Minimum Support Distance?  

RE: Minimum Support Distance?

If you are looking at the preview, it is only an estimate of what really happens. Try printing it.

Regardless, if you feel there is a defect in Prusa Slicer, document the issue very clearly so the devs can reliably repeat the issue you are seeing, and then post the issue on Github.  Dev's don't generally watch the peer to peer forum looking for defect reports.

And, your post of the part never made it: you need to place parts files in a ZIP to post.

RE: Minimum Support Distance?

https://github.com/prusa3d/PrusaSlicer/issues/2620. From three weeks ago

Those are the coordinates that the printer uses. It's not an estimate. I'm not trying to see "how close I can set it", I'm trying to get good support. Before my MMU arrived I did a Lot of experimenting, with posts and pictures. Everything was ugly.

I'd just like the option to get slow, close printing on difficult supports.

RE: Minimum Support Distance?

https://forum.prusa3d.com/forum/original-prusa-i3-mk3s-mk3-how-do-i-print-this-printing-help/rounded-overhangs/

More, from when I thought really making an effort to present my case would help.

RE: Minimum Support Distance?

Reading between the lines and skipping all the exciting engineering work, I'm understanding that the goal is to print 3mm spheres with good undersides. Is that correct? If so, I've yet to see good results using current consumer-grade FFF printing with Cura, Simplify 3D, Slic3r, KISSlicer or PrusaSlicer. I certainly haven't done exhaustive testing, but I've done some and followed several discussions in support groups for each slicer, each reaching the conclusion that supported surfaces will look poor due to lack of anything the adhere to immediately underneath, yielding some unavoidable degree of stringiness.

All run into the issue of support gaps being large enough to facilitate removal requiring the supported surface to literally print on air.

Someone posted the results of using dissolvabe supports with 0 Z gap and the undersides were still quite rough, and certainly not pleasing for a 3mm part. It's always a balancing act. I understand the issue you're describing but I have to think it's been looked at before and deemed not worth the effort for the quality of result. A different technology is probably a better choice for delicate parts such as this case. Much like we have to live with Z seams now, supports suck for finish with the current tech. 

RE: Minimum Support Distance?

In your post above you are blaming supports for lack of perimeters and under-extrusion.  You see underlying layer infill because the perimeter count is too low to fully "cover" the exposed region. Plus, the layers are not being filled as the should.

Ideally, the best you can expect when perimeters don't "hide" infill is what a good top surface looks like. If you inspect the right leg and nose, you can see infill exposed by too few perimeters.

And a better example of perimeters not hiding infill:

RE: Minimum Support Distance?

But that said - FDM printing has limits. Here's a head I was printing. Getting the underside curves to print properly was a headache: but the issue is not unexpected since some angles simply exceed the printers abilities to lay filament on prior layers and it droops uncontrollably. The only option in those cases is soluble supports and 100% contact.

RE: Minimum Support Distance?

Posted by: bobstro

Reading between the lines and skipping all the exciting engineering work, I'm understanding that the goal is to print 3mm spheres with good undersides. Is that correct? If so, I've yet to see good results using current consumer-grade FFF printing with Cura, Simplify 3D, Slic3r, KISSlicer or PrusaSlicer. 

All run into the issue of support gaps being large enough to facilitate removal requiring the supported surface to literally print on air.

Someone posted the results of using dissolvabe supports with 0 Z gap and the undersides were still quite rough, and certainly not pleasing for a 3mm part.

Posted by: bobstro

Someone posted the results of using dissolvabe supports with 0 Z gap and the undersides were still quite rough, and certainly not pleasing for a 3mm part.

I don't know if that was me, but I've gotten some VERY nice results with TPU and PVA, using zero support distance. Little, wonderful spheres and clean, arbitrary angle overhangs.

I run up against a 4 color limit, and would like to get the process down.

The reason I bring it up isn't just random complaining - the supports ARE done right without "detect bridging perimeters" on. 

What I'm asking for is if the supports can be treated the same as when Detect Bridging is turned off.

"Don't expect too much" won't advance anything. I see a solvable problem and I've not had someone explain why this leads-to-defects behavior is intentionally left in.

Thankfully the stuff I'm printing doesn't require 'bridging perimeters', so turning that off makes the support behave. It would be nice to have the "z-contact-distance" actually do what it says - even if you just change the tooltip to say "This is NOT the z-contact distance, but a factor in how it is calculated. To exactly set Z-contact distance, disable Detect Bridging Perimeters".

Is that really asking for too much?

RE: Minimum Support Distance?

Still can't get my head around why detecting bridging perimeters is or is not an issue?  Infill changes a bit, and the slice reduces speed and flow of perimeters (or not, depending on USER settings).  But the perimeters are still placed where they are supposed to be. Infill is still placed where it is supposed to be.  And since you can turn off Detect Bridging Perimeters ... 

You need to be able to explain precisely what the problem is - and so far, you haven't done that.  Neither in words nor examples.

Here's your own words:

The reason I bring it up isn't just random complaining - the supports ARE done right without "detect bridging perimeters" on. 

What I'm asking for is if the supports can be treated the same as when Detect Bridging is turned off.

Maybe I am especially dense today  -  but doesn't turning off the setting satisfy your request?  After all, supports will "support" what would have been bridges "right" with the setting off, which is what you're asking for...  

RE: Minimum Support Distance?

That's a reasonable point - I've been wondering it myself (it is easy to get on a path on forget why). Detect Bridging DOES effect wall and bottom quality, though mostly on genuine bridges. For the purely organic shapes I'm printing, it's less of an issue. For items that have bridges you care about, it can be pretty annoying, the quality improvements with Detect Bridges on are quite noticeable... If I could use them in conjunction with supports I think I could get some really nice prints. 

With all that said, if the software incorrectly describes what it does (support distance only works when detect bridges is off), that's a BUG, and should be fixed even if the behavior doesn't change. Additionally, I think allowing support distance to be user selectable with the support distance variable is not unreasonable - indeed it is how the program works with the option off. Imagine if the left turn signal in your car only worked in 2nd gear, but not in 4th.

RE: Minimum Support Distance?

Posted by: abraham.m
I don't know if that was me, but I've gotten some VERY nice results with TPU and PVA, using zero support distance. Little, wonderful spheres and clean, arbitrary angle overhangs.

[...] The reason I bring it up isn't just random complaining - the supports ARE done right without "detect bridging perimeters" on. 

What I'm asking for is if the supports can be treated the same as when Detect Bridging is turned off.

"Don't expect too much" won't advance anything. I see a solvable problem and I've not had someone explain why this leads-to-defects behavior is intentionally left in.

Sorry, didn't mean to suggest you shouldn't ask. I was just pointing out that it was confusing to me as to what you were asking for. "How to print 3mm spheres?" versus "what is up with supports?".

If you have seen better results using your approach with other slicers or can demonstrate that it will be superior, I think you'll get a lot more traction with the Prusa developers. Reading through, I had the impression that you wanted specific changes without clearly demonstrating that they would provide benefit. Don't take this as argumentative. I'm just suggesting some points that might improve your case.

Thankfully the stuff I'm printing doesn't require 'bridging perimeters', so turning that off makes the support behave. It would be nice to have the "z-contact-distance" actually do what it says - even if you just change the tooltip to say "This is NOT the z-contact distance, but a factor in how it is calculated. To exactly set Z-contact distance, disable Detect Bridging Perimeters".

Agreed.

Is that really asking for too much?

No, not at all. I've followed similar discussions with other slicers and while there have been many ideas, nobody seems to have cracked the "good visual results with supported FFF printed surfaces" nut. I'd love to see improvements as well, but whatever comes next in terms of tech may be the ultimate fix.

RE: Minimum Support Distance?

While waiting for my first MMU to show up, I looked into supports pretty closely, with lots of pics and descriptions. But without interaction or feedback from the developers, it is a complete waste of my time and the community's.

PET-G supports for TPU - https://photos.app.goo.gl/94HYQ638cMG5aWxG7 Didn't work great, it still stuck a bit too well.

Pet-G supporting PLA worked great! https://photos.app.goo.gl/aLv11t1sE6bBfRmD8

This is petg (on bed) vs PLA (with gap, obviously), vs PVA on another printer https://photos.app.goo.gl/E2ACikuPDuHS9jvc9

A bunch of unsorted pictures of my quest to get nice bottoms:
https://photos.app.goo.gl/DfHjb2HPSzzWrm94A

Either you get nice bottoms, or you get the parts to stick. 🙁

Not having the best luck finding my pictures, but just use PVA/TPU/PetG and see how it goes. I wrote quite a bit more but I'll skip the future flame wars and just say CURA and S3D both took a number of suggestions and fixed a lot that was wrong in their software.

RE: Minimum Support Distance?

Posted by: abraham.m

[...] A bunch of unsorted pictures of my quest to get nice bottoms:

https://photos.app.goo.gl/DfHjb2HPSzzWrm94A

Either you get nice bottoms, or you get the parts to stick. 🙁

You're printing cranberries!

[...] CURA and S3D both took a number of suggestions and fixed a lot that was wrong in their software.

I'm also on the quest for better supported surfaces, so am quite interested. Did you find a Cura or Simplify 3D solution for the specific issue of getting good(er) undersides using supports? I spent time with all 3 plus KISS and while you can make it look a bit better, it's not really a good result appearance-wise. Of course, rounded top surfaces don't look great with FFF either.

I've obviously got no idea on what your interactions with the Prusa dev team have been, but they do seem to be a dog easily distracted by squirrels. I can't really blame them given the pace of development. The trick does seem to be to get their attention, and showing a result "what" rather than describing an algorithm or specific "how" seems to work best.

Good luck with it. Looks like you've got an interesting project underway.

RE: Minimum Support Distance?

Posted by: abraham.m

With all that said, if the software incorrectly describes what it does (support distance only works when detect bridges is off), that's a BUG, and should be fixed even if the behavior doesn't change.

But I just ran through a part and with bridging on, perimeter detection on, the spacing DID work and WAS adjustable. From too far away to full contact.   Check this thread for context: https://forum.prusa3d.com/forum/original-prusa-i3-mk3s-mk3-how-do-i-print-this-printing-help/bridge-infill-versus-support/#post-156192

I was able to adjust the contact Z to get the support to lightly kiss the part (0.01 mm) so supports were still easily removable.  At times it was tough to do because the preview slicer gives is just plain wrong (and a known issue) - but after a few prints what was happening clicked and I found numbers that did what I wanted them to do.  Mainly, melt droop during bridging was affecting the "final" distance, but that was no fault of the application.

From this (forced no bridging): 

To this (controlled bridging and minimal contact):

And what the support looked like:

RE: Minimum Support Distance?

Tim: I like what you're doing! Thanks for taking the trouble. I'm a touch caught up right now and want to look at this in the depth it deserves when I get a little more time. Could be hours or days.... 🙂

Bob! Honestly, I'd just been pursuing the PVA thing, and there's any number of issues with it. I took a CR-10S and added a second extruder, and had lots of issues with purge towers that don't work (in PrusaSlicer, I can't get the purge tower to appear for dual NOZZLE printers - but I haven't done anything but tick boxes yet), or skirts that don't stick. Basically I'm having issues with PVA/PLA adhesion. I redid the printer as a direct drive + bowden hybrid and it's working... but still has issues with dribbling, and then I saw my dream cheap IDEX machine and I'm giving it some testing to see if I like it.

Not really internet ready, but: https://photos.app.goo.gl/AWSWt6zh94UXTdF47