5000 winds to 8K - something's not adding up

[Meowy]

Most reports, anecdotes, published interviews, etc state that a "typical" vintage PAF coil was wound to ~5000 turns. Using 42 AWG and an approx resistance value of .74 ohms per turn works out to a ~3.7K coil or a ~7.4K pickup, probably a touch higher to account for lead / hookup wire, but not pushing 8K or more.

So either your typical > 8K PAF is wound beyond 5000 winds per coil (average) or the coil wire used was marginally smaller than 42 AWG.

Discuss / Comments?

[black_labb]

there are plenty of irregularities in the wire. could also have stretched the wire with excess tension.

[archtop]

42awg

not all 42AWG are created equal. Use a good micrometer to measure the magnet wire. This is THE MAIN/CRUCIAL KEY that many in this forum tend to overlooked. 5000 turns of unstretched 42AWG can range from 7.2 to 8.5 per HB. You can slowly and carefully stretch .0027 wire down to .0024 wire (anything beyond, the wire will break), but the tone you get will not be the same as if you used natural unstretched .0024 wire.

[Possum]

wrong...

you missed the point there, its 5,000 turns PER COIL, total 10,000 turns same as they used for P90s.....

[archtop]

OK..Let me do the calculation

from 7.2 / 2 = 3.6k per coil to 8.5 / 2 = 4.25k per coil with 5000 turns per coil...

[Possum]

micrometers.....

YES, if you make pickups you MUST have a micrometer. If you deal with Wirenetics and you don't have a micrometer you're going to get screwed and sent the wrong wire and you'll never know it unless you mic the wire. MWS wire is really tightly tolerance and they have mic measurements on their spools so no problems there. Your micrometer should be digital and read to five or six places. I still have a 2lb spool from five years ago from Wirenetics marked 42 gauge, it reads .0030" if its 42 its heavy build :-)

[J S Moore]

5,000 turns per coil is how Seth Lover designed the humbucker. Literally a P-90 divided by 2. And 7.4k is about right for those specs if you don't stretch the wire significantly and if you're using bobbins with the correct string spacing. There will be a slight difference in the values between 49.2, 50 and 53 mm.

There's been a lot written about the inadequacies of the Leesona winder so I don't doubt that the specs varied widely for the original PAF's. The wire will have had variations in thickness depending on the age of the die used for drawing it as well, just like today.

[archtop]

I still have a 2lb spool from five years ago from Wirenetics marked 42 gauge, it reads .0030" if its 42 its heavy build :-)

1 way to find out and I find this method is the most effective. Take measurement of the wire before you quickly run an inch or 2 thru an open flame to burn off the coating. Take measurement and compare.

[Meowy]

you missed the point there, its 5,000 turns PER COIL, total 10,000 turns same as they used for P90s.....

5,000 turns per coil is how Seth Lover designed the humbucker. Literally a P-90 divided by 2. And 7.4k is about right for those specs if you don't stretch the wire significantly and if you're using bobbins with the correct string spacing. There will be a slight difference in the values between 49.2, 50 and 53 mm.

There's been a lot written about the inadequacies of the Leesona winder so I don't doubt that the specs varied widely for the original PAF's. The wire will have had variations in thickness depending on the age of the die used for drawing it as well, just like today.

I am aware that it is 2 coils of ~5000 turns. But with the wire I have used (Stew Mac / Schatten and Essex) I need to wind > 5200 and approaching 5400 turns to get into the range of 4.0K. 5000 gets me more like 3.70 - 3.75

So as I said in the original post. 5000 is a low quoted number, the #42 Gibson used for vintage PAFs was finer than 42 or they stretched it and increased resistance - or many original PAFs were well below 8K.

Interstingly TTops and those pickups that immediately succeeded the PAFs are quoted at 7.75K - a nearly perfect spec for 5000 winds.

So it seems to reason that many (if not the majority of) original PAFs were overwound as compared to publicized specs

[Possum]

...

Probably the core of a PAF bobbin and the core of a vintage P90 were not exactly the same measurements, tiny differences there will show up noticeably the more winds are put on....

[Stan H]

I thought that the turns on a PAF weren't actually counted (no counters). Were'nt they just "wound until full" and the 5000 turns were assumed? And that the "8K" was more of an average DCR for PAF's.

[Meowy]

I thought that the turns on a PAF weren't actually counted (no counters). Weren't they just "wound until full" and the 5000 turns were assumed? And that the "8K" was more of an average DCR for PAF's.

Yes I believe that is part of the "lore", but I can put 4600 or 5600 on a bobbin and it looks full. That's a 20% variance.

If 8K is an average then most PAFs were wound above 5000.

I guess I'm just wondering out loud if 5000K is misinformation (either intentional or just a "rounded number") and the more common # of winds is on the order of 5200-5400.

The best pickup I've made to date (granted I have not made all that many) is pushing 6000 winds per coil and meters out at around 8.6K

[kevinT]

The best pickup I've made to date (granted I have not made all that many) is pushing 6000 winds per coil and meters out at around 8.6K

did you use 42 AWG PE for that?

[Meowy]

did you use 42 AWG PE for that?

yes Essex PE 42

[J S Moore]

Plus the more winds on a P-90 bobbin and the further the outside winds have to travel. In other words the outside winds have significantly more wire per wind.

If I use 6000 winds per bobbin I'm a little over 9k.

In order to get the "typical turns" on a PAF you would have to un-wind every single one of them and get an average. And since I'm fairly certain that hasn't been done I think what is said to be typical is a mix of original design specs and whatever actual measurements can be taken. I don't think there is anyone in the world who has un-wound enough PAF's to be considered a good sample size in statistical terms to make absolute statements about turn counts. Everything I have read seems to be of the "We planned it this way but who the hell knows what happened" variety. Seth Lover himself is vague on the whole issue.

On the other hand it could be correct and the great sounding ones that we all want to copy are atypical. So we get a skewed view of the entire picture.