Nice except you've done nothing to solve your Z-axis problem, you've only made it worse. Sure you have triangles between the towers but you don't have any triangles stabilizing the towers. Look directly at any face and all you see is a square, a square that will quickly turn into a parallelogram when the bit touches down. Ask anyone who has built an oversized delta printer where the greatest challenge in the design was and most will tell you it was overcoming sway in the frame. Trying the same thing with a CNC only amplifies the problem tenfold. Consider this concept instead. It does what you are seeking and offers considerably more usable area than a triangular based solution.
That guy has 2 problems, one which I already mentioned, one which I didn't. 1st) it moves a variable amount of weight. Depending on the weight that is placed on the board, the motor will be unable to move it without losing steps. This design doesn't run into that problem. 2nd) the enclosed sides mean that I can't use my stupidly-long A axis. The open sides in this design take that into account, though I didn't mention in. I get the problem that you're trying to solve: we wouldn't want the Z axes to tilt along one of the other ones. It seems that in this case, that might be well-served by a suspension system (like a suspension bridge) applied to each Z axis instead of trying to find internal rigidity. Or we could turn one of the faces (the one parallel to the X axis) into a few triangles, which would ensure the rigidity of all those faces. Thanks for taking a look! If you have other suggestions, I'll consider those, too!
Oh, I checked that thread on my laptop without signing in, so the picture was tiny and I didn't realize that one of the walls is open. That's actually a great idea for making it more rigid! I agree on the square being simpler in theory, but it wouldn't be simpler in practice once I have the lengths figured out for the short sides. Then it's just slapping parts together one way or the other. I'm avoiding a square because I can easily run 3 motors on one axis, but adding a fourth one would double the work that I need to do/money that I need to spend without adding rigidity.
if it works or not it's good to see out of the box thinking. you could use 1 motor to spin 3 or 4 towers, just use belts and pulleys between them... Bob
I am no engineer nor will I claim to know anything about any of this because I am still trying to figure this stuff out as well, but shouldn't the router mount be on the inside of the triangle not the outside? I almost appears to me that this frame design is highly restrictive, that it would be better served as a laser engraver or a 3D printer and not used for a cnc router.
Good news! Got this thing built and working (finally)! Will be updating the build post soon with a log and notes on what I'd change now.