Hexapod Leg Design: Optimizing a 3D Printed Femur Assembly

[theycallmevirgo]

The below is a leg assembly from the Antdroid open source hexapod robot (https://github.com/antdroid-hexapod/antdroid). The "femur" assembly connects the servos at S1 and S2. It is comprised of two side plates connected by cylindrical struts.

Being fundamentally lazy, I prefer to redesign this to be 3d printed as one unit. Obviously, the most likely way to cause destructive deformation is to rotate one end about central axis A_1 while holding the other end fixed. Strain along coplanar vertical axis is not in my opinion worth considering.

The question is, simply, how should I approach the design of the strut? Should I create a solid on a plane between the servo mounts, or should I create a solid that conforms to the plate arc? Either way, can I safely semove some part of the solid to save material? If so, at the center or at the edges?

If my explanations are unclear I will post images of alternatives later.

I have access to some decent software to study the stress reactions. I'm asking here so I can understand which results are "reasonable". Of course I will post outputs here.

 

[Lnewqban]

You could copy the fundamental C shape of cast members of car's suspension struts.
Please, do a search for "cast strut arm"

If the printer could do it, a closed rectangular cross section would give the most torsional resistance about axis A1.
The core of that rectangle can be empty, as that material does not contribute to much resistance, saving some weight.