Automated Supports not consistant

#1

Using a customized automatic support generation to support flat surfaces. (paralel to baseplate).

On parts with a relativly large surface area, the dense supportgrid is generated correctly.
On smaller flat surfaces almost nothing is generated, with exactly the same automated support settings.

To illustrate the problem I joined a screengrab: I just X-scaled the object x2.2times in the Xdirection, so it is basically the same object, but with a larger surface.
I think i tried all possible automatic support parameters, with no difference.

However, if I go to “Edit Critical features” and select the flat surface (turns yellow) and ask to resample the selection, it indicates the correct grid positions for supports. Sadely, when hitting “apply” , no supports are generated…!

supports01
supports01.jpg1145x752 317 KB

editcriticalfeatures
editcriticalfeatures.jpg664x660 128 KB

#2

Hi,

Interesting. Are you able to email us the part at our info@ address?
And what version fo the software are you using?

kind regards
Elco

#3

Are the z height locations of the parts similar?

#4

Thanks Elco, for fast response !
We’re using v1.1.4.5

I’ve asked permission to send the part, so expect it in your info box soon…

Just to clarify: both objects are the same and at the same Z height. The only difference is the scaling in X-axis.

I bumped into this strange problem because I previously had adjusted the settings for other samples .
All had closed bottoms and all worked fine, until I started loading parts that were hollow. (made in designer software, not the hollow feature from Formware).
Then my settings refused to work properly and couldn’t figure out why… Slanting them 1° helped a bit but did leave the starting edge bare.

I personally am well skilled in 3dprinting to complete and make it work manually, however i’m setting up the system for non-skilled people to use, hence my focus on the settings with automated supports…

#5

Hi,

When I open the part and run auto supporter (version 1145) the result looks like this:

image
image.png1610x595 190 KB

So i’m suspecting some kind of parameter combination or floating point error that produces this.

  • What support settings are you using? (you can export the support settings in the machine config)
  • Do the supports get generated if you move the just part vertically? Say +25mm (unrealistically high)

kind regards
Elco

#6

The fact they are generated if you do it manually on the surface means the math logic there is OK.

They are just filtered out in a later stage. The question is which parameter combination causes this…
I think for some reason they are filtered out

The single support type has a ‘cap’ build in that if it gets generated below the build table when the first beam will get to long; it will lift itself up. And make itself shorter. So that is happening in this case, and working correctly.

  • does it happen when you increase or decrease the density?
  • and when you just copy the part?
#7

Elco,

Changing height does not make any difference…

diff_height
diff_height.jpg679x556 112 KB

I’ll export the support settings and send it to your info@

A higher density (175%) does result in a slight increase of supports, but by no way what it supposed to be…

higher%20density
higher density.jpg679x588 121 KB

Copying the part results in exactly the same supports…so no difference there…

#8

Hi,

So after some investigation; this is indeed a bug caused by parameter combination. Thanks a lot for reporting it.
I’ll correct it right away for the next release.

If you set the first beam length very short (1 mm in this scenario)
A check of intersection of the first beam with the part fails. (because it flips a normal, and thus finds an intersection inwards of the model).

So what parameter to use to avoid it for now:

First beam length should be larger than 2x the pole diameter.

In your example the pole diameter was 1.2
The first beam length (only) 1.0.
So if you make your first beam length 2.5 it generates.

There is no correlation to the scaling of the model. I have the same on the scaled model.
It could be the support profile you generated that with was different. Not sure. But this is definitely it.

kind regards
Elco

#9

Tested and positively confirmed !
Setting ‘first beam’ to 2.5x the pole diameter solved the missing supports!

Thanks for the help!

Although i’m experienced in 3dprinting, I am new to this software, i’m still discovering the more intricate details and trying to understand the parameters and their usage.