Hey so you have about three listings for this printer on thingiverse, could you make it only one? I can never really figure out which is the current, right page.

On the wobble issue, as long as your joints are tight (a big if), the stiffness of the square part is related to the third power of beam depth in the direction paralell to the load, linear with respect to the width of the beam, and linear with respect to the modulus of elasticity of the beam material, all other things being constant. Stiffness is proportional to E*I

Steel: E=30,000,000

ABS: E=320,000

Square tube: I=(a^4-b^4)/12 (a is outside dimension, b is inside dimension)

round bar: I=.25*PI*r^4

Threaded rod is not as stiff as smooth rod, so I would approximate the M8 threaded rod as about 7mm OD rod for an I of 0.0028 in^4 and EI=8400

The printed frame is 12.7mm square, and will be two filaments thick (estimate at 0.5mm) for an I of 0.00146 in^4 and an EI=466

To get the same stiffness with .5 mm walls you need a 32.5mm square beam, but only a 19mm on a side solid beam. I-beam style construction uses material more efficiently, but you still need something pretty big. You can get a lot more improvement by bracing the frame and reducing the effective beam length (deflection is related to the cube of beam length). Adding some corner braces in the space above the build volume could help quite a bit. Another interesting idea would be to make all four of the top bar supports into triangles and extend the horizontal threaded rods outward; this could radically improve the stiffness without adding too much more material.

Regards,

Rod

Nice concept!

However, i've found that even the standard "M8" Mendels exhibit a great deal of wobble along the x-axis when printing, especially as the x-axis moves furter up. In my experience, Printed parts are nowhere as rigid as a M8 threaded rod, so i expect the Ronthomp to woblle even worse. How are yor experiences on that issue?

More background on the wobble issue:
First of all, triangles are the only structures that introduce rigidity by themselves when you are using long bars / rods. When you look at the basic Mendel geometry, those triangles only support the Y/Z plane, which doesn't even need a rigid structure (the x-c
arriage never moves along the y axis, therefore introducing no forces along Y). The X/Z-plane however is only a rectangle, wich will turn into a paralellogram under load. When you look at the machine from the front, the structural rectangle consist of
- the upmost horizontal rods
- the two "front" h
orizontal rods
- the two RP parts connecting the rods.
Same on the back. So for the perfect rectangle to turn into a paralellogram, either the front/back/top rods or the RP bars need to bend or the part where they are screwed to the frame vertexes shifts. This issue was solved in the Sells Mendel by
using 6mm MDF sheet as an electronics holder across that rectangle. Because it was boted down twiche on both sides, it introduced a total of four supporting triangles (count 'em!).

But i think this is more of a generic problem with the Prusa Mendel than it is one you introduced with the redesign.
I'd love to see you solve it!

OH WHAT

Looks awesome ronthomp! How big mis the print area? Could it build it's own build platform too? Keep going man, I love it! More plastic!

hi!

might I suggest you start a git repository for this project?

also, I think http://www.thingiverse.com/thi... may be better for tensioning Y axis than the swing motor mount, those swing mounts tend to slip with lots of vibration and the kind of tension we subject our belts to!