Annealing Compensation
Hi there,
I was watching a video on annealing PLA prints, for those who don't know, annealing is basically heat treating / cooling something in order to stengthen the part . . . . this can be done in a standard domestic oven.
The benefits with annealing PLA are an increase in strength by as much as 40%, and an increase in stiffness of around 25% (after 1 hour at 110°C).
The downside is that the process causes the treated part to shrink and/or expand (as should be expected when you put plastic in an oven!), typically the PLA part will shrink ~5% in the XY axis and expand 2% in the Y axis . . so not great if you need dimensional stability.
So, my idea for an improvement would be to include annealing compensation in the Prusa slicer, if you need the extra rigidity and strength and are going to be throwing the parts into the oven (obviously for a fixed time / temperature) and you need your 140mm part to be 140mm long and not 133mm . . . then you would select 'annealing compensation' in a menu (or whatever) and set it to PLA (or whatever you are using), maybe even specify the brand and other specifics that might influence the result, and the slicer would scale the part accordingly.
So, if you know your brand of PLA shrinks ~5% in the X/Y axis and expands 2% in the Y axis when annealed (at a known time / temperature), then the slicer would scale your part up by X/Y 105.26% and Y 102% to compensate.
Of course all this can be done manually, it's very straightforward maths - once you've done a few tests to find out the shrinkage/expansion of a particular brand of PLA or type of filament, but one advantage to having annealing compensation built into the slicer would be that it would (hopefully) encourage users to add their own profiles for various filaments, which everyone could benefit from, it'd be great to pull up a profile for - for example - Matterhackers Green PLA - and without having to run your own tests, have the slicer subtly scale your parts to compensate for annealing.
Let me know what you think !
(warning, never place PLA parts directly on top of roast chicken, remove roast chicken from the oven first).
. . . . . . . . . . . .
The video I was referring to:
Re: Annealing Compensation
You can scale your part with a different number for each axis.
Something more is needed?
Re: Annealing Compensation
I don't think there's any way they could ever come up with one value for PLA. Everything from your bed temp, ambient temp, and fan speeds will affect how much and how fast the part will shrink. Even the orientation you bake it at (or even if you're boiling instead of baking) will change those %s.
I think you'd be much better off doing your own tests, and scaling each axis accordingly. Or stocking up on drillbits and sandpaper.
Re: Annealing Compensation
I don't think there's any way they could ever come up with one value for PLA. Everything from your bed temp, ambient temp, and fan speeds will affect how much and how fast the part will shrink.
Yes, that's kind of the idea - to take as many variables into account as is viable, so if you load up X brand PLA, with a certain bed temp, ambient temp and fan speed, the calculation would take all that into account, as well as the target annealing time/temp, I'm not saying this would be - or needs to be - an absolute guarantee of dimensional accuracy, but it would be great if a system like this put you in the right ball-park, you could then make small scaling adjustments if needed.
Re: Annealing Compensation
I don't think there's any way they could ever come up with one value for PLA. Everything from your bed temp, ambient temp, and fan speeds will affect how much and how fast the part will shrink. Even the orientation you bake it at (or even if you're boiling instead of baking) will change those %s.
I think you'd be much better off doing your own tests, and scaling each axis accordingly. Or stocking up on drillbits and sandpaper.
Yes, that's kind of the idea - to take as many variables into account as is viable, so if you load up X brand PLA, with a certain bed temp, ambient temp and fan speed, the calculation would take all that into account, as well as the target annealing time/temp, I'm not saying this would be - or needs to be - an absolute guarantee of dimensional accuracy, but it would be great if a system like this put you in the right ball-park, you could then make small scaling adjustments if needed.