Automatic Coil Winding Machine

Update 2/24/2022: I've uploaded the STL files for the machine. Note that not all of these parts are used in the final iteration and I didn't sort them yet

Update 1/17/2022: Schematics and code are linked below. Currently I only have the values for 26 AWG in the code so you'll have to extrapolate for other wire gauges

In this video I'll show how i made a machine to automate the process of coil winding. It can be used for electromagnets, tesla coils, loopstick antennas, transformers, generator coils or indcutive chokes - all things I use a lot of.

Code, Schematics, STLs:
I'm still making a few tweaks. Check back in a few days and I'll update this field with links to all of these.

Hardware:
-3A / 12V Wall wart with 5.5mm barrel jack
-A4988 Stepper Motor Drivers
-NEMA17 Stepper Motors
-3d printer leadscrew (8mm pitch?)
-20mm V-slotted rail
-M5 screws + T-nuts for railing
-M3 screws for motor mounts / set screws
-16x2 LCD with I2C bus driver
-LM7805 For logic supply
-ATMega328P-PU as the brain

Music:

Kevin MacLeod - George Street Shuffle
Serge Pavkin - Tech Research
Serge Pavkin - Futuristic Documentary

[HyperspacePirate]

I've added links to code & schematics in the description

[boulos]

The issue you tried to solve with limit switches is what we call in yarn winding "lip breaking". the reason is simple : as you lay one layer over the other, the motor basically stops to reverse, which automatically means that it lays x2 on the particular point. Also the speed of the winding affects it because the motor basically never stops on the spot (even with high end servos). The solution is to break the speed a couple of mm before the ends and to implement the lip breaking : you lay a couple of layers at full length and then others at length - x . The number of winds on full and those on full-x need to be programmable

[artbyrobot]

I am blown away by your channel. You are doing such a great job on these videos. I appreciate what you are doing here and am learning alot!

[Meccanokinematics]

Great vid. If you want to avoid any electrical noise by your stepper motors don't use A4988 or DRV8825 drivers, but TMC drivers, like TMC 2226. These are socalled silent steppers. After substituting my DRV drivers by silentsteppers I was absolutely astonished! Steppermotors were silent. No vibrations either whatsoever. Philips has written a very interesting paper about orthocyclic coil winding in the early 60's. They were able to make perfect orthocyclic coils after more than 2 years of research. If you want I can mail the paper to you.

[Mikhael]

Very insightful and creative as usual, and I like how you come up with creative solutions to achieve desired outcome. One thing I might suggest is that the wire that is being fed to the winder needs more tension points (ex. wires go around pully gears with spring tension), and the Feedpoint should have a tension of sorts which presses the wire down or up depending on the vertical height difference between the Spool and Feedpoint.

[alocin]

I have been watching your videos for few weeks. I accidently hit your channel but I am glad I did. Your videos are very inspiring, real world design, and well explained. You are a true engineer who knows how to implement ideas into workable results. Thank you very much for the details, schematic, and explanations on solving the issues. Great teaching skills you have. I watched it with great interest. Thank you for sharing. I liked your channel and also subscribed to it.

[ChiefBridgeFuser]

I find myself needing a solenoid winder, sort of like you've done here. Thanks for the ideas!

[anthonybaransky]

All I can say is WOW. I used to work in a cable manufacturing company and I ran a machine like you built but the machine that I ran could take up to an 84" spool. And the had rewinder machines that could run a 10' spool

[wtechboy]

I have a suggestion!
Instead of using a fixed clamped feed point, what about using a free-swinging feed arm that has a tiny roller on the end, so that the feed point is always basically sitting directly on the surface of the existing coil and just "rolls" against the surface? You could have your feed distance be almost zero with that setup.

[jarrahoyen]

you're a champion. i'm literally doing 2 things right now. one is making a machine to do this. and the other is upgrade my laser etcher to a larger frame. i can make some adjustments to this concept and get the build done with parts laying around. i'll be looking at adding some tensioners to the design to keep things just a little tighter (v bearings) if it turns out as well as i hope i'll shoot another message so you can take the design and fix anything and give it away :)

[LilBigDude]

Definitely needed one of these. Thanks for the files.

[PrinceAlhorian]

A must have for building a DIY Gauss Cannon

[amosdecarmo]

Sorry to bother you Hyperspace Pirate But, I am Having a senior moment here. I have down loaded your code & schematic and at a quick look over the pinout on the code does not match the schematic.

[gfr]

9:40 you have to look at textile machinery to find the simple solution for correct wire alignment. Yes the feed point must be near the bobbin you are winding but the spool from which the wire come from must be offset aside from the feed point. basically you want the wire touch always the same edge of the hole. If not the wire will go left and right random. Tensioning the wire is very simple and precise if you copy design from textile machinery, your method does not guarantee constant tension because the spool changes diameter as the wire is consumed.

[txkflier]

Excellent work. It could very easily be adapted to winding guitar pickups. However, uniform-wound pickups have more distributed capacitance due to the wraps of wire being mostly parallel to each other. To decrease the distributed capacitance, the wraps need to be laid down randomly. The stepper motor that moves the wire guide could be programmed to do that. Or, it could just be sped up so that each wrap is further away from the previous wrap.

[simonlaker]

Generally the winding machines I've worked on don't use the spool weight to get the tension.
They stand upright and de-spool from the top through a cone /funnel.

The wire then goes through two rollers.
First the wire goes around 420 deg (over one turn and comes out at 90 deg from entry point)
This is the tensioner roller that has a felt pad at the end that is tightened for appropriate tension.
Wire leaves at 90 deg from entry.
Then goes up to second roller, does a 90 deg turn, then goes through a v roller.
This keeps it center of the moving carriage.

[nerddub]

Aweosme project! Looks like you could improve the compactness of the coils maybe by replacing / modifying the sliding mechanism with a linear rail or a carriage with v-slot wheels and maybe adding a wire tensioning system, the motion of the wire outlet moving down the track looks pretty wobbly, but for ease of use and ability to make many different kinds of coiis within your frame, this is great!

[whatelseison]

This is awesome! I've been wanting to build one of these for some time now. Lately I've come into quite a few nice little steppers from label printing machines I found so I've started gathering supplies and thinking of a design for a sort of general purpose desktop CNC rig for laser, router and hopefully 3D printing but I may just have to take the plunge and buy a commercial rig for the latter. Regarding this thing though, you might be able to get even better performance and open up the possibility of using less-than-pristine magnet wire if you use a wire guide based on a series of brass V pulleys arranged in 2 rows with the wire zig-zagging between them. With a 90 degree V in the pulleys the arrangement naturally straightens the wire in both transverse axes simultaneously and can even be driven to precisely regulate the wire speed regardless of the upstream tension. I've been messing around with using that method just to straighten salvaged wire (seriously, who _buys_ wire? lol 😜) with good results.

[GaryMcKinnonUFO]

Additionally, the new loop wants to sit in between the 2 loops below it rather than being right on top of one of them.

[nigelhungerford-symes]

Nice build and great explanatory video