Hello all! I’m printing a sealed cylinder that is hollow. The top and bottom “lids” are 3 mm thick and the walls are 6 mm thick. The rest of the center is hollow to preserve filament (doesn’t need to be super strong with lots of infill). However, when printed vertically, the top portion has nothing to support it underneath. I printed this way and it “seems” fine, but I know that bottom most layer of the lid was floating when printed. I don’t think using support material would be good, because it would be in the cylinder forever and if it breaks, I would have pieces rattling inside. Just wanted to get people’s take on this. I could, of course, just print a solid object with infill, but I plan on having different infill percentages in other parts of the print and I would prefer having the same infill settings everywhere else. In other words, no need for this cylinder to have 90% infill, for example. If no one seems any instability issues with a 3mm thick lid with the lowest layer printing in space initially, then perhaps I’m fine
You didn’t mentions the size of your object so one can only offer superficial response.
If I understand you correctly, you have a model that in an ideal world, you want to be hollow. That said, the lightning infill is likely to be ask close as you can get as it is designed specifically to provide the least amount of filament while still providing some level of structure. I would start with 3% infill and experiment. Again, if we could see your model it would answer a lot of questions and you would get a higher quality response. As an example printing a hollow tube of 10mm versus 100mm versus a tiny figurine versus a larger model would have very different answers. Details matter.
Here are some more details. I’ve attached a picture. It is overall about 100mm tall. It is 67mm in diameter and 6mm thick around the cylinder. The top and bottom are 3mm thick. The center is hollow.
So armed with that additional data–thank you–my suggestions would be twofold.
If you have control over the CAD file, design in manual supports such as an X in the middle of the model that bisects the model in the Z axis. This need only be 0.5mm in width.
Along with internal bridge, experiment with lightning, support cubic or tri-hexagon as these are specifically designed to do what you wish. Again start at 3% and work your way up.
Note: With Lightning supports the supports only accumulate towards the top of the model and are ideal for models with vertical walls like your example. This provides support for the top overhangs only where needed. You will have to experiment but in this case it was done with 10% infill.
We are assuming in your case since you mentioned it, that you are relying on the 6mm wall thickness for rigidity. That is a totally acceptable and common approach for a hollow object.
As an alternative, you could make the model in two pieces, which is my preference. I take the bottom half and draft a lip to hide the seam. Then I build the top half with a bottom layer height of zero using the techniques I described above. This creates a hollow body shape, and the thin interior supports rip out easily with needle nose pliers. I then seal the bottom using acrylic cement, which chemically melts the two components, resulting in a welded join rather than a glued one—bonding the materials at both a molecular and mechanical level.
If you need a pictorial example please let me know.
With Lightning at 10%, the sparse infill takes about 2% of the total filament used, so if the goal is to save filament without a complicated design, this certainly does the job very well.
That is one of the things I was unclear on with the OP. I interpreted the goal not so much as a filament saving move but more as a desire to maintain a clean inner container. By the looks of the shape, I’m assuming it’s design intent is to remain as hollow as possible but that’s the downside when someone posts a question and doesn’t want to reveal the details, they get half an answer.
Thank you for the details and advice. Yes, I’m designing this in CAD. So, your recommendation is to both put in an internal “X” type of support structure as well as use Lightening? I’m assuming Lightening by itself, since it is support material designed to be removed, would not be sufficient on its own, correct? My goals were two-fold:
Save filament usage
support the top lid of the cylinder from underneath since this is a sealed shape. This would have to be a permanent support since the container is sealed.
No, Lightning support is not designed to be removed. It’s not “support”, it’s “infill” just like any other infill, it’s permanent. It’s specific purpose is to address the problem you’re worried about - a top surface that needs some internal support in order to print cleanly, but where the supports are intended to be permanent. It satisfies this objective while using the minimum amount of filament it can.
If you use this, you do not need your own supports.
Easiest way to understand, try it and see what it does in the slicer preview.
Now that I know you’re using CAD, for this shape I would say that you have a much better option by designing the removable supports in the upper 10% of your shape. Think if it like removable trusses that are deliberately weak in the XY direction but strong in the Z axis. They could be torn away if you make the model in two pieces and would leave a cleaner surface.
Here’s a quick example of what I mean. Note that the top parts are thicker than need be to illustrate that they are the break away portion that one would grip. If one thins out the bottom to 0.40mm it would break away clean. Of course this presumes a two part construction to enable removal of the supports but one could easily just draft in a permanent support.
Supports are great if you want something quick and want the slicer to do the heavy lifting but the fall short on how the algorithm calculates the support. If you’re already in CAD, there’s no sin in designing the supports on your own. In the example I just showed, in Onshape that’s a total of 11 features and I was being lazy.
I’ll also point out that by making the supports separate features, you have the added bonus that when you import the STEP file, you can turn off specific supports as you experiment, you can’t do that with the slicer supports as easily.
It’s a sealed cylinder, he doesn’t want the supports to be removable. If he’s got to have them, he wants them to be solid so they can’t break loose.
Really, I’d just bridge the opening. Yeah, the bottom-most layer inside won’t be pretty, but you can’t see it. Add an extra top layer or two, to make up for it. The bridged layer will act as supports for the subsequent layers, no additional internal structure should be necessary.
Set the slicer to reduce flow slightly (a few percent at most) for bridging, make the speed nice and slow, run the part cooling fan at max.
I think you can simply print it as is. The usual term for what you call floating lines is “bridges”.FDM Printers are surprisingly good at bridging such gaps. The first layer above the air will sag a bit but within a few layers, you have solid layers again. As you won’t see it, the aesthetics are of no concern.
I would set the nr of bottom layers a bit higher than the default. Maybe 5 or 6.
If you want to completely avoid the bridging, you could instead do overhangs by letting the top of the hollow space become gradually narrower towards the top instead of a flat surface.
E.g. if your outer cylinder is 80mm and the inner hollow space is a cylinder with 70mm, you could add a chamfer with 45° and 35mm at the top. Then there are no lines above air at all.
edit: oh didn’t see Rocketsleds response, I fully agree.
Thanks! Yeah after a few prints of the same, the bridge seems to be holding up just fine even though the lowest layer isn’t supported. The top of the cylinder is 3mm so that seems plenty to get a solid top set of layers. If I ever run into issues, I can always do my own internal support or do something like a chamfer towards the top on the inside to provide additional support.
