printing a mold of an STL file

[lycan]

Gday All,

Ok was just wondering if you have already created an object and saved the stl, is there a software program you can run the file through that actually creates a printable mold of your object so you can reproduce without having to reprint the item ???

Just curious.

Cheers

[curious aardvark]

Well you can use openscad. Create a cube and then difference the stl file from it.
I would imagine some of the more high end cad programs would have a create mould function.
No clue as to which ones though.

[3dkarma]

There's an OpenSCAD two-part mold generator that will generate a mold from an STL file on Thingiverse: http://www.thingiverse.com/thing:31581.

[Wolfie]

I am just dubious of this. Yea, it would work. But doesn't it take way way way longer to 3D print the mold than to print the object? Why not print the object and use inexpensive and fast materials to make a mold of the printed object? End result is the same or better and it takes less time. And I would think it would be easier to do post processing cleanup on the object than to do it on the interior of a mold.

[chris@radeuslabs]

We ended up using a mold for something we are developing, basically due to the overall design / shape and the fact there was an epic fail on the printing of the original part. The mold turned out really nice after a little finishing work (sanding/waxing). As far as software goes, we just created the mold as a new part.

[curious aardvark]

Wolfie printing a mould is faster and uses less material than printing the actual part. Think about it - you're basically creating a space shaped like a part.

[Wolfie]

Wolfie printing a mould is faster and uses less material than printing the actual part. Think about it - you're basically creating a space shaped like a part.

Ahh, and there's the rub. You are creating a space shaped like the part. THEN you surround that space with an even larger box to support it! So, not only are you creating a shape LARGER than the part (ie on the outside of the part's wall, not inside), you must also fill the space around THAT shell and then print another shell in the form of a box. So, the actual print volume must therefore be larger than the part itself.

[curious aardvark]

well only if you insist on making the shell really thick. Plus the mould doesn't need to be particularly strong so you'd use a lot less infill.

[Wolfie]

Even at the same thickness as the shell of the object, its still MORE plastic. Take an apple. Pretend that saran wrap (cellophane) is the same thickness as the apple skin and wrap the apple in the saran wrap. There is more volume area of the saran wrap than there is of the apple skin. Why? Because it, by definition must be bigger than the apple. The apple skin starts at its surface and goes inwards. The wrap starts at the outer surface of the apple skin and goes outwards. Its ALWAYS longer around the outside of a circle than the inside. Ask any truck driver (lorie) or track runner athlete.

So, even if the infill is 0%, there is more plastic in the mold than the object.

Lets say the object is roughly a ball (to make calculations easier but the principals hold true for any non-cubic object)... lets say its 2cm in diameter (1cm radius).
V = 3/4 * π * R³ = 4/3 * π * 1³ = 4.188 cm³

So the object is 4.19 cubic CM volume.

Now the mold must be at least a 2cm cube since thats the object diameter. And since it must have room for the inner wall and outer wall lets bump it up to 2.1cm each side (5mm on each side for wall and infill). Bare minimum I should think.
V = a³ = 2.1³ = 9.261 cm³

So the volume of the mold must be 9.26 cubic CM. Subtract the volume of the object (4.19) and you have a volume remaining of 5.07 cm³ outside the object but inside the cube of the mold. So already it contains 21% more volume to be infilled than the object itself.

Now, the outside must have a surface so add (A=6a²) another 26.46 cm² times whatever the thickness you make the skin. Even if there is zero infill, the volume of the skin of the mold around the object will always exceed that of the object it contains. Then add in the volume of the cube's outer skin and you guarantee it will contain more plastic.

Therefore, in any case where the object is roughly more spherical than it is cubic, the mold must contain more plastic than the model. And if it contains more plastic than the object, it therefore will also take more time to print at the same resolution as the original object. The higher the infill percentage, the wider the gap in print times becomes.

If the object is roughly cubic in nature, then its possible that the interior of the object can exceed the infill volume of the mold. But this is the only case I can think of where the mold would have less volume itself than the object its a mold for.

And finally, why would the mold have to be less strong than the object? I should think the mold would have to be stronger than the original piece, no? The mold must support not only the original object (out of whatever material is being molded) but also its own weight and structure.

[LambdaFF]

... Who says they intend to make only 1 object ?
In that case a mold is a smart move.