I’ve tried a bunch of different ironing test models on the H2S (because the setting I use on P1S and A1 don’t work on the H2S) and I find a setting that produces great results, try it on a model and end up with stringy “burnt” looking sections, sections with poor coverage etc. I don’t get it. Anyone else having issues dialing in ironing on their H2?
1 Like
I’ve found this popular ironing test to be very good. I’ve done it on my P1S as well as my H2D.
I can get “clean” ironing on tests, but when I try to put it on an actual model, it turns to garbage.
Test:
Exact same settings on a model:
This only happens with my H2S. The P1S’s and the A1’s all iron just fine.
What’s your result on the P1S or A1 for the same model?
AFAIK, it’s basically futile to iron any surface that has a couple holes on it. Because the path of ironing would always break and then your surface would look blocky.
The slicer always tries to make less travel moves, so that it tries to print the ironing surface with longest continuous paths. WHICH IS DUMB. It should just scan through the surface regardless of how many travel jumps it might need to do.
I iron surfaces with holes and buttons and everything and they seem to come out pretty good. The only time I have trouble with ironing and the H2D is with wood pla. It never does a good job as my P1S.
Here’s a miniature light-up radar machine I designed and ironed the surfaces of the top 2 control consoles. One would think something like this would not be flat enough to iron but it worked.
It gets much worse when you have larger surfaces
i can agree, i have bad ironing in large flat surfaces with bl pla and default settings
SIGNIFICANTLY cleaner. There are some trails, due to the “breaks” for the letters, but on the P1S and the A1 (and the mini too) the finish is smooth, there’s no discoloration and “grittiness” like I’m getting on the H2S. I don’t have a print handy that was done on the other machines to grab pictures of, but, there’s a marked difference in the finish.
I’m having the same issues, I’ll do swatch tests, even printing 100% same sized swatches like below, but once I make the actual print it comes out with the telltale gaps again. It seems very context dependent, and the infill underneath seems to be playing a bigger part than it ever did with the P1/X1, case in point:
what irks me the most is I already went through this exact process with the H2D and had to GIVE UP because nothing I did mattered, it always printed with gaps on large areas. I went back to using my X1Cs for all of the ironed items.Tried again with the P2S (because the X1C is discontinued and I needed another unit) and now I don’t know what the future holds for my farm anymore. I might have to switch brands.
You need a higher ironing flow ratio.
For large parts it’s always required to test ironing based on the actual LARGE surfaces with a certain height below (at least some layers of sparse infill included). The small patches of test patterns won’t do a good enough job.
Sounds like the classic ironing quirk on these machines. What helped me was lowering the ironing speed a bit and bumping the temp up by 5–10°C so the filament stays “melted” enough to smooth out. Also make sure the top layer flow isn’t too low - if it’s under-extruding even slightly, ironing won’t have much material to work with. After tweaking those, mine started looking way cleaner.
1 Like
same sized surfaces already tested in the photo above it, same height also tested - without the walls surrounding it, which is what is mostly affecting the issue, it’s not the EXACT same as the final print, but this is a 29 hour print, and that’s 17 hours of it to get to the ironing stage. Vs the X1C where I could test ironing on a swatch no matter what size and it’d faithfully print the exact same sheen no matter how big the final print, every single time, repeatedly, for years, literally I’m using the same settings I spec’d on my first X1 (pre-X1C) from 3 years ago, on 6 X1Cs to this day all using the same settings per filament, never having to change or re-calibrate. but with the H2 and P2 I have to create an EXACT DUPLICATE of the entire model to calibrate ironing? sigh. I’m on the hunt for old-new stock of X1C’s, I just got an email from MatterHackers, they’re sold out due to demand since the discontinuation. I might just have to pick up some P1S’s
I hope someone from @BambuLab cares about the print farm owners who use Ironing - since we’re about to skip buying ALL of their new models and start buying old ones because production will always supersede endless experimentation.
Update: Spent the last 2 weeks going back and forth with Support, They surprisingly cut to the chase right away and had me try some settings I’ve never done before: changing the spacing and some settings tarting at 10mms/25% speed, had terrible results, from there we did calibration swatch tests that I balked at because I’ve done them dozens of times on many filaments… then they asked me to do a needle unclog and cold pull… I was insenced, of course these brand new nozzles I’ve tried, two of them, wouldn’t need this, but I did it anyway…
Dirty AF, took 3 retries of the cold pull to get white filament to come out clean. Then I started over with both calibrations, and then did the swatch test again, then my own large swatches… and it’s working now.
So the culprit for me must’ve been some residue in my nozzles, and/or some nozzle related issue throwing my calibrations and swatch tests off.
This is my current result now: 30mms / 30% speed, 0.1mm spacing is what worked perfectly for me now, but those same settings before produced terrible surface texture.
Thanks so much to the un-named @BambuLab Support agent who walked me through all the steps.
2 Likes
Update to the update: no.
I’ve never gotten this to work in real world printing outside of swatches, I finally developed a “slice” of my final design to continue testing without waiting 8+ hours per print. I should amend that this if anything Gives me hope that it’s possible. But has not yet proven possible. It’s so frustrating so g d frustrating. I have 38 test prints, and sometimes printing the SAME SETTINGS Yields DIFFERENT results. It’s like the printer is gaslighting me. With the X1Cs I can vary the settings in a huge range: 30/25 being the average for all Bambu lab pla with some needing 25/30. I can’t get blue or white to print remotely as well as my x1Cs on the P2S nor the H2D which are identical in nozzle and problem.
1 Like
I regret purchasing both the H2D and P2S which both sit unused while my 5 X1Cs are printing nearly 24/7 and I can’t buy more. I’m very upset with Bambu Lab right now.
2 Likes
I know this isn’t the topic at hand, but I’ll throw this tip your way, especially if you’re mass producing. It’s not worth it for everyone, but once you reach a certain quantity: acetone smoothing - or PVB filament and isopropyl alcohol (IPA).
As mentioned, this is not a recommendation for everyone; it is only worthwhile once you have reached a certain volume.
Then you also switch off ironing and can use the printers more effectively.
1 Like
Ironing is just like this. You tune your settings to perfection by printing 1000000 swatches then in the real prints you get fked over again.
And also, in this case, your settings are wrong because for ironing you need to do both SMALL and LARGE swatches and find the settings that work in both situations.
And also even if you’ve fine tuned your settings, when you have patterns like this:
It’s still going to suck because the ironing toolpath isn’t “monotonic”. It’s always filling the lines in PATCHES, especially when you have holes on the surface, like these characters, and that makes the surface look PATCHY.
In short: Ironing is the answer for a wrong problem. Forget about it. As @RetroSharky has mentioned, ABS+Acetone, or PVB+IPA, or do sanding and smoothing and painting is the real answer. The other option you might try is to tune your top surface line widths and flow ratios to avoid ironing.
5 Likes
I am not talking specifically about OP’s problem, but about ironing in general.
In addition to what you said: Many people have completely wrong expectations when it comes to large surfaces and ironing. No wonder: That’s how 3D models are advertised - which are actually 3D renderings - enough user who don’t know that.
…and this is what it actually looks like:
4 Likes