Linear stepper actuator for automating winding lay

 :

9/20/2002 12:21 AM

R.G.

Linear stepper actuator for automating winding lay We touched on the topic of automating the lay of wire by using a stepper-fed mechanism to control the placement of the wire inside the bobbin. If you'll go to     http://www.alltronics.com/motors.htm     You'll find:     LINEAR ACTUATOR   24V, 6-wire, 0.35A, 1.8°/2005-rev. l6TPI lead screw, travel ~1" max. 2 1/4"sq. face x 2-1/8" long. Eastern Air Devices #LA23ECKA-12A1.   20M002 $7.95 each     Looks ideal. A 1" max travel and fine resolution should do a neat job. If I read that right it's 200 steps per turn and each turn is 1/16 inch, for a resolution of 0.0625/200 or one thousandth of an inch per three steps.     You could do something similar with 1/4"-20 threaded rod and a standard stepper, of course.     Here's a thought.     Universal wound/ progressive wound coils can easily be made with a stepper running the bobbin turns directly and something like the linear actuator running the traverse. However, "scatter wound" just means that the winding is put on the bobbin somewhat erratically, usually by hand. If you program the automated lay-er to do the wire lay in a pseudo-random pattern, you can **replicate** identical "random" wound pickups.     Here's the idea. Get a largish stepper with about 100-200 steps per turn, and mate its shaft with a bobbin holder to do the main turning of the bobbin. Then use something like the linear actuator to do the bobbin lay automation from side to side. Run the steppers from the printer port of your PC, and do the bobbin design on the PC, then fire it up and wind the pickup.     I may have to do some mechanical drawings of how to set that one up. Should work out very nicely. I bet two stepper motors, some plexiglass (and a couple of years of tinkering would turn up a really nice automated pickup winder.     Shoot, can you replace the DC driver motor on that Grizzly wood lathe with a stepper?     R.G.

 

9/20/2002 2:26 AM
Jason Lollar	There is another stepper control offered as a kit here http://www.ozitronics.com/ .     I like to use a cam though, very simple and works fine.

 

9/20/2002 4:24 AM

Dr Strangelove

RG wrote: [stuff about stepper motors controlling wire traverse]     An automatic winder is not a substitute for a thoughful experienced person doing hand winds.     It is a tool that will let you try out different winds quickly.     Making an automatic winder traverse means addressing several points. For calculation's sake, I assume a Strat pickup bobbin is roughly .5" inches wide and 6.5k. Both those figures are slightly off, but near enough for starters.     1) traverse speed -- the motor needs speed as well as fine resolution. In a worst case Moebius wind, the wire moves across the entire bobbin width in 1/2 revolution. At 600 rpm, that's .500 inch displacement or 1500 steps in .05 sec. This is beyond most steppers as even the good ones rapidly taper off above 2000 steps/sec. Hobbyist ones range ~2 - 4 ms/step.     What falls out is that you must control the winding speed for difficult patterns and still fall within the 30 minute wind limit for a rational production schedule. You can't always go full speed at 1200 or 5000 rpm on a guitar pickup. 9000 winds at 600 rpm takes 15 minutes for our generic Strat pickup.     2) traverse position -- at the very least, you need limit switches at the right and left extremes of the bobbin position. Each bobbin needs a calibration run if the control software has to count steps. You could use the quadrature detector from a hi-rez mechanical mouse to tell you absolute location.     3) traverse resolution -- the worst case is a "perfect lay" helical wind. For our Strat bobbin, the necessary resolutions are:     #42 -- .0028" dia -- 179 winds/.5 inch width   #43 -- .0025" dia -- 200 winds/.5 inch   #44 -- .0022" dia -- 227 winds/.5 inch     I'll settle for 200 steps per inch on the stepper and the position detector. That's manageable.     4) bobbin position -- the traverse program needs to know when the bobbin is at top dead center and bottom, too. A little bit of closed loop control makes for reproducible winding.     5) programmable winding patterns - random winds aren't really random in a deterministic machine. Different pitches, patterns, quasi-randoms must be changeable on the fly over the course of a bobbin. You need this to _control_, not minimize the pickup reactances for your unique sound and preferences.     Note: A Strat pickup has roughly 60-70 layers by the above calculations. John Suhr of Suhr Guitars claimed to use 60 pattern changes for winding his V60 pups on a Tanaka winder probably like this one:   <http://www.tanac.com/ax2/ax2.html>     The AX2 traverse speed is 80mm/sec (3.2"/sec) and takes 159 milliseconds to traverse a Strat bobbin. It ain't doin' no Moebious wind at 600rpm. 180 rpm is more like it. A 50 minute wind? Ack.     5a) Recording mode -- the traverse computer needs a "learn" mode that records your hand-wound pattern while you do it. Not all hand winds are equal, even by the same person, so it's very important to have that accidental "great" wind pattern recorded for posterity and your livelihood.       -drh   --

 

9/20/2002 7:17 AM
Jason Lollar	In addition to the above: changing direction of the traverse too fast at higher bobbin rotation RPM (600+) can create unacceptable loose winds.Check it out.

 

9/20/2002 7:25 AM
Regan	I think that a combination of steppers and maybe a motor with a shaft encoder for the bobbins would work, and would be easy enough to do. I think that maybe another stepper controlling a tension arm might be cool to.   I think that this is getting into the fairly high production realm though   Regan

 

9/20/2002 7:02 PM
WolfeMacleod	Hehehe, didn't you and I just talk about "learn mode" the other day?   j00 b3 g

 

9/20/2002 7:57 PM
Dr. Strangelove	Wolfe wrote:[QUOTE]Hehehe, didn't you and I just talk about "learn mode" the other day?   j00 b3 g

 

Page 1 of 4

Next>

Last Page>>