Re: [CAD_CAM_EDM_DRO] CNC coil winder traverse

Hi Graham,

I recently read a post on someone winding ignition coils for model IC
engines. He wrote a Gcode program to do the winding!

So I'd say YES! Just use something like TurboCNC or EMC to run the two
steppers/servos. Geckos will be fine, and the microstepping will give
you more steps/mm.

Do you really need ballscrews? I'd just use a 20 TPI (eqv. mm)
leadscrews (think Sherline). Backlash should not be a problem, as
you're laying down turns in the same direction, until the ends, where
you have adequate time to "step out the backlash" for the direction
change.

I'd be interested in hearing a little more detail!

Alan KM6VV


Graham Stabler wrote:

>
> I have been collecting parts for a CNC air coil winder for a long
> time now and have come to the part that has been stopping me, the
> wire traverse.
>
> I am using very thin wire down to 0.02mm and want to wind as
> accurately as I can, preferably perfect wound coils. To do this and
> to make the coils I want to make, at the speed I want to do it, I
> guess I need a system accurate to perhaps 0.01mm (half the wire
> width) or perhaps less and capable of say 10mm/second traverse.
>
> I had originally assumed a stepper motor system with a fine ballscrew
> but for the price I don't think I will get the accuracy I desire.
>
> What I am contemplating is a servo system using a Geckodrive, a small
> servo motor, an IKO rail (which I have) and a belt drive. The
> feedback to the servo would be via a short linear encoder.
>
> Questions:
>
> 1. Does this seem feasible?
>
> 2. Any suggestions for sources of small servomotors?
>
> 3. Any issues with driving small servos with a Gecko?
>
> 4. Will a belt drive be good enough in terms of backlash, I guess yes.
>
> 5. Suggestions for the encoder, sources, lines per inch whatever
>
> The load will be low as it will be a simple wire guide, this machine
> will be optimised for thin wire.
>
> Cheers,
>
> Graham
>