Commercial Winders - who's using one?

[ReWind]

I know some of you are using modern commercial winders. Whatcha got? Whatcha like/dislike about it? There's many out there. Half of them seem to have controllers made by the same company. I've got a number of old winders, one modern one I built, and I'm on the list for one of those UK CNC winders (they look impressive).

[SonnyW]

I know some of you are using modern commercial winders. Whatcha got? Whatcha like/dislike about it? There's many out there. Half of them seem to have controllers made by the same company. I've got a number of old winders, one modern one I built, and I'm on the list for one of those UK CNC winders (they look impressive).

I may be the only one here that uses one, but I have been using the small JLW CNC model for a couple of years now and I really like it once I finally figured out the Engrish manual. I have modified it (with a fair amount of effort) to include a tailstock, (which I rarely need) and added a glass wire guide made of a short piece of thermometer capillary tubing. Really that nylon wire guide was the only thing I didn't like. Not precise enough. Other than that, it is rock solidly made, heavy duty, easy to use-if you learn to think in asian terms, and (knock on wood) so far it has been very reliable and does anything at all I need it to. I can wind a single coil unattended in under 6 minutes at 40% of top speed and they come out within an omm or two. Similar for humbucker coils. They also have a better larger model that already has a tailstock installed. For the $ I think it can't be beat. You can program up to 999 layers but I usually only use 4 or 5 and control TPL, bobbin dimensions, rotation, speed, start slow, etc. They are good to work with too. I think I remember emailing back and forth with Josephine Wong who is the J in JLW. I think they make transformers mainly. When I got mine I think I was only the 4th in the U.S. to buy one. You can see mine as I have modified it on my website in the Shop Tour section.

Small model
Small Automatic coil Transformer CNC Winding Machine | eBay

Large Model
Fully Automatic coil Transformer CNC Winding Machine | eBay

I would also be happy to make some of the capillary tube small diameter glass wire guides for anyone on the forum who needs one, I got a large supply of the tubing at the time.

[ReWind]

I just checked out your site, Sonny. You've got some fun toys! I especially dig your charging/decharging station.

So how's that controller to program? Is it intuitive for a pickup maker? I thought I read in their specs that those winders can only go down to 40AWG. Perhaps that was just at full speed or with the nylon wire guide?

I love my old machines, they have soul and character. ...but they can be animals with minds of their own sometimes. It's almost a full-time job keeping them in check and operating properly. They get moody. I'd like something that can take a somewhat complex program or that I can program in "margins" rather than hard figures to allow a little loose-spec slop to mimic my old mechanical winders. A tailstock is absolutely essential to me. I want to wind four or five identical bobbins at once, and fast.

[SonnyW]

I just checked out your site, Sonny. You've got some fun toys! I especially dig your charging/decharging station.

So how's that controller to program? Is it intuitive for a pickup maker? I thought I read in their specs that those winders can only go down to 40AWG. Perhaps that was just at full speed or with the nylon wire guide?

I love my old machines, they have soul and character. ...but they can be animals with minds of their own sometimes. It's almost a full-time job keeping them in check and operating properly. They get moody. I'd like something that can take a somewhat complex program or that I can program in "margins" rather than hard figures to allow a little loose-spec slop to mimic my old mechanical winders. A tailstock is absolutely essential to me. I want to wind four or five identical bobbins at once, and fast.

Well, I wouldn't say it is exactly intuitive, but it makes sense they way it was done once you start using it. The manual was pretty sparse and at first I did have some difficulty. The biggest thing I wasn't getting at first, is that when you program it you set the start and end steps, then instead of programming each step, you program the parameter for all the steps, before going to the next parameter. Which makes sense because a lot of the time the parameter doesn't change and you can use a copy key to fill in the data. There are about 10 parameters for each step. It automatically calculates the traverse for a perfect wind from the wire diameter, but it is easy to control TPL just by telling it a different wire diameter. After using it for a while I can believe it was designed by or with lots of input from someone with plenty of winding experience. Mainly you tell it the start position wire diameter and bobbin width to control the traverse, which is nice if you need to move the start position because of a bobbin that is maybe a little warped or not the same as the last one you wound. You can just move the start position and it recalculates the end position from the bobbin width. All the programming is done from the front panel and it is all labeled in English, the units are metric. I leave the programs in and have a little book that I write them down in and to change to a stored program I just change the start and end steps.

It wasn't made to go down to the wire sizes smaller than 40, but there aren't any programming restrictions preventing it. I have wound 44 awg bobbins on it fine with no trouble. I suspect the only reason they spec'd 40 awg as the limit was because of the wire guide as you mentioned. The one supplied is made of a couple of pieces of nylon, clamped together by thumbscrews with different sized slots cut in it. There isn't a slot small enough to handle the small wires well and repeatably. My answer to that was to get some glass capillary tubing and make a couple of guides by cutting off inch long sections of the tubing and fire polishing both ends to get a smooth bearing surface, and so it is easy to thread. Then I glued the pieces of tubing in the slots. I have two so I could wind two bobbins at once, but it is fast enough that I never do. With the guide, I think I could go down smaller than 44 if I needed to. That also makes it very repeatable on the wire position, because the hole in the guide is small. I forget the bore on the tubing I got it might be 1mm. It will let you tell it a lot smaller wire diameter than 44, half gauges etc. It will let you input wire diameter of 0.001mm for example. I haven't tried running it at top speed. I usually program it for slow speed of 8% and top speed of 30- 40%. That's plenty. 30% = 1480 rpm. I remember there is a lower limit on the slow speed and 7% might be it. The supplied felt tensioner works well for small gauges. It comes with a tensioner for heavier wire I have never needed. I forget the top gauge limit but it is big.

Here is an example of one of my program sheets if that helps at all.



Edit: I should make a correction to my first post on this, I said in effect (not counting misspelling ohm as omm) that winding two identical bobbins with the same program comes out within an ohm or two. I wasn't speaking properly. Actually my ohmmeter doesn't give that resolution. I should have said they come out to within a significant digit or two, which would be 10 or 20 ohms, in this case. Usually they will read the same or within 1 digit displayed using the multimeter function on my Tek scope.

[Fluoroscope 5000]

That machine looks really interesting. Good idea to use the glass guide. I wonder if the tensioner wears out of spec over time? Hope getting parts and repair/maintenance is doable. Can it interface with a PC?

Are you sure the min spec is 40 AWG, because the ebay page says 0.04mm. That's more like 46 AWG. I think they mistranslated that. It'll probably do 46 AWG. 44 AWG is small enough to get good midrange phase accuracy from average wind count Hi-Z coils. You may notice the difference in your results. You should be able to get no eddy current losses with quality annealed poly wire. Of course, you need to make sure tension doesn't cause insulation stretching when the wire goes around the bobbin ends. I'd guess you'd have to wind a little looser than recommended to achieve that. Inductance will drop a little, but at least you won't get increased capacitance, or dreaded high end damping eddy currents. I don't know if you can do this, but it's important to check the series inductance, so the pickup's are in spec with planned goals. HP makes a meter for a few $100 that will give accurate series readings. I can find out the model if you want. There's also a way to check for eddy currents, so you can fine tune tension for the most consistent results. If you can make hide, or hair of it, this post explains about how to do it, and why:
DC resistance don't mean diddly - Telecaster Guitar Forum

I'm not a winder, but have learned a good deal about the process in the last several years. I have talked extensively with Bill Lawrence and his team members, and have read through all his writings and patents. I know it's a point of controversy in the winding community, but I wouldn't bother with winding patterns, or scattering. Essentially, it just moves more of the coil away from the core, compromising efficiency. I'd focus on coil dimension/density/volume within a given magnetic field shape/density/power to maximize phase accuracy and efficiency, but that's me. I'm curious how your new experiments go. More power to you, Sonny.

[SonnyW]

That machine looks really interesting. Good idea to use the glass guide. I wonder if the tensioner wears out of spec over time? Hope getting parts and repair/maintenance is doable. Can it interface with a PC?

Are you sure the min spec is 40 AWG, because the ebay page says 0.04mm. That's more like 46 AWG. I think they mistranslated that. It'll probably do 46 AWG. 44 AWG is small enough to get good midrange phase accuracy from average wind count Hi-Z coils. You may notice the difference in your results. You should be able to get no eddy current losses with quality annealed poly wire. Of course, you need to make sure tension doesn't cause insulation stretching when the wire goes around the bobbin ends. I'd guess you'd have to wind a little looser than recommended to achieve that. Inductance will drop a little, but at least you won't get increased capacitance, or dreaded high end damping eddy currents. I don't know if you can do this, but it's important to check the series inductance, so the pickup's are in spec with planned goals. HP makes a meter for a few $100 that will give accurate series readings. I can find out the model if you want. There's also a way to check for eddy currents, so you can fine tune tension for the most consistent results. If you can make hide, or hair of it, this post explains about how to do it, and why:
DC resistance don't mean diddly - Telecaster Guitar Forum

I'm not a winder, but have learned a good deal about the process in the last several years. I have talked extensively with Bill Lawrence and his team members, and have read through all his writings and patents. I know it's a point of controversy in the winding community, but I wouldn't bother with winding patterns, or scattering. Essentially, it just moves more of the coil away from the core, compromising efficiency. I'd focus on coil dimension/density/volume within a given magnetic field shape/density/power to maximize phase accuracy and efficiency, but that's me. I'm curious how your new experiments go. More power to you, Sonny.

A couple of comments - The tensioner is just felt pads and thumbscrews, so it does wear, but there isn't a spec really. I put a small folded piece of thinner craft felt in the first tensioner in addition to the one supplied. It builds up a residue over time, but I can move the wear spot around and it is easily replaced.

There is no interface to connect to a pc. Everything is done from the front panel. It does have a dedicated key to copy which makes some of that keying in easier.

The min spec does say .04mm but on the same line it also says 40 gauge. With the improved guide 44 awg isn't any problem. I do measure tension accurately with a separate gauge and adjust the thumbscrews.

There is a lot of information on the forum about measuring inductance. Be careful on the meters - most of them aren't made to be accurate on Low Q devices. We are talking about a Q factor in the range of 2-4 here. Sometimes a LCR meter made for higher Q can be off by 2x. I have demonstrated it with comparisons between several meters I have. Look for all the threads about the Extech meters. The Extech 380193 meter is what most of us use. It gives information also that can be used to measure eddy currents. Eddy currents if limited are not all bad. They can be part of what makes up the tonal character of a pickup.

Bill Lawrence has some great accomplishments and lots of experience, and some innovative ideas. He also has lots of opinions. But I can tell you definitely from my own personal measurements that the winding pattern does make a difference in characteristics. It changes the distributed capacitance, besides the things you mentioned. I test all my pickups with a variety of gear including a HP 3580A Audio spectrum analyzer. I can see differences in the winding pattern reflected in the spectrum plots, and though my hearing isn't perfect I can hear tonal differences. So can my customers. I'd bet Bill would admit it too if asked in the right context. For example discussing the effect of the number of turns per layer. In my opinion, one of the biggest reasons that cheap pickups sound bad is that their coils are perfect wound. Compromising efficiency is part of moving the tonal spectrum around. So are all those other things you mentioned, i.e. coil shape, magnetics etc. Just my thoughts on it and thanks for the comments and suggestions.

[Fluoroscope 5000]

Hey, thanks for responding to each inquiry in kind, Sonny. I don't often see that in these forums.

The Extech was the other meter I couldn't remember. It sounds like you know what you are doing regarding all that. Good to hear, sir.

Bill does have a lot of strong opinions on things. That's for sure. Regarding eddy currents: yes, they can be made use of for tuning purposes, like what materials are used and how they affect the field. Pickup housings and Aluminum pickguard shields, etc. Just as some winders experiment with partial wax potting and all. I can see how that might be useful in some ways. I was referring specifically to the higher frequency phase canceling eddy currents created by cracked, or over-stretched insulation, or even random dips from stretched wire. I actually don't think the former is generally an issue with poly wire, but excess tension on annealed wire could cause wire stretching around the corners. I'd think that would alter the Re at certain frequencies? Either way, it sounds like you have it under control.

If you have found that using certain wind patterns does give you repeatable and desirable results, then I see no reason to abandon the practice. Bill has concentrated on calculations involving materials and dimensions to achieve his results. Things like phase coherency, random peak/notch elimination, and increased fundamental harmonic ratio all add up to produce full, clear and sweet sounding pickups that don't suffer from high frequency losses. The phase accuracy of the pickup means the bass is never muddy, and the mids/highs are "sweet", instead of the thin, or "phasey" sounding.

Here in the states, we are all accustomed to Hi-Z pickups. EM build up occurs in the larger Hi-Z coils, which causes cancellations with the core field, or even with other portions of the coil. I think such coil phase anomalies are relatively nonexistent in Lo-Z coils. That's why I'm interested in how your thinner wire pickups end up sounding, possibly even with 46 AWG? With most Hi-Z pickups, players just use a longer cable to roll off the "phasey" upper frequencies, and then adjust pickup height to somewhat reduce muddiness in the bass. You don't have to do that with Bill's pickups, especially with the new ~46 AWG MicroCoils. Scott L. at Zexoil also understands the benefits of thinner wire. His pickups utilize 44 AWG wire. I don't know what affect increased distributed capacitance might have, but along with slightly decreasing output and note fundamental, pattern/scatter winding may actually reduce EM build up, reducing, or even eliminating the phase cancellations of which I spoke?

As far as cheap pickups sounding harsh: I think that has a lot to do with the things I mentioned above, and that the proportions and/or types of magnetic/core materials in some cases don't really "play" well together, thereby creating further phase issues. It strikes me that pattern, or scatter winding has developed as more of a work around to the inherent issues of larger coils. I encourage you to experiment with thinner wire, but there's certainly no reason to abandon whatever works for you. I'm curious to hear how things go. Thanks again for responding. Shine on, Sonny.

[SonnyW]

Enjoy the forum, we argue these fine points endlessly. Now and then we learn something new. Well, some of us do.

I did read through the thread you posted, several of the ones that posted there are regulars here too.

I have never really had a problem with cracked insulation and shorting causing eddy currents, but I control my tension to within 5g. when machine winding, and don't have a way to measure it really when hand winding except by feel but it is not a problem for me. I think I would see the effect of the eddy currents in the ACR and spectral analysis if that was happening. I so far haven't had a need for any finer wire than 44awg, but then again I am doing more of the vintage toned pickups. Some of my best results for full and clear sound have been with larger wire, even 41awg, but I am open to trying new things.

I do see the results of winding patterns in the location of the resonant peak. I can measure that very accurately using the zero phase difference method and my Tek 2236 scope in X-Y mode. With my test setup I get an elliptical lissajous plot that collapses to a straight line at the resonance (XL = XC) and I can read the frequency accurately on the digital readout of the scope. I measure every pickup I ship, and I commonly see variations as much as 1 kHz in handwound models that are otherwise identical. I have also done controlled experiments varying only the turns per layer of the winding pattern on my CNC winder and can repeatably move the resonant peak that much on some strat designs. The location of the resonant peak is a good indicator of the tonal qualities of the pickup. I don't scatter a lot when handwinding, and don't advertise that I do.

You are right about all the other things that make the cheap pickups sound harsh, I wasn't intending to imply that it is only the winding pattern, just that it is a factor. I think another big thing is the cheap materials.

Thanks for your comments, happy to respond.

[Fluoroscope 5000]

Ah, I'm not surprised to hear that the resonance peak changes with wind pattern. Capacitance and inductance are likely to change, so it makes sense. Re could be affected, too. The phase anomaly thing may alter the sound, as well. Thing is, it wouldn't show up when running an AC signal through it, because it would be dependent on the magnetic reactance with the core. You'd have to do a physical test with strings, or something. Follow?

Speaking of following things, I don't know much about the physics, or math at all. I've picked up some basics, but really just know some concepts. I'm not completely sure about the large coil phase issue I mentioned. I'm extrapolating from Bill's writing, but it makes sense. Something to listen for, in any case.

Since I'm not in touch with Bill lately and he doesn't post on the Internet, I wonder if you could clear up a few things when you get a chance:

-Does the Q factor go up or down when you ad windings to an inductor?

-Q factor is the bandwidth at the resonance, like on a parametric equalizer, right?

-Does the Q factor increase as the resonant frequency drops when you increase capacitance load on the inductor?

Take care, and thanks again.

[captcoolaid]

I run a 1961 pre CNC Gorman spiin winder and am currently looking for another. i do not want a cnc controlled one. I love everything about it.

[jason lollar]

I use winders I have built- i use old mechanical winders that work similar to the leesona and someone gave me a big old tanac that weighs probably 200 pounds. I used the tanac on one project winding several hundred Fender rhodes pickups which are round coils and well having used all these winders that thing we hear over and over how "machines wind each layer neatly next to the last which kills tone" typically isnt true. Everyones been saying it for so long no one questions it
For one thing ypu would have to be winding cylindrical coils like a rhodes and your wire guide would have to be right up on the coil or what youll get is the wire will dance back and forth on its own which it tends to do on oblong coils anyway- but anyone thats ever wound a bobbin on a sewing machine with the bobbin winding attachment will be familiar with the feed being a distance from the bobbin will make whatever you wind oscillate without any help with the traverse. Couple that with if there is any slop in the machine like on the old mechanical winders and you would be suprised how sloppy of a traverse you can get if thats what you are after. A "perfectly wound" machine coil is not typical, i think its more typical that bad sounding production pickups are due to maybe the people making them dont even listen to what they are making and obviously through observing alot of cheap import pickups materials quality and construction has very little consistancy- thats a huge deal with cheap imports.
Regardless what you useyou have to be smarter than the machine- there is always some trouble shooting whatever you use and there is always some way to make whatever tool you use perform the way you want it to.

[SonnyW]

Ah, I'm not surprised to hear that the resonance peak changes with wind pattern. Capacitance and inductance are likely to change, so it makes sense. Re could be affected, too. The phase anomaly thing may alter the sound, as well. Thing is, it wouldn't show up when running an AC signal through it, because it would be dependent on the magnetic reactance with the core. You'd have to do a physical test with strings, or something. Follow?

Speaking of following things, I don't know much about the physics, or math at all. I've picked up some basics, but really just know some concepts. I'm not completely sure about the large coil phase issue I mentioned. I'm extrapolating from Bill's writing, but it makes sense. Something to listen for, in any case.

Since I'm not in touch with Bill lately and he doesn't post on the Internet, I wonder if you could clear up a few things when you get a chance:

-Does the Q factor go up or down when you ad windings to an inductor?

-Q factor is the bandwidth at the resonance, like on a parametric equalizer, right?

-Does the Q factor increase as the resonant frequency drops when you increase capacitance load on the inductor?

Take care, and thanks again.

Well, I am probably not going to be much help on those questions. I am no expert on the math and physics of this, though there are several here on the forum that are. To me the Q is more of a constant to be dealt with for a given design, than something to be manipulated. I don't focus on Q much at all, and I don't think of a high Q as being any indicator of quality. If anything, the reverse might be sometimes the case.

Starting with your second question, the Q factor can be thought of as the degree of sharpness of the resonant peak. In a plot of signal current vs frequency, an infinite Q would be a spike at the resonant frequency and a Q of zero would have no peak. The lower the Q the broader the bell shaped curve of the resonance peak. Mathematically, Q is the ratio of the inductive reactance to the circuit resistance. So measuring Q is affected by the frequency. As far as phase, I believe at frequencies higher than resonance the current lags and at frequencies lower than the resonance the current leads. So depending on where in the frequency spectrum you are measuring, the amount that frequency is off resonance affects the phase. The math can get complex, but in my experience the Q factor doesn't change in any significant amount when adding turns. I looked through a few examples in my data sheets of pickups I've made and could see very little pattern. Adding turns will increase both the numerator and denominator. I was looking at stratocaster pickups and for most of the measurements I had (measured at 1Khz) the Q was between 2.1 and 2.7. For example adding 200 turns to otherwise identical pickups in one case increased the Q from 2.16 to 2.26, but in another case the Q was 2.28 without the added turns. Another example from a different model is four pickups of the same design with a standard wind, and 5, 8, and 10% overwinds. The Q increased from 2.29 for the standard wind, to 2.48, 2.52, and 2.65 for the overwinds. So to answer your first question, it might increase very slightly based on this small sample but the amount could be less than the variation from other factors. To the third question, I just don't know about that one.

[SonnyW]

I use winders I have built- i use old mechanical winders that work similar to the leesona and someone gave me a big old tanac that weighs probably 200 pounds. I used the tanac on one project winding several hundred Fender rhodes pickups which are round coils and well having used all these winders that thing we hear over and over how "machines wind each layer neatly next to the last which kills tone" typically isnt true. Everyones been saying it for so long no one questions it
For one thing ypu would have to be winding cylindrical coils like a rhodes and your wire guide would have to be right up on the coil or what youll get is the wire will dance back and forth on its own which it tends to do on oblong coils anyway- but anyone thats ever wound a bobbin on a sewing machine with the bobbin winding attachment will be familiar with the feed being a distance from the bobbin will make whatever you wind oscillate without any help with the traverse. Couple that with if there is any slop in the machine like on the old mechanical winders and you would be suprised how sloppy of a traverse you can get if thats what you are after. A "perfectly wound" machine coil is not typical, i think its more typical that bad sounding production pickups are due to maybe the people making them dont even listen to what they are making and obviously through observing alot of cheap import pickups materials quality and construction has very little consistancy- thats a huge deal with cheap imports.
Regardless what you useyou have to be smarter than the machine- there is always some trouble shooting whatever you use and there is always some way to make whatever tool you use perform the way you want it to.

I guess you are right, about everyone saying it for so long that no one even questions it. That's a good point. I didn't question it. I just repeated it without thinking about it like everyone else seems to. So I for one, am guilty as charged on that.

But, not to be defensive here at all, just to be clear on this, when I mentioned the "perfect wind" on the cheap pickups I was thinking more about the cheap pickups winder operator not manipulating TPL mostly, not that they actually achieved a "perfect wind". And I was also thinking in the context of my particular winder's design, which seems to assume that type of wind as a goal. I personally don't think that winding neatly kills tone. Even when handwinding, I don't scatter much. But I do see the value in manipulating the turns per layer. In the case of the machine I have, which was the subject of the original questions, it is designed to always calculate the traverse based on the input wire diameter, so that in theory each wire should lay tight up against it's neighbor. That is probably not ideal for all pickups. So to control TPL I have to trick it and input a different wire diameter than the actual number.

[rjb]

...Mathematically, Q is the ratio of the inductive reactance to the circuit resistance.
....but in my experience the Q factor doesn't change in any significant amount when adding turns.

"Theoretically", Q should go up with the number of turns.
Resistance is linearly proportional to the number of turns.
Inductance is proportional to the square of the number of turns.

[SonnyW]

"Theoretically", Q should go up with the number of turns.
Resistance is linearly proportional to the number of turns.
Inductance is proportional to the square of the number of turns.

Good point. That's exactly what it did, but there is a fair amount of variation. I didn't do much research or math on this. Just going on my data.

"In theory there is no difference between theory and practice, but in practice there is." Yogi Berra