Persistent Curling
The photo shows serious warping (or curling) of my print from my old PEI sheet which was printed on my MK4 with my old heatbed. All four of the corners warped (only the worst corner is shown).
This has been a persistent problem for the last couple of years. Smaller models printed closer to the center of the bed are fine. Only bigger prints which approach the edges of the bed curl up.
I recently upgraded my MK4 to the MK4S. While I was at it, I replaced the heatbed and I bought a new PEI sheet. And even though the sheet was brand new, I did wash it thoroughly with warm water and Dawn dish soap before using it.
After running through the full calibration and reslicing my model for the new high-flow nozzle using the latest stable release of PrusaSlicer, I reprinted it with high expectations.
Unfortunately, I was disappointed. One of the corners warped. It wasn't quite as severe as before, so that's a slight improvement. The corner that did warp happens to be the same one shown in the pic.
Print Settings: Prusa MK4S printer (no enclosure) 0.4 mm high-flow nozzle 0.2 mm resolution 15% infill Prusament Army Green PLA
-- Edit --
Wow, there are lots of good ideas here. Thanks everyone!
The things I'm going to try (in no paticular order).
- Covering the printer with a cardboard box or insulation sheets to see if an enclosure would help.
- Cutting a tiny relief across the sides on the first layer.
- Adding a brim.
- Slow down the first layer.
- Spraying 3DLac on the corners
34 replies
If you look at the edge above the lifted corner, isn't that supposed to be straight, or are the walls not meant to be flat?
My guess is that you need to print slower and/or increase nozzle temp. If the filament isn't properly melted that also leaves tension in the material, which can cause corners to be pulled off the bed. I.e. cooling issues is not the only thing that can cause warping. What are your current slicer settings? I assume your high-flow nozzle isn't PTFE-lined, so printing hotter should be safe.
Those kind of uneven layers can also be caused by a mechanical issue on the vertical axis, but I don't think that would increase the risk of warping.
Lots of great advice in this thread. To add to it:
Warp happens because of uneven cooling causing long straight lines of plastic to pull stronger than the bed adhesion. Sticking it to the bed better has had lots of focus in this chat. But you can also try to influence those first two parameters.
Other than breezes, Uneven cooling might be because those corners of your plate aren't heating as efficiently---possibly they aren't as insulated or the heating element just doesn't reach there as strongly, causing cooler spots. The part itself may also be cooling too quickly there because of holes or channels, where a more evenly solid part might cool slower and more consistently.
My favourite hack for reliable designs though, especially if you wanted to share them and don't want other people to worry as much about the warp, is creating relief features to address the long straight lines issue. Basically, on long lines heading into corners like you have here, a couple of centimeters from the corner, create a one layer high thin cut across the long line such that the slicer will interrupt the continuous line there. Make it as thin as you can get the slicer to register. Usually 0.1mm is sufficient. This is usually small enough that it will still end up closed over in the final part, so it won't be apparent as an actual gap, but it will result in significantly less strain pulling on that corner.
Warping being directly proportional to long straight runs was a major educational moment for me for modeling.
Thin cuts in large flat sections and "wrinkles" in long walls have eliminated any threat of warping. You can also change the first layer fill direction to have shorter runs.
only on the first layer?
Never tried deeper, could probably do more. But first only ensures it'll never affect the part strength. Good question, theoretically the higher layers will still pull on the end; but at least the first layer isn't de-adhering purely because of its own horizontal shrink
I need to make two more prints of this model. I will give the Thin Cut idea a try!
This is gold
do you have an enclosure? that was the only way I was able stop bed pull even with a very heated bed (tho this was a cr-10 so ymmv)
Ah, no. I forgot to mention I do not have an enclosure.
you could build a temporary one out of sheets of insulating foam - literally just build a cube with tape and cut a tiny window for observation - if that fixes it, replace with something fancier... if you feel like it.
I had these types of issues a lot before I built an enclosure for my printer. It's like night and day difference being able to keep the ambient temperature consistent throughout the print. A brim can also help with this but the enclosure is the real answer.
My mk4s is in an official prusa enclosure and I still get warp on large parts. Depending on the part I will take one or more of these additional measures, they all help:
is there air moving around in the build chamber? I suggested an enclosure above, before seeing this comment. argh, that really should fix it...
I do have a circulation fan but I only turn it on for materials that want more cooling like PLA. For higher temp materials (mainly PETG but I also print some ASA, PA, PC, TPU) I keep it off. Which is a little frustrating because I got it originally to circulate air through a carbon filter for fumes from those nastier materials.
that is so weird. as soon as I put up an enclosure my pla curling just went away.
Plus no dust, so when you get the machine dialed in, it just stays that way
Surprised nobody's suggested adding a brim. That's always been the final missing ingredient for me even after following all the other advice present in this thread.
Brim and also ensure that the room doesn't have any active cooling.
A fan blowing on the print will cause corners (with a higher surface area) to cool faster and the contraction will result in curling. Another alternative would be to put a cardboard box over the printer to ensure the print stays warm while it's printing.
Throwing a cardboard box over it is a cheap and easy test. I'll give that a try!
One thing I've had luck with above and beyond all the other suggestions in this thread, is to lay down the first layer really slowly. I think I have mine set to 25 mm/s.
Hm. That sounds easy enough to try!
3 things.
bevelchamfer on the surface that touches the build plate. Doesn’t meet to be large. Just something. It helps with cooling shrinkage and retraction of pla on the first layers.Last thing I would do is maybe adjust your z-offset. It looks like it might be experiencing some elephants foot at the bottom. I would try raising the head just by about .10. Really though that is going to be fully determined by your first layer of course.
I'll experiment with the bevel idea. That's one I haven't heard of.
Hey OP, just so you do not waste time. I mis-spoke and told you the opposite of what you should do in the program. No bevel, I meant chamfer!
I know it doesn’t make a lot of sense to make the surface area touching the build plate smaller but it is worth it since it provides relief from the layers above pulling when it is shrinking.
I'm not convinced this is helpful. These tests even suggest it's harmful
Re: no glue
Indeed don't use glue sticks - or if you must, PREHEAT the bed before applying the glue, the heat helps form an even layer you won't really get with regular glue sticks.
I know it's expensive but I've had extremely good results with 3DLAC's bed spray. It applies evenly, doesn't stink like hairspray, isn't as gunky and sticky to get rid of, and even the worst offenders against bed adhesion have printed like magic after a single coat.
Of course none of this alleviates washing the damn thing with basic dish soap. I do spray mine afterwards with some isopropyl alcohol, but that's mostly to speed up the drying, not to clean it.
Long straight lines are prone to curling. Assuming your bed temp is already 60c and you regularly wash your build plate with soap and water. Your options are some sort of adhesion aid like a brim or mouse ears or slow your print down. You may find a combination all these works best.
Modern slicers have settings for slowing your first layer as well as setting a number of subsequent slow layers that transition into your normal speed. I would start by increasing the number of slow layers to something like 10.
The number one enemy of any 3D printer is physics. In this case it's the physics of rapidly laying long straight lines of molten plastic on to one another. As those layers cool they shrink and pull against the previous layers. If the previous layer has sufficient time to cool then that reduces the shrinkage and the forces applied to the previous layers.
I have these issues with PETG on my mk4s, but usually not pla.
Brim is a must for the wide based prints.
Consider printing a draftshield if possible as well.
If this fails try a gluestick (I like elagoo)
Cleaning with isopropyl? Try a different brand (for whatever reason the cheaper generic drug store 90% ones seem to work best imo)
Not sure it would help with this, but satin sheet has been a game-changer for me.
For adhesion issues clean with soap and water, not alcohol. Take your build plate to the kitchen sink and wash it.
Would reducing wall thickness to a single width help. That's a whole lot of plastic mass in those corners
Yeah, it's part of a shoe rack. It doesn't need to be that thick, but you know it's just a matter of time until someone uses it as a step stool.
Would dual thinner wall with a gap and lattice of ribs help?
Especially the spray is 100% worth its cost (also a must if you print e.g. ABS/ASA).
I found adjusting the z-axis to put the nozzle closer to the bed to help a lot on my prints when I had that problem.