Potting electrical coil in steel

[Harold_V]

Heh! That was a good source, assuming I was filthy rich.
I paid about $4,500 for my lathe, new. It contains six of those clutches.
I realize that times have changed, and that the dollar isn't what it used to be, but I'd have to rob a bank to afford their offering.

- Electric clutch for Sag 12 € 680 Ready

- Bank and Ship € 90

Back to rewinding.

H

[Bill Shields]

supply and demand being what it is....that price does not totally surprise me.

I was guessing $500

in reality, you probably cannot make one that cheap....proper potting being the key to keeping a newly made coil functioning for more than a few days.

it may well have the layers of windings separated by 'paper'...only cutting one apart will tell

[Harold_V]

Bill,
I have one apart now. I know it's wound with a wire size of .018" (appears to be #26 American wire gauge). I have split the coil and taken a photo, which I will crop and enlarge so I can count the turns.

I can wind the coil, using the lathe (it still operates, just without one of the speeds), but I'll have to determine a way to count turns. Potting won't be too much of an issue, assuming I can procure the proper medium. I have a vacuum setup, including a bell jar, so I can pull a vacuum to remove air and ensure a full enclosure (most likely critical to cooling the coil). It's obvious from the removed coil that it was potted fully, likely the same way. I found no paper separating layers, and that's likely why there's failures (they are known to happen). The original coil has a few wraps of some kind of tape, but only in a few (maybe three) places, which, I assume, is to restrain the windings until the coil is potted. I did find a layer of paper that formed the ID, which is likely the reason the coils don't short to ground. I am entertaining the idea of adding a spray insulating layer between each layer of wire. The material required is readily available, with a .001" layer resisting 1,500 volts. As the clutch operates @ 24 V, that may offer the slight edge required to ensure the clutch doesn't fail from internal shorting, as the others have.

I'm moving slowly on this project, assuring that I will achieve success.

By the way, I just received a second response from the company in Italy. The are willing to forgo the shipping/handling cost, but that's still a rather expensive repair, especially for a machine that sits idle the majority of the time. I do not use it for gain, needless to say.

H

[Harold_V]

but I can't see why the bearings couldn't move

That's quite simple. The washers you used to form the sides make contact with both the inside and outside races, clamping them together. I realize they are a necessity, but the sides must NOT contact the outer races.

The washers can be easily relieved if you have soft jaw capability. A simple face turn on one side, creating the necessary relief. A thou or two will do it. Soft jaws make that easy, assuming you machine the jaws with a shallow cavity with a shoulder, to accept the washer, keeping it perpendicular.

H

[John Hasler]

I suggest looking for magnet wire with a high temperature rating rather than a high voltage rating. At 24V voltage breakdown is not going to be the limiting factor.

Also make sure the wire insulation is compatible with the potting compound.

I don't think the spray is a good idea. It could interfere with the flow of the potting compound.

[Harold_V]

Thanks, John. All very good tips. I'm somewhat out of my element with this project, so any guidance I can get is welcome.

H

[Bill Shields]

if you are going to wind this, you need to figure out a way to be sure that all the coils lay side by side.

if you know wire size and amount of wire (finished ID / OD / HEIGHT) of the coil..i would suggest looking around to find someone who is / has been in the coil winding business and see if they can help. I spent the best part of 10 years helping my father wind toroidal transformers...that there is a lot more to this than immediately meets the eye.

google / youtube -> coil winding

might suggest that you consider using your cnc mill as the winding machine. mount the coil on the quill

program feed down on the quill in inches / revolution where the feedrate is one wire diameter.

when you get to the 'bottom' of the stroke, program feed up in inches / revolution.

repeat as needed for as many layers as required.

this way, the wire feeder stays stationary while the coil moves / rotates.

should give you a good wind....

[John Hasler]

I have a turns counter around here somewhere that I've never used. If you want I can try to find it and send it to you. I like Bill's idea better, though. You'll know how many turns per layer you have so let the machine count layers. Note that the wire gauge refers to the bare wire while the diameter for the purpose of winding calculations must include insulation.

Is that measurement of the wire precise? .0179" would be 25 AWG but as the coil was made in Europe the wire might be 26 SWG which is .0180". If it is 26 SWG it would be ok to substitute 25 AWG but odd gauges can be hard to get (transformer and coil manufacturers do use them, though. Be glad they didn't use square wire).

When you section the shorted coil and count the turns try to figure out the style of layup they used.

https://en.wikipedia.org/wiki/Coil_winding_technology

[Bill Shields]

you can do this as a macro / DO WHILE loop

if you do not know how, I can write a sample for you and you can plug it into the control

suggest you get something to tension the wire but not rub off the insulation...you won't need a straightener...

on the order of wrap the wire around a plastic spool several times then 'brake' the spool lightly along its rotary axis. Guide the incoming wire (to the spool with a donut like on a fishing pole.

If you need a picture, i am sure I can find something.

A magnetic brake is the idea way...but doubt that you have one laying around under your bench..

ever seen a machine for making electrical wire that has a braided shield on the OK?

while you are at it...make several...you may need it in the near future. Once you are setup, making 5 is no more work than making 1

try to find someone to do the potting and send it off to them....you will be $$$ ahead

...and who knows? you may be able to sell more than a few to others and break even (or make a little) on the exercise

note...when you pull the wire off the purchased spool...unwind it by rotating the spool. suggest you NOT set the spool stationary and 'pull' the wire off the end...less chance of a tangle...

suggestion:

take a good working coil, disconnect the wires and measure the resistance of the coil with the most accurate ohm meter you can find.

this gives you a standard to work to...if your new coil has the same resistance ...good chance that it will work.

another reasonable test would be to measure the current flow through the working coil with the 24v on it...again, this gives you a baseline.

[armscor 1]

Hi Harold, years ago I had to wind a coil for my Honda MC, the sensor coil in the crankcase to control the ignition, I was too cheap to buy one considering the silly price.
3000 turns, a mission to count and unwind, I mounted a magnet on the spindle of the Lathe and fitted a digital counter, cheap from Ebay.
After winding potted it with super strength Araldite and worked faultlessly for years.
Had to be careful to feed the thin wire on the spool to get a consistent overlay and achieve the amount of wire on the small spool.
Good luck with your coils.

[Harold_V]

Ok! A little progress to report.

Early this year I posted on this project and received some excellent guidance. I have all but concluded the project of rewinding the clutch. If you're interested in reading the details, here's a link to the thread.

https://www.chaski.org/homemachinist/vi ... 4&t=110112

H