153 replies to this topic

[PyroSam]

Yeah, Python standalone executable is possible:
http://www.pyinstaller.org/

[Peret]

Yeah, Python standalone executable is possible:
http://www.pyinstaller.org/

Ah, thanks for that. I only knew of py2exe, and that only works with py2.6 and hasn't been supported for several years. I'll try it tomorrow.

[PyroSam]

Post back how it works! I'm not nearly to that point yet, just started working my way through "Learn Python The Hard Way" at:

http://learnpythonthehardway.org/

[Peret]

More thoughts.

It seems to me this should be halved? It would take 14.24 strips but a wrap is really two strips - front & back?

PyroSam, you're right - each rotation lays down two wraps so the number of wraps should be halved, and the number of steps to turn by a tape width should be doubled. Also I meant to say the tape width term is the tape width divided by the ball CIRCUMFERENCE, not diameter. Sorry for the errors, it was late at night. All these formulae need to be verified by experiment anyway, to see how the rounding errors add up, but as a starting point they're close.

As you can see from the pictures above, the WASP leaves a hole at the end. This is to stop a thick buildup of paper at the "poles". The hole should ideally be one tape width in diameter and you get it by starting to turn the ball a little short of a full wrap. The exact point can be worked out from ball diameter and tape width. Turning on the spot makes an abrupt turn that folds the paper and doesn't lay flat, so for best results, radius the turn by going M steps X and N steps Y, repeated until the required number of Y steps is completed. You want the mid point of the turn to be nearest the pole. I don't have a tested formula for M and N but using your example above, corrected for the errors above, the skew needed in Y would be 22 steps, so as a starting point I would start the turn 11 steps short of a full diameter and end 11 steps after, and (M,N) = (1,1).

To compensate for tape thickness, the additional circumference per layer is 2.pi.(tape thickness), which can be expressed in stepper motor steps. Since it's unlikely to be a whole number, and for a single layer it may be less than 1, it's probably best to work this out every new layer to avoid a buildup of rounding errors.

4 inches is probably not the best size for the drive rollers but they are available off the shelf. They cost about $50 each, however, so if you have access to a lathe it's economic to make your own. I suggest 2 inches, 50mm. The machine will run slower but only needs half the torque from the motors, which can also be smaller and cheaper, and the overall size of the equipment is dramatically reduced too. One way to make them is to turn a cotton-reel shape from aluminium stock rod, 1.75 inches diameter, 1.875 inches diameter over the flanges, with 2 inches between the flanges. Center drill to suit the motor shaft diameter and drill two radial holes tapped for set screws to secure it. Fill the center section with sixteen #224 neoprene O rings - these are 1/8 inch diameter, giving a final surface diameter of 2 inches. If you mount the motors so that the ends of the rollers are an inch apart at the closest point, it will fit all sizes from 2 inches up to more than 8 inches without having to move the motors.

[PyroSam]

Hi Peret,

Thanks for taking the time to read through the math and catching the errors, I'll add your corrections to my notes.

Thank you so much for the tip on the drive rollers. After doing a LOT of research on the net it seems the motors stalling is the one recurring complaint. It's not a common complaint though and it seems it's most often associated with the Stinger, the Super Stinger I don't think I've ran across a complaint on yet. The Super Stinger uses the highest torque NEMA 23 motor I've ran across so far - a 570 oz-in (about 4 Nm). These motors seems to be the best bang for the buck, they run about $50 US (I haven't looked for a source for them over here yet). Once you leave the NEMA 23 frame size the next size up, NEMA 34, starts at around 1,000 oz-in so it's almost double the torque but it's also double the money. Though it's nice to use off the shelf parts, sometimes a custom part is the better choice. I would think for the difference in motor cost it would pay to have a machinist turn the spools to make your own drive wheels. Beyond the cost savings, making the machine smaller and more compact by staying with the NEMA 23 motors is probably something most of us would like. (for those non-machinist types out there, the cotton-reel shape is just a spool, picture a spool of sewing thread). The need to not have to re-adjust the drive wheel spacing could also be a bonus. They may have discovered that also with the new Mini Stinger as it's motor mounts are non-adjustable.

[PyroSam]

Hi Guys,

I did a spreadsheet with the above algorithm info, nothing really productive, I was just playing and thought others may want a look too.

Peret, if you wouldn't mind, have a look to make sure I got it right.

http://pyrotechnics....orithm Info.xls

[Peret]

I'll have a proper look at it tomorrow. Your comment that the Y axis rotation (in steps) between wraps doesn't change with shell size seems counter-intuitive, but it's correct. Suppose you have a shell that takes 10 wraps, then the rotation needed is (360/10) degrees, or 36 degrees. A bigger one that takes 20 wraps needs (360/20) degrees, or 18 degrees. But we're not working in degrees, we're working in steps, and because the contact points are always one radius apart, a bigger shell takes more steps to turn it through a given angle. The angle needed is actually arctan(tape_width/2.pi.shell_radius) and when we work out the full equation all the way back to motor steps, the radius terms cancel and the number of steps turns out to be independent of size.

[PyroSam]

Hi Peret,

Thanks! Like I said, it's not a really productive worksheet but it does help one think about what's going on.

Along these lines, I just found out I'll be doing a bit more overtime again so though I'm enjoying learning Python I'm afraid my study time will be hit and miss for a while.

To get up and running quickly I"m thinking I may revisit the Excel option, though doing some net searches for serial communication via it are an excercise in patience (you'd think this would be a well covered thing, but it appears it's not).

I found some reasonably priced software that can read the serial port though and then use DDE to interface to Excel, writing back to the port doesn't look too difficult either. It's at: http://www.billproduction.com/ and costs $35 (US).

I haven't taken the time to research whether it would work with Open Office or not.

I would still like to persue a more open source control software option but sometimes one just needs to get moving and can revisit things later.

I've found two motors I like - both NEMA 23 570 oz-in like the Super Stinger. One is a uni or bipolar motor but only has a 1/4" shaft (420 oz-in in unipolar mode). The other is bipolar only but has a 3/8" shaft. I like the idea of a unipolar as there's more controllers out there for them (and using a smaller drive wheel would compensate for the loss in torque) in case the one I found doesn't work satisfactorly, but 1/4" is getting kinda spindly - I'd prefer a 3/8" shaft. What are your thoughts - do you think the 1/4" would be hefty enough?

[parabolic]

pyrosam,

stick with unipolar stepper motors, cheaper and easy to control, I would say 1/4 spindle is plently. I wouldn't think you would need massive torque anyway and you would be surprised what they can cope with.

[parabolic]

I was thinking about 3" or 4" shells!, I take it your gona make some big ones!

[martyn]

I don't want to poo poo a great thread, but I think we need to be just a bit careful we don't make Wayne look like a plonker when he's up before the gods .
We all know what might go on in private, and the enforcers aren't stupid, but to talk about making shells on an open forum is taking the piss a bit, and yes, I have probably been guilty of it in the past.
Even a 4" can weigh around 400g apparently and is also an item for practical use.
I know the views of forum members are not necessarily those of the UKPS but some things are possibly best left to the imagination or discussed in private.
On the other hand perhaps I'm just being a bit precious?
What do others think?

[PyroSam]

PyroPDC,

What was entailed in converting your stinger to a super stinger?