Wind Toroidal Cores - Tips for 20 Gauge Magnet Wire

[Jdo300]

Hey Everyone,

I recently got a hold of a coupe of toroidal cores that I want to wind some wire on (see pic). They are roughly 2.5" OD, 1" ID, 1" thk. I plan to wind two coils of 20 gauge magnet wire onto them but I was wondering if anyone here knows how the factory-made ones are wound? Do they have some kind of special jig to do it or are these things still hand-wound? If there is a winding jig, how does it work, I am thinking about trying to make one but I'm not sure how I would do it.

Thanks,
Jason O

 

[Jdo300]

I found this video here showing a toridal winding machine:

http://www.gormanmachine.com/videos/proii/proii.htm

but I am still having a hard time seeing how they are getting the wire off of the spinning spool without it tangling on the spool along the way... any thoughts?

 

[Danger]

It won't play for me.
Your link gives a blank screen, with the text below. When I went back to the home page and clicked on the item, it gave a still picture, but returned to a blank screen when I tried to run it. It might just be a Mac-unfriendly site, so I'll try again when I get to work. I certainly want to see this thing in action.

 

[Jdo300]

Hi Danger,

I hope you can see the video, it is very cool how they figured out how to do that . While I was looking around for someone who could explain it to me, I came across a comment someone made about the winding mechanism being similer to that of a sewing machine. So I went to HowStuffWorks.com and checked out this article here: http://home.howstuffworks.com/sewing-machine.htm.

The lock-stitch mechanism seems to be closest to what might be happening with the toroidal winders as far as the wire comming off the spool is concerned. Any thoughts? Ideally, I would like to try to make a simplified version of this for myself.

- Jason O

 

[Danger]

Okay, the PC at work plays it. Unfortunately, I can't really see what the thing is doing. It certainly looks very cool, though. What I can't figure out just now is how it can use something like the lock-stitch without the second wire messing with the field lines from the first one. Maybe they just connect them in parallel and treat the two as one big wire?
Now I'll have to find a commercial toroid somewhere and take it apart to see what's what.

 

[Jdo300]

Hi Danger,

The winding machine doesn't actually use two wires like the lock-stitch uses two threads. I was concentrating on the bottom half of the sewing machine with the rotating bobin that the lower set of wires is fed off of. I was saying that the big metal spool that the wire is fed onto the toroid from might work in the same way that the bobbin feeds the thread off. I noticed in one of the videos that you can see a black thing that was spinning with the metal ring holding the wires, I liken that to the shuttle-hook mechanism that spins the thread off of the bobbin.

- Jason O

 

[Danger]

It's actually a lot easier to see what's happening in the video of the taping machine; it's moving slower. Still can't quite make it out, though. I'll have a go at designing one that you can build yourself. It might not work the same way as theirs, but it'll do the job. Don't worry about manufacturing capability; I always design my stuff using common pre-made parts such as from VCR's, typewriters and cars. I just have to sketch it up the way that I want it first, then modify it to fit what I have lying around.
This might take me a few days.

 

[berkeman]

The winding machine we use (Jovil Universal) first puts the wire onto a circular shuttle ring, and then that is spun around the toroid to wind the core. The wire is kind of spiralled inside of the shuttle (which is a hollow ring about 3cm in overall diameter and about 2mm in inner diameter size), and the shuttle ring has a split so that it can be put over the toroid after the wire has been spiralled into it. Here's their website, but they don't seem to show any closeups of the winding head and shuttle:

http://www.jovil.com/

For your hand wiring, you need to put the length of wire onto some sort of spool that is small enough to fit through the center of the toroid. That way you can keep passing the bulk of the wire through the center of the toroid as you wind it.

 

[Jdo300]

It's actually a lot easier to see what's happening in the video of the taping machine; it's moving slower. Still can't quite make it out, though. I'll have a go at designing one that you can build yourself. It might not work the same way as theirs, but it'll do the job. Don't worry about manufacturing capability; I always design my stuff using common pre-made parts such as from VCR's, typewriters and cars. I just have to sketch it up the way that I want it first, then modify it to fit what I have lying around.
This might take me a few days.

Hi Danger,

I'll be happy to see what you come up with, I'm still pondering a way to do it as well (besides manually of course).

@berkeman

That’s great that you have access to one of those things. Though I must say that I am still having a hard time visualizing the mechanism at work. How can you put the shuttle ring through the toroid once the wire is fed through it? Is there anyway you could make a sketch of how it is supposed to work, or take a few close up photos of the actual machine? That would help tremendously.

Thanks,
Jason O

 

[berkeman]

That’s great that you have access to one of those things. Though I must say that I am still having a hard time visualizing the mechanism at work. How can you put the shuttle ring through the toroid once the wire is fed through it? Is there anyway you could make a sketch of how it is supposed to work, or take a few close up photos of the actual machine? That would help tremendously.

I'll try in words first. Imagine a bigger version of the shuttle ring, say you took a drinking straw and could bend it into a circle with a small gap left. You take the magnet wire, and form it into a tight spiral the same size as the inside diameter of the straw, and push the spiral of wire into the straw from one end. You end up with a long spiral of wire through the length of the straw, with the final end of the wire sticking out of one end. You are limited in how much wire you can pre-stage in the shuttle ring by the thickness of the wire and the size of the shuttle. This determines the max number of windings you can get on the toroid.

Once the shuttle ring is filled with the spiral of wire, you open the split in the shuttle ring slightly so that you can put the shuttle ring over the toroid, with the split positioned in the center opening of the toroid. You then attach the spinning rubber wheel things to the shuttle ring, hold on to the exposed end of the wire, and start the machine. The rubber wheels spin the shuttle ring, while the toroid is slowly rotated by another set of wheels to evenly distribute the windings.

So the whole trick to this particular toroid winding machine is how they spiral the wire into the inside of the shuttle ring prior to doing the winding step. As I said, winding any toroid involves putting the wire on something that you can then pass through the middle of the toroid multiple times. In your case, that would probably just be a small spool.

 

[zoobyshoe]

berkeman, excellent job explaining that. I can visualize the whole setup without confusion.

I'm impressed by the cleverness of whoever figured out how to do it that way.

 

[Jdo300]

Hi Berkeman,

Thank you for the detailed explanation, that cleared it up for me :-).

- Jason O

 

[Danger]

Well... this is turning out to be a bit stickier than I'd anticipated, but I'm still working on it. Beginning practical experiments to test out a couple of ideas. Stayed tuned.

 

[Jdo300]

Alrighty

 

[Danger]

Plan 'A' -- dismal failure
Plan 'B' -- dismal failure
Plan 'C' -- showing promise, but harder to make

 

[NoTime]

Plan 'A' -- dismal failure
Plan 'B' -- dismal failure
Plan 'C' -- showing promise, but harder to make

If you're not making mistakes, you're not doing anything!
The key is to not keep making the same mistake

 

[Danger]

I almost never make the same mistake twice, but I have some kind of record for the number made the first time.

 

[NoTime]

Busy guy

After you beat me with your dinner table, I don't think I'm going chalange your claim

 

[Danger]

I'd forgotten about that.

 

[berkeman]

Um, it's just wire guys.

 

[Jdo300]

Hi Danger,

Could you run your idea past me? I still have more toroidal coils to make (on larger diameter non-conductive cores so any ideas will definitely be helpful. As much as I enjoy spending insane amounts of time manually winding these things, it would me much better if I could do it automatically.

- Jason O

 

[Danger]

Hmmm... a bit hard to describe since I haven't actually done any sketches for it except on paper.
Basically, it's sort of like the spool that fishing line comes on, except inside-out (ie: a 'conveyor belt' made of channel material with the open side in, configured similar to the 'hoop' in the video). It opens like a bracelet to insert the toroid, then is closed.
A 'slider' is then threaded onto the free end of the wire from the original spool and snapped into a track in the channel to maintain tension and prevent tangles. That's the part that's giving me fits right now. I haven't been able to find anything lying around that meets the requirements. Right now, all that I can come up with is to use a mandrel-bent tube and grind the inside part off to create the channel.
The end of the wire is then taped to the inside of the channel, and the channel spun up to load it with as much as it will hold.
Anyhow, once it's ready, you do a couple of winds by hand to start it off, then run the motor in reverse to unwind the wire onto the toroid.
A slow gear train, ratchet, or stepper motor increments the orientation of the toroid each time a turn is completed. I haven't spent any thought on that yet, since it's the easiest part.
As you can see, I'm nowhere near a workable design yet.

 

[Danger]

Got a bit more figured out today at work. Was going to do up a sketch in Illustrator and post it, but W intercepted me with tacos and beer. Anyhow, I've got the slider worked out, and am thinking toward using the ribbon-advance mechanism from a typewriter for the toroid staging.
There are two things that I need to know in advance. The first is what kind of tools/shop you have available. The current design unfortunately requires brazing or low-intensity welding, plus a shop vise and metal-cutting saws. I'm looking for a way to do it with plastic rather than metal, but that can be worked out later if the initial design proves workable.
The second thing that I need to know is what tolerance there is between adjacent coils. This thing might be a bit sloppy, so toroid staging might become critical.

 

[Jdo300]

HI Danger,

I have access to a small shop with a manual lathe and mill. I also have access to both plexiglass and aluminum stock to work with. I can't weld anything though I do have a soldering iron and/or a small butane torch at home to braze with (not so good at it though). Maybe if you can run the design past me, I could come up with something to work with. As for accuracy, I'm working with wire sizes ranging from 16 to 24 gauge, and toroids ranging in size from 2" OD to 6" OD floral rings :-).

Thanks,
Jason O

 

[Danger]

Hey, Jdo;
I've got the hoop and slider worked out. Am just about to start sketching it in Illustrator. Will post as soon as I'm finished. This might take a while, because the wife is quite irritated about me paying more attention to the computer than to her. (That's not true, but you know how women are. ) On the bright side, she's gone to bed and doesn't expect me too soon.

edit: I've just spent quite a while on it, but am having some scaling problems. Also, our new budgie is loose in the house, and I have to catch her before going to bed. I am, therefore, going to have to leave this for tomorrow.

 

[Danger]

Okay, I'm going to do something ass-backwards here to save time. I'm at work with no access to my drawing programme, so I'm going to go ahead and post the assembly instructions that go with it before the drawing itself. Of course, it won't make much sense until you can see what I'm referring to. It'll be up tonight if nothing goes wrong.
To start with, the hoop is composed of 4 x 1/64" aluminum rings, two of which (the 'inner' rings) are 12" outside diameter. That is also the size of the 2 'spacer' rings, which can be composed of plastic or whatever and are 1/32" thick. The 'outer' rings are 12 1/4". All have an inner hole of 11 9/16" diameter. The perimeter of the hoop is composed of 3/8" wide steel banding material such as is used to package lumber. You can get scraps from any lumber yard or hardware store, but will have to remove the rustproofing coat (easy). Just make sure that there are no kinks or creases in it. It needs to be 37 1/2" long. You also need 4 'studs', which are 3/32" bolts about 1/4" long with the heads ground down to 1/32" thickness, with 3 matching spring washers and nuts. The hinge plates and latch can be any fairly strong, thin material such as 1/32" aluminum.

PROCEDURE

1) Tape the band end-to-end in a 12" outside diameter circle (there'll be a small gap) and set it on one of the 'inner' rings. Place the other 'inner' ring on top, so the band is sandwiched between them rather than wrapped around them. Clamp solidly and braze the band to both rings from the outside. Once that's started, you'll probably have to keep moving the clamps and positioning the band as you go. Make sure to keep the outside flush with the rings. At some point, remember to remove the tape. You might also need to heat-sink the aluminum, but maybe not if you're careful.
2) Clamp the 'inner' assembly, 'spacer' rings, 'outer' rings, latch and hinge plates in place. The hinge plates should be centred over the gap in the band, with the latch diametrically opposed. Drill the stud holes. The ones for the hinge plates go all the way through the whole assembly (to ensure alignment). The latch ones just go through one 'outer' ring and spacer, although you can go farther without hurting anything.
3) Unclamp the parts and enlarge the holes in the spacer rings to fit around the stud heads.
4) Put the studs into place from the inside of the 'outer' rings and braze the heads to the rings.
5) Clamp the 'inner' assembly, 'spacer' rings and 'outer' rings into position and braze the rims of the 'inner' assembly to the 'outer' rings with as small a bead as possible. This might burn the spacer rings, but it doesn't matter too much. (Watch for fumes, though.)
6) Cut the 'V' notch for the hinge. You might have to cut more of the banding away if the original gap isn't big enough.
7) Install the hinge plates, using lock washers and nuts.
8) Grind the hinge plates down to match the curvature of the 'outer' rings. (Important!)
9) Cut the latch opening and install the latch using one lock washer and nut on the pivot end.
That's it! (For the hoop, that is.)

 

[Danger]

Okay, here is the preliminary sketch for the loop section. Please forgive the quality; I'm very tired and somewhat inebriated, but I hope that it gives enough info in conjunction with the assembly instructions. The part that isn't included is the 'V' notch mentioned in the instructions, but I'll look after that tomorrow after work.
http://img155.imageshack.us/img155/5204/toroid1ji1.jpg
Note that the sketch is not to scale. That's the problem that I was running into yesterday; when properly scaled, you can't see the details.

edit: Now that I've seen the thing, I'll enlarge the text and repost tomorrow. That's pretty hard to read.

edit#2: Never mind. I found another bottle of beer in the back of the fridge, so I stayed up and reposted the picture. If it's still unclear, I'll try again tomorrow.

 

[NoTime]

Do you really need a hinge on this?
At 12" I'm thinking its probably flexible enough to get it into the toroid core without kinking it.

 

[Danger]

That's a good insight, NoTime. Since at the time of designing this, I didn't know what sort of facilities were available, I set it up for 'worst case scenario'. I thought that he might have to use higher gauge steel rather than aluminum for the rings. Not too much flexibility there. Also, I didn't know the toroid size other than going by the original posting, so was trying to give the most possible turns with something that would still work within a half-inch (ID) toroid. Keeping in mind that the clearance will decrease with every full rotation of the toroid, I was looking for the smallest possible device that might still fulfill the requirements. That implies to me some universal approach, which doesn't allow for flexibility. My design might flex enough to allow placement around the toroid, but I doubt it.

 

[Danger]

Here's a more close-up sketch of the important parts.
http://img104.imageshack.us/img104/9203/toroidenlarged1fg1.jpg
I'm currently working on the slider sketch.

edit: WTF?! That's not what it looked like when I uploaded it. I'll take another shot at it.

edit: Something's screwy here. Maybe it's just this computer. I uploaded the image from the Mac and it's fine on the ImageShack site. Just now, I entered PF through the Mac and clicked on my own link. It came up fine. On this thing, it's just a bunch of barely visible lines. Can the rest of you see it?

 

[berkeman]

I can see it fine. Man, you've been busy Danger!

 

[Danger]

Not really. I devote a couple of hours a day to it. Probably more today, because W is up in Lloydminster for a wedding and won't be back until tomorrow morning. I like to have something to do while I'm drinking, and there's nothing good on TV today.

 

[berkeman]

Wish I was home with a cold one as well. Unfortunately I'm at work on this sunny Sunday. Got an important deadline that I'll just barely meet. Have one for me!

 

[Danger]

I don't know what you drink, so I had one of everything in the house for you. Pretty much settled into the Canuk and Corona now, because I'm out of Keiths. Saving the Havana Club and Bacardi Breezers for an emergency. The liquor store is at least 100 yards from here, and I don't feel like walking.
Anyhow, here's the slider sketch. Again, it's not to scale and I didn't spend a lot of time trying to 'polish' it.
http://img56.imageshack.us/img56/7863/slider1ju7.jpg
It should be made out of some tough plastic such as nylon with a very 'springy' nature and low friction. Delrin might work. It's made out of a flat strip, and so wants to straighten out. This keeps it spring-loaded against the wire in the hoop. The configuration is designed so that the tension of the wire during the toroid winding phase also tends to expand the diameter. It's width is a snug fit within the channel of the hoop. The 'anchor ring' is there to stick a wire or something through during the loading phase, because the slider has to be restrained from rotating with the hoop. I'm planning to incorporate such a device into the capstan system.
Okay, I'm off to start sketching the aforementioned capstan thingy.

 

[Danger]

Sorry, guys. I took time out to watch a couple of movies and hoist a couple out on the lawn with the neighbours. Anyhow, here's the initial capstan sketch. Since I'm just finishing off beer #18, my input for tonight is probably at an end. Catch you tomorrow.
http://img179.imageshack.us/img179/668/capstan1sq6.jpg

edit: Might screwy, indeed. I'm at work now, and can't see the sketches on this machine either. When I go to ImageShack--My Images, though, they show up fine.