If you want smooth top surfaces on your 3D printed parts, a common technique is to turn on ironing in your slicer. This causes the head to drag through the top of the part, emitting a small amount of plastic to smooth the surface. [Make Wonderful Things] asserts that you don’t need to do this time-consuming step. Instead, he proposes using statistical analysis to identify the optimal settings to place the top layer correctly the first time, as shown in the video below.
The parameters he thinks make a difference are line width, flow ratio, and print speed. Picking reasonable step sizes suggested that there were 19,200 combinations of settings to test. Obviously, that’s too many, so he picked up techniques from famous mathematician [George E. P. Box] and also used Bayesian analysis to reduce the amount of printing required to converge on the perfect settings.
Did it work? Judging from the video, it appears to have done so. The best test pieces looked as good as the one that used traditional ironing. Compared to ironing, the non-ironed parts saved about 34% of print time. Not bad.
Of course, there are variations on traditional ironing, so your results may vary.
Saves 34% of print time?
Only if the print has two layers and the optimal speed is unchanged from the control.
Otherwise the result wouldn’t be so ’round’, when expressed as a fraction.
Shenanigans!
Yeah that’s what i was thinking too…my current setup is ‘pretty good’ but i definitely see signs of the internal geometry on the top layer. all top layers are not equal.
Not that i want ironing anyways. I save a lot of time by not caring :)
Sometimes I want a piece to friction fit with something else, and i do wind up using a knife to level off the tiny peaks at each U-turn
…tiny peaks? You might have your pressure advance out of whack
nah i don’t agree and i think it’s illustrative why i don’t.
I engineer my parts to perform certain functions. And i’m pretty successful at a broad range of projects. That doesn’t mean perfecting my printer’s behavior according to a platonic ideal of fidelity — something i don’t really think is feasible, and certainly isn’t a good investment of my time. It means characterizing the process (the whole process!) and using that knowledge to inform my designs.
Trimming my parts after printing is a part of my process. It doesn’t take much time and i get a result that is rarely surprising and almost always functional. The process i’m in is called engineering, and this is what engineering looks like.
If i really need a flat surface, i’m much more likely to buy an actually flat surface at the hardware store than to invest effort into tuning my 3d printer for something that it’s fundamentally not gonna do a good job of. When we say that the internal structure is often visible on the top layer, that’s not just anecdata, it’s something i take into account when matching processes to projects. Knowing the weaknesses of this process is key to engineering. Even after i go through the iterative process of generally tuning my printer, for each low-tolerance part i produce, i will have to (at best) repeat some subset of that iterative process. That’s a lousey process, in my judgement. I use it for what it’s good at.
I mostly agree, however one trick you may not have considered for when all you want is A flat surface – put that face down onto the print bed. Even the worst printers will then produce a pretty darn flat (but maybe textured*) result unless your filament is really warp prone and you have the wrong settings. Which lets you get the best of 3d printers producing complex geometry, even if its just some drill guide divots so you can get something like decent CNC mill accuracy out of manual/light framed 3dprinter etc AND still get a pretty darn flat surface – I tend to make anything with large flats print that way with a living hinge edge or two to fold the usually box like structures up. Folded once and solvent welded shut in the end and its as durable and smooth a surface finish as you can get without polishing (if you print on glass anyway).
*Worth remembering the surface finish will be inherited from your build plate – some folks have created holographic build plates to deliberately pass that structural colour type stuff into their prints, so the much larger texture of the more normal grippy buildplates will definitely show.
haha i should say too, though…anecdotally, i think you’re correct. My direct extrusion printer has much smaller peaks than my old bowden printer did! Pressure advance definitely defines a lot of error factors. I just don’t think it’s worth tuning something that’s never gonna be ‘perfect’ for long. Good enough!!
Given their test pieces appear to only be a few mm tall at most I’m guessing the top layer is a significant proportion of the print time so any time saved looks impressive.
If they did this on a test cube or a larger more complex print I’m betting time saved would be much less as a proportion.
The link to makerworld in the yt description literally talks about how they hope to combine this with ironing and so the fundamental title of this article is not just incorrect but also proves that whoever wrote it did little to no research.
A lot of people need to use ironing because they simply haven’t calibrated flow and bridging well. I’m not saying this is YOU, dear reader, you have your printer dialed in perfectly.
Mine is dialed in well but not perfectly because I couldn’t be arsed, but good ironing with an adequate filament results in completely smooth top layers, essentially like stamped plastic. Good, non-ironed top layers can be very nice, but lines are still visible.
Ok I know nothing about 3D printing but talk of ironing brought my daughter’s Hama beads to mind, so here goes… could you actually iron (with a clothes iron), through greaseproof paper, A flat 3D printed surface to smooth it out?
Not sure. It might work, but the act of ironing those beads was to get the tops to stick to each other. Melting plastic enough to force it into all the small voids so it looks perfectly uniform may not work the way you hope. Someone try it out.
If your object has one perfectly flat top surface, sure.
But slicer ironing works even on objects that have complex top surfaces. I.e. imagine trying to use a traditional iron on the bottom of a bowl.
Eventually… yes just depends on the size of said 3d object 😉
I’ve used this procedure and it worked great. The results after tuning were actually better than I ever got with ironing.
This sounds like the kind of trial that would have to be repeated on each separate machine every time you switched filament material or changed layer height or even when the ambient seasonal temperature changes. So only worth it it you were basically mass producing one piece.
The beauty of ironing is that it never needs to be re-dialed. It just works.
If I can get smooth surfaces by sacrificing 30% print speed or by doing periodic manual tuning, I’ll sacrifice the speed.
That might be appropriate with the right materials and designs. Generally though, I think the vast majority of testing can be cut down even further to give more generically applicable settings and as a simple step in a normal calibration procedure.
I don’t see the logic here. I have watched the video, and as I am still new to 3D printing, I don’t see (mechanically) how to enter in this data to achieve this so called success