Different layer setting per object and Array creation

I am new to SLA technology, but so far i have been blow away with SLA part quality versus FDM technology.
i would like to see some similar software features used on current FDM software especially Simplify 3D

1- it would be nice having to assign different layer thickness for different objects, this way it can dramatically reduce printing time when parts are bundled together, i understand the layer thickness has to be a multiple of each other, in Formlabs case it is already the case since the resolution is at 25, 50 and 100 micron, i don’t see why the laser couldn’t do 4 cross section for the 25 micron version for 1 section of the 100 micron cross section

2- i found it a little bit tedious to use the machine for a production environment since an array feature is not available, i understand i can make copies but it is not the same as an array, obviously there are ways around it by making the array in CAD and export the whole thing as one piece. however i have to rely on the software suggestions for 1 piece and try to copy it in the CAD before creating the array, so far it works but takes too much time.
it would be nice to create an array from a seed part, when it gets re-oriented all the array follows it

i got the machine last week, for me as an engineer it is absolutely incredible, i was able to create miniature snaps and almost use the parts for low production quantities, something i could never do with my FDM machine.

Great work Formlabs team, you created a great product.


As to Array… You can do arrays rather quickly by duplicating a part and then selecting both parts and duplicating them, and then selecting all four parts and duplicating them… you can duplicate 10, 20, 30 parts at a throw just by selecting them all.

It takes me about 40 seconds to cover the entire built surface with parts using this technique.

I wouldn’t call that taking a lot of time.

i agree with you regarding making copies rather quickly, however the parts placement are almost picked by the software at a random direction and distance, at least this is what it looks to me.
after the parts are created it might seem a small orientation tweak to the seed part might yield better fill result.
so now you have to manually move them around or re-do the copy all over again.
i understand this option makes sense to my application and may not be a main stream feature for the majority of users, i will be using the machine for low production volumes and i want the build plate to be as full as possible.

as mentioned earlier, the way i do it now and it works just fine
1- import a single part into PreForm
2- figure out the best orientation using PreForm
2- create a CAD array using virtual envelope identical to the machine’s (Solidworks )
3- try to get close orientation to the PreForm one (all parts of the array adjust automatically too),
4- adjust the array distance so all parts fit in the CAD virtual envelope
5- save the entire CAD file as one .STL file
6- import the array as is into PreForm and double check the fit
7- print

thank you for your comments.


1 Like

If I position one copy relative to the first, then duplicate it, the duplicate retains the spatial relation of the copied pair.

I then MOVE the duplicated pair to the relative position I want to the first pair. Select all and duplicate that group of 4.- Position that group of 4 relative to the prior 4… select 8- duplicate…
I quickly end up with a ROW of parts straight across the build platform along one side. Select all and duplicate that and they will paste right next to first row- keep hitting duplicate and each row will come in moved over until the platform is covered.

Again… it doesn’t take any serious time. given the wildly varying sizes of parts, orientations, and support structures… any one click ARRAY feature would either need to have a bunch of settings you had to monkey with to control the layout of the array… or it would just as often layout parts in a way you did not like and had to monkey with, anyway.

Oh- and in case you didn’t realize- one reason why I do not bother with creating arrays in a modeling app and then importing is because when I create an array of duplicates in Preform, I can select all of them and ROTATE them, or change their orientations, and they STAY in the ARRAY- each duplicate rotates around its OWN center.
I like being able to adjust the array like this to nuance peel forces or part interference.

thanks for information

That is far from always being the case :

Group management in Preform is still barebone, adding some feature would be of tremendous help in certain cases and I don’t see it being a big issue for Formlabs to code it in… it certainly is more of a manpower than a technical problem.

Another popular idea which has been suggested several time before is to add the ability to have variable layer thickness across a single part. Oftentimes you only need a .1mm layer height on most of the part but there is one part where some feature would resolve better at a lower layer height… a simple example is when printing a sphere, you’d need .25mm at the very bottom and top, .5mm right after that for a few layers and all the rest could be printer at .1, this would significantly reduce print time while still avoiding the aliasing issue when printing faces/surfaces that are normal to the Z axis.

i wasn’t suggesting it should not be improved. I was suggesting that it its not that time consuming to generate an array of smaller parts… And pointing out that any more capable ‘array’ would take just as much time setting up the parameters for an array- OR correcting an auto-generated array that did not lay out the parts as you think would be better.

For example- Auto-Orient - its pretty useless. YES it will automatically orient a piece… but it has never oriented a piece in the way that would give me the best result.
Since I stopped using it I have never once had a print failure, because I can always out think an algorithm on how best to orient each new and entirely unique print.

Given that I am waiting from 4 to 36 hours for a print that will use $200 per liter materials…its not that big a deal to invest 5 or 10 minutes in ensuring a good result.
but then… I am old- I am accustomed to software that didn’t do everything with a single click.

I don’t think varying the layer thickness for different parts in the same print will be all that useful. We already see layer artifacts when the machine pauses (thanks to the stupid resin cartridge removed errors).

These bands would likely show up in a print that mixed 25 and 100 micron layers in one print. The 100 micron print will “pause” for several layers while the 25 micron print catches up.

In the end you’re probably not going to play that much with layer height settings…

For very big parts, z-resolution is usually less important so they are printed at 0.1mm(and for many resins, 0.1mm is really nice, can’t compare with FDM). For smaller parts we always use 0.05mm. As prints take a long time anyway(on average more than a working day) they are printed over night so not losing time that way.

However, a setting that could vary layer thickness automatically within a single part would be cool. Many tall parts have long walls which can print at 0.1mm and some details that need 0.05mm. It could save on printing time. But I think it was patented by autodesk.

i have been using Simplify3D for couple of years now for my FDM machine, the variable layer thickness within an object is available along with assigning different layer thickness for different objects ( it even has a wizard tool).
i am showing an example of 2 spheres where the one in front is using .1 mm at the poles and .2 mm in the middle and the second one behind is using a .3 mm along the entire height.
i find this feature very useful especially when i print large fixtures where critical areas a very limited

regarding FredB comment “layer artifact” caused by relatively long stops between layers, it seems like an aesthetic issue, i wasn’t aware of it

regarding fantasy2, i would love to have an automatic layer adjustment as mentioned, the closest example i can see often is the mesh creation in FEA analysis, typically you set a general parameter and the software will automatically reduce the element size to accommodate for smaller features.

to be honest my new Formlab 2 blew my mind when it comes to small feature creation and surface finish, i hate to see it limited or lagging behind due to some relatively easy software modification.

again it just happened that i decided to use the machine for low production runs alongside prototyping as the first reason for purchase, these features are useful to me and maybe not for a typical user, in my opinion adding an advanced tab for what i suggested earlier is nice to have without sacrificing the simplicity of the current software.