Ok, so I'm basically going for an Prusa i3 like machine that is close to 100% printed. I would like to avoid using linear rods or bearings, as good rods and bearings aren't cheap.
In a worst case scenario, if my printed rails fail I could always fall back to linear rods but I'd really like to avoid it if at all possible.
In theory, if I can get a printable axis that has 0.8mm error in linearity in it, and it can position to an accuracy of 0.6mm then I should be able to have the machine self replicate.
I've been doing some designing and here's what I have at the moment:
I needed a machine to print on. Since the Icepick Delta wasn't printing accurately enough for my purposes I rebuilt my repstrap into a proper reprap (this was some weeks back) and ordered parts for Reprap Nuemann. Ever so much better than the my original repstrap:
Features a robust belt clamp and adjustable tensioner:
It works nicely, though, there is an issue: the rail is printed with the top face down. When the part is removed from the build platform it warps. With each rail that I printed I attempted to fix the problem by adjusting my heated bed temperature, less fan, very good first layer adhesion to no avail. End result is that the carriage motion isn't linear and that the rollers have slop near the ends.
In the next iteration I'll redo the rails so that the face of the rail facing the camera is position upward on the print bed and is bolted to the frame. this way, if, when warp occurs, it will be straightened with help of the frame and not affect the linearity of the motion of the carriage, since the opposing side also had the same warp, but opposing. That is, if the frame hasn't already pulled it all straight.
I'll leave it as it is right now, and get to work on the Z and X axis.