Why SD and home-made are different?

[csjang]

Hi folks,

I found quite interesting stuff while I was trying to do network analysis using CGR101. If you see the snapshot, there is a big scoop with home-made pickup beyond 10kHz while there is another peak with SD SH-2N. So I checked the slug side coil response. SH-2N slug has the second but higher peak at around 12kHz while home-made slug (1018 alloy) shows smooth decline.

The home-made one was made according to the vintage PAF spec and materials (SPN42 wire, 5000 turns, 1018 alloy slug). As an experiment, I exchanged the magnet (AlNiCo5) between these 2 pickups but the result did not change. So I guessed the secret might be with slug material but am not sure... I also checked SH-4 and Burstbucker Pro pickups and they showed similar second peaks with the slug coil though not as dramatic as SH-2N, and smooth decline without a scoop from the whole humbucker response.

Can anyone answer where this difference came from?

[David Schwab]

What kind of magnet wire did you use?

How did you wind it? It's not just the materials that matter.

I wonder if you have any shorts in your coil?

What's your baseplate made from? Cover or no cover?

All pickups should have those resonant peaks before the response drops off.

[Joe Gwinn]

First question is how exactly the measurements were made. Without that information, it's hard to assess the data.

[csjang]

What kind of magnet wire did you use?

How did you wind it? It's not just the materials that matter.

I wonder if you have any shorts in your coil?

What's your baseplate made from? Cover or no cover?

All pickups should have those resonant peaks before the response drops off.

I tried SPN42, SPN43 and PE42 and they were all hand-wound. No shorts. German silver baseplate was used and there was no cover. I also tried a couple of StewMac humbucker pickup kits as well. No matter what I tried, the response has the scoop after 10kHz, while SD and Gibson pickups don't...

[copperheadroads]

wHY isn't there details about the "turns per layer" mentioned
are you hand winding you pups or machine winding them ?

[csjang]

First question is how exactly the measurements were made. Without that information, it's hard to assess the data.

I attached the measurement setup. CGR101 generates wave signal with varying frequencies from 1kHz to 40kHz and measure the response at once. I cannot comment on the accuracy but can tell how consistent it is -- all home-made pickups had the scoop but all the commercial humbucker pickups did not.

[csjang]

I hand-wind pickups. The turns per layer should not be a factor, I guess, because I also tried different bobbins with different winding spaces but it did not make much difference...

[jonson]

So you guess the secret may be the slugs and you've tried all else but them so swap the slugs.

[Mike Sulzer]

Second peaks are often the result of coupling between the coils, but of course you only see the effect if the coils are different. So it is possible, as a guess, when you measured the slug coil alone you left it in place next to the other coil and the coupling to its resonance (coil inductance and capacitance) resulted in your plot. One of those SD pickups might use coils with different number of turns, that I do not know.

With the capacitance of the guitar cable, the resonant frequency comes way down, and these sedondary peaks arevery small, especially given the 5 KHz limit of the guitar speaker.

Hi folks,

I found quite interesting stuff while I was trying to do network analysis using CGR101. If you see the snapshot, there is a big scoop with home-made pickup beyond 10kHz while there is another peak with SD SH-2N. So I checked the slug side coil response. SH-2N slug has the second but higher peak at around 12kHz while home-made slug (1018 alloy) shows smooth decline.

The home-made one was made according to the vintage PAF spec and materials (SPN42 wire, 5000 turns, 1018 alloy slug). As an experiment, I exchanged the magnet (AlNiCo5) between these 2 pickups but the result did not change. So I guessed the secret might be with slug material but am not sure... I also checked SH-4 and Burstbucker Pro pickups and they showed similar second peaks with the slug coil though not as dramatic as SH-2N, and smooth decline without a scoop from the whole humbucker response.

Can anyone answer where this difference came from?

[Joe Gwinn]

[ATTACH=CONFIG]29176[/ATTACH]

I attached the measurement setup. CGR101 generates wave signal with varying frequencies from 1kHz to 40kHz and measure the response at once. I cannot comment on the accuracy but can tell how consistent it is -- all home-made pickups had the scoop but all the commercial humbucker pickups did not.

That helps. But, it's hard to discern the circuit diagram from the photo. I think I see a drive coil stuck to the pickup under test. A sketch of the circuit diagram would answer most questions.

[csjang]

Second peaks are often the result of coupling between the coils, but of course you only see the effect if the coils are different. So it is possible, as a guess, when you measured the slug coil alone you left it in place next to the other coil and the coupling to its resonance (coil inductance and capacitance) resulted in your plot. One of those SD pickups might use coils with different number of turns, that I do not know.

With the capacitance of the guitar cable, the resonant frequency comes way down, and these sedondary peaks arevery small, especially given the 5 KHz limit of the guitar speaker.

Thanks for the comment, Mike. I will try a different signal coil.

[csjang]

Hi Joe, it is not too complex.
- driver coil is connected to the signal generator of CGR101
- pick is connected to the oscilloscope of CGR101
- I put a piece of plastic between the driver coil and the pickup (they are not stuck to each other)
- the rest of the measurement is done by the software, like sweeping frequency measuring response, etc.

[csjang]

yeah, maybe i will need to swap slug coil between the two pickups to check if it is the smoking gun.

[jack briggs]

[ATTACH=CONFIG]29176[/ATTACH]

I attached the measurement setup. CGR101 generates wave signal with varying frequencies from 1kHz to 40kHz and measure the response at once. I cannot comment on the accuracy but can tell how consistent it is -- all home-made pickups had the scoop but all the commercial humbucker pickups did not.

I don't know if this makes a difference, but are you using magnets in the driver coil or just slugs, and why? I can't get a signal from the test coil(s) if the driver coil is parallel to the test coil(s) - only perpendicular. Is this why you have poles in the driver?


cheers,

[Mike Sulzer]

Thanks for the comment, Mike. I will try a different signal coil.

I am referring to coupling between the two coils of the humbucker.

Your driver coil has fewer turns of a larger wire size; its resonance should be high enough so that it is not an issue. Some effects such as this might be easier to analyze by measuring the impedance of the pickup rather than exciting it with a coil. An impedance measurement is more fundamental because it has a definite meaning in circuit analysis. Exciting the pickup coil with a coil does not have a well defined relationship to exciting it with a vibrating string, but is sometimes useful, especially when you want to see the effect of connecting the pickup to an external circuit.