Hi,
I want to share my last print. It took me about a week to print ( it’s composed of 22 parts printed a 100micron ). The entire print are about 50*40*35cm.
The ground was made with plasticine. I’m really happy with the quality.
I just have few warping parts, that i handle by bending them when i glued the parts together.
Amazing effort : )
It is brilliant! You’ve made a new record from Form1. Ihave two question:
1, What did you use to glue them together?
2, Did you plan to print them all straight up without using the auto orient to make sure they can match each other easier? I mean if print each part in different orientation will cause the deviation/wrap even more?
thank you 
Lock, i used exposy. I think it’s really better than superglue, because i have 5 min to try to fit the pieces and bend them together (not 20 secondes). And it’s progressive, i could bend and replace after 20min. and after 1h the bond are really strong. ( : http://www.cdecomania.com/FR/Photos/30601294%20Sader%20EPOXY%20Rapide%2025ml%20seringue1_1.jpg )
I never auto orient, i prefer to do it myself for few reasons. Because for complex geometries, the auto orient will never find the best orientation. I can save on time and resin and all the supports will be on the hidden side of my piece. But the auto orient can be used as a guide…
For this print i only had 1 fail part because of a wrong orientation ( but i learned from that )
Form what i see, this is not because you orient the part in a different way that will cause more warping.
Because all the part are differents you have to find the orientation that will cause less warping.
I saw that just position the part at an angle is not enough. It’s really important to have less curing layer from the bottom of the build platform to the top. ( because the force cause by the peeling process will be less important )
I made a quick drawing to show you what i’m talking. It’s avaible for all kind of print ( flat, round, complex geometrie)
Because i follow thoses rules i have less/none warping with my new prints.
not only is the sculpture amazing but your feedback on orientation is very informative and important in helping us to get good prints. I hope that Formlabs include this in their guide. Just to make it clearer, could you put the peel direction on your diagrams? The left and the middle ones could be construed as being the same but one is obviously good and the other bad.
Once again, brilliant and encouraging work.
Thanks Alan for your feedback. here is the new picto :
Hi Gilles, those diagrams are definitely money saver! The warping issue is the most problem on Form1 and for those who want to print big part, your experience is very helpful.
Heelo Gilles,
really complimets for your wonderful print. You set the bar so high!
Question: I see you have a thin and costant thickenss of the tree’s mesh. Did you hollowed by yourself? If yes, what’s your technique?
For complex triangulated meshes, I use ZBrush: I duplicate the object, recalculate it as Dynamesh and reduce the traingles amount, smooth it and apply an inflate modifier (inner direction). Then I smooth again and recalculate a Dynamesh. Finally I subtract with boolean operation the inner voulme to object mesh. And subtract some cylinder in stratic points to create connection holes outside/inside and make the air and resin flowing during printing.
This way is little tedious and can’t allow me to keep the control over the inner mesh at 100%.
So I’m searching better way to hollow a trianguklated complex mesh (that have undercuts and complex volume).
Thanks for any advice,
Mattia
Thanks Mattia !
For hollowing meshes i tried different techniques. And as you said i didn’t use zbrush because it’s too tedious.
My first bet was to give a try to Meshup because you have a feature that allow “volume shelling”. That mean that it’s not a simple extrude. The shell take in account the thickness of the mesh and will fill certain area if it’s really thin, instead of just pull the polygon in/out. But the software are still in development so i searched for an other solution.
After few tests i now use Meshmixer. It doesn’t have the volume shelling but it’s free, really fast and accurate.
I have the same steps as you : do an extrude, smooth the inside, reduce the poly. With meshmixer you can put an exact value for the hollow ( i always put 2mm because i find it’s a good compromise between rigidity and the possibility to bend it ).
And you have features to select the inner mesh and quickly erase/smooth the excess in order to have the same result as “volume shelling”.
It took me less than 1min for hollowing each mesh of my tree with import/export.
I really encourage to test it
Hello Gilles-Alexandre,
Many thanks for the suggestion. I’m going to test Meshmixer software.
Best regards,
Mattia
Wow, this is beautiful! Really excellent work.
The patchwork dividing and gluing work is really nice. The final piece looks like 1 print-- but that last photo shows how many pieces went into it and how big it really is. Wow! Did you model in any alignment or snap features to help you fit it together?
Also, I agree that 2mm is the best default wall thickness for shelled objects.
Also, I just realized this is the first single part printed which is bigger than the Form 1 itself in every dimension! That’s really awesome.
Can you post a photo of the tree next to the printer to really show off the scale?
Thanks Jason !
I sculpt the entire mesh within zbrush. I used maya to project curves on it and split my model
That way i was sure the parts will be perfectly aligned.
I also made a template of the build volume to check if all the parts will fit into the build volume of the form1.
The form1 is 10cm higher than my print, but yes my model is much larger and longer.
That’s awesome! It’s like “printing outside of the box”.
Hello Gilles and thank you for your thoughts on how to better orient models. Which in short is meaning ‘Do not overhang toward peeling side’. Correct? I don’t understand your wording explanation on the other hand.
But on your message with four pictures it seems pic 1 and pic 3 are showing the same case but first is marked as right and third as wrong. May you please comment on this?
I’d like to second Andrey’s request. This looks amazing and I’m more interested in your tips that you have learned as i am about to attempt the same thing using zbrush and “cube slicing” the mesh to print in multiple parts. I’m not sure I understand your tip about the peeling and orientation but it sounds important so I’m wondering if you could clarify it some more.
I’m sorry if i wasn’t enough clear. I made it really quickly… and I should have separate the drawings.
Like you said, Pic 1 and 2 show the problem off putting the higher supports toward the peeling side. That will lead to more failures because the peel process are strong and the supports will break/ shift easily.
At the begining i didn’t wanna make those 2 drawings because if i remember, this has already been explained by some other users on this forum.
The main thing i wanna explain are on drawings 3 and 4 ( just considers it as a separate issue )
I’m not a native english speaker but i will try to explain it differently.
It’s better to reduce the peel effort on the print by having a smallest section toward the build platform. The suction will be less important and so the warping also !
Here is my question to formlabs : why did you make the resin so viscous ? I think all of those problems are because of that. A more liquid resin will solve all the problems…
I agree with Gilles-alexandre regarding the less viscous resin. However, there is a trade-off with a more viscous resin. The pigment will settle during long prints and you will get even more failures. For short prints (3-5 hours) it will be great. A clear less viscous resin with enough photo-blockers to achieve high detail and resolution and direct-castable will be more than welcome though.
I meant there is a trade-off with a LESS viscous resin. Sorry.
