Wednesday, October 17, 2007


The long and the short and the tall

I have been experimenting extruding different aspect ratio shapes with HDPE :-

Hopefully you should be able to see a general improvement from left to right. My conclusion is that HDPE is not really suitable for extruded layered manufacture because of its high melting point and mold shrinkage.

More details in my blog: scaling-new-heights


I am in the process of ordering 100 Kg of polycaprolactone for the RepRap shop. The lower melting point and reduced shrinkage should get rid of your (excuse the phrase) curly bottom problem. But let me not detract - those pictures are most impressive.
Is there going to be a UK distribution point for this as shipping heavy items for the US is not economical?
i dunno... those look like a good start. if we could find a way to compensate for the shrinkage in software, then HDPE starts to look alot better.
Damn! That looks familiar! Great work Nop! :-D
I can't see how the shrinkage can be compensated for in software. Sure, if it all shrinks together then you just have to make it a bit over sized. The problem is that it is shrinking while you build it so each layer puts the squeeze on the layers below leading to curling.

Maybe we could calculate the resulting distortion and warp the model in the inverse way so that after it curls it is the right shape but I am not sure whether that is even theoretically possible, let alone practical.
Maybe take reprap to the sauna? Don´t know is the steppers are going to like 80 degC on the long run, but it will sure help against the "curly bottom problem". You can lower the extrusion temp, too, then, so the last layer is solid before you print the next one on it.
If we have to create a heated environment FDM machine it's pretty much "game over" insofar as anybody being able to afford to run such a system in their home. It would be like paying your electric bill when you were baking bread 24x7.
It´s not that bad, I think.
If we find out 60 degC is warm enough, thats just twice the difference in temperature to my room as from my kitchen to the inside of the fridge. Cooling is much less efficient than heating, so it wouldn´t be more expensive than having a second fridge at home. And the reprap doesn´t run 24/7. All you need is good insulation.
Anybody got an old fridge to spare?
I'm trying to set up a UK & Europe distribution for RepRap parts to get over the shipping problems
Great idea, Ian!
One could implement a neural net then extrude 40-50 standard pieces as a training set, comparing each to the ideal and back-propagating the errors until the warping diminishes to an acceptable level. Using a genetic algorithm would entail way too many test extrusions although one could be used to optimize and smooth the extruder head path.
Don't the professionals use support material to prevent warping like this? I seem to recall they tend to lay down thick, buttress-like layers of support material that end up completely enclosing the final product in a much larger block, which can then be dissolved away or otherwise disintegrated. I've been reading up on this a bit; apparently a mixture of 75% plaster of paris with 25% potato starch makes a decent temporary material that breaks up in water for other moulding and forming jobs; the very rapid set time might be a problem, however. Perhaps just ordinary clay could be extruded and allowed to dry out, then dissolved in water and reused?
Post a Comment

Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]