Pages

Friday, August 24, 2007

RepRaps by the ton

We (and especially Zach) have been devoting quite a bit of thought to how to get people started with RepRaps - when a lot of people have them, then they can all make parts for each other, but when hardly anyone has them then they can't.

In Bath we've had a visiting student over the summer: Nishad Sohoni. He has really come up with something special. Here are the parts for the standard RepRap polymer extruder (quick-swap version) stuck to some sticky tape:




The sticky tape will form a mould split line. Here are the same parts in a box with some risers and runners added:


And here's what you get when you pour silicone rubber in under a vacuum (to eliminate bubbles):



Then Nishad cut the silicone to the split line with a scalpel and took the original parts out:


Note the use of a few silver steel rods as cores for deep through-holes in the parts to be made.

Then he put the mould back together with a load of elastic bands and poured in a resin (no vacuum - the resin stays liquid enough for the bubbles to float out of the risers before it sets). The resin we used was the "water-clear polyester casting resin" from e-bay here. Here is the result:




We're obviously going to make it into an extruder and test it...

The neat thing about this process is that, while you need a vacuum chamber to make the original mould, the making of the actual parts can be done in ordinary (well ventilated...) conditions. So we can make moulds and send them to people, who can then make kits for the RepRap store (or elsewhere).

11 comments:

  1. I really like the clear-plastic look. Is it safe to assume that Zach will be getting a set of molds for rrrf?

    ReplyDelete
  2. If he wants them, yes. We're still experimenting with parts that need more complicated-shaped cores, but the extruder is solved, I think.

    ReplyDelete
  3. I have trouble seeing this method extended to many of the parts, especially those with trapped nuts and other, similar large internal cavities. Am I missing something?

    ReplyDelete
  4. Wouldn't it make more sense to put the extruder parts back into AoI and then use the boolean ops to design rigid breakapart molds which you could print with the Stratasys?

    Once you had breakapart molds you could spray them with a release agent and use them to cast something like methyl methacrylate (marketed under the names Lucite, Perspex and Plexiglass)

    ReplyDelete
  5. > I have trouble seeing this method
    > extended to many of the parts,
    > especially those with trapped nuts and
    > other, similar large internal cavities.

    That's what the cores are about. You cast a solid in the shape of the complicated interior, either in silicone or in something water-soluble. Then you place that in the mould for the outside of the object. Then you cast the resin. Then you pick out (or dissolve away) the core. It's straightforward, but we just haven't done it yet.

    > Wouldn't it make more sense to put the
    > extruder parts back into AoI and then
    > use the boolean ops to design rigid
    > breakapart molds which you could print
    > with the Stratasys?

    We could do that, but it would be quite a bit of extra effort, and would give us two sets of designs for each object. Moulding in silicone (or "soft tooling" as it's called) has the great advantage that you can make multiple copies of almost arbitrary complexity just given one original of the object you want to mould.

    ReplyDelete
  6. I seem to recall a process of building silicone molds that I saw described once where you painted layer on layer of silicone onto the object for which the mold was to be made thus building up the mold. That technique didn't require a vacuum chamber.

    ReplyDelete
  7. That would be neat. The problem I envisage is getting enough thickness for rigidity. The mould we made builds accurate parts because it doesn't distort under its own weight or the weight of the added resin.

    I guess you could use the painting-on trick with a rigid armature that stayed embedded in the mould, though.

    ReplyDelete
  8. A common way it is done in metal casting is to make a thin layer of a material with good surface recognition and then to pack around that material with stronger rigid materials.

    If you can model the interior cores in AOI then you can make molds for THEM. Cast them in wax and then suspend them inside of the part mold prior to casting. After removing the part then all you need do is bake the material hot enough to melt the wax. Some semi permanent molds will use ballons inside of a collapable plastic or wooden core so that they can be deflated enought to allow the removal of the core.

    For captured nuts what is sometimes done is that cavity of the right size is embedded into the mold. Prior to casting a bolt is covered in release agent and placed in that cavity with the nut screwed on. After the part is removed the bolt can be unscrewed leaving the nut and an appropratly sized cavity in the casting.

    Ohiomike

    ReplyDelete
  9. hi Adrian, is there any chance u could help me,, i am after a set of mendel parts so i can start making one,, but up to now i cant find a source to get my parts from....

    is there anyway you could help me or make me a set of parts and i will give you the money for them ?

    please email me at ridingg@hotmail.com as soon as possible

    thank u

    gary

    ReplyDelete
  10. hi Adrian

    i am wondering if you could help me in anyway ... i am after a set of mendel parts which i am finding it very hard to find a source, and i was wondering if you could help me in anyway

    or could make me a set of mendel parts which i would give u the money for

    please email me asp at ridingg@hotmail.com

    thanks

    gary

    ReplyDelete