Tuesday, July 22, 2014

AVI-TAR System Design Part 4:Plastic Prototyping

This post will dive into the design of the plastics used in the prototype. For someone without any real experience in the process of plastics fabrication I found the process to be pretty painless. The introduction of simple CAD tools and the revolution in 3D printing made this project possible.

Concept Comes First

This can be the hardest part of the process. Sometimes the design concept just falls into your lap, other times you try to force something and still other times you just feel lost. Since I'm starting with a base design (the TAR AEG) the conceptual process is exploratory: how does this look? How does that work? How do the two relate and can it be made cleaner, less complex, more robust?

In the end, I used some paper to scribble out ideas until I came up with this:




Aside: Yes, I keep a paper project binder. It's so much less structured than a computer and usually more available to capture quick bits of info. Eventually everything makes its way into files but I can always find that one measurement, coding tidbit or just my original train of thought in that binder. Your results may vary...

Now that I have new parts identified, time for measurements. I simple slide caliper did the job, giving me critical dimensions for replacement parts:



Design Comes Next

I've got to take that pile of numbers and turn into something that I can touch. I'd already decided that 3D printing was the way to go for quick prototyping so it's just a matter of creating .STL files (stereo lithography file format...used by a lot of printing systems).

I've dabbled with various 3D modeling software, mostly for animations (Lightwave, Blender and some stuff that doesn't exist anymore) but I needed something that was geared towards solid modeling for CAM. I didn't want to deal with procedural or parametric modeling; I write enough code already. I ended up using Autodesk's 123D Design. It's free, pretty basic and spits out STL files. Since it's pretty simple the learning curve isn't too bad. With about 12 hours of fiddling, watching tutorials online and letting a few cuss words leak out I started ending up with complete objects like this:



Once I'm happy with the dimensions and look it gets exported to an STL file for printing.
Time to Make the Doughnuts

I've been following the advances in 3D printing via MAKE and similar places. I'd seen some of the first commercial UV-laser/liquid polymer stereo lithography systems in the early '90s and was excited for the future. 

I have access to an Affina H-series printer. It's got a fairly small 6" x 6" x 6" print volume but that's more than I need for the parts I'm interested in. The printer is pretty easy to set up and operate. The only gotcha I've seen so far is in the feed system. Your filament spool sits on a hanger off to the side of the printer and I've found that some spools don't rotate properly, causing knots or snags during printing.

So, as I designed parts, I'd run them on the printer for try-fitting:




If you've not had the chance to use a 3D printer like the Affina, there are some kinks to be aware of:


  • The parts are layered so there is a grain to the printed parts. If you're going to have parts that need to bend, like mounting tabs, print the bending component "with the grain". That is, print so the bending component can be represented in one layer. Otherwise, the first time you bend it you'll snap the part at a layer interface.
  • Unless you set the printer properly, solids aren't really solid. Most STL slicers build hollow shells with internal webbing to save time and materials. If you need something to be thick and solid, be aware of the printer's settings.
  • Be aware of rafting and support requirements for more complex shapes. Sometimes you can print a part that seems to be mostly support material. Cutting that stuff out from the part is a pain.

Parts In Hand...Do They Work?


It's always nice to compare your replacements against the original parts just to see if you got the basic form correct:



In the above photo, the original top scope mount/rear sight assembly is at the bottom. The new display mount and housing (with the LCD display) is at the top. Looks like we're close to proper fitment. The only thing to do is try it:



In this case, it took 3 iterations to make a display mount that fit the existing top rail assembly. If you look at the original part you'll see a small bolt; the new part uses the same mounting nut and a longer bolt to secure itself to the top rail.

Parts Inventory

To complete the prototype I printed the following:
  • Bottom grip cap (holds the selection switch). You can see it in other posts.
  • Top-rear rail, which is the base for...
  • Display housing (the big curved thing that holds the display)
  • Display Bezel (locks the display in place)
  • Magazine switch mount. This holds the microswitch used to detect magazine state. It's glued inside, just below the gearbox, where there's a little room.


What's Next

More field testing is underway. Time constraints have kept the prototype off the field but I should have some field videos soon. Until then, keep winding those mags.




No comments:

Post a Comment