More clever supports idea


No it would not make the struts like springs at all. The laser determines the strut NOT the position of the tank. The tank just lets the laser go to the resin. Its position is completely irrelevant.


You are right !
I never thought of that. It makes complete sense.
Just a software trick.
Not sure you can offset the tank along 2 axii though…


So far as I can see its currently limited in the Wiper Axis. Don’t know what axis this is but indeed it would just be software. Just offset each layer by 0.1mm or such and increase the lifetime by a tenfold or something


Alright I didn’t get what you were saying at first. That’s a clever idea !

One thing to note though : the tank isn’t flat, it’s slightly concave due to surface tension and capillarity when the PDMS is poured. Also the linear axis on which the tank moves may also not be perfectly level and I’m pretty confident they rely on the fact that the rails are calibrated to that the tank is as flat as possible in one position.

Also as Olivier wrote the tank can only be moved along one axis not two.

I’m not sure I would use this feature if it were available, I have enough trouble fine-tuning the Z offset so that prints are easy to remove but still stick when I use the center portion of the tank, and that’s without speaking about the build plates which are also different from one another. Adding a seemingly random position of the print relative to the tank seems like a recipe for unreliability. I would rather have a reliable print experience and change the tanks, that plus the fact that the LT tanks are very resistant to clouding…


What a great idea. I’m a E.E. but I get “Space Frames”. I suspect that one issue may be peel resistance of the supports before they are securely woven together? Once complete and tied together it shouldn’t be an issue. I’m very glad that I don’t have to develop the algorithms to build them!


Pretty sure the peel force works out to be a vector that is roughly 45 degrees from vertical.
At least that is what I recall from discussions about it when the F2 launched.


from back in the day…


Hello Christopher,

You might be interested by this approach which is a geometry-based, structural form finding algorithm.



Really wonderful stuff! Thanks for sharing that.


What you are saying makes a lot of sense from an engineering point of view.

But, I doubt that the bean counters at Formlabs would ever want this, after all it would save resin and help the PDMS in the tank last longer. Unfortunately that would lose them revenue over a period of time.

Lets be fair we can reduce the standard base thickness from 2mm to 1mm with absolutely no adverse affect, other than (from Formlabs point of view) it uses less resin per print, yet what is the default - (2mm).?


You have a point here.
Maybe when Formlabs has a direct competitor, this would change.
Regarding the base thickness, I don’t think it would be wise to make it too thin because it would lose stiffness and crack when trying to peel it off the aluminium base.
Yet, it might be interesting to investigate an isogrid design…