[yorttroy]

I am in the process of building a cnc guitar pickup winder. It will have two steppers, one to spin the pickup, and one to guide the wire. Neither stepper will see very much resistence. The stepper to guide the wire will only see the resistence of the lead screw, all thread, spinning on the plastic material used in cutting boards, I forget the name. The stepper to spin the pickup will see far less resistence then that. For my prototype I have been disconnecting the hookup wires from my cnc router and using my spare stepper. I would like to use a separate controller, power and stepper setup.
My question is what steppers and controller and power supply to use. I need steppers that can have a reasonable amount of torque, enough to turn the lead screw, at low speeds, 4 ipm max, and be able to turn 300 to 600 rpm, to spin the pickup. The controller needs to have micro stepping. I would like to have a small, compact, power supply as possible to make it a one piece unite.
I am using a Xylotex controller, a large power supply and steppers on my cnc router, this would be overkill for this project, and way more then I would like to spend. I would like to know what size of steppers I could comfortably get away with, and if anyone knows of a less expensive controller out there that would meet my needs. I would not be against making my one, but I would need a simple design, or a printed circuit board.

[alnicov]

I started to research a CNC coil winder last spring but shelved it due to more important projects. Something the size of the shuttle motor in an ink jet printer would be sufficient for the wire quide and maybe a 25 oz.-in. stepper for the coil driver. Any ideas for the tensioner encoder?

There are a couple of programs that are available for software development, like Vbasic. I've yet to find any shareware for coil winders.

Xylotex would no doubt be a nice controller for this type of application. Or, maybe something open source.

-al nicov

[greybeard]

I'll join in the search.
I'm thinking of a small coil winder for my first wind turbine generator. I've got a fair number of small steppers and was thinking along the lines of a simple pulse generator (555) to drive the spinner, then a pulse dividing chain to step the wire guide.
However I realised that the "scanning" of the guide would be a little more complicated, as the turnaround is non-linear, nor rotational, for want of a better word.
An old sewing machine in the family had a sort of nephroid cam to do the business.
Perhaps a look-up table could be written to map the guide to the spinner ?
I shall be using 22 swg(0.71mm) wire and looking at a spin rate of less than 100 rpm.
Are your needs for the guitar pick up very different ? I guess you are looking at much finer wire, and a lot more turns.

[greybeard]

I've just googled "simple coil winder" and it produced an interesting link to http://www.circuitceller.com/renesas...20abstract.pdf
(I hope that works !)
May be the sort of thing you were looking for.
John

[greybeard]

I've started from basics -

Stepper motor for spinner directly coupled to the coil bobbin.
Requires S pulses per rev.
Stepper moter for guide directly coupled to the lead screw with nut moving the wire guide.
Requires G pulses per rev.

Lead screw pitch = P
Thickness of wire = T

For A pulses fed to spinner motor, feed (A/S x G/P x T) pulses to the guide motor.

When A/S =N x L (where L = Number of turns per layer, and N = 1,2,3,4....),

reverse sign of "dir" until N x L = total number of turns required.

How to produce a circuit that will generate that relationship I don't know :frown:

But I expect someone on the forum will be able to produce the answer.

[alnicov]

I've just googled "simple coil winder" and it produced an interesting link to http://www.circuitceller.com/renesa...%20abstract.pdf
(I hope that works !)
May be the sort of thing you were looking for.
John

I wound my first strat coil with a B & D 1/4" hand drill and it inspired me enough to get a motorized hand guided winder from COWECO. I'll probably draw a plan off of that.

Typical gauges for these coils are AWG42 to 44. Some of the '60's guitar mfgs. used 45. Wraps can go above 10K on some single coils.

-al

[yorttroy]

Thank you for all the input, I will try to respond to the responses as the came in. I am thinking of scraping the separate power supply and controller and run it off my cnc router for now. I don’t want to put out the extra cost on an idea that may not work, later I can add the extras to make it a stand alone unit. For now I will ebay and get an extra stepper and use the leads off the router.

For tension I will run the wire through felt with a manual tension screw. I think if I build it with an area below the winder for the wire bobbin, as part of the unit, I will only haft to run it through the felt tensioner once for each wire bobbin, instead of each time I go and use the winder (I hope that makes sense).

I have already wrote a vbasic program. You input # of turns total on bobbin, size of bobbin, # of turns per pass, and feed rate. I am working on a second program that randomly selects the # of turns per pass from #’s the user inputs, for scatter winding. If anyone is interested I can make the program available when I finish it.

I built my first winder with a sewing machine using a adjustable cam for the wire guide, it was a nightmare trying to adjust it the cam and keeping it adjusted. You use 42-44ANG wire, thinner then a hare.

Luckily, in pickup winding you want an uneven # of turns, you don’t want the wire to lay perfectly even in rows like in most application. Getting the wire to build up flat is hard enough with out having to worry about evenly spaced turns. Tomorrow I plan to go on my first part hunting adventure, I will keep you posted on the progress.

[greybeard]

Yorttroy - thanks for starting this thread. I can see how different our needs are, but it has helped me focus on a detailed solution to my own setup.
In the midle of the night, it occurred to me that as most of my motors have the same pulse per rev, then the guide feed problem can be simplified to matching the leadscrew pitch to the desired wire "pitch".
In my case, with much thicker wire, that's quite easy, and the final simplification can come from using a simple reversing switch linked via an adjustable lever to the guide.

A short length of fine threaded bolt out of the junk box coupled to the shaft of the guide stepper, with a knife edge resting in the threads as a lead nut should give me my wire guide.

[alnicov]

yorttroy,
Yes, as I understand it, laying the wraps down evenly as opposed to scatter winding changes the capacitance of the coil which results in a change in the tonal characteristics of the pickup.
My rig also has felt lined blocks on the de-reeler and works fine. At some point I'd like to incorporate or perhaps build a tension controller like the Azonic 3000-F. here's the URL. http://www.azonicproducts.com/30-40.htm
With only .0001 to .0002 inches for the insulation thickness, I wonder if a rolling guide wheel would be needed to prevent deforming or stressing the insulation.
Thanks for the offer to share the program, I'm definietly interested.
-al

[yorttroy]

I see what you are saying about the tension controller. My biggest reason for wanting to build a cnc winder is for consistency, I can wind pickups that sound good, but not two that sound alike. I think if the new winder works the tension will be my next week point. I also wonder if the Azonic 3000-F could handle pickup wire. Maybe a spring loaded wheel that had an indicator attached to it that the wire passed over. You would still have to use the felt, but you could make a reference point to achieve the same tension each time and as the machine speeds up. I saw plastic wheels used for sliding door repair kits that might work.

[alnicov]

I don't know if sliding door wheels will have the close tolerances needed for a wire guide. Those would be easy enough to make with delrin.
The Azonic 3000-F handles wire gauges 40 to 47 which encompases most high impedance pickups. It's scaled at 30 to 40 grams tension. However, I wouldn't know where to begin to find the diagrams for the pulleys and scale indicator. It might be easier overall to just get the 3000-F at $377 US and put that R & D time to better use.
What do you have for a winder?

-al

[lwalker]

I started a CNC coil winder a while back. You don't really need two steppers: I used a 200 steps/rev size 23 motor for the traverse (which used a short pitch lead screw) and a geared down DC motor to turn the bobbin. I think it turned at about 1000 rpm. The bobbin holder had a rotation sensor so I knew when to advance the traverse on each turn. I was able to get it to lay down a smooth first layer, but ran into problems reversing for the second and subsequent layers: there was usually a hump where the wire direction changed. I think it was a combination of inertia and backlash and slowing down the drive motor would have helped.

It should be a pretty easy machine to complete: I stopped working on it due to time, since the coils were only needed for a hobby anyway.

[CJL5585]

I see what you are saying about the tension controller. My biggest reason for wanting to build a cnc winder is for consistency, I can wind pickups that sound good, but not two that sound alike. I think if the new winder works the tension will be my next week point. I also wonder if the Azonic 3000-F could handle pickup wire. Maybe a spring loaded wheel that had an indicator attached to it that the wire passed over. You would still have to use the felt, but you could make a reference point to achieve the same tension each time and as the machine speeds up. I saw plastic wheels used for sliding door repair kits that might work.

The sound is determined by the magnets used, and the wire size. The Mosrite pick-ups used broken medium power magnets and #54 wire (13,000 turns). The reason I know this is because I built Semi Moseley a computerized winding system with turns counting, auto shut-off etc. We used a pre-fab assy like a sewing machine bobbin threader. Requires only 1 DC motor or stepper.

The secret to coil winding is to start the motor out slow, then slowly increase the speed without any jerking motion. The coil can be wound in any form. It does not have to be neat and side by side. Strength of magnets, magnet placement, and wire size are what gives pick-ups their distinct sound.

Just my 2 Cents.
Jerry

[yorttroy]

I finished the pickup winder to this point. Let me know if the photos don’t show up. I used a printer for the linear bearing and made the bobbin moveable as opposed to the wire. I wounded two bobbins for a humbucker, one came out to 4.46 and the other to 4.47. I wounded them to 6000 windes at 150 rpm, about 40 minutes each. I had to stop about every 2000 turns to make a slight adjustment on the y axes, the wire would start to build a little high on one edge, but not too bad I probably could have left it alone. I used the VB program I wrote to run the machine. Unfortunately, I had to reformat my hard drive this week and did not save the program. All I have left is the 6000 turn 150rpm g-code test program I will rewrite it this week, since I can not run the machine without it, and make it available to anyone interested.

[alnicov]

Yorttroy,
Moving the bobbin is an interesting solution but could it be the change in inertia that's causing the build-up on the ends? Perhaps there's an algorhythm to adjust for that. Anyway, keep up the good work. I'd like to make some progress on my design but I'm still in the process of moving and furnishing the shop along with keeping the workload from backing up any more than it is now!
Happy new year and let us know how things progress.

-al

[unterhaus]

I never would have thought of using a printer. I usually take out a few pieces and then smash the printer up so it takes up less room in the trash can. More of the next one will escape that horrible fate.

I'm guessing the problem with the ridge comes from the wires crossing over when you switch directions. If that's true, you probably could make some very small changes in the program and get it to lay down a little better.

I've wanted one of these for a long time, and I even have two little stepper drivers that would work great for this.

[jimini]

I started to research a CNC coil winder last spring but shelved it due to more important projects. Something the size of the shuttle motor in an ink jet printer would be sufficient for the wire quide and maybe a 25 oz.-in. stepper for the coil driver. Any ideas for the tensioner encoder?

There are a couple of programs that are available for software development, like Vbasic. I've yet to find any shareware for coil winders.

Xylotex would no doubt be a nice controller for this type of application. Or, maybe something open source.

-al nicov

Picostep www.picobotics.com/ makes an individual controller similar to the xylotex @ $49.95 ea.
You would only need 2 right?

[smarbaga]

you might want to put a dancer system in line to ensire even tension.
these are adjustable weights that take up backlash and ensure even
tension.
600 rpm's seems fast to be winding a oblonged shape coil.
but if u say so.
perhaps a open faced fishing reel to evenly disperse the layers
may be of help as well.

[kenski]

Two things will help with build up at the ends.Staggerring the reversal point around the circumference is one. Accelerating off the reversal instead of just reversing direction at the traversing speed will also help.
Dancer control is probably the best way to control tension on this material.Basically you want to control the position or the extension of a spring loaded pulley.When you control the springs extension to some constant length, you're wire tension will be constant. Does this make sense?

-Kenski

[zigmart]

What I am thinking about is a semi-cnc pickup winder (no computer required)!

Only needs 1 stepper motor.

Use a regular motor to drive the bobbin spindle. On the faceplate attach a magnet that will trigger a reed switch. The reed switch has the dual purpose of triggering the counter and providing the step pulses for the stepper controller. I would like to have the ability for both full (1 step per rotation of the bobbin) and half steps(1 step per two rotations of the bobbin). For direction control use adjustable limit switches to change the direction of the stepper motor at the each direction. Also have a manual direction change switch. This is not my idea, it is already in use. To see more about it go to
http://designed2wind.alphalink.com.au/RDLab2.html. He doesn't provide schematics, but I think we can work it out if we try hard enough.