Okay guys. Here is the long winded manifesto. Feel free to savage me. I'm sure some of this some people already know but I think there is some good info here.
To begin with I just want to state that this post is not an effort to pass judgment upon those that wish to hand wind in their quest for PAF tone. I myself used to offered a hand wound PAF clone. But over the last two years I have changed my approach in favor of machine winding via computer winding machine of my own design and now via my own Leesona 102 winding machine. I recently purchased one of the few complete and functional Leesona 102 winders left. Now having learned exactly what the Leesona 102 does it has clarified some of the mysteries of the PAF for me. I thought this would be of particular interest to pickup winders and vintage Gibson enthusiasts.
So what is the Leesona 102?
In the late 40's Gibson purchased two Leesona Model 102 coil winders to use for winding guitar pickups. These winders remained in use at Gibson until Gibson moved from Kalamazoo. It is widely accepted that the Leesona 102 winder was used in the manufacture of PAF pickups. It is also widely accepted that PAF's were exclusively machine wound. Here are some pics of mine fress of the truck.
Why do PAF's vary so widely in resistance?
The Leesona 102 is a 3 station winder. Each of these 3 stations have separate start levers. The manner in which it is usually run is for one station to be set up and started and then the operator moves on to the next station to set it up until all 3 stations are running. If the winding is timed as Gibson apparently did for most of the PAF era you end up with 3 bobbins with staggered turn counts for each run of the machine. Throw the wound bobbins in a box and it is anyone's guess what you end up with for resistance of the finished pickup. The Leesona 102 did in fact come with manual counters with auto shutoffs. However the gears that runs the counters are fiber gears and not very durable in a day to day use situation. I believe that by the time PAF's were being wound it is quite likely that all of the Leesona 102 counters had failed and Gibson had long since abandoned then and gone to the timing method of stopping the coil wind.
What accounts for the widely varying winding patterns of PAF's?
The Leesona 102 was never exclusively used for any one pickup model. Consequently it would require a set up change for each different model wound on it. And each time it was reset up to do a PAF the parameters of the winding pattern would have to be changed. Some of these adjustments are very crude on the Leesona 102 and operators would only be able to do a ballpark adjustment at best which would mean wide variations in winding pattern between set ups. I have retro fitted my own Leesona 102 with dial indicators in order to make set ups repeatable and to allow for accurate experimentation between them. But the original machine is not capable of having some of the parameters fine tuned beyond a trial and error set up without this retrofit. So good enough would be all that was aimed for by the operator resulting in widely varying winding patterns.
What factors account for the erratic and sometimes hand wound character of PAF coils winding patterns?
The above factors in user setup account for much of the variation. But another important factor is the tolerances inherent in the machine itself. The Leesona 102 is designed to do coils as large as 3". At 3" the tolerances of the cam and other mechanisms of the machine are negligible. But if you ratchet the traverse down to 1/4" for a PAF bobbin these same tolerances now become a very large percentage of the now very small 1/4" traverse of the wire guide. The result is a sometimes very non linear wire traverse when compared to a modern computer guided winding machine or even compared to the same machine at a 3" traverse.
Why does the wire diameter vary with PAF pickups and why do the bobbins warp?
The first simple answer concerning wire diameter is that the difference between maximum and minimum tolerances of 42AWG plain enamel magnet wire. But this is not the whole story. The Leesona 102 came with tensioners that were only designed to accommodate 40AWG magnet wire. This means that when you use the thinner 42AWG wire it is very easily is stretched. My Leesona 102 has the original functioning wire tensioners and I can attest that they are a pain to set up. The only other functional Leesona 102 I am aware of uses a whisker disk for tensioning and I completely understand why as it is a breeze to set up. But if you don't use the original tensioners this higher tension factor can be lost. The additional tension makes for a very tightly wound coil which encourages the warpage of the already soft butyrate bobbins.
Other facts about the Leesona 102.
The Leesona 102 came with a standard gear set from the factory that would give several very specific turn per layer numbers. This turn per layer count is independent of the traverse travel adjustment. So the same gear set would give the same turn per layer count for a 1/4" bobbin or a 3" bobbin. In my experience there is only one PAF turn per layer number and this is one that the Leesona 102 would do with the stock factory gear sets. I have heard of another turn per layer count for PAF's but cannot confirm this personally with my own PAF research. But interestingly enough this other turn per layer count does jibe with the factory gear sets of the Leesona 102. So perhaps the two Leesona 102 winders Gibson had were set to two different turn per layer counts. But this again is speculation on my part which I have not confirmed to my satisfaction.
The Leesona 102 had one cam for all traverse sizes and turn per layer counts. The ingenious design of the Leesona 102 means that one cam covers the territory for all winding scenarios of the machine. Knowing this when I first got my Leesona 102 winder I feared that the cam might be significantly worn. But the Leesona 102 is a mammoth machine that is way over engineered for what it does. The cam itself sits in a pool of oil as do most of the other moving parts of the winder. This meant the cam and in fact every other important mechanism on my Leesona 102 showed wear that was nearly non existent.
So what effect does all of this have on the tone of the PAF?
Well that is a question for every maker to decide for themselves. But make no mistake all of these variable effect the final tone of a PAF style pickup. I'm in the process of quantifying all of this for myself. As a hand winder much of the things I learned from experiments with my computer controlled winder and my Leesona 102 were entirely contrary to what I learned hand winding. Machine winding has a different set of rules but ultimately results in a coil and tone that I feel cannot be gotten any other way other than machine winding. The PAF is a compilation of parts many of which have a direct effect on the final tone of the pickup. But perhaps the most abstract of these variables is the wind if the wire. Placing this abstract within the frame of the Leesona 102 has given the winding of a PAF a mental, visual and auditory clarity that I had been searching for.
To begin with I just want to state that this post is not an effort to pass judgment upon those that wish to hand wind in their quest for PAF tone. I myself used to offered a hand wound PAF clone. But over the last two years I have changed my approach in favor of machine winding via computer winding machine of my own design and now via my own Leesona 102 winding machine. I recently purchased one of the few complete and functional Leesona 102 winders left. Now having learned exactly what the Leesona 102 does it has clarified some of the mysteries of the PAF for me. I thought this would be of particular interest to pickup winders and vintage Gibson enthusiasts.
So what is the Leesona 102?
In the late 40's Gibson purchased two Leesona Model 102 coil winders to use for winding guitar pickups. These winders remained in use at Gibson until Gibson moved from Kalamazoo. It is widely accepted that the Leesona 102 winder was used in the manufacture of PAF pickups. It is also widely accepted that PAF's were exclusively machine wound. Here are some pics of mine fress of the truck.
Why do PAF's vary so widely in resistance?
The Leesona 102 is a 3 station winder. Each of these 3 stations have separate start levers. The manner in which it is usually run is for one station to be set up and started and then the operator moves on to the next station to set it up until all 3 stations are running. If the winding is timed as Gibson apparently did for most of the PAF era you end up with 3 bobbins with staggered turn counts for each run of the machine. Throw the wound bobbins in a box and it is anyone's guess what you end up with for resistance of the finished pickup. The Leesona 102 did in fact come with manual counters with auto shutoffs. However the gears that runs the counters are fiber gears and not very durable in a day to day use situation. I believe that by the time PAF's were being wound it is quite likely that all of the Leesona 102 counters had failed and Gibson had long since abandoned then and gone to the timing method of stopping the coil wind.
What accounts for the widely varying winding patterns of PAF's?
The Leesona 102 was never exclusively used for any one pickup model. Consequently it would require a set up change for each different model wound on it. And each time it was reset up to do a PAF the parameters of the winding pattern would have to be changed. Some of these adjustments are very crude on the Leesona 102 and operators would only be able to do a ballpark adjustment at best which would mean wide variations in winding pattern between set ups. I have retro fitted my own Leesona 102 with dial indicators in order to make set ups repeatable and to allow for accurate experimentation between them. But the original machine is not capable of having some of the parameters fine tuned beyond a trial and error set up without this retrofit. So good enough would be all that was aimed for by the operator resulting in widely varying winding patterns.
What factors account for the erratic and sometimes hand wound character of PAF coils winding patterns?
The above factors in user setup account for much of the variation. But another important factor is the tolerances inherent in the machine itself. The Leesona 102 is designed to do coils as large as 3". At 3" the tolerances of the cam and other mechanisms of the machine are negligible. But if you ratchet the traverse down to 1/4" for a PAF bobbin these same tolerances now become a very large percentage of the now very small 1/4" traverse of the wire guide. The result is a sometimes very non linear wire traverse when compared to a modern computer guided winding machine or even compared to the same machine at a 3" traverse.
Why does the wire diameter vary with PAF pickups and why do the bobbins warp?
The first simple answer concerning wire diameter is that the difference between maximum and minimum tolerances of 42AWG plain enamel magnet wire. But this is not the whole story. The Leesona 102 came with tensioners that were only designed to accommodate 40AWG magnet wire. This means that when you use the thinner 42AWG wire it is very easily is stretched. My Leesona 102 has the original functioning wire tensioners and I can attest that they are a pain to set up. The only other functional Leesona 102 I am aware of uses a whisker disk for tensioning and I completely understand why as it is a breeze to set up. But if you don't use the original tensioners this higher tension factor can be lost. The additional tension makes for a very tightly wound coil which encourages the warpage of the already soft butyrate bobbins.
Other facts about the Leesona 102.
The Leesona 102 came with a standard gear set from the factory that would give several very specific turn per layer numbers. This turn per layer count is independent of the traverse travel adjustment. So the same gear set would give the same turn per layer count for a 1/4" bobbin or a 3" bobbin. In my experience there is only one PAF turn per layer number and this is one that the Leesona 102 would do with the stock factory gear sets. I have heard of another turn per layer count for PAF's but cannot confirm this personally with my own PAF research. But interestingly enough this other turn per layer count does jibe with the factory gear sets of the Leesona 102. So perhaps the two Leesona 102 winders Gibson had were set to two different turn per layer counts. But this again is speculation on my part which I have not confirmed to my satisfaction.
The Leesona 102 had one cam for all traverse sizes and turn per layer counts. The ingenious design of the Leesona 102 means that one cam covers the territory for all winding scenarios of the machine. Knowing this when I first got my Leesona 102 winder I feared that the cam might be significantly worn. But the Leesona 102 is a mammoth machine that is way over engineered for what it does. The cam itself sits in a pool of oil as do most of the other moving parts of the winder. This meant the cam and in fact every other important mechanism on my Leesona 102 showed wear that was nearly non existent.
So what effect does all of this have on the tone of the PAF?
Well that is a question for every maker to decide for themselves. But make no mistake all of these variable effect the final tone of a PAF style pickup. I'm in the process of quantifying all of this for myself. As a hand winder much of the things I learned from experiments with my computer controlled winder and my Leesona 102 were entirely contrary to what I learned hand winding. Machine winding has a different set of rules but ultimately results in a coil and tone that I feel cannot be gotten any other way other than machine winding. The PAF is a compilation of parts many of which have a direct effect on the final tone of the pickup. But perhaps the most abstract of these variables is the wind if the wire. Placing this abstract within the frame of the Leesona 102 has given the winding of a PAF a mental, visual and auditory clarity that I had been searching for.
Comment