Electrical or magnetic polarity swap changing sound within a single humbucker?

[uOpt]

Re: Electrical or magnetic polarity swap changing sound within a single humbucker?

Supposed to be according to who? How did you measure 20 nF? Have you also measured the inductance?

Here are some readings of my humbuckers:

                        Kohm    H       nF
Jazz n                  7.15    3.96    35.5
aph1n w/cover           7.31    4.27    34.7
aph1n wo/cover          7.35    4.34    35.0
PG Pearly Gates n       7.37    4.17    34.8
Ant neck                7.38    4.45    34.5
59n                     7.44    4.23    34.2
59nz                    7.45    4.31    34.4
skatter neck            7.49    4.55    34.2
T-top                   7.50    4.23    34.0
aph1b                   7.86    5.11    32.8
mccarty neck            7.92    5.33    32.4
Burny2                  8.21    4.22    30.1
Burny1                  8.26    4.24    29.9
My Explorer neck 496R   8.30    5.26    31.5
Skatter birdge          8.30    5.51    31.2
Ant bridge              8.39    5.47    30.7
59b w/cover             8.46    5.09    30.3
59b wo/cover            8.49    5.06    30.4
Sunburst DM PAF neck    8.53    5.06    29.9
Sunburst DM PAF bridge  8.64    5.23    29.1
mccarty bridge          8.92    6.60    28.8
Demon                   9.37    4.34    25.3
Hyb1                    10.56   6.81    24.7
CC                      13.36   8.48    19.5
My Explorer bridge 500T 14.63   8.60    17.9
FSTB Full Shred Tremb.  15.14   8.48    17.3
L500XL                  15.33   8.75    17.2
JB                      15.94   8.58    15.9
Invader                 16.16   9.68    15.8

This also approaches the values given in Helmut Lemme's book.

The readings, according to "my method", are not wrong. DC resistance of my CC is on the lower side for a CC, 13.7kOhm, and I measure the inductance and capacitance to ca. 8.5 H and 85 pF, respectively. This gives me a theoretical resonant frequency of 5900 Hz - a res. freq. a bit higher than the CC listed on the SD Tone Comparison Chart (5200). Something I attribute partly to the lower DC res. and me just using a simple model/equation. It's fairly close to what I measure with the RCL meter though.

Adding your volume and tone pots as well as a cable changes the frequency response a great deal - but doesn't kill anything (flattens and with treble reduced though). I've tried to model it theoretically and show you the results on the figure attached (may not be entirely correct, but should be in the ball park). (Volume pot, tone pot, elec. and cable cap. included, but not any amp/effects input impedance.)

Effect of adding more capacitance: lowers the resonant freq., but increases the resonant peak.
Effect of adding lower value pots in parallel: lowers the resonant peak. (and resonant freq)

We might never agree on this topic though

I think the way that you are trying to measure the resonance peak make you come up with capacitance value that's are just capacitance value as if you used the pickup as a capacitor.

I think that maybe what happens is that your testing equipment uses that capacitance to build the circuit to test for the resonance frequency. Your resonance frequency values and the graph look good for the curve with no electronics, but there is no way that you end up entirely without any resonance peak just by attaching the electronics. The green curve is definitely not correct. If that was the case we'd all be running preamps before the pots. Again, Lemme's book has frequency response curves that are more realistic.

However that might be, your pickup cannot have a straight capacitance of 85 pF. If it did, then a 600 pF cable would kill it and you'd get a huge advantage out of using a 300 pF cable.

You are doing something there that is using a wrong capacitance as a base and the wrong green curve for with-electronics is a result from that.

 

[Ozram]

Re: Electrical or magnetic polarity swap changing sound within a single humbucker?

Thank you for the table. I assume the values are measured on the pickups alone? Can you please tell me how you did the measurements (what wires, instrument/meter and method)?

I don't have Helmut Lemme's book yet, but looking at his table at http://buildyourguitar.com/resources/lemme/table.htm I can compare your measurements of the '59 vs his. You agree on the inductance (ca. 5 H), but not the winding (or parasittic) capacitance (Lemme: 120 pF, you 30 nF). Using the RCL electrical model of the pickup, the resonance frequency is found using the equation f = 1/(2*pi*sqrt(L*C)) (e.g. see http://en.wikipedia.org/wiki/RLC_circuit).
Using Lemme's values I get a res. freq of ca. 6500 Hz. Using your higher capacitance values, I get a res. freq of ca. 400 Hz.
Which one do you think best matches Seymour Duncan's Tone Comparison Chart?

I'm sorry, but I just can't see the capacitance being in the nanofarad range the way I think of it used here.

 

[uOpt]

Re: Electrical or magnetic polarity swap changing sound within a single humbucker?

I use a multimeter that has capacitance and inductance settings. Not sure of the brand. I measure coils in series and the ground isn't connected.

Hmm, Lemme's book indeed says 130 pF (and 3.8 H) for the "Gibson Humbucker".

Seymour seems to use a similar meter to mine since he comes to 8.3 H for what seems to be a awg44 JB like coil:
http://www.seymourduncan.com/products/custom-shop/artists-signatu/george_lynch_su/

%%

I'm not sure what to make of this. I think we agree that the capacitance of the pickup and the load capacitance (cable etc.) are just added together. And that the cable is about 600 pF.

If so, then the capacitance of the pickup must be much higher than that. If the pickup is in the 100 pF range, then it would just die from a cable with 600 pF. Everybody would use preamps and vintage non-capacitance cables with 2 inch diameter would go for horrible prices on ebay. If the pickup is in the 30 nF range then it makes sense that good and bad cables between 500 and 1200 pF make an audible difference, but not a mind-altering one.

I still think that there are two different ways of measuring here. I think whatever instrument you are using there is equipped to measure the resonance peak and obviously it needs variable frequency for that. I honestly don't know how exactly one measures capacitance, but if the pickup's capacitance is frequency dependent then that would be one possible explanation.

Can you post a photo of the measuring gear you use? I'll get one of mine on the way out.

I'll try to corner one of the EE guys at work today, maybe there's some easy answer here.

 

[Ozram]

Re: Electrical or magnetic polarity swap changing sound within a single humbucker?

I use a multimeter that has capacitance and inductance settings. Not sure of the brand. I measure coils in series and the ground isn't connected.

Hmm, Lemme's book indeed says 130 pF (and 3.8 H) for the "Gibson Humbucker".

Seymour seems to use a similar meter to mine since he comes to 8.3 H for what seems to be a awg44 JB like coil:
http://www.seymourduncan.com/products/custom-shop/artists-signatu/george_lynch_su/

Bit of a digression from the original topic, but understanding is always important

First of all, I believe your inductance measurements are correct. (And higher DC resistance indicates more windings and thus higher inductance.)

However, I don't think measuring parasittic capacitance in a coil is straightforward. I used a Philips/Fluke PM6304 RCL meter. It's a calibrated, fairly expensive meter I borrowed from work capable of measuring the complex impedance over different frequencies - from that I can deduct parameters as coil inductance, parasittic capacitance and resonant frequency. See e.g. http://www.teknetelectronics.com/Search.asp?p_ID=23251&pDo=DETAIL for a picture of PM6304. More importantly: PM6304 gives me values which makes sense when I model the pickup as a RCL circuit - e.g. theretical resonant frequency matches actual measured one.

When you said you used a multimeter I tried some of mine to measure the CC capacitance:
Escort EDM168A: 25 nF
Dynascan 830 autoranging capacitance meter - autorange mode got confused, but using manual mode: 0.9 nF
Now we are more in your ball park. However, I have to dig into how the various devices actually measure capacitance (they are surely different). I think the "simpler" devices works ok with actual capacitors and cables, but gets into trouble in a RLC circuit (e.g. an actual coil) where the inductance of the coil interfers.

 

[Ozram]

Re: Electrical or magnetic polarity swap changing sound within a single humbucker?

I'm not sure what to make of this. I think we agree that the capacitance of the pickup and the load capacitance (cable etc.) are just added together. And that the cable is about 600 pF.

If so, then the capacitance of the pickup must be much higher than that. If the pickup is in the 100 pF range, then it would just die from a cable with 600 pF. Everybody would use preamps and vintage non-capacitance cables with 2 inch diameter would go for horrible prices on ebay. If the pickup is in the 30 nF range then it makes sense that good and bad cables between 500 and 1200 pF make an audible difference, but not a mind-altering one.

A 100 pF pickup would not die from a 600 pF cable. But yes, the caps add together and changes the frequency response of the pickup.

Keep in mind that the pickup only transfer the sound from the guitar - it does not generate any frequencies. A flat frequency response means you don't color the sound of the guitar itself. The resonant peak has the effect of boosting some frequencies (usually in the mids). Adding resitive load in parallel with these capacitances (e.g. volume pots) has the effect of reducing the height of the resonant peak (smaller resistance means more attenuation of the peak). And adding more capacitance increase the height. The resonant peak colors the sound of your guitar. But you don't want the peak to be too high (as it is with no load) - that would be like having a bandpass filter seeming to let only a small portion of your signal pass (kinda like a wah wah).

The green line in the frequency response I showed earlier does also have a resonant peak - but the height of it is much smaller than the others. The effect of adding more capacitance in the form of a cable boosts (but also lowers in freq.) the resonant peak.

Note that the effect of letting the pickup have a 30 nF winding capacitance shifts the frequency response very much to the left, with resonant freq. at 340 Hz (height of 3-4 dB), -3dB treble cutoff at 550 Hz. Needless to say, not a pickup (frequency response) you would want to use - it's just like turning down the tone knob. The good thing though is that the you can hardly see any effect of swapping cables...

 

[Kimock]

Re: Electrical or magnetic polarity swap changing sound within a single humbucker?

For a humbucker I always though that flipping the magnet or switching the "hot" (black) and "ground" (green) electrically would not change the sound of that pickup alone. Only when used together with another pickup - i.e. to get the Peter Green out-of-phase sound.

But: I have tried switching either electrically or flipping the magnet and both definitely changes the sound of my bridge position Custom Custom (for the time being with an A8 magnet). Compared to the normal configuration (south screw side, black = hot), this alternate position gives a fuller and sweeter kind of sound.

I actually prefer this "new" sound, but hate not knowing what causes it. Theoretically this does not make sense to me.
If it only was when I flipped the magnet it could have been other magnetic fields or rest/"leftover" magnetism in the slugs/screws causing a decrease in the magnetic field strength. However, same effect occurs with electrical polarity change without magnet flip. And in case anyone wonders, changing polarity both electrically and magnetically changes nothing (which makes sense).

Is it only an anomaly in my setup or is there anyone with similar experience?

Can anyone give me some clues to this effect?

Hi. First post here for me. Apologies for the zombie thread resurrection.
Yeah, you'll notice a little change with a polarity flip, but it's not as a result of anything electrical.
It's due to the altered loop characteristic of the acoustical interaction between the guitar string and the speaker.
It's feedback basically.
In one polarity you're favoring one coil, in the other polarity the other coil will be favored.
They're not in the exact same location, so one will always be driven a little positive, the other a little negative wrt to the acoustical coupling between the string and the speaker.

It's easy to miss, but if you're used to it one way and you flip it, you'll know something's changed.
Anyway, you'll get the same effect if flip your speaker polarity, so that'd be the test.
If you notice the polarity swap with the PU, the speaker polarity flip should restore it.

Doesn't matter where you do the 180 flip, you're going to change the feedback loop characteristics between the string and the speaker.
It's not an electronics "AC motor" issue, it's acoustical.

Ok! HTH. .