Designing a CNC Winder and Tensioner

[Phil m]

Hey guys,

I don't post often over here. For those who might not know me, I'm a hobbiest bass builder located in Brisbane, Australia. I'm slowly getting on the pro side of building though and have been building my own pickups since day one. (photos of my current winder available in another thread somewhere down this forum)

Two years ago I decided to go back to school to study Mechanical Engineering, I'm in my mid-30's. This is now my final semester of school, it just started this week. For our last semester we need to design a product to the point where it can be manufactured. This means I need to design the whole thing, calculate all forces and moments, choose the right components, draw this up in 2D and 3D CAD and maybe even do a CAD animation of it.

You guessed it, I'll be doing a CNC winder My teacher has just approved the idea so now I have to start researching this and drawing up my ideas.

The point of this thread is to get some ideas and feedback from you guys once in a while. My semester runs 17 weeks so over the course of those weeks I'll post drawings, designs and maybe pics of where i'm at.

My ideas so far:

The CNC will run on two Axis. X will be the rotating bobbin and Y the traverse. The CNC software will be running the slow start and stop of the X axis. The counter will be part of the software. At this point all of this will be done "theoretically" I'm not sure I'll be able to physically build the thing at the same time but the goal of the project is to design something that can be manufactured with my documentation at the end of the semester.

Questions:
How fast can I run my X axis in rpm's during the wind without screwing up 43AWG wire and still maintain decent tension?

Is there any tensioner devices out there that will let you tension 42-44AWG wire in "set steps" i.e. you get a readout telling you how much force is on the wire?

[RedHouse]

You really might want to ping good 'ol RG, he will likely have some great thoughts on this subject, and always willing to discuss them with you.

If you don't know who I'm refering to, RG Keene (hope I spelled it right) is a regular here and other forums, great guy, always thinking, has loads of cranial fortitude, but his best attribute is that he is quite willing to share concepts and ideas across the internet, give him a shout, you won't be sorry.

In fact he has a post here in this forum recently, see the thread:

http://music-electronics-forum.com/s...ead.php?t=3725

[dpm]

Hey Phil! What an interesting project. I certainly haven't wound enough coils to claim a great deal of knowledge, but if you want to bounce some ideas around that would be cool. It might even be something I'd be interested in getting a working version of if it's relatively low cost for what it is.

As for your q's..... I'm able to get up to around 2200 rpm on a hand guided, hand tensioned wind using 44AWG SPN. Formvar has a lot more grab, it annoys the hell out of me I imagine a machine with accurate tension sensing and control could get pretty damn fast, to the point of reaching the spool limitations. So maybe the spool design and wind characteristics would be the limiting factors?

I saw an electromagnetic tensioning device a while back.... just googled 'electromagnetic wire tensioner' and cam up with this.... gonna have a browse... http://www.synthesisindia.com/produc...tensioners.htm

In all seriousness, it would be really cool if someone could come up with a relatively advanced but affordable winder in kit form. Something that gives programmable control for little outlay.

[Phil m]

Hey thanks for the replies guys. I was beginning to think nobody was remotely interested in this

I had read the thread recently on RG's tensioner. I checked the pdf he has in the thread very interesting but I'll have to do more research on tensioners. I checked out a few patents, I'll check out a couple more. I'm more interested in making a mechanical tensioner i.e. no electrical parts. It should be doable with just a few pulleys springs and a dancer arm. I just need to calculate the forces and angles of wrap well enough.

Thanks for the link Dan, I hadn't seen that site yet, its got nice devices on it. I doubt the end product will be cheap though, I would expect to pay about $500 in parts to build it. The more expensive stuff being the steppers, drivers, power supply and then all the mechanical parts. There's also a license of Mach3 (CNC controller software) needed to run it.

I spent the last few days modeling the winder in Inventor. Since there's some interest here's a few pics of it. I'll probably spend some time tomorrow modelling the CNC traverse . I need to do more research before I start modelling the tensioner though. My next "project" class is Wednesday. I plan on showing my teacher the 3D drawings and ask him for his advice and opinions on this and ask him exactly which parts he wants me to calculate and "prove" to him.

The first pic is the front of the winder. The stepper motor has an XL timing belt with 2:1 reduction (didn't really have an idea, wil have to check on motor speeds) its also attached to a handle through a coupler to "zero in" the CNC or unwind manually. The faceplate has an M3 threaded hole in the shaft. I hold my bobbins by screwing them into the faceplate, I still use double tape under it though, just want to make sure it doesn't go flying off.

The second pic show the back of the winder "yet" it will likely change once I stick the second motor for the traverse in it. You can see the power supply and driver PCB's plus the on/off switch and 3 prong PC type power jack.

[ATTACH]3706[/ATTACH]
[ATTACH]3707[/ATTACH]

[Achiles]

Have a look:
http://www.youtube.com/watch?v=oUXmVx3xn94
http://www.youtube.com/watch?v=uSB8ajg6x8Q
http://www.youtube.com/watch?v=Q_Cd_...eature=related
http://www.youtube.com/watch?v=mb_6X...eature=related
http://www.youtube.com/watch?v=Tcl7o...eature=related
http://www.youtube.com/watch?v=IbvMm...eature=related

[David Schwab]

Hey thanks for the replies guys. I was beginning to think nobody was remotely interested in this

I'm very interested!

[Phil m]

Thanks for the links Achiles. I had already seen Wilser and Scott's CNC winders before though but its pretty interesting to see again. There's also a few other interesting machines on Youtube but I don't recall seeing other CNC traverses. I like the look of the Chinese winder dpm bought some time ago (I would love to check it out Dan)

The main difference with the one I want to build and the two above mentionned ones are the fact that these two machines have got the traverse hardcoded on chips. From what I can see on the videos the traverses are moving simply from left to right then right to left. Its pretty much a modern way of using a cam traverse. As an aside to this, Wilser's eletronics knowledge is way beyond what I'll ever know. I could never program all those parts like he does. However, I do know the software part of it (enough).

My CNC traverse will be controlled from the software. The reason for that is because I want to simulate my own hand winding and maybe even program several different winding patterns. I had a class on CNC operation and CAM programming last year and we learned to write G-code. A winding pattern is dead easy to program in G-code once you know what all the codes mean.

I even got to program and cut out some pickup drilling templates in steel from the school CNC, how cool is that?

[ATTACH]3708[/ATTACH]

i wish I had made more different types of drilling templates before the end of the class though

[Phil m]

I just finished drawing up the traverse. The traverse itself will likely me made of a high impact self-lubricant plastic that can be threaded easily like Delrin. If you look at the two top pictures you can see that the traverse is guided by a threaded rod. The lower rod is hardened steel and there's a linear bearing fitted to the bottom of the traverse. The handle is to zero in the traverse and do small adjustments.

As it can be seen, the traverse is behind the winder (from the operator), the tensioner and wire bobbin will also be behind the winder on the table.

[ATTACH]3709[/ATTACH]
[ATTACH]3710[/ATTACH]
[ATTACH]3711[/ATTACH]

I've also designed a second traverse arm that is attached to the first one. That's to let the winder wind two bobbins at a time. That second traverse only is attached to the lower rod with a linear bearing and is directly screwed to the first traverse which is then guided by the leadscrew through the first traverse. The screw between the two traverses needs to be adjusted to select the right distance between the bobbins. That will all depend to the thickness of the separating piece between the two pickup bobbins on the faceplate.

[ATTACH]3712[/ATTACH]
[ATTACH]3713[/ATTACH]

That's the first part for now.

Feel free to critique this design and throw more ideas around. That's the reason why I posted this in the first place.

[dpm]

The traverse is the part I'm most interested in, along with the tension device. I've been thinking of something similar to use with linear stepper motors (just got a couple of different types on ebay for about $30 each) - using a fixed guide bar paired with the screw of the stepper. I've got a stepper board to power them and was kind of hoping to be able to rig up a mechanically triggered direction pulse. Obviously this wouldn't tie in with the X rotation, but with variable speed adjustment it would allow a wide array of turns/layer... if it works. CNC control certainly crossed my mind though!

[Phil m]

I'm not sure what you mean about the linear step motors. I wouldn't mind having a look at those Ebay auctions you mention.

I have to design the whole thing for the project (winder, tensioner, traverse) but I did think of just buildng the tensioner and traverse at some point. My winder may look like crap but it winds well enough. I definitely need/want the traverse and tensioner for it. It could be interesting to come up with an easy and cheap enough CNC traverse that could be retrofit to most winders out there.

[dpm]

This one is a true linear stepper - the shaft travels through the centre of the motor.

The other one is a standard rotary stepper with a lead screw and flange.

I agree that a retrofit traverse and tensioner would be nice. I'm hot and cold with the CNC aspect. I'd kind of like a simple stand-alone programmable unit, with travel limits and speed control, but I also see the appeal of something fully programmable for more varied patterns but that seems to bring with it the requirement of co-ordinated X and Y axis. Either one would be infinitely superior to my current traverse arrangement

[R.G.]

When I was working out the setup on my winder, I was focused on the goal of making it entirely out of things that could be bought at a hardware store; no machining at all. The idea was that there are enough precision widgets manufactured today that some combination of them could be used and eliminate the machining. I almost succeeded.

There is a lot less work needed to make a leadscrew traverse if you bolt both the traverse arms and the leadscrew nut onto the top of a ball-bearing drawer slide.

[wilser]

hey Phil! that looks very similar to the new winder I'm working on. It's also CNC (microcontroller) driven but a lot simpler than yours. I'll post pics of the progress so far later today.

[wilser]

Here's what I've got so far:
BLDC motor with integrated controller that outputs TTL pulses. 4x4 keypad. Rotary encoder. Matrix Orbital intelligent display. AVR Mcu. Linear stepper for the traverse. Bruce Johnson (of Johnson's Strange Musical Instrument) was kind enough to machine the faceplate for me out of aluminum. The bobbins will be fixed via 4-40 bolts to the faceplate. The software will have soft start and stop. Eventually the traverse direction will be triggered via lasers, but for the initial versions it'll be pre-positioned using the keypad, encoder and software. 100% stand alone and automated, no computer needed.

[ATTACH]3716[/ATTACH][ATTACH]3717[/ATTACH][ATTACH]3718[/ATTACH][ATTACH]3719[/ATTACH]

[David Schwab]

That looks very cool!