Help with magnet wire selection

Barncat

H-M Supporter - Gold Member
H-M Supporter Gold Member
Joined
May 14, 2018
Messages
335
I want to try and repair a blown variac from a welder. After trying 3 different tools to measure the wire, I kept coming up with .0380 diameter wire, which is between 18 and 19 awg wire. Then I see that there actually is a 18.5 awg magnet wire that is .0381, but can’t find it in small affordable quantities. If I were to use 18 awg, the 100 or so turns I need to replace would end up being about 94 turns due to the increased diameter(assuming I can keep the turns as tight as the machine that wound it originally). Will having a couple fewer turns significantly change the working ability of the variac?DA3EC0D6-EA4F-4326-B0D0-D11C67759B14.jpegE87D90A4-6707-4B93-949A-610D40631096.jpeg
 
I recently did some work on a Powerstat, although smaller than the one you have. In my case, I split the windings of a 240 volt Powerstat to make an isolation transformer.
I see several problems with you project. From the photo, you have a three phase unit and using a different wire size will put the Powerstat out of sync with the other two. The reduced number of windings will also increase the current in the transformer and increase core heating. Finally, the construction technique used was to overlap windings on the inner diameter in order to have them lay flat where the wiper travels and maintain the winding density. Winding replacement wraps will require using a shuttle and it will be a challenge to obtain a tight winding with uniform wraps. You would also have a slight difference in winding height.
I see some possible approaches. I have in the past, stretched copper wire to straighten it, reducing the diameter in the process. It would take a 12% increase in length to reduce 18AWG diameter to .0381". It work if the insulation were able to stretch that much without failing. Another possible would be to wind with the 19AWG, living with the increased resistance of the 19AWG compared to 18.5AWG.
The 18.5AWG wire would have a cross section of 1452 circular mils and a resistance of 7.14 ohms/1,000ft, compared to 1288 and 8.05 for 19AWG and 1624 and 6.38 for 18AWG. You are probably dealing with in excess of 100 ft of wire.
You might want to check actual diameters of replacement wire. My experience is that nominal AWG sizes seldom have the nominal diameter and usually run smaller.
 
The other thing to be sure of is that you are actually measuring the conductor dia and not also measuring over any insulation. Even if you have the correct size conductor there are still differences in the thickness of the insulation system that will make a difference as to weather it will fit or not.
 
Wire is probably 19AWG (0.0379"), some variations may be due to the insulation thickness. Not a simple task of just rewinding it. I would replace it rather than trying to wire it.
 
Thanks for the help everyone! I agree, winding it getting the overlap was going to be tricky. I did burn off the insulation to measure in one spot, and scrape it off in another. I have been looking for a replacement for about 2 weeks with no luck. I guess mksj eBay skills are just better than mine. It is a no brainer to just replace it, so I bought that one he linked to. With shipping and tax, that unit was $60, it was going to be about 30 in materials to try and fix mine, so easy decision since I don’t know how well the fix would go.
 
Last edited:
Good, you made the right decision. Rewinding a toroid (donut) shaped core would be darn near impossible by hand, with that many turns.
-Mark
 
I received the replacement in the mail today and it looks in much better condition than either of the two remaining from my welder. Mine are covered in a gray dust that doesn’t really blow off, makes me wonder if it is a fine metal dust. I have thought about a gently washing for the two I have to get any metal grit out from between the windings, and then a couple days next to the wood stove to dry out. Good idea or bad? I have also seen people on YouTube spraying the outside of the coils with a clear spray paint to “protect them”. good or bad idea?
 
The insulation on magnet wire used to wind the Variac is fairly resistant so unless it is burned off, it should be adequate. Coating with epoxy would be better than spray paint. Clean as best you can before painting on a coating. As to cleaning, I would avoid using detergent as any residual could be corrosive. Plain water followed by thorough drying should be OK. Isopropyl alcohol should also be OK. Try a little patch first, preferably not on the winding. Experiment on your blown Variac. I wouldn't use something like lacquer thinner or acetone for fear of damaging the insulation.

I am curious as to the cause of the catastrophic failure. Possibly a piece of metal dropped in and shorted the windings?
 
I can try some alcohol on the blown one and see what happens, test with an ohm meter after and see if the coating comes off. The inside of the welder along with all three variacs And every other component were coated with a very fine gray powder/dust. I think it was metal dust. Maybe the welder was positioned next to a grinder? I assume there ended up being enough build up to short it out, thus causing the failure. But that is just a guess. The pictures are after some work with the air nozzle blowing all the crud off the inside. Hopefully the stream of air didn’t hurt the coating of the windings on the two non-blown variacs.
 
I hope you know... no coating on the end where the carbon brush runs ;-)

Sent from my SM-G892A using Tapatalk
 
Back
Top