Syndicate This Blog |
Pushing the limitsWednesday, February 04. 2009In which your narrator pushes tests the envelope of both Art of Illusion and himself in pursuit of a printable stepper motor... I spent a week's worth of spare time translating Tigertaler's vision of a printable stepper motor's thrust plate into a VB app with the intention of translating that into a Java toolscript for Art of Illusion. By the time I got finished I understood his design well enough to be dissatisfied with it.  Tigertaler's design is brilliant for a fine resolution stepper motor. I found, though, that when I translated his tiered thrust path design, which had the lower tier for one direction and higher tier for the other, to the much lower resolutions that I needed that the difference in the force needed to move the stepper in one direction was very different than that required for the other. As well, assuming a constant velocity for the solenoid, the rate of movement in one direction would be very different than the rate of movement in the other. As Tigertaler designed his, the differences would not be terribly significant. Anyhow, having used Tigertaler's design for a learning exercise, I sat about to try to design something more suited to what I want to do. In doing so, I developed a very healthy respect for what Tigertaler is capable of. That's HARD work, for me anyway. For starters, I designed a 45 degree step angle thrust path. Instead of a two-tiered thrust path pattern, I went for a design that had a mirror image of the path for one direction on the other side of the thrust plate. My design selects direction by setting down the solenoid thrust nub into the thrust path of the appropriate side.  The thrust path looks deceptively simple. It isn't. It took me over a week's worth of evenings to generate the code to properly describe it. As well, it took moving Heaven and Earth to get Art of Illusion's boolean option to work with it. I've come to the conclusion that I am going to have to grit my teeth and admit that the critics of AoI like Nophead and many others have got the right of it. It just isn't powerful enough to do what I want to do. In any case, I finally got an assemblage made in AoI that I can use with the milling options in my Slice and Dice app. Here you can see a shaded expression of the milling assemblage.  To give you an idea of just how nasty this thing is, I'm including a wire frame view of the same assemblage.  I've tried running it through Slice and Dice and I'm going to have
to a little more cleaning of that code to get a good set of milling roads,
too.Trackbacks
Trackback specific URI for this entry
No Trackbacks
Comments
Display comments as
(Linear | Threaded)
Crazy idea... Have you heard that there's a slicing tool for POV-Ray to generate Gcode?
POV-Ray doesn't have a gui. Designing a scene is an exercise in coding. For some of us, I suspect that's familiar territory. For others, it might be more difficult.
What I'm trying to say is that not only would POV-Ray likely handle what you're trying to do better, it would also let you do auto-generation much more easily.
Not sure how well it would work, but something to think about for sure.
Brendan
It's been a couple of years since I looked in on POV-ray. I'll give it another look. Thanks!
Check out the distribution license.
http://www.povray.org/distribution-license.html
Hmmm... it appears that some sort of source is, indeed, open. That is a quibble that I'm not too interested in just now. Last night I downloaded current versions of both PovRay and BRL-CAD {I already had Blender}. It does not appear that BRL-CAD has got over being user hostile and frankly, I'm not at all interested in using something that requires joining a cult, more or less, to use. I had the same hostility to SDSS and any number of other apps. PovRay looks like it might be able to do the job and I am going to invest some hours getting used to in before too long. Right now, however, my task at hand is to prove out this printable stepper motor concept. When I increased the step angle to 72 degrees, AoI decided that it could deal with that. Once I generated a five step/rotation thrust plate design I discovered a hitherto unsuspected limitation in the paths generation code in Slice and Dice. I'm going to work on that this evening. I've got good clean STL's for a change. That makes me confident that I can pull this off.
Hi Forrest,
Why are you using a rotating design? Tigertales scheme seems to work as well "unrolled" as a direct linear motor, IIRC he even has a Youtube video with something like that. Also there really seems an issue with making progress when there is a counter force (other than the restive force).
"Why are you using a rotating design?" Because I am acquainted with the issues surrounding rotary steppers. "direct linear motor" Nice as that looks, I can't see how I can get sufficient resolution using that approach. "there really seems an issue with making progress when there is a counter force (other than the restive force)" Yes, you are correct. In my case, I don't think I have a real problem until I get to the vertical axis.
|
