Now that the motor mounts are complete I need something to hold them together. I'm trying and make the most of my limited build volume with the 3D printer so I've chosen a 152 mm (6") diameter circle. I still can't print 152 mm x 152 mm with my printer even though that's what Lulzbot advertises. The Cura generated g-code first prints an outline of the object and that would be beyond my printer's build volume. So, unfortunately, I have to break this up into pieces.
The first version (1.0) was going to be three parts (120 degrees per part). The lower tier parts will be bolted together with the flanges at the seams and welded together with acetone (a solvent). This didn't work out as planned and even after half an hour of sanding I could not get the parts to meet up for proper welding. I decided to leave it as is to see if it was good enough. That was a bad idea. As weight started to accumulate with more parts (batteries, speakers, etc.). The inside of the chassis would sink very easily and the wheels would no longer make full contact with the ground. As a result the wheels would bounce around as they moved since they weren't all level with the surface. Time to redesign the lower tier of the chassis!
On each motor mount connection there are three bolt holes. One hole is for negative camber adjustment so the wheels can adjust to 33 degrees so they make total contact with the surface inside the sphere.
With version 1.1 I've sliced the lower tier into two parts instead of three. The two parts would be welded together with solvent along with a strip on top and on the bottom to help hold it together. This turned out to be much stiffer than the previous design. The motor mounts can still move around a bit up and down, but the degree of movement is a lot less. I still need to improve the motor mount connections (bolt holes) to eliminate the movement. I've thought about using more bolts or larger bolts, but for now this will will have to wait. I need to move on. I should probably add that the motor mount wiggle is not a big deal when I adjust the camber to -33 degrees (so it's against the surface of a sphere). It's really only a problem on flat surfaces (only for testing).
The next part I need is somethign to connect the lower tier to the next tier, for now I'll call this a flange since I can't come up with anything else to describe it. I've also added a spacer to go below the flange so that if I add parts that bolt on with the same bolts, the other flanges stay level with each other (so long as the added part has a 4 mm thickness). This part is suppose to be universal and will fit in various other places on the chassis, but unfortunately some teirs may have to be taller than others so I've made different variations of the flange (some taller some shorter).
This is what it looks like all put together and I think this is far enough along that I can start playing with the motors even if I don't have an Arduino ready for it yet. The total print time is something like 9 - 10 hours. The gray parts seem to be quite a bit stronger with HIPS filament. The orange parts are made with ABS (which is much more common). The bolts used with the flanges are #6 and the bolts used for the motor mounts are smaller #4 bolts. I really wanted to use metric nuts and bolts, but my local hardware stores only carry these strange SAE sizes. Maybe I should just accept this like a true American?
After weeks of work on what seems so little I'm eager to see if when I adjust the camber to negative 33 degrees will it fit properly in my 500 mm (20") sphere? It does! When moving the chassis around inside the sphere it makes great contact and gets good traction. I'm impressed!
So how does it perform when I hook a battery up to the motors and get it spinning? You can probalby hear some of the bouncing around from the wheels, this was before I stiffened up the chassis.