additional series inductance in the equivalent circuit. But why don't you just measure and compare? Call it loading down or not, in result the (high) frequency response will change.

The PCSU200 shows "output impedance: 50ohm" https://www.velleman.eu/products/view/?id=407512 , so the field coil's impedance would need to be well below 50 ohms, otherwise series resistance must be added?
You have to add the field coil's resistance to the series resistance for total circuit resistance Rtot. What matters is the Rtot/L ratio. It should be well above 150kOhm/H. In other words the corner frequency should lie well above the frequency range analysed and is given by f=Rtot/(2pi*L). This can be achieved by increasing series and/or coil resistance as well as by decreasing field coil inductance.

Also a question, does anyone know these plots typically show a slope that is much lower than 6dB/oct below 200Hz?

v2adfer.png
My impedance plots of PUs without anomalies show almost perfect -6dB/octave (i.e. capacitive) behaviour above ca. 100kHz. To stay ahead of noise floor I recommend max. generator voltage and automatic voltage scale.
Anomalies indicate that the PU's behaviour is not purely capactive (but disturbed by the interaction with a smaller separated part of the inductance) in the corresponding frequency range, thus no clear -6dB/octave slope.
 
Helmholtz 4/25/2018 2:34 PM
Quote Originally Posted by Mike Sulzer View Post
In either case, the coil resistance places a lower limit on the magnitude of the coil impedance.
But only below resonance. As the whole thing is shunted by the distributed capacitance, impedance tends to 0 for very high frequencies.
 
Mike Sulzer 4/25/2018 3:22 PM
Quote Originally Posted by Helmholtz View Post
But only below resonance. As the whole thing is shunted by the distributed capacitance, impedance tends to 0 for very high frequencies.
I believe Antigua's question was for below 200 Hz where the pickup coil resistance sets a lower limit on the magnitude of the pickup coil impedance.
 
Helmholtz 4/25/2018 3:46 PM
You are right, I am sorry. Did not read carefully. Impedance always starts horizontally with DCR from 0 Hz. But the bandpass transfer response is different and must have 0 signal at 0 Hz.
 
Mike Sulzer 4/25/2018 4:19 PM
Quote Originally Posted by Helmholtz View Post
And here is the more important part of my comments, dealing with measuring the PU's transfer response:

Measuring PU transfer response requires access to an input port. Inserting a signal voltage source in series with the inductor part as typically done in simulations is not possible in real life. Instead, the well accepted method is to use the PU coil as secondary in a current transformer arrangement. The idea is to inject a current into the PU coil (inductance) via a coupled external coil driven by constant current and measure the resulting voltage across the PU terminals. Mind that driving the external coil directly by a (low impedance) voltage source would load down the PU and change its frequency response.
The induced constant current in the PU coil produces a voltage across its inductance, rising proportionally with frequency and consequently the PU shows a typical bandpass behaviour.
The main requirement for the external primary circuit is that the drive current must stay constant for all frequencies to be measured. This means that not only the self-resonance of the field coil has to lie far above the highest frequency of interest but also that the impedance of the field coil stays negligible compared to the total series resistance (279 Ohms in the example). With the values given in the article the corner frequency for this requirement is around 1.2kHz. Above this frequency the drive current drops with 6dB/octave and distorts the measured frequency response as can be seen in the PU responses of figure 5. The cure is to increase the L/R ratio by a factor of 20 or more.
The current in the field or exciter coil creates an ac magnetic field that induces a voltage, not a current, in series with the pickup coil. This follows from Maxwell's equation or the law of magnetic induction. Current flows if there is a load on the coil, such as the coil capacitance or the loading caused by eddy currents in the cores, etc. (A so called current transformer is a tightly coupled transformer driven from a high impedance so that the current in the secondary is related to that in the primary by the turns ratio. The very loosely coupled situation here does not behave that way.)

You describe one way to make the current through the field coil independent of frequency: make the inductive reactance low compared to the dc resistance of the coil across the whole useful frequency range. A better way is to drive the coil with a current source, that is, a circuit with an output impedance much higher than the impedance of the coil at any useful frequency.
 
Antigua 4/26/2018 12:36 AM
Quote Originally Posted by Mike Sulzer View Post
The current in the field or exciter coil creates an ac magnetic field that induces a voltage, not a current, in series with the pickup coil. This follows from Maxwell's equation or the law of magnetic induction. Current flows if there is a load on the coil, such as the coil capacitance or the loading caused by eddy currents in the cores, etc. (A so called current transformer is a tightly coupled transformer driven from a high impedance so that the current in the secondary is related to that in the primary by the turns ratio. The very loosely coupled situation here does not behave that way.)

You describe one way to make the current through the field coil independent of frequency: make the inductive reactance low compared to the dc resistance of the coil across the whole useful frequency range. A better way is to drive the coil with a current source, that is, a circuit with an output impedance much higher than the impedance of the coil at any useful frequency.

Thanks for your last few posts, you cleared up several things that were not clicking for me before, especially about there being a voltage, but not necessarily a current unless a load exists on the pickup. I know that's a basic idea, but I was slow to connect the dots.

And the other point, about the resistance preceding the reactance at low frequencies, I modeled this with LTSpice. The different plot lines indicate three steps of resistance, 10k, 20k and 30k ohms, with the flat, low frequency portion extending further for each increase in step (as well as lowering the Q factor).

[ATTACH=CONFIG]48644[/ATTACH]

Something I'm confused about though, are the circumstances under which +6dB...-6dB/oct slopes emerge, as seen in "raw" bode plots, and the above LTSpice model, but only when the AC voltage source has been placed outside of the pickup.

When a pickup is modeled as with the AC source inside the pickup, as shown below, there is a 0dB/oct line, then the resonant peak, and then a -12dB/oct slope:

[ATTACH=CONFIG]48645[/ATTACH]

In practical testing, both the exciter / field coil method, as well we putting the pickup in series with the function generator, both yield +dB...-6dB/oct slopes, as seen in the first simulation. Driving the pickup with a series voltage obviously puts the voltage source outside of the pickup, but shouldn't the exciter coil method place the voltage inside of the pickup? Why does this testing method not result in a 0dB/oct..-12dB/oct plot, as is seen in the second screen shot?

Another issue with the model which has the voltage source inside the pickup, is as seen in the second screen shot, that increasing the series resistance from 10k to 20k to 30k has no apparent impact on the 0dB/oct slope, as it does with the-6dB slope in the first screen shot.
 
Antigua 4/26/2018 12:58 AM
Quote Originally Posted by Helmholtz View Post
I have no values for the coupling factor. Generally a coupling factor below 100% introduces additional series inductance in the equivalent circuit. But why don't you just measure and compare? Call it loading down or not, in result the (high) frequency response will change.

You have to add the field coil's resistance to the series resistance for total circuit resistance Rtot. What matters is the Rtot/L ratio. It should be well above 150kOhm/H. In other words the corner frequency should lie well above the frequency range analysed and is given by f=Rtot/(2pi*L). This can be achieved by increasing series and/or coil resistance as well as by decreasing field coil inductance.
In that past, I had measured a pickup using an exciter, both with and without a resistor in series with the exciter. It made no difference in the plot lines, but the added resistance made the exciter coil weaker, reducing the S/N ratio. So if the coupling is lower, the series inductance is higher, but apparently not high enough to interfere with measurements.
 
Helmholtz 4/26/2018 5:59 AM
Quote Originally Posted by Antigua View Post
In that past, I had measured a pickup using an exciter, both with and without a resistor in series with the exciter. It made no difference in the plot lines, but the added resistance made the exciter coil weaker, reducing the S/N ratio. So if the coupling is lower, the series inductance is higher, but apparently not high enough to interfere with measurements.
I cannot comment on your measurements. I gave you all the info necessary to make sure you have constant exciter current over frequency. When the exciter current falls with increasing frequency, so will your PU output voltage and thus frequency response will deviate. This effect can be seen in figure 5 of the document, where the responses show a pseudo plateau starting around 1kHz, just as predicted by the formula. In reality the PU bandpass response gets (increasingly) steeper towards resonance.

You may verify the frequency (in)dependance of the drive current by measuring the voltage over a series resistor, via the second channel of the Velleman. The effect of a decreasing drive may be partly masked by the resonance of the PU.
 
Helmholtz 4/26/2018 7:18 AM
The current in the field or exciter coil creates an ac magnetic field that induces a voltage, not a current, in series with the pickup coil. This follows from Maxwell's equation or the law of magnetic induction. Current flows if there is a load on the coil, such as the coil capacitance or the loading caused by eddy currents in the cores, etc. (A so called current transformer is a tightly coupled transformer driven from a high impedance so that the current in the secondary is related to that in the primary by the turns ratio. The very loosely coupled situation here does not behave that way.)

You describe one way to make the current through the field coil independent of frequency: make the inductive reactance low compared to the dc resistance of the coil across the whole useful frequency range. A better way is to drive the coil with a current source, that is, a circuit with an output impedance much higher than the impedance of the coil at any useful frequency.
True, the primary effect of induction is electric field and voltage. But this causality rarely matters in real life situations, where there are current paths and the currents produce counterfields.
Any pair of coupled coils can be descibed as a (non-ideal) transformer. And a transformer is able to transform voltage and current as well as Z,R,L,C. Loose coupling causes reduced voltages and deviations from the ideal turns ratio relations. But the principle works nevertheless.
The idea behind the method is to generate a frequency-independent current in the inductance part of the PU. Constant current through the inductance produces a voltage across the inductance rising linearly with frequency, just like in real PU operation, where the voltage is induced via dPhi/dt.
This constant current is the input test signal and must flow through the load consisting of DCR and capacitance. The output signal is the voltage developed across the capacitance/terminals.
The method is best descibed via the current transformer principle. And if done carefully it works just fine, as can be most easily seen by the straight horizontal line behaviour of the integrated output voltage. My explanations are in line with Lemme and Zollner.

Your are right, generating the constant current via the constant current driven field coil is nothing but a constant current source. And it could be replaced by an (active) wide band CCS if there were direct access to the inductance part of the PU, which is not.
Feeding a constant current to the (output) terminals of the PU inevitably yields the two-pole/two-terminal impedance response. This is different from the quadripole transfer response revealed by the descibed method.
I don't pretend, though, that the transfer response reveals information not available from the impedance response.
 
Helmholtz 4/26/2018 7:33 AM
oabx00z.png

In practical testing, both the exciter / field coil method, as well we putting the pickup in series with the function generator, both yield +dB...-6dB/oct slopes, as seen in the first simulation. Driving the pickup with a series voltage obviously puts the voltage source outside of the pickup, but shouldn't the exciter coil method place the voltage inside of the pickup? Why does this testing method not result in a 0dB/oct..-12dB/oct plot, as is seen in the second screen shot?
Placing a constant voltage source in series with L does not correspond to the exciter method, where a constant current through L is generated. You should use a swept current source in parallel with L instead. This will give you the bandpass transfer response with +/- 6dB slopes and 0 output for 0 Hz.
You did simulate the low-pass transfer response instead.
 
Mike Sulzer 4/26/2018 8:30 AM
Quote Originally Posted by Helmholtz View Post
The idea behind the method is to generate a frequency-independent current in the inductance part of the PU.

That does not happen. The constant current in the exciter coil generates a magnetic field with amplitude independent of frequency. This generates a frequency dependent voltage in the pickup coil. The current that flows is frequency dependent because of that and because the load on the pickup is frequency dependent. At very low frequencies where the reactance of the coil capacitance is very high and the effect of eddy currents is negligible, this current is very small, only that which flows in the input resistance of the amplifier that senses the pickup voltage.

As long as the current in the exciter coil remains constant, the excitation part of the total magnetic field does not change, and so there is no interaction between the pickup coil and the exciter coil, except for the capacitance of the exciter coil. But if you use a tiny exciter coil, this is negligible. In my set up the constant current is a result of the 8 ohm resister, which has an impedance much larger than the tiny coil. Any coil resistance just adds to the resister, and the reactance value of the coil inductance is too small to change the current significantly at audio frequencies.
 
Helmholtz 4/26/2018 8:47 AM
That does not happen. The constant current in the exciter coil generates a magnetic field with amplitude independent of frequency. This generates a frequency dependent voltage in the pickup coil. The current that flows is frequency dependent because of that and because the load on the pickup is frequency dependent. At very low frequencies where the reactance of the coil capacitance is very high and the effect of eddy currents is negligible, this current is very small, only that which flows in the input resistance of the amplifier that senses the pickup voltage.
You are wrong. The current circulating in the PU will be independent of frequency if done correctly. This is a precondition for measuring the true bandpass response of a filter circuit. It can be verfied by simulation.
 
Antigua 4/26/2018 10:50 AM
Quote Originally Posted by Helmholtz View Post
I cannot comment on your measurements. I gave you all the info necessary to make sure you have constant exciter current over frequency. When the exciter current falls with increasing frequency, so will your PU output voltage and thus frequency response will deviate. This effect can be seen in figure 5 of the document, where the responses show a pseudo plateau starting around 1kHz, just as predicted by the formula. In reality the PU bandpass response gets (increasingly) steeper towards resonance.

You may verify the frequency (in)dependance of the drive current by measuring the voltage over a series resistor, via the second channel of the Velleman. The effect of a decreasing drive may be partly masked by the resonance of the PU.
I don't know how the "figure 5" plot came to exist, with the huge "hump", I've never witnessed that myself, except in the case of extreme eddy current losses. For example, it looks suspiciously similar to a Fiolter'tron:

[ATTACH=CONFIG]48650[/ATTACH]

Both my exciter coil and "driven" plots show the same curves, so I think I'm good to go. The only difference, as mentioned earlier, is that the exciter coil will also reveal eddy current losses with respect to the guitar string, where as the "driven" method with a series resistor only shows eddy current losses with respect to the pickup's coil.
 
Mike Sulzer 4/26/2018 11:13 AM
Quote Originally Posted by Helmholtz View Post
You are wrong. The current circulating in the PU will be independent of frequency if done correctly. This is a precondition for measuring the true bandpass response of a filter circuit. It can be verfied by simulation.
Can you describe the physical process by which a constant magnitude ac magnetic field, varying over the required frequency range, can excite a current in a coil independent of the frequency varying load on the coil? Even if you could, it would not be what you want to do. Filter circuits generate the correct bandpass when the source and load have the impedance for which they were designed. For example, rf passive filters might be designed to work from a 50 ohm source into a fifty ohm load.

The field generated by the coil, if small enough, is very similar to the ac field generated by a vibrating string. If you have any doubts that that generates a voltage, read MacDonald (Princeton). Very similar application of Maxwell's equations. So if the string generates a voltage in series with the coil, then we want our test to generate a voltage.
 
Helmholtz 4/26/2018 1:30 PM
Can you describe the physical process by which a constant magnitude ac magnetic field, varying over the required frequency range, can excite a current in a coil independent of the frequency varying load on the coil?
Yes, I can but don't bother to elaborate. No sense discussing with people who don't even consider my arguments.

The field generated by the coil, if small enough, is very similar to the ac field generated by a vibrating string. If you have any doubts that that generates a voltage, read MacDonald (Princeton). Very similar application of Maxwell's equations. So if the string generates a voltage in series with the coil, then we want our test to generate a voltage.
I never denied that the ac field generates a voltage. In fact, the voltage induced across the inductance of the PU images the voltage across the inductance part of the exciter coil (which is not directly accessible, though), namely a voltage rising linearly with frequency. And this means a constant, frequency independent current.
I definitely don't need private lessons on physics by amateurs.

This is my last post on this subject. I have wasted enough time trying to convince people who don't want to learn.

You may try to direct your questions to GITEC e.V.
 
Helmholtz 4/26/2018 1:39 PM
Both my exciter coil and "driven" plots show the same curves, so I think I'm good to go. The only difference, as mentioned earlier, is that the exciter coil will also reveal eddy current losses with respect to the guitar string, where as the "driven" method with a series resistor only shows eddy current losses with respect to the pickup's coil.
Well, if you don't care, I don't either - but will rely on my own measurements. Your "eddy current losses" are only partly real.
 
Mike Sulzer 4/26/2018 1:51 PM
Quote Originally Posted by Helmholtz View Post
Yes, I can but don't bother to elaborate. No sense discussing with people who don't even consider my arguments.



I never denied that the ac field generates a voltage. In fact, the voltage induced across the inductance of the PU images the voltage across the inductance part of the exciter coil (which is not directly accessible, though), namely a voltage rising linearly with frequency. And this means a constant, frequency independent current.
I definitely don't need private lesson on physics by amateurs.

This is my last post on this subject. I have wasted enough time trying to convince people who don't want to learn.

You may try to direct your questions to GITEC e.V.
No, the voltage is generated in series with the pickup coil, and this voltage is not directly accessible. The inductance of the pickup coil is the series leg, and its capacitance is the shunt leg of a voltage divider. This makes a resonant low pass filter. Thus the voltage across the pickup coil is the output of the filter, and it is flat at low frequencies, rises below the resonance and falls above it.

In addition, the voltage rises with frequency because the induced voltage is proportional to the rate of change of the flux through the coil. This has to be accounted for. Finally, the pickup coil couples to metal parts like a transformer with high leakage flux. This places a load across the coil that mostly affects the response in the region of the resonance.

The current that flows through the pickup coil is a function of these loads on it. It is very small at low frequencies since the load is a very high resistance, the preamp input. The current peaks at resonance because of the circulating current in the parallel resonating circuit, and then it falls at higher frequencies.

I am not an amateur.
 
Mike Sulzer 4/26/2018 2:03 PM
Quote Originally Posted by Helmholtz View Post
Well, if you don't care, I don't either - but will rely on my own measurements. Your "eddy current losses" are only partly real.
So are you saying that a thick conductive cover cannot reduce the signal reaching the pickup coil? It can. Both types of eddy current losses can be significant, and that is why it is necessary to make both kinds of measurements if the pickup has a thick cover.
 
Helmholtz 4/26/2018 2:24 PM
So are you saying that a thick conductive cover cannot reduce the signal reaching the pickup coil? It can. Both types of eddy current losses can be significant, and that is why it is necessary to make both kinds of measurements if the pickup has a thick cover.
I can only advise to make sure/control that the exciter current stays constant over the whole frequency range. Otherwise you might find unreal sag below the resonance. But I am repeating myself. This said, some PUs with strong eddy effect do show some real sag. But it is suspicious if this effect depends on the drive circuit.
 
Antigua 4/26/2018 2:32 PM
Quote Originally Posted by Helmholtz View Post
But it is suspicious if this effect depends on the drive circuit.
One example would be the base plate of a pickup, it's very close to the coil itself, so it figures prominently with respect to the coil, but it's very far away from the moving guitar string, so it would have little interaction with the AC magnetic field of the guitar string. A cover would be the other way around; impeding the AC field of the string more so than the AC field of the coil.

Going back to the transformer analogy, wouldn't eddy current losses be greater if the conductive plane was in between the coils, rather than being placed at far side of the transformer?

Within the next week or two I will compare driven versus externally excited plots of pickups with covers and substantial steal cores in order to quantify any difference.
 
Helmholtz 4/26/2018 2:37 PM
Quote Originally Posted by Antigua View Post
One example would be the base plate of a pickup, it's very close to the coil itself, so it figures prominently with respect to the coil, but it's very far away from the moving guitar string, so it would have little interaction with the AC magnetic field of the guitar string. A cover would be the other way around; impeding the AC field of the string more so than the AC field of the coil.

Going back to the transformer analogy, wouldn't eddy current losses be greater if the conductive plane was in between the coils, rather than being placed at far side of the transformer?
Yes, of course. But where is the link to the exciter coil circuit?