Support material settings

[jlvter]

I've done a few prints now with our new V3. Most of them good. A couple of big fails. I have noticed that the support structure is difficult to remove from prints-what settings will make it easier to remove? I am using Mattercontrol 1.5.-Larger or smaller interface layer? Pattern spacing? Thanks for any information.

[Polygonhell]

PLA is notoriously difficult to remove support from, ABS is a lot better.

I've messed around with support in slicers for years and always been dissatisfied with the results, I do still occasionally print things with suport.

People will rave about Simplify 3D's easy to remove support material, but it gets there by using a structure that results in poor quality at the interface.
Of the Slicers I've used with support I prefer KissSlicers, but it's far from perfect.

I picked up an Up Mini 2 a couple of weeks ago because I'd heard good things about it's support and I wanted to experiment with it.
I've been messing around with support settings in various slicers for the last couple of weeks and even with ABS I have yet to find a setting on any of the free Slicers that's comparable to the support the proprietary UP software produces, and the difference is night and day, it's both easy to remove and doesn't leave a poor interface.

The way the UP does it seems similar to KissSlicer with the 100% interface setting, but I've yet to manage to tweak the KissSlicer settings to get comparable results.
Simplify 3D's support is just less than ideal, being able to edit the locations is really nice, but the support interface layer doesn't match the shape of the supported material, so the filament droops between the contact points.

I'm still playing with settings in both slicers, if I find something better I'll report it here. Either way I'll do a comparison when I've finished fiddling with settings.

[Jimustanguitar]

Yeah, support is always feast or famine. I've been fighting a part this week with an elevated mounting tab that's making me crazy... Finding that right mix of support and contact will prove to be tricky, depending on the part.

[img]http://i.imgur.com/1P8mxIa.jpg[/img]

[mhackney]

For critical parts the only way (imo) to get good support is to design it into the part (and only after you've explored other design options that do not require support). Jonathan Rocholl's "zipper" support for the mini kossel ball bearing carriages is a great example. It zips off a critical overhanging part cleanly and was key to the success of that design. No automated support generation (at least today) could do it.

[jlvter]

OK,maybe a better question to you guys with lots of experience printing would be, do you have a rule of thumb when you look at a part that you know right off that you will need support material? Say for instance if the part has thin sections or small diameter weak features? I'm sure this would vary by material as well. Just looking for a starting point. Perhaps I am using support material and don't need to......

[mhackney]

"overhangs" is all I care about for support.

[IMBoring25]

Jim, I've had some success with similar features designing in single nozzle width walls supporting the perimeter and the edge of the hole, then removing them with scissors, diagonal cutters, needlenose pliers, files, or whatever other tools are convenient.

The ultimate in good surface finish on supported surfaces is, of course, a dual-extrusion upgrade and soluble support material. That also allows some configurations that can't be made any other way because integral support would be impossible to remove.

Sharp overhangs and long bridges are the primary applications for support material.

[lightninjay]

If you feel like configuring a new slicer, you can try craftware. It has customizable supports similar to s3d, but also has some control over how the support actually gets laid down.

From what I can tell based on the settings, it modfifies the extrusion rate and travel speed to make sparser support structures , which in my experience break away pretty easily and cleanly, while also making a good interface layer.

I have some base slicer settings if you wanted a starting point. I posted them in another thread recently.

[jlvter]

For critical parts the only way (imo) to get good support is to design it into the part (and only after you've explored other design options that do not require support). Jonathan Rocholl's "zipper" support for the mini kossel ball bearing carriages is a great example. It zips off a critical overhanging part cleanly and was key to the success of that design. No automated support generation (at least today) could do it.

I see, that's an interesting idea to add support to design. Thanks.

[jlvter]

If you feel like configuring a new slicer, you can try craftware. It has customizable supports similar to s3d, but also has some control over how the support actually gets laid down.

From what I can tell based on the settings, it modfifies the extrusion rate and travel speed to make sparser support structures , which in my experience break away pretty easily and cleanly, while also making a good interface layer.

I have some base slicer settings if you wanted a starting point. I posted them in another thread recently.

Thanks, I'll check them out.

[mhackney]

You can use an STL editor to add support to STL models you download. I forget which one (open source) actually has pretty good tools for doing it.

You can really tweak the support to get the best results. For instance, to support a long bridge I'll add in a zipper like this screenshot. Notice that it does not touch the side of the opening. It is also exactly 1/2 of a layer height shorter than the opening. That also minimizes interlayer bonding. These are just a few tricks you can do if you roll your own. The little time spent up front in learning is more than recovered with consistent and great results later.

[Polygonhell]

For critical parts the only way (imo) to get good support is to design it into the part (and only after you've explored other design options that do not require support). Jonathan Rocholl's "zipper" support for the mini kossel ball bearing carriages is a great example. It zips off a critical overhanging part cleanly and was key to the success of that design. No automated support generation (at least today) could do it.

Not true, the original version of that had no builtin support it was designed to be sliced with KissSlicer with a maximum support height set. so it didn't generate support in the interior of the bearing.
I have been stunned by the UP's support, its redefined my belief in what is possible for same material support,
I printed this https://www.thingiverse.com/thing:1783512 for my wife on who wanted a "cat" model, I loaded into the UP software pushed print and walked away, I wasn't actually expecting to get a decent result, but the support popped off cleanly and the supported surfaces were close to perfect (some slight marring where the support was removed).
I've been using the same model to test other slicers at various settings and my initial attempts were pathetic by comparison, the worst quality area being the overhang on the rear legs.
I've been refining settings and I can get comparable surface quality from KiSS, but most of the time the support is so hard to remove I end up destroying the print,
I'm still iterating through settings and it's gradually improving, unfortunately the print time for that model even at 60+mm/s is over 3 hours, so the iteration rate has been a little slow.
I'll post some pictures when I get through with the tests.

[jlvter]

I use Camworks solids so it's easy enough to convert an .stl back to a solid model, edit and save back to .stl format. Good idea, thanks.

[jlvter]

PLA is notoriously difficult to remove support from, ABS is a lot better.

I've messed around with support in slicers for years and always been dissatisfied with the results, I do still occasionally print things with suport.

People will rave about Simplify 3D's easy to remove support material, but it gets there by using a structure that results in poor quality at the interface.
Of the Slicers I've used with support I prefer KissSlicers, but it's far from perfect.

I picked up an Up Mini 2 a couple of weeks ago because I'd heard good things about it's support and I wanted to experiment with it.
I've been messing around with support settings in various slicers for the last couple of weeks and even with ABS I have yet to find a setting on any of the free Slicers that's comparable to the support the proprietary UP software produces, and the difference is night and day, it's both easy to remove and doesn't leave a poor interface.

The way the UP does it seems similar to KissSlicer with the 100% interface setting, but I've yet to manage to tweak the KissSlicer settings to get comparable results.
Simplify 3D's support is just less than ideal, being able to edit the locations is really nice, but the support interface layer doesn't match the shape of the supported material, so the filament droops between the contact points.

I'm still playing with settings in both slicers, if I find something better I'll report it here. Either way I'll do a comparison when I've finished fiddling with settings.

Great info, thank you.

[jlvter]

Jim, I've had some success with similar features designing in single nozzle width walls supporting the perimeter and the edge of the hole, then removing them with scissors, diagonal cutters, needlenose pliers, files, or whatever other tools are convenient.

The ultimate in good surface finish on supported surfaces is, of course, a dual-extrusion upgrade and soluble support material. That also allows some configurations that can't be made any other way because integral support would be impossible to remove.

Sharp overhangs and long bridges are the primary applications for support material.

Are you using the dual extrusion method?

[IMBoring25]

My Mendel has triple extrusion capability in its current configuration but I've only been using one extruder for most of my recent projects. I have in the past supported PETG with water-soluble PVA for a couple of projects that demanded it.

[Polygonhell]

Before thinking about dual extrusion as some sort of panacea, you should go and read some of the older dual extrusion threads on the board.
We are starting to see some viable dual extrusion solutions now, but the classic put two nozzles next to each other solution is more pain than it's worth unless you come up with some mechanism to move the unused head out of the way during printing.
Solutions like the prometheus system/Prusa system seem promising, but I'd like to see how reliable they are, the BCN3D Sigma dual carriage design is apparently very good, and Ultimaker released the UM3 yesterday with a nozzle lift system, not a lot of people have used it yet, but given their reputation for print quality and the sample prints they were showing I'd be surprised if it didn't work well.
There finally seems to be a push for workable dual extrusion, and experimentation on the soluble support space (normal PVA is hideous to work with), it'll be interesting which techniques survive over the next 12 months or so, and which prove to be reliable.

[jlvter]

Before thinking about dual extrusion as some sort of panacea, you should go and read some of the older dual extrusion threads on the board.
We are starting to see some viable dual extrusion solutions now, but the classic put two nozzles next to each other solution is more pain than it's worth unless you come up with some mechanism to move the unused head out of the way during printing.
Solutions like the prometheus system/Prusa system seem promising, but I'd like to see how reliable they are, the BCN3D Sigma dual carriage design is apparently very good, and Ultimaker released the UM3 yesterday with a nozzle lift system, not a lot of people have used it yet, but given their reputation for print quality and the sample prints they were showing I'd be surprised if it didn't work well.
There finally seems to be a push for workable dual extrusion, and experimentation on the soluble support space (normal PVA is hideous to work with), it'll be interesting which techniques survive over the next 12 months or so, and which prove to be reliable.

I have a friend who has one of the first Replicators with dual and it was quite a nightmare according to him. He ended up using only one.

[adarcher]

I've keep going back to Cura 2.3 for supports. I hate that I can't edit them out in certain places, but I've gotten to the point where it almost always comes off easy. The interface settings are great, I have been able to turn them off completely for simple things, and add them in for delicate things. The trick is to get the density and spacing right--though it isn't too much of a trick since the default settings seem to work just fine for me.

Cura can't connect to my Rostock though, so I just slice it there and save the gcode for octoprint.