After spending the last couple months tweaking, tuning and generally goofing around with my Prusa Reprap 3D printer, this weekend I figured it was finally time to start printing robot parts.
I’ve had a minisumo design kicking around for the last year or so that was intended to have the lower half cut from aluminum and then some sort of bent sheet metal for the upper portion. The electronics were put together and blogged about on Sept. 27th, 2011. Ouch! July 26th had a shot of the 3D model I was working on for the base.
I printed the original design on Saturday and it turned out pretty good. I still needed the upper part though. And then it hit me… I didn’t have to make it in two pieces any more! I could simply print the entire thing. My brain still works in the 2.5D mode of my CNC machine… with the CNC it always came down to having to fixture the part under design to do undercuts. That is a real pain in the ass… cutting a piece, moving and rotating the work piece, cutting again… and so on. It generally meant that I made the pieces as simple and as 2D as possible.
Well, not any more. I whipped up an initial full 3D design this morning.
Seemed ok. The cavity inside the body is where the LiPo battery sits. The cavity in the back is where the 2 motors sit. There are 4 holes in the case where the sharp IR sensors poke through. They sit under the PCB so I’ve got stand-offs to hold the PCB off as well as bolt down. Initial seat-of-the-pants figuring seemed like the PCB could slip in over the standoffs, drop down and then slip in the rest of the way.
I printed with support material. Essentially it prints with lightly bonded material which can be torn off easily after the job is done. It means that overhangs can be easily printed. It also means that I can treat the design as full 3D as long as I leave room to grab inside and rip out the support material.
So I fired up the printer and headed out for mother’s day supper with the family. When I got back, everything was done. The next 5 pics show what the print looks like with support material still in place.
After about 20 minutes with some pliers and a bit with an exacto knife, the support material was mostly removed and the chassis could finally be seen.
Not too bad at all. Only two little hiccups. In my rush to get out the door for supper, I hurried the warm-up phase on my printer. It takes about 30 minutes for it to stabilize I’ve discovered. The heated bed only takes about 10 minutes to reach temperature but I need to let it sit for another 20 minutes to let all the various bits and pieces physically expand another fraction of a millimeter. If I rush it, I end up printing about 0.05mm higher than expected. For smaller prints it doesn’t matter too much aside from the bottom of the print not being as pretty. As I discovered with this print though, it also means that larger prints have the possibility of pulling off the table and warping. Whoops… Two of the corners of the base pulled up half a millimeter or so. I could likely heat it again and put it back in place, but it was just a test print so it doesn’t bother me much.
The second hiccup was my mounting standoffs. In particular, the ones inside. the sharp sensor could have gotten over as intended, but I forgot that there was a very large capacitor on the PCB right near the front edge. Doh! No way to slip the PCB in with large things on the top and bottom of it.
Ah well… Even if it had been perfect, I still needed to print it again in the right colour and with open areas on the bottom in order to fill it with lead for weight anyhow. It would have been nice to test fit the PCB though. I may cut out the standoffs for a test fit anyways. Another day.