Gimbal arm Mk I

As softly indicated in our Rendering post, our current focus is on designing a gimbal arms. These will serve to extend gimbals configuration away from gondola and align them with main thrusters.

To be able to print such a monster (60+ cm end to end) it got separated in three sections.

1/ The gondola clamp

I think that pictures will tell more than explaining here what and how.

Overall this clamp is pretty stable and does the job, however joint with that mid-section is too klanky and needs re-designing.

2/ Mid arm section

How simple it looks, it took several hours to design a shape which starts with ellipse cross-section of one size (6cm) and skews sideways into another ellipse of a different size (4cm). I am not going to spam here with a code for the others, but I am actually pretty proud of this part so here it comes!

fn=32;

#parameterised 2D ellipse
function ellipse(r1, r2) = [for (phi = [1 : 1 : 360]) [r1 * cos(phi), r2 * sin(phi)]];

module tube(r1, r2, rr, R, th) {
  D() {
    skin([for(i=[0:fn]) 
    transform(rotation([0,80/fn*i,0])*translation([-R,0,0]),
      ellipse((r1+(r1-r2)/fn*i)/rr, r1+(r1-r2)/fn*i))]);
    
    
    assign(r1 = r1-th,r2 = r2-th)
    scale([1,1.2,1])
    skin([for(i=[0:fn]) 
    transform(rotation([0,80/fn*i,0])*translation([-R,0,0]), 
      ellipse((r1+(r1-r2)/fn*i)/rr, r1+(r1-r2)/fn*i))]);
  }
}

module mid_section() {
  D() {
    Ry(10) {
     skew([0, 0, 45, 0, 0, 0])
     tube(r1 = 3, r2 = 1.5, rr = 6, R = 25, th = 1);
    }
  }
}

mid_section();

Design above allows printing all that monstrosity quite efficiently “laying flat”

It clearly was our longest print so far – it took 18hrs – and seeing how that worked out it already needs redesign. An internal reinforcing rib is needed and joints on both sides needs some sort of clamps to be more solid.

3/ Gimbal adapter section

When having that above mid-section in hands, designing this part was much easier. Practically the same, but smaller + bit of fun when scaling cylinders to make them fit that ellipsoid cross-section. Problematic part came when attempting to join it with the gimbal as there was not enough material to make me confident about it withstand some reasonable sheer forces (400N). Simple “test by Ondrej” demonstrated that I was right.

Video from printing here:

Anyway, at the end it all clicked together (with bit of sanding, heating and persuasion) to the point where we could demonstrate its first prototype!

As you can see above, there is quite a sag on that arm which is caused by multiple design weaknesses, but mainly the joint between the main clamp and the long arm. That needs to be surely reinforced. I am still liking it as it really looks pretty close to that original rendering, just this is real! 😀