The 3D printing process is cool, but it’s also really annoying at times. Specifically, when you want to get a part printed, and no matter how you orientate things, what adhesion aids you use or what slicer settings you tweak, it just won’t print right. [David Malawey] has been thinking a little about the problem of the edges of wide prints tending to curl upwards, and we believe they may be on to something.
Obviously, we’re talking about the lowest common denominator of 3D printing, FDM, here. Other 3D printing technologies have their gotchas. Anyway, when printing a wide object, edge curling or warping is a known annoyance. Many people will just try it and hope for the best. When a print’s extreme ends start peeling away from the heat bed, causing the print to collide with the head, they often get ripped off the bed and unceremoniously ejected onto the carpet. Our first thought will be, “Oh, bed adhesion again”, followed by checking the usual suspects: bed temperature, cleanliness and surface preparation. Next, we might add a brim or some sacrificial ‘bunny ears’ to keep those pesky edges nailed down. Sometimes this works, but sometimes not. It can be frustrating. [David] explains in the YouTube short how the contraction of each layer of materials is compounded by its length, and these stresses accumulate as the print layers build. A simple demonstration shows how a stack of stressed sections will want to curl at the ends and roll up inwards.
This mechanism would certainly go some way to explain the way these long prints behave and why our mitigation attempts are sometimes in vain. The long and short of it is to fix the issue at the design stage, to minimize those contraction forces, and reduce the likelihood of edge curling.
Does this sound familiar? We thought we remembered this, too, from years ago. Anyway, the demonstration was good and highlighted the issue well.
Thanks to [Keith] for the tip!
So whats the conclusion? What can we do to prevent this?
according to the video, part design.
But specifically how? “Design it differently” is by itself not very helpful. Are there any concrete suggestions in the video or does it just come up with a good hypothesis for why the prints warp, with anything else left as an exercise for the viewer?
Maybe just watch the 51 seconds long video?
He says it very clearly:
split up large surfaces and think of force vectors.
i dont know what else you are asking for.
thanks – that was exactly what i was asking for!!
Solution: Don’t do the thing that causes warping.
Honestly why did none of us think of that before?
In essence it comes to this: “use less material”. Prevent long continuous areas/stretches of 3d printed material.
If you use a plastic that (for instance) has a shrinkage of 1%, it will shrink 1mm over 100mm, but only 0.01 over a distance of 1mm, although both is the same 1%, the effect of a part wanting to be 0.01mm smaller has less impact then something that needs to shrink 1mm. And this works in all directions, but mostly only causes real problems in the X-Y because there all the plastic will cool down at the same rate but in the Z direction the difference in temperature and therefore effective shrinkage can almost be neglected as the layer benath is already cooled down when the next layer is added, therefore the layer below has already shrunk and therefore the shrinkages do not accumulate over the total distance (unless you have a very small but tall part that is not allowed to cool down before the next layer is added).
But this doesn’t mean you can’t do large prints, just allow for the shrinkage to happen (allow for the diversion of the shrinkage forces) or prevent the buildup of these forces by breaking up large areas into smaller areas by adding holes. Using honeycomb fill patterns instead of a solid fill certainly helps and it decreases you print times. I hope this answers your question.
“use less material”
Have you printed with ABS? I print lots of long thin parts at work and you can clearly see the walls pulling in where the infill is. Solid layers/solid infill = no wall warping.
Will – i wonder if you’re describing a separate phenomenon caused by something like surface tension?
it seems to me like the most important part is that as you’re building, you address the problems you meet…i don’t imagine it’s possible to have an accurate mental model of every source of deformation before you start playing with it.
but also maybe i’m feeling smug with PLA :) which does warp but in a way i’m used to. i had a weird PLA blend that would do stuff like that — where i could see the slicer infill choices affecting the surface dimension — and when i realized i was trying to anticipate that in my designs, i switched to a different brand of filament!
Thank you, the text is almost useless in itself.
The video is really good to explain the problem, but really bad at explaining the solution.
When you told me about holes, my brain actually clicked and I understand the problem and solution in itself.
Thank you
that’s explained at the end of the video. split up long runs to create separate planes of contraction so that each section is only subject to the contraction force within its run. that, according to the presenter, results in forces similar to those of a line of individual small parts.
Basically, long straight sections are going to lift so don’t make long straight parts.
Of course that’s not practical, but implies that adding notches to the edges of a straight part eases lifting because the length is reduced.
The alternative way to deal with it is that each layer following the base layers must have less mass than the layer below it so that it cools more quickly generating less forces.
If there’s curling, instead of using four base layers, try two. Reduce the infill percentage. It would be a similar effect to stacking two pieces of tape on each other and then following pieces of tape are mesh tape or have holes followed by two normal pieces of tape in top.
So yes, it’s a part design issue.
That sounds more like a print configuration issue…? I dunno, I haven’t seen the problem but I’m still printing PLA.
Everything you have suggested will make the part weaker.
Breaking up long straight sections with notches will lower their strength and introduce a weak point.
Same with your suggestion to lower curling, using less base layers and less infill will lead to a much weaker part.
So in a lot of cases your suggestions just aren’t practical.
Correct. One hardfast rule applied to every design will not work everywhere. In one design you might have a large flat surface that is hard to print without warping, can be helped by adding notches and doesn’t need to be super strong.
The notches could even be made to be decorative.
Another part might need a lot of strength but not be a shape that incurs so much shrink force. So no notches needed.
Another part might need be big and flat AND need to be strong. In that case… does it have to be flat and solid? Maybe some sort of honeycomb shape would work.
Or if you really just need to print something flat and notch-free then you are perfecting your enclosure and temperature, scrubbing your bed, tweaking settings, etc.. until it works.
There is a lot one can do with a 3d printer even as a beginner so no one needs to be put off by this. But there is also a LOT of room to learn more techniques to better design for 3d printing as one becomes more skilled.
This is just another tool to stick in your box and use when it fits the job.
The best way to prevent this is to lower your bed temperature after the first few layers. You can do this in stages. The warping is caused by the temperature gradient in the z-axis throughout the duration of the print.
PEI+PVP gang here.
I bought an empty marker with a 50mm wide felt “brush” attached to it from China, ordered a bunch of PVP glue sticks from China. I made thin slices of the glue, put it in a sauce pan with water until it was all liquid, then added it to a mason jar. I fill up the marker with that glue water when empty, and wipe my entire bed with it before printing. Made thousands of prints without warping, never bothered to clean the bed in between, just put a new layer on before printing. Haven’t had a warped print since I started with it.
I always clean the build plate with denatured alcohol before prints and check my e-steps and bed level once a week. Only time I get wappred parts is big parts after the print when the bed cools down. That only happens with low infill like 12 to 20 percent. To fix that I just do 50 to 100 percent infill.
I hate reddit but sometimes it has good solutions. I’ve been using the mix here with amazing results
https://www.reddit.com/r/3Dprinting/comments/g07ndb/recipe_for_magigoo/
in case the link doesn’t load, and i use denatured alcohol instead, brush it on with a cheap foam brush from walmart, works amazingly well, and insanely cheap.
Mix 1:2:2 clear school glue, isopropyl alcohol and water.
I found that fan speed helps prevent curling if you find your prints are curling try up your fan speed and see if that works I did that and now no more issues with that, hope it helps others
I have tried cutting slits in long straight edges for this very reason, but the results have been inconclusive since I never tested it methodically.
I use very small notches (smaller than the line width), so I didn’t end up with visible gaps, but long lines are weakened every couple of cm. Basically I assume a print to be a bundle of fibers stuck together, more than a solid mass of plastic, and that contraction works accordingly.
your description made it click for me…really what you need is a way to ensure that there is a void somewhere specific, without generating a new outer wall the way a bore-hole would. a novel slicer feature :)
That’d be something to interesting to test in a more scientific way because it is a technique that slicers could implement if it bares fruit.
OK, good to know the forces behind it so you can sometimes make design changes to compensate. But other times you need a big flat bottom as part of the design. And even if you can add some grooves or ridges on the underside, wouldn’t those same forces on the (attached) upper layers still cause it to lift, curl, and peel?
Besides general adhesion, bed temp is a big issue here. Perhaps keep the same bed temp for the entire print, perhaps even raise it on upper layers? And the same trick used for ABS should work for PLA, PETG, etc. which is to enclose the printer to avoid drafts, and perhaps even heat the chamber.
Warping has been an issue for me too so with this new knowledge in mind I think I need to run some experiments…
so this ‘fix’ will only work if you can alter the model shape but that sounds very impractical on downloaded models…
Or print in another direction. We found with resin printing that putting stuff on an angle really helped surface finish. If your problem is long faces mean adhesion issues, tilt it and use supports, maybe.
We went from searchable referencable forums, to ephemeral chat rooms full of memes, to brain-death inspiring, goldfish entertaining, engagement chasing 50 second videos.
Every step away from fora is willingly giving up the usability of a technical reference.
This, right here, is encouraging the enshitification of the Internet and our ability to have usable documentation.
Shame on this “shorts” poster and HaD for patting them on the head.
yep….
Bring back blogs for technical documentation
So in theory, a pattern of voids through the part to break up stress concentration would do the trick. A simple way would to be projecting alternating perpendicular cut lines across the part every few mm around the perimeter of each layer, so that the part is printed as a series of islands. The islands would overlap vertically but be separate horizontally, and then the stress would be trying to pull each island apart in all random directions instead of concentrating along the longest axis of the part. Something tiny like a 0.05mm gap, or however much is needed for the horizontally adjacent extrusions to not fully bond.
I scrubbed my glass bed with a diamond polishing sponge in a fit of rage when faced with this problem, and it helped a little too much. When it cooled it popped off the bed with a chunk of glass stuck to the part. Now I have to compensate for the hole in my bed during part design.
This is one of those circumstances that highlights how much of a harder time you’ll have if you skip out on really getting to grips with the fundamentals of what you’re doing.
Knowing what forces your part is subjected to during different phases of fabrication really should be part of the design process.
This is not new to me at all, it was obvious to me more or less from the beginning of by 3D printing “career” that this is the root cause of warping. There are more ways to fix it, all have “strings attached”. Typically you will want to try all of them so the downside effects can be kept low:
Keep temperature difference between printed part and nozzle as small as possible.
This calls for a heated bed, an enclosure (maybe even a heated enclosure), and the highest possible bed temperature (and enclosure temperature) together with the lowest acceptable nozzle temperature. Keep in mind that low nozzle temperatures can cause bad layer bonding, so these are conflicting requirements. Also high bed temperatures can cause low bed adhesion and bad precision if the part does not cool down properly, although that might be attacked together with the second point.
Cool the printed parts before the layer is finished.
If you can cool strong enough such that the printed material is significantly cooled down before the layer is complete, this effect gets reduced because only the part of the layer which is cooling down is contributing to warping. So in contrast to many believes, stronger cooling helps against warping and does not cause it. Keep in mind that strong cooling might cool down your nozzle creating again bad layer bonding or even seemingly jammed nozzles, so use silicone socks or so. Also the cooling needs to be uniform around the nozzle. Finally keep in mind that you don’t even have to increase cooling, if you instead print slower (in terms of material flow per time unit, not X/Y axis movement).
Modifying the part design is – similarly to using adhesive helps – just fighting the symptom and not actually the root problem, but of course it is still a good approach together with the above points to improve your result.
I am printing nowadays even ABS with maximum cooling, and I plan to even increase my printer’s cooling power, because I think the printing speed is limited by cooling power. Also I typically print very thin layers (0.05mm), since I anyway cannot print faster. Thin layers might even cool down faster….
your advice about cooling reminds me of a dumb hack i use on my printer…
i have a script called “airout.pl” that transforms Gcode to recognize any layer that takes less than 20 seconds to print. at the end of that layer, it moves the head off to the side and tells it to wait the remainder. that way every layer has had at least 20 seconds to cool before the layer above it is deposited.
absolutely necessary for printing towers on my printer. and i use it a lot for objects that might be a good size all around but the top 1cm of it might be just a few protrusions that wouldn’t cool enough.
It’s great to see practical solutions for the problem- I have a tough time with ABS and can go back to it with new tools. But isn’t this best fixed in the slicer? In the same way I don’t include supports in my model I would expect the slicer to look at my material and the geometry and make the appropriate tweaks.