universal sleeve generator detail 2

Comment

Comment by martyn hogg on January 14, 2015 at 1:07pm

You could just make the parts hollow but include a step at the right depth inside each cylinder section that the struts would butt up against?

I would expect that it would be better to stick to a minimum wall thickness of >1.0mm but I'm not sure about aluminium sintering. We have an Objet 3d printer at work that jets plastic media and we try to keep above 1mm wall thickness on that. It will print much thinner but it gets very hard to remove the support material without damaging the part.

You could create a tag attached by a thin section that protrudes from a discreet position on the inside of a corner piece? This could easily be removed during assembly.

Chris: looks like another good project! I'm currently trying to design moulds for all my corner pieces since there are 6 different corner pieces on my geodesic dome, 3 of which are split in half for ease of assembly, making 9 moulds! 3D printing was a bit out of my budget!

Comment by Chris Hanley on January 14, 2015 at 12:48pm

One of the lessons I learned when doing a similar project, is that the placement of the  identification/labeling of the finished parts is tricky, particularly the joints.  For my project, I didn't want any visible marks on the 3d printed parts and hoped to rely on the "model number" as I uploaded to my 3d print provider.  What I learned is that, (as with most large scale printers), all the parts are placed in the same print bed in an optimized/packed orientation.  With parts like this, each one is unique, but similar, so sorting is not ideal.

 

One suggestion was to place a "rope" through all the parts.  If I were to do it again, I would probably put some type of tab or maybe a numbered wireframe bounding box that could be cut off after printing.

Comment by Max Hans on January 13, 2015 at 3:35pm

yes, that's also a thing i still have problems with or don't exactly know, where to start - the most simple thing would be adding fillets to the knots where the cylinder meet. but - as you already mentioned - rhino has a hard time adding fillets on such "complex" geometry.

that's why i thought about using a mesh instead. i  don't have a lot experience with mesh editing, but there seem to be a lot of people who had similar problems, so i'm quite confident about that.

and yes - you are right with the un-sintered aluminium. i talked to a guy at a 3d printing workshop and he proposed the same thing: holes so the unsintered powder can go outside. maybe "bone-structure" is a bit too specific. it's more like a thin net-structure - like a 3d-spiderweb - so there won't be closed cells where the material might get trapped.

acutally as i'm thinking about it, the easiest (and probably statically most unproblematic) way would be placing the holes at the front of the pieces that will be sticked into the pipes.

tomorrow i can make a quick detailed screenshot of the sleeves without the pipes so you get an idea.

i will have to make some experiments with this since i couldn't find anything like this on a short search on the web. minimum thickness is 0.7mm, so i'm curious how this could turn out and how much weight can be saved.

Comment by martyn hogg on January 13, 2015 at 12:32pm

I originally started with pipes but had problems adding fillets and I was aiming for a more organic shape with a flat base so I could print them on a cheap 3d printer. When I tried adding fillets around the piped geometry, rhino always had problems and I just couldn't get the shape I wanted so ended up (with a lot of help) creating 3d Hulls and generating meshes.

Then I realised that I might be creating the world's worst 3d jigsaw that couldn't in fact be assembled so I ended up splitting the corner pieces in half so they clamped either side of the panes and struts (you can do this with geodesic corners).

For 3d printing in aluminium (laser sintering I presume) you should watch out for minimum wall thicknesses and how the support material effects the final part. If you add a bone structure to the insides of the corners, won't there still be un-sintered aluminium powder trapped inside the bone structure? You normally need to leave decent sized holes to be able to get the un-sintered powder out I think.

Comment by Max Hans on January 13, 2015 at 8:42am

yes, IDing them will be essential, since there will be about 50 pieces. i kind of forgot about this, since i'm currently still working on the insides (i'm planning on generating a bone-like mesh structure on the inside to save weight and material).

adding text in grasshopper seemed a bit strange for me, but now i found a good article about it:

https://sites.google.com/a/umn.edu/digitalresources/tutorials/creat...

right now, all the elements are still primitives like spheres and cylinders, but i'm planning on incorporating minimal surfaces.

this is all still at a very early point of the project, there are still a lot of details to look at.

i will probably put an ID on every part, so it's easier to assemble. for the pipes i already have that

Comment by martyn hogg on January 13, 2015 at 6:16am

I did something like this but corner-by-corner since it was for a geodesic dome so had some commonality.

Are you going to somehow ID each corner so you can print them and assemble them?

Also, have you created meshes or surfaces/solids for your corner pieces?