How does your support generation strategy change when using CPE vs ABS?

[mattgriffin 28]

Hey folks!

Running a few experiments with CPE in various colors, attempting to determine between CPE and the new Ultimaker ABS formulation, which is better for quick fit-test prototypes of mechanical parts.

My theory was that I would like CPE better, despite the lower warp ABS formulation. However, while testing i've noticed that the support structure generated in Cura 2.1.2 for CPE doesn't seem to behave like support structure for ABS. And I realized that I have no idea how to think about support differently for polyester materials. I suspect that the fact that most PLA and ABS materials have some sorta silica elements as filler materials -- which makes it easy to snap them easily if you score them. But the PET family materials tend to stretch before the break instead.

Have any of you come up with some adjustment to support settings when using CPE that really makes you happy? The material itself, I really really love when I don't have tough support removal issues.

[mattgriffin 28]

So far, the suggestions I'm receiving offline (and elsewhere in the forums) include dropping fan speed to 50%, increasing the Support Z-Dimension gap a bit, and thinning out the top skin a touch. I have tried a few configurations -- would love to hear mastery from other users of CPE or other co-polyester variants.!

[Labern 775]

Hey Matt

Good to see you on the forum

If I have to use support then I usually print with the .25 nozzle.

I normally print a .1 layers with a .4 nozzle, so if you print with .1 layers with the .25 nozzle the increase in print time is only on flat surfaces. So as long as there are no large flat surfaces then the extra print time is minimal but the support comes off so much easier. I then leave the support settings on default.

Edited July 21, 2016 by Guest

[mattgriffin 28]

Oh, this is handy! And you are loving this for CPE? I can imagine this strategy working great for PLA because you can snap and break away these tendrils so much more easily. But I have to admit to not printing much CPE with a 0.25mm nozzle.

Do you have a go-to CPE mechanical demo print that comes to mind that I should try your method for? While I wait for your reply I'll pick an old standby calibration object, but I'll bet you have some favorites. ;-)

[Labern 775]

With PLA and 80% support roof this method makes for very smooth overhangs.

I have a few objects but I don't have the files with me and it will be 3-4 days before I could dig them out. Still traveling.

[Labern 775]

 

HERE: is the one that I first learnt this technique and was impressed with its overall results.

[mattgriffin 28]

Heh, perfect. I had that in my queue to print, so I'll give that a go!

[Labern 775]

It was in PLA not CPE thought. But I found that it still works well for CPE but I normally don't use support roof.

[mattgriffin 28]

Ah, so you would say that Support Roof is much more a PLA strategy than a CPE (or ABS) one, huh. I just saw your note and went and canceled my test print. ;-)

[Labern 775]

I just found some of the rings of roof support stick with CPE and have to be cut off, where Pla they just peel away.

But this may just be a temp thing on my part.

Maybe one of the pieces for the cube puzzle would make a good test. They are small and you could rotate it in a way that requires support.

[mattgriffin 28]

Oh, that's a brilliant idea for support testing -- especially because watching for those corners to get creamed top or bottom from support at too great a gap should be a great test.

I grabbed Part C from the Half Hour Cube from Michael QueallyMichael Queally here. I rotated over to be all about overhang, and I'm going to use this to explore the various settings with the various materials and see how that goes. This is a great suggestion, Labern, because the total height of the model is pretty much only pertinent to the support test.

My goal will be to find a solution with the minimal amount of support, that is easy to remove, and doesn't leave those corners to droop out of shape. And then when I think I'm close I'll switch to a "drunken" UltiBot (tipped on its edge) and start running mechanical pieces that are are critical in more than one axis. Would love a suggestion for the last one ... if you happen to have some really small mechanical projects that work in more than one axis.

[ultiarjan 1,223]

My gut feeling says a good test piece should have mid air angels at various angels AND a horizontal area..... should be only a few minutes to draw....

[mattgriffin 28]

Okay, project underway here....

I'm using this piece as my publicly hosted lab book. I'll check in and run each test when I have machine (and time) to do so. Which won't be as often as I'd like, but I'll be able to refer back to photos and resource there -- and over time get some science done. ;-)