I’m attempting to print with 3DXTech 90A TPU. I’m feeding it directly into the right nozzle with the glass cover removed.
The filament loads properly. I’m starting with a flow rate calibration. No issues during the filament change/purge step at the beginning of the print.
The print seems to be going well. I’m not hearing any skipping of the extruder and the filament is extruding properly, print quality looks great. Then after a few seconds I get an error saying that the extrusion resistance is abnormal. First time I unloaded the filament, didn’t look bunched or anything like I’ve seen when printing too fast on the A1 mini.
I lowered the print speed and tried again -same thing. I tried resuming and filament flowed normally even after the error and then a few seconds later the error popped back up. I originally tried 3.6mm^3/s volumetric flow rate, then went down to 2.8mm^3/s. Temp is 225C (mfg recommends 220-240 and I’ve been successfully printing at 225 on the A1 mini -albeit only at 1mm^3/s).
Any ideas what is causing the error despite all signs pointing to there being no physical issue?
Edit: Tried with the 0.4mm HF nozzle as well, no other settings were changed. It got much further and I thought the issue might have been solved, but after about 35 minutes or so the error came up again.
It’s most likely not a mechanical blockage but a temperature and flow balance issue. The printer’s sensor detects higher extrusion resistance when the melt inside the nozzle becomes slightly too viscous — usually because the nozzle temperature drops below what’s needed for your chosen volumetric flow. This can happen if cooling air hits the nozzle or if the flow rate is near the melt capacity limit for that temperature. Using a 0.4 HF nozzle helps because it has a larger melt zone, but if cooling remains too strong, the issue will still appear after some time.
I print TPU 95A HF, TPU 90A, TPU 85A, PEBA 95A, and TPU Air directly from my AMS HT via bypass through PTFE without any issues.
However, whenever I’ve seen exactly what you describe, it’s been caused 100 % by too aggressive part cooling, which cools down the nozzle. That makes the melt thicker and increases back pressure, triggering the extrusion resistance abnormal warning even though the extrusion itself still looks perfectly fine.
Three quick things to try (since you can’t use the top cover when feeding directly):
Lower part-cooling fan speed to 0–10 % and disable “force cooling” for overhangs/bridges.
Raise nozzle temperature to 235–240 °C to stabilize melt flow.
Reduce volumetric flow to around 1.5–2.0 mm³/s and keep acceleration moderate.
Thank you for the suggestions! The cooling was set to 100% full-time, surprising since it was based on BBL’s own generic TPU profile. I implemented your suggestions and I still had the same problem during a PA calibration print. One thing I noticed was that the print seemed to be going much faster than I expected based on the filament settings, but with the way Bambu Studio sets up calibrations I could not verify since there’s no preview.
I got enough data from what worked of the PA cal to ball park and I’m running a test print right now (just a 10% gyroid infill cube without walls) and the print speed (@2.8mm^3/s) seems much slower and more what I expected to see than what I saw in the calibration print. 40 layers in and everything looks good so far.
It’s almost like the Bambu Studio calibrations ignore the some of the filament settings? I don’t know of a way to verify this, but that’s the only thing I can really think of based on what I’ve seen.
Edit: Still can’t confirm anything re: my suspicions about the calibration function, but the cube printed flawlessly -146 layers, 6.5g, 40 mins, no errors.
The PEBA I’m currently using is Siraya Tech PEBA 95A.
It exhibits significantly higher toughness compared to TPU and, in my experience, can be easier to print once the process parameters are dialed in. However, it requires a higher bed temperature, stronger part cooling, and generally tighter environmental control to achieve optimal layer adhesion and surface finish.
Although I haven’t completed extensive testing yet, according to the manufacturer’s data, PEBA provides higher tensile strength, lower density, and maintains its mechanical properties down to –60 °C. It also demonstrates an energy rebound of up to 78 %, depending on the infill pattern and overall model geometry, which gives it superior elastic recovery compared to typical TPU formulations.
TPU Air produces a satin-like surface finish, while materials such as Bambu Lab TPU 95A HF and similar high-flow TPUs tend to result in a shinier, glossier appearance.
If you experience excessive back pressure in the nozzle, it can trigger this type of error.
High back pressure can result from a nozzle that is too cold, printing too fast, incorrect pressure advance ( PA ) settings, or minor blockages restricting flow.
As a test, try increasing the nozzle temperature by 5–10 °C below the filament’s specified maximum to see if it improves stability. It could also be something as simple as a partially clogged extruder — even a small contaminant can cause flow resistance significant enough for TPU 90A to trigger a load or resistance error.
With PEBA, I can reproduce the same behavior when printing at Siraya Tech’s recommended temperature of 240 °C with full part cooling. That works fine for low-density or non-solid prints, but when printing solid parts the system quickly throws a resistance or overload error.
I can share my working settings for TPU 90A later today so you can compare. Keep in mind that I print directly from my AMS HT (actively heated during printing) and through the PTFE feed system, so your optimal settings might differ slightly.
My current setup works reliably with TPU 85A, TPU 90A, TPU 95A HF, PEBA 95A, and TPU Air. I haven’t yet tested PEBA Air, since I haven’t been able to get any spools of that material.
I’d love to see that. And if you have a picture of your setup that’d be great too. I’ve been thinking about getting an AMS-HT as well. If I keep printing a lot of soft filaments it seems like it’d be worthwhile. Thanks again for your help!
I use PTFE Tough, pretty hard to find outside US I think and now with the crazy taxes they will be expensive as hell too.
But until I can snap a photo…
My AMS HT is on top of the printer with the PTFE Tough running from the AMS HT to the 4-in-1 PTFE splitter and then there is another PTFE Tough running to the printhead. I switch the PTFE at the print head and have no fully permanent setup since I wanted to reduce drag as much as possible.
Keep the PTFE as straight as possible, every turn will create drag and might interfere with the feeding of soft filament. Seen other setups that also works so just try and have the PTFE without unnecessary bends and you will probably be fine.
I fought with this and I’m no expert but the two things I came away with that fixed it for me were:
Cold nozzle pull with PETG (not PLA). PLA doesn’t seem to stick to TPU well enough to properly clear it. Could have just been the particular TPU I was using.
Buy HF nozzles, they have identical print quality to non HF nozzles from what I can see but help sidestep whatever the issue is.
There’s likely a ‘proper’ answer to fix yours and my issues but that worked for me and if you’re not getting anywhere at least it’s something to try.
So 1.8mm^3/s did end up working for me. I swapped back to the standard 0.4mm nozzle since BBL says not to use HF. I don’t know why, but figured if that volumetric flow rate works on the standard it’s also cheaper to have a dedicated nozzle for TPU.
Why do you adjust print speeds? Doesn’t the max volumetric flow automatically cap speeds to Max Vol. Flow/Nozzle Dia./Layer Height? Or is there a reason to adjust speeds in the process settings?
MVS will cap everything, it’s the max amount of filament that can exit the nozzle per second I believe. If you try and push more than it can cope with then it’ll strain the motor and you’ll get that error.
Setting all speeds to 40mm/s is probably what they should be anyway for TPU. It’ll also help with keeping the finish nice as there won’t be speed changes that can cause an obvious change in finish.
You might get 50 or 60 if you’re lucky though. If it’s printing as expected you could slowly work up the MVS and print speeds to see where you can get to before the error returns or quality starts to suffer.
I’m no expert and am also learning along the way so take what I say with a pinch of salt.
Do you have a link to the BBL advise on not using HF? I’d be interested to read it so I can be more informed and maybe switch back to normal nozzles if it’s not good.
Sorry I may not have been clear on my question. I understand why you want to cap the maximum speed to 40mm/s, my confusion is why you need to set those values in the process parameters. If you set max volumetric flow rate in the filament that should cap the speed automatically. For example at 1.8mm^3/s the maximum linear movement speed (for .24mm layers) should be capped at ~19mm/s. What does changing the speed in the process parameters to 40mm/s do?
I’m not using their TPU either, but assume it applies to other TPUs of the same shore hardness. But in any case, I was not able to achieve any more speed with the HF nozzle and it’s over twice the cost of the standard so it seems like there’s not much reason to use the HF even if it does work.
True, it likely does apply to others too. I’ve not found any issues with the filament I’m using and HF nozzles. I wonder what the problem with HF nozzles are and TPU, then I’d be able to see if it’s something that affects my setup or not.
Definitely no point paying out for HF if there’s no benefit. I have them paid for and installed now, they work for my setup and ‘appeared’ to solve my extrusion issues.
I’ll try switching back to the normal nozzles as there is probably benefits to standard over HF if they’re recommending not to use HF.
HF nozzles tend to generate higher back pressure when printing flexible filaments, which is why they’re generally not recommended for those materials. I still use HF nozzles ranging from 0.4 mm to 0.8 mm for flexible filaments, but you need to fine-tune the temperature, cooling, and other parameters to suit your specific setup.
The reason is that a standard nozzle has a longer melt zone, allowing flexible materials to soften gradually and flow evenly. HF nozzles, on the other hand, have a shorter melt zone designed for high-flow printing, which can cause partially solid filament to meet molten material, increasing resistance and back pressure.
