Shrinkage

[Synchron]

Hi,

i thought it would be better to create a new topic instead of discussing that in the "Bigest print to date - update #54"-Topic

What i understood so far (not much ):
- The Resin has a shrinkage of 8%.
- Best results if the object is symetric in x/y

Open Questions:
- mike_bidell asked: Is the shinkage of a resin deterministic and repeatable? could this be automaticly considered by the print-software?
- When does the shrink process take place? in the moment where the resin cures completely/the laser hits the resin?
- What about filling a model up with colored resin. Would it shrink inside and be free, or would it stick to the shell?
- Does the z axis shrink as well as x/y?

Things to test:
- Objects in multiple Positions to see how the position changes the result

Thanks for your great work!
Synchron

[Chayat]

Is the shrinkage uniform or does it introduce noise onto an otherwise flat area? If it is uniform and can be predicted then maybe this is a way to hit a higher level of detail?

[rylangrayston]

Hey Synchron thanks for stating and new thread! That should probably happen alot more often around here
and gathering up all the questions! nicely done.

First thing I have to say is that whatever I have to say about shrinkage is unfounded .. I have not done shrinkage specific testing yet.
So word of warning every answer you are about to read is dangerously intuitional and plausibly incorrect.


- mike_bidell asked: Is the shinkage of a resin deterministic and repeatable? could this be automaticly considered by the print-software?

I think it is repeatable esp in smaller prints, and could be acounted for in software for smaller prints without much effort.

- When does the shrink process take place? in the moment where the resin cures completely/the laser hits the resin?

I think shrinkage takes place at the very moment that light hardens the resin. having said that you can definitely set your printer up to print soft partially
cured prints, and in that case I would expect more shrinking to occur if you were to place a soft print in the sun. Finishing the curing process after printing
could be a way to lessen the effect of shrinkage during printing.

- What about filling a model up with coloured resin. Would it shrink inside and be free, or would it stick to the shell?

Ive done this with stiff resin only, with our current formula it sticks to the shell so well you cannot tell there ever was a shell, even if you pick the print apart with a knife. Also of note is that if you do fill up a print, it needs to be cured slowly, otherwise little cracks can sprout from the uneven internal stresses of shrinkage. Putting a resin filled print in direct sunlight can make it behave a bit like a prince Rupert's drop..... so if your going to do that be safe and where safely glasses.

- Does the z axis shrink as well as x/y?

This is my favourite thing about this problem..... I think the resin dose shrink in all directions. however if it shrinks in the z then the print will become shorter and submerge its self in the resin by the amount it shrinks. Any layer that shrinks in the Z will Immediately cause the layer to be printed above it to be slightly thicker. So I if shrinkage occurs as quickly as curing, then I theorize it will govern its self to have practically no effect in the Z.


-Is the shrinkage uniform or does it introduce noise onto an otherwise flat area? If it is uniform and can be predicted then maybe this is a way to hit a higher level of detail?

I think it is fairly uniform on small aria or perhaps on a large honey comb aria. however if you were to print a single wall that was a foot long, I wory that shrinkage would make the wall go wavy in an unpredictable way. As for using it for detail.. I don't know enough to comment.

Quite some time ago I thought up a solution to this and was chatting to someone who had already implemented the same solution!
Im going to search my email so I can give credit to the right guy..... searching bannanas...... ....

Ah It was Pete!
Pete im sure you wont mind me quoting you in explaining your solution:
Pete wrote-
"
I have a little experience with FDM in this respect, when you print a bead of hot stuff and it shrinks you end up with a big banana, I have a dirty script I wrote somewhere which altered GCode for the 3D printer we have at work it looked for long lines on some layers and when it found one broke it up into short sections with a micro break between each one, this then creates lots of micro bananas which basically relieve tension in the print and you end up with a much better print. I can see exactly how you would hit the same problem.
"

Perhaps Pete will share his script with us some day, since it alters Gcode, It should play very nicely with the peachy printer software.

Here is an animation I did quite some time ago, I think is shows how Petes script worked.
Or perhaps it works by stacking the micro bananas like bricks?

[MemorianX]

if we asume shrinking is instant that would mean that any figure you draw in x/y would become smaller and the path that laser took would be free space (partially) what would happen if you redrew the layer essentially making a thicker wall by putting new cured resin in that free space left behind by the first run.
Would this lead to a final product closer to the desired output ust with a slightly thicker wall and higher material usage?

[mike_biddell]

As Rylan expects (and I think he is probably right) that the Z shrinkage is self-compensating, just print x and y 8% larger from the outset.

[rylangrayston]

if we asume shrinking is instant that would mean that any figure you draw in x/y would become smaller and the path that laser took would be free space (partially) what would happen if you redrew the layer essentially making a thicker wall by putting new cured resin in that free space left behind by the first run.
Would this lead to a final product closer to the desired output ust with a slightly thicker wall and higher material usage?

Good point!
yes I agree that this is very plausible, In a way the x and y could be somewhat self compensating for the outer dimensions of a symmetrical print.
I would think the thickening would occur toward the centre.
I did a supper ruff animation of your thought MemorianX