Simplify3D User Forum

How is in the name of Ritchie and Kernighan this bridging supposed to work?

Is anyone even looking at these bug reports? The bridging bugs are starting to really prevent Simplify3D from being usable

rotating and moving object around does not help ..
I even tried setting 100% infill and no perimeters.. this is even more ridiculous

here's the original STL if someone want to give it a go and use it as test case to fix a bug

In the name of Ritchie and Kernighan,
turn the box 15 ° around the Z-axis and use support, 1 mm columnes, no dense, no extra inflation, only 2 upper and 2 lower separation layers , horizontal distance ~ 0,8 - 1,0 mm

3 - 5 top- and bottom layers,
2 -5 perimeters.


Regards
horst.w, GER

Hi,

Rotating don't help much (I tried 20, 15, 40, 45degrees and number of others trying to make it work ok..)

With 2-3 perimeters it don't work (as you can see in the pictures I uploaded), looks the same at 15 degrees too
Top/Bottom layers don't affect it at all.

With 4+ perimeters it calculates the bridge in right direction but it's not usable since it prints "2 lines of bridge" and rest it prints as perimeters and not as bridge and those are printed at "wrong settings" so of course it does not work .. For some reason (in my opinion it's a serious bug) Simplify3D prints perimeters around bridge and, of course, they don't work

I had to get Slic3r back up to print this part .. not a big issue since the top part I planned to print with Slic3r anyhow (prints under 10min with Slic3r, over an hour with Simplify3D) but ... yeah, never mind, it's a bug, a serious one, easy to reproduce, so if anyone is looking at the bugs section and dev/sustaining team still exist I'm sure fixing this bug would do more for Simplify3D userts then adding a 100 new obscure printers and 20 new languages

horst.w wrote:use support

The whole point is to not use support and to do this as a bridge (Slic3r manages to do this properly, Simplify3D was doing it just fine on previous 10 versions of the object, this version is slightly different dimensions, same shape, and then the bug hits and the wrong bridging is calculated, when I change dimensions again Simplify3D again works ok, problem is this is the dimensions I need). Support works with Simplify3D, bridging is riddled with bugs. This time I need bridging, not support.
p.s. note that I'm mentioning Slic3r here just as "other slicer that is rendering this properly" so example that it's something that can be solved... not as "Slic3r is better then Simplify3D" as if I think that I would not be trying to do it with Simplify3D in the first place

p.p.s. I found it weird that both slicers that have "good support system" (Simplify3D and Craftware) both have problem with bridging

Do not use their names in vain!

upsm wrote:

horst.w wrote:use support

The whole point is to not use support ....

Bridging is one of the poorest function of S3D and improvement thousend times requested.

Live with it or help yourself. And that is using support, I can't see any other solution. than Cura, Slic3r or others which can do it better.


H.

Yep, whether bridging in S3D works is rather random. It seems to me the main problem is miscalculation of the bridging direction, as demonstrated also in this thread. The other main issue is the lack of observing the set bridging speed for the perimeters of the bridge. We can only hope that this will be addressed in the next version of S3D, which we all hope will be here before long. At least it should be, judging from the previous release shedule.

Looking more closely at your STL file, I think the issue is that there actually isn't anything to even bridge between! If you look at the top-down view, the region where you are doing the infill has absolutely zero support below it. There's no foundation to even bridge between. With that being the case, I would actually PREFER that it goes in this shorter direction, since that is much more likely to succeed if there is absolutely nothing below it to support it. So I for one prefer the way it works in your original screenshot, and I'm honestly not sure why the software would ever choose to go the direction you are showing, since again, there's nothing below it for those long extrusions to grab onto.

If you want it to go that direction, reduce the number of perimeter outlines for that layer to either 1 or 2, and I bet it will work (since now there will actually be a ledge for the infill to bridge between)

horst.w wrote: Bridging is one of the poorest function of S3D and improvement thousend times requested.

Live with it or help yourself.

I help myself using Slic3r and I try to help us all by posting a "reproducible test case". Fixing a bug when you have a reproducible test case is usually simple; always lot simpler then when you don't have a reproducible test case.

brian442 wrote:Looking more closely at your STL file, I think the issue is that there actually isn't anything to even bridge between!

If you look at the code produced by the slic3r you see that it can be done. Also I can snap a picture of the printed part that came out perfectly (and it's ABS so not a material as easy to bridge as PLA) so you can see no loose wires at all - perfect strong bridge.

Also, I printed few versions of the same object with slightly different overall dimensions (height and width) but with identical design and walls with identical width and that "ledge" with identical size and Simplify3D sliced it properly and printed "almost ok". "Almost ok" because it prints perimeters as perimeters and not as bridge so there were 2 loose wires (perimeters) on the bridge but everything else was perfect. The way it calculates it now is impossible to print.

brian442 wrote: If you look at the top-down view, the region where you are doing the infill has absolutely zero support below it.

That's the point of bridging - bridging a gap - no support underneath. There are 2 parallel walls between whom you can print a proper bridge. It would be better if those walls are thicker then 1.8mm but it is enough. It was enough for the same part for Simplify3D with slightly changed object (the bridging part was not touched) and it's enough for Slic3r. Bridging is a real thing, it works, and it's often required for a proper print. Support is good to have, also often required, but in this case bridging is a better option. Adding support underneath it to solve Simplify3D bug is just not solution.

brian442 wrote: There's no foundation to even bridge between.

?! there is a wall on one side, and wall on other side. A classic bridge, you anchor line on one side, run a line from one side to the other and anchor line on other side. Just like Slic3r did, just like Simplify3D did for that same object if you change things "a bit". What "foundations" are you talking about. This is a clear bug in Simplify3D, nothing else.

brian442 wrote: With that being the case, I would actually PREFER that it goes in this shorter direction, since that is much more likely to succeed if there is absolutely nothing below it to support it.

How can it succeed ?! you go from edge to "nowhere" and back and except filament to? levitate?
You think you are doing a short bridge between wall and that hanging perimeter? That's the core of the bug. Simplify3D don't understand that perimeter is a bridge too and in this case it printed perimeter across the walls. This perimeter will fall down as it's not printed using bridging settings. Then Simplify3D calculates the bridge and sees that perimeter (that already fell down) as "solid edge" to attach bridge to and it bridges using wrong direction. So the core bug for most bridging issues I seen is that Simplify3D prints perimeter on the bridge as perimeter and not bridge and then calculates that the perimeter printed is "solid" that bridge can attach to. Properly printing the perimeter on the bridge would most likely solve most of the bridging issues of Simplify3D

brian442 wrote: So I for one prefer the way it works in your original screenshot, and I'm honestly not sure why the software would ever choose to go the direction you are showing, since again, there's nothing below it for those long extrusions to grab onto.

what is it grabbing on to when doing short extrusion?