Finished my 36 hour TPU duct print

Nov 21, 2026 Last reply: 19 hours ago 6 Replies
Jump to replies (6)

So a while back I asked here how to print this curved duct part and honestly I should have just trusted the H2S from the start. Ran it slow on Bambu 90A TPU, 36 hours total, and it came out good enough to use. Not the prettiest thing, some under extrusion in spots and there's still support stuck under the base, but it works.



I dried the filament for 2 days before printing and still got some stringing, though I actually think it helped hold some of the really thin supports together. Next print I'll cut most of those since the part doesn't seem to need them.



I could probably go 95A but I'd rather keep it a bit more flexible than the original part, that stuff is a pain to work with. Getting it off the plate was a whole thing though. I nearly bent the part in half wrestling it loose and the plate may not have survived the experience either. Round two is on the way with some design tweaks and a few ideas from the helpful folks here.


I've been in that same wrestling match with TPU on the textured plate, nearly ruined a part myself trying to pry it loose. Two things made round two painless for me. While the part is still on the plate, spray or douse some isopropyl alcohol around the rim and let it wick under the edge for a minute, the part lifts off so much easier. Also try a glue stick wiped on the plate before the print, it works as a release agent and the piece separates instead of welding itself down. Wish I'd known that before I flexed my first one nearly in half too. In your photo that duct looks genuinely usable for a 36 hour run, the ribs came out clean and the finish is nice even with the wisps visible around the interior overhangs. The support stuck under the base is annoying but understandable on a part like that.

Since you already dried the filament two days and still got stringing, printing straight from the dryer for the next one seems worth a shot, that's the direction the replies here lean and it matches what I'd try.

I'd stick with 90A like you're planning. You wanted more flex than the original part anyway, and 95A sounds miserable to work with.

I'm with you on sticking with the 90A. I tried a stiffer TPU on a ducting job of my own once and regretted it, working something that rigid onto a flange was miserable. A bit of extra squish beats fighting the part every time you fit it, even if the harder stuff prints prettier. The ribs on yours came out looking cleaner than I'd have guessed, honestly. Those whiskers tucked up inside the curve near the angled opening don't look like they'd do anything worse than shed a little fuzz, and it made me grin that the stringing ended up holding your thin supports together. I've had supports snap off mid print on smaller parts before I even got a chance to remove them, so I'll take the sticky ones.

The patch of support stuck under the flange is the bit that would bug me most, since you can spot it through the opening. Curious what the design tweaks are for round two. Are you thickening the thin sections so most of the supports can go, or changing the orientation so they're not needed in the first place?

Agree with keeping the 90A. I fought the same battle with a stiffer TPU once, tried to fit it where I needed some give and regretted it the whole time. Flexible beats pretty for a part like that. That glossy finish in your photo came out nicer than I expected for TPU too, even with the wisps up in the overhangs.

That flange at the bottom in your photo looks like it came out nice and solid, I'd never have guessed the stretchy rubbery stuff could print that clean. I'm still working up the nerve to try anything that bendy, my one go at it was a blobby mess. The bit about the stringy hairs actually holding those skinny supports together is kind of funny, like the flaw doing you a favor. What kind of design tweaks are you planning for round two? I'm curious if you're changing the shape under there so it needs less of that support stuff in the first place.

Those patchy spots where the extrusion dropped out, are they anywhere that actually matters for sealing or airflow, or just cosmetic? I'd be tempted to leave well enough alone if the part does its job, messing with settings after a run like that risks swapping one flaw for a worse one. In your photo the finish reads way better than 'good enough to use' suggests, that glossy black hides a lot. Genuinely curious what the design tweaks are for round two, whether it's the supports or the shape of that curved section itself. Long slow prints like that are a nerve test, I'd be checking the camera every hour.

Update: back on the thread to say round one actually came out usable, curly bits and all. Gonna try the isopropyl trick from earlier before I even touch the next one, that plate deserves a break. Second print is planned with the supports mostly trimmed out, cheers for the tips.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required