Simplest Mechanism for wrapping coils and/or heaters

[minime72706]

I don't want to get overly complex with this, but what I'd like is probably something that merely turns an object (like a ceramic rod) while also applying adjustable tension to wire that is being fed toward the object. Having better control over things such as having a linear actuator move the object along its widest dimension under computer control to get semi-precise spacing would be nice, but I think simple is best at this stage. Anyone have any pictures of systems they've thrown together or just general tips from experience?

Thanks.

PS - I want to be able to wrap E-core and torroidal transformers/inductors at some point, but right now my priority is for a resistive heater. Apparently no one makes an aquarium heater that doesn't completely suck and I'd like to try my hand at it.

[calexanian]

Some things from experience. Whatever mandrel you use make sure the spring tension of the material you are wrapping is such that it can be easily removed after the coil is complete. It may be easier to close wrap the coil and then stretch it to final length afterwards. What kind of material are you using? Nichrome? Kanthal? Tungsten? Each one has different winding characteristics. Round wire or ribbon? Wire diameter plays a huge part too.

[minime72706]

The most immediate application involves ~20 AWG nichrome wire and my plan was to wind it under tension and then heat it with either a power supply or a torch followed by a quench. I was playing with some very thin nichrome wire that was in the form of springs that came out of a cheap aquarium heater and saw how well it held shape if heated and quenched.

[qno]

Here is my experience from watching a heater coil being manufactures.
I have worked for a year for this company.

Take a rod preferably stainless steel slightly smaller than the coil you want to use.
Wind a coil made from Kanthal wire.

The trick they used is put power on the heating wire coil so it lights up until red hot and let it cool.
This sets the coil shape of the wire.
When you remove the coil from the rod it stays in coil shape.

Put it in a stainless steel pipe with some tention so the heating coil does not touch the tube and fill the tube with magnesium powder.
Vibrate the the tube so the magnesium powder compacts.

Then both sides of the tube are sealed with a ceramic feed thrue plug.
The diameter of the tube is now rolled into a smaller diameter by feeding it thrue some rollers.
Then you can bend it into the desired shape.

Than insulation tests are done and when OK it can be used.

You can learn a lot by studying heater elements of coffee makers, cookers, ovens etc. etc.

[minime72706]

My intention was to wrap the wire around a stainless steel rod and then transfer it to a ceramic rod of the same size. I would then use thermal epoxy assuming I determine that the stuff I have on hand will allow expansion without cracking things. Since this is for an aquarium, I'd then place the assembly inside of a titanium tube ($$$). There will be temperature sensors, of course.

EDIT: and of course I intend to wrap it hot, or, in the very least, apply power to it periodically as I wrap it.

[calexanian]

line the inside of the tube with something electrically insulating. Mica paper if you can find it. glass cloth or tape (Such as used in fiberglass) works well also. You can get a little bit of it from any fiberglass shop or boat builder. It will also help with spreading out the heat transferance.

[minime72706]

line the inside of the tube with something electrically insulating. Mica paper if you can find it. glass cloth or tape (Such as used in fiberglass) works well also. You can get a little bit of it from any fiberglass shop or boat builder. It will also help with spreading out the heat transferance.

I do have thermal epoxy which is meant for potting live electronics like ballasts and the like. I currently intend to use that but might change my mind later or add something.

[minime72706]

Anyone think that there's a practical way that I can wrap the coil on a threaded rod and then successfully remove it? I suppose if the rod was oiled/greased, I could "unscrew" the coil; however, I also plan to heat up the coil to make sure it holds form. Those ideas seem incompatible unless oiling after the fact is able to get under the wire and into the threads.

[JackOfVA]

Anyone think that there's a practical way that I can wrap the coil on a threaded rod and then successfully remove it? I suppose if the rod was oiled/greased, I could "unscrew" the coil; however, I also plan to heat up the coil to make sure it holds form. Those ideas seem incompatible unless oiling after the fact is able to get under the wire and into the threads.

You could use a low melting temperature alloy (these are alloys of bismuth and indium) and cast it into a rod then thread the rod in a lathe for your mandrel. Wind the coil over the threads and then dunk it in a pot of boiling water to melt the mandrel. Or use a wax rod and melt it - wax is way too soft for any appreciable wire tension, of course.

Not the simplest approach, to say the least.

There are also expanding sleeve mandrels and I suppose one could take a piece of material (metal or plastic) and split it in two pieces. Temporarily join the two pieces and bore a hole in the center suitable for an expanding sleeve mandrel. Chuck the combination in the lathe and turn to the desired diameter.  Then wind the coil on the former and when done de-expand the sleeve mandrel which should allow you to remove the two former halves without disturbing the coil. This only works if the wire is sufficiently strong to be be self supporting.

As far as turn counting and the like, look on E-bay for "coil winders" and you will see a lot of products.

I normally wind inductors on the lathe, running slow in back gear, usually guiding and tensioning the wire by hand. However, you can also engage the thread cutting mechanism and use that to position the wire with a very simple tube through the toolpost. The tube needs to be soft and smooth to avoid wire damage - Delrin or Teflon is a good choice. You can also use a hand crank through the headstock and not run the lathe motor, which gives you excellent control but sometimes it seems that you need three hands to do that right.

I've also wound coils on a paper former - wrap a couple turns of standard computer or copy paper around a steel rod and tape the paper to the steel rod at both ends. Chuck the rod in the lathe and wind away. You can mark the paper with the turns location to help you guide the spacing. After winding, slide the steel rod out leaving just the paper former and the winding - you could burn the paper out with a match leaving just the coil or you could carefully remove the paper.

[calexanian]

I was just playing with some nichrome wire about .015 diameter. It pretty much stays where you wind it. It has a soft temper. I would just grab any wound item about the diameter you want and wind her up.

[minime72706]

Anyone think that there's a practical way that I can wrap the coil on a threaded rod and then successfully remove it? I suppose if the rod was oiled/greased, I could "unscrew" the coil; however, I also plan to heat up the coil to make sure it holds form. Those ideas seem incompatible unless oiling after the fact is able to get under the wire and into the threads.

You could use a low melting temperature alloy (these are alloys of bismuth and indium) and cast it into a rod then thread the rod in a lathe for your mandrel. Wind the coil over the threads and then dunk it in a pot of boiling water to melt the mandrel. Or use a wax rod and melt it - wax is way too soft for any appreciable wire tension, of course.

Not the simplest approach, to say the least.

There are also expanding sleeve mandrels and I suppose one could take a piece of material (metal or plastic) and split it in two pieces. Temporarily join the two pieces and bore a hole in the center suitable for an expanding sleeve mandrel. Chuck the combination in the lathe and turn to the desired diameter.  Then wind the coil on the former and when done de-expand the sleeve mandrel which should allow you to remove the two former halves without disturbing the coil. This only works if the wire is sufficiently strong to be be self supporting.

As far as turn counting and the like, look on E-bay for "coil winders" and you will see a lot of products.

I normally wind inductors on the lathe, running slow in back gear, usually guiding and tensioning the wire by hand. However, you can also engage the thread cutting mechanism and use that to position the wire with a very simple tube through the toolpost. The tube needs to be soft and smooth to avoid wire damage - Delrin or Teflon is a good choice. You can also use a hand crank through the headstock and not run the lathe motor, which gives you excellent control but sometimes it seems that you need three hands to do that right.

I've also wound coils on a paper former - wrap a couple turns of standard computer or copy paper around a steel rod and tape the paper to the steel rod at both ends. Chuck the rod in the lathe and wind away. You can mark the paper with the turns location to help you guide the spacing. After winding, slide the steel rod out leaving just the paper former and the winding - you could burn the paper out with a match leaving just the coil or you could carefully remove the paper.

I like the paper idea a lot, but a threaded rod would still be nice because I have to really ensure the wire isn't touching like in winding inductive coils. That's a bit tough when you are guiding the wire by hand. I may still find a way to use that idea!

[JackOfVA]

One method used to close wind a coil by hand using a lathe uses two spools - one of wire and one of twine with the twine diameter picked to match the desired turn-to-turn spacing. You feed the wire and twine in parallel with a bit of an angle so they stack up adjacent to each other. When completed, unwrap the twine and discard it.

It's surprisingly easy to do this with a slow running lathe after a bit of practice.

And, of course, as I said it's used for close spaced turns where the twine is similar in diameter to the wire. Not going to be so good with a large disparity between wire and twine diameters.

Image below is stepper motor based coil winder I designed and built a few years ago. Uses two steppers, one to rotate the coil form and the second to drive, via an Acme threaded shaft, a carriage with the wire guide. Both steppers under the control of PIC microcontrollers.  This allows universal winding as well as normal solenoidal winding, etc.  Got it working reasonably well but never could get the wire guide and feed tension mechanism to the point where I could produce consistent inductors.

[calexanian]

I love the shotgun shell! looks good.

[minime72706]

One method used to close wind a coil by hand using a lathe uses two spools - one of wire and one of twine with the twine diameter picked to match the desired turn-to-turn spacing. You feed the wire and twine in parallel with a bit of an angle so they stack up adjacent to each other. When completed, unwrap the twine and discard it.

It's surprisingly easy to do this with a slow running lathe after a bit of practice.

And, of course, as I said it's used for close spaced turns where the twine is similar in diameter to the wire. Not going to be so good with a large disparity between wire and twine diameters.

Image below is stepper motor based coil winder I designed and built a few years ago. Uses two steppers, one to rotate the coil form and the second to drive, via an Acme threaded shaft, a carriage with the wire guide. Both steppers under the control of PIC microcontrollers.  This allows universal winding as well as normal solenoidal winding, etc.  Got it working reasonably well but never could get the wire guide and feed tension mechanism to the point where I could produce consistent inductors.

Impressive little jig there, my friend. I guess I'm going to hope that I can get a threaded rod the proper size for the ceramic rod I got. The spacing isn't super critical in this application, so this should work fine, but I do want to make inductors as well someday. With a threaded rod I can just screw the coil off of the rod, assuming that enough tension was applied to keep its shape. I'm thinking of covering the rod with some sort of varnish to make it temporarily non-conductive. That way I can apply current to the coil after I'm done to make it hold shape. I have received both the 20 AWG nichrome wire and the 1/2" alumina rod today. I'm also playing with some "knob" from classic knob and tube wiring. A 1/4" coil fits perfectly inside and then I plan to pot it and see what happens when I apply power to it.

[JackOfVA]

I love the shotgun shell! looks good.

Just make sure the shell is expended before applying power to the inductor ...

[calexanian]

I love the shotgun shell! looks good.

Just make sure the shell is expended before applying power to the inductor ...

It's that little addition of the game of chance.

[Jay_Diddy_B]

Hi,

Have a look at the pictures in this thread. I have one of these AVO Douglas coil winders. If you search 'AVO Douglas Coil Winder' you will find more pictures of this wonderful machine. The  ratio between the lead screw and the main spindle is controlled by the position of the rubber wheel between the two the metal disks. There is a turns counter. It also has a mechanism that cause the direction of the lead screw to reverse when the end stop is reached.

http://madmodder.net/index.php?topic=5484.0

In the modern day equivalent the main spindle and Lead screw are driven by separate stepper motors. This how JackofVA machine works.

With both of these techniques the pitch of the winding is not restricted to the pitch of the lead screw.

Jay_Diddy_B

[JackOfVA]

I built one of the AVO Douglas inspired designs - the one described by Dave Gingery and it left a lot to be desired.

The original, now antique, coil winder was the Morris and it used gears to determine the ratio between coil rotation and wire movement. No possibility of slipping and good repeatability. Gingery's design substituted a disk and rubber driver arrangement for the gear shaft and that was its failing - next to impossible to get the same setup twice and even if you could, slippage was a problem.

Photos of my version of the Gingery winder and some resulting universal wound (but failure) coils.

How it Works (numbers are keyed to the photograph).

Place a hollow tube coil form on the threaded spindle (9) and tighten the conical mandrells to hold it in place. Put a spool of magnet wire on the wire holder ( thread the wire under the spindle, over the top of the wire guide and attach it to the coil form with a bit of tape or glue.

Turning the crank (1) revolves the drive shaft (2), which in turn drives the spindle through the rubber tire and disk mechanism (3) causing the coil form to revolve and take up the magnet wire. Disk mechanism (3) is held against the rubber tire by spring (4). Simultaneously, cam (5) causes cam follower rod (6) to oscillate back and forth. The wire guide (10) tracks the movement of the cam follower rod (6) thereby causing the wire to follow a universal winding path on the coil form. 

With every revolution of the spindle (9) cam (7) trips a lever that advances the digital counter, thereby displaying the number of turns wound on the coil.

[Jay_Diddy_B]

JackofVA and the group,

My AVO Douglas winding machine does not wind wave wound coils at all. The ratio between the wire feeder and the main spindle is such that it will lace the wires side by side or with spacing. It will not move the feeder fast enough to generate wave wound or self-supporting coils.

Wave wound coils are best wound with cotton covered wire or Litz wire. This has the friction that is needed to keep the winding from collapsing. When I was younger SCC single cotton covered and DCC double cotton covered copper wire was available. This was probably very common in the 1930s when the AVO Douglas machine was designed.

Jay_Diddy_B

[JackOfVA]

Looked at the photos of the Douglas winder - very nice machine indeed.

In addition to cotton covered (SCC = single cotton covered, DCC = double cotton covered) wire, many universal wound RF chokes were wound with silk insulated wire. The classic National Radio 2.5mH RF choke, for example.

New universal wound chokes from Hammond look to be wound with some sort of enamel or polyester coated enamel wire. 

As you note, the wire to wire friction is important in getting a universal wound choke, even with the correct coil winder, to come out right.  I experimented with liquid rosin and some high tack glues but could not equal the cloth insulated wire friction.