Overshooting set temperature whilst drying filament

Bambu Lab AMS AMS 2 Pro
1

The past few months I have noticed the AMS 2 Pro having a “thermal runaway” at least on 3 occasions.

Could have been more as there are occasions where I set it on drying (PLA 45°C) for 12 hours and don’t pay attention to it for the rest of the day.

But as stated before I have noticed on a few occasions that the temperature surpasses the 45°C, by a lot.. without the AMS doing anything about it.

Attached you’ll see 2 separate occasions where it reached 60°C, however I have caught it once at 72°C even.

Anyone else experiencing this?

What concerns me about this is, well for one it can mess up your rolls of filament but mostly, that the AMS nor the printer warns you about it happening.. The display just happily shows you set it to 45 and it is current 60/65/70/75 degrees. How is there not a (software)limit built in to prevent this? and if it happens, to just kill the heater?

2026-06-11 19.18.01
2026-06-11 19.18.011920×1079 219 KB

2026-05-19 22.42.14
2026-05-19 22.42.141920×1079 130 KB

AMS 2 Pro over temp for drying
2

are you sure that reading is correct ? already tried to put another device into the ams to verify the temp

3

Looks like an urgent case for a ticket. Even if the temp was actually lower, this really should not be happening.

Does the filament bind slightly on the spool? After 6h at 60°C, I’d expect some binding between strands of PLA.

:four_leaf_clover: & :crossed_fingers:

4

PLA got weak and limp and the AMS 2 Pro lid felt warmer than usual during PLA drying. The photos of the printer screen were in both occasions taken after the lid was opened to air out some heat quickly. So yes.. I am sure they were 60+ degrees.

And even it was a misreading on the sensors part, should the firmware not intervene and shut down the heater?

5

Weirdly(and luckily) enough no binding, but definitely softened the PLA and on that >70°C occasion, the filament that was in there was deformed, and no longer straight strands on a spool, but a mess of spaghetti.

1 Like
6

Definitely file a ticket.
Support | Bambu Lab

I did not need to do one for a while but the experience can be a bit frustrating. Nevertheless, with an issue like that, Bambu should prioritize it reasonably quickly.

:four_leaf_clover: & :crossed_fingers:

8

To be honest I feared their support, judging by this forum, before buying Bambu.

I have had my fair share of problems, with a brand new machine.. which should not happen. But that is a different discussion.

Support always has been fair and fast with their help/replies.

Haven’t filed a ticket yet since I only noticed it happening a 2 times before today.. but perhaps after 3 times, I should no longer ignore it or label it an incident.

9

Well, my single experience started at the worst time of the year: X-mass 2024 (or 23?). And I got the full load.
However, when it finally became constructive, support was really helpful and professional and I got my printer replaced in May, deposit returned within 2 weeks. Fortunately, the printer was still within warranty. That is also a good reason to file a ticket quickly. If I had filed mine 2 days later, it would have been a very different thing.
You should also keep in mind that many who have issues with support post while those who have a good experience don’t.

Not sure I’d be able to cool my cool after that. Two incidents ending at 60°C, one at 70°C, …

:four_leaf_clover: & :crossed_fingers:

10

So my question is why you still kept using it?

File ticket please, and stop using it until after you get a reply.

Also not sure “runaway” is the correct word as that implies catastrophic loop.

Like, if you were concerned, why would you just carry on? Were you actually concerned?

1 Like
11

Yes he’s using “thermal runaway” when he means “it gets too hot”, probably annoying only to engineers. Like everything is “layer shift” to beginners, I think we know what he really means.

1 Like
12

What a nice place this forum is turning into.. My goodness..

Not that there is any need to justify myself, but FYI I am an electrical engineer that has been 3D printing since the early 2000s.
The only reason the AMS is not spiraling out of control is because the heaters don’t have the power to go much further than the 72 degrees I encountered that time.

Excuse me for “misusing” the word “runaway”.
Please grace me with your forgiveness, for I have sinned.

I hope you superior guys feel so much better now for correcting my terrible mistake.
Enjoy finding another topic to troll, instead of having a meaning full discussion.

2 Likes
13

As some background, People are a little sensitive to the words thermal runaway because other product lines [A1] have had versions that are having a legit AC board thermal runaway, to the point of definitely smoldering and allegedly ignition.

And on topic, make a ticket, something like this you probably want their input on. Assuming the temp readout is functioning and is legit way higher than the setpoint, its likely some sort of firmware failure and/or a real esoteric hardware failure

1 Like
14

To be fair, you said you had concerns and you already had a ruined spool of filament. That user suggested (see quote below) that you file a ticket and stop using it until hearing from BL. The user was engaging in meaningful discussion.

The question on the term thermal runaway was raised secondary to the main suggestion. Based on what I can see, your last post was a bit dramatic and was only about their questioning of that term.

15

Thank you for adding context @Bullocks

@DWdesigns I get what you are saying, but just pointing someone to support is not really constructive in my opinion. No need for a forum for that. Parallel to a support ticket we could exchange ideas here what the cause could be, giving other user a place to pitch in. Maybe share a similar experience and what support told them etc. Especially with people claiming to be engineers, why not talk about what technical means could be the root of the problem (if any). In my opinion labeling someone as a beginner or not an engineer or whatever is not really helpful. Would be nice if “beginners” and non engineers feel free to have a discussion on here as well.

In too many topics on the forum people are being bashed for asking the wrong question or stating the wrong facts/experiences already. Not very constructive.. Sure a question or topic might feel dumb to you, or asking/stating something obvious to you. Fine.. Move on, go read something that does interest you. I am sure there are plenty topics for you. No need to make your point and skip a substantive response, just for the sake of posting. (Not referring to you here, just in general)

So as a bit of an update on the matter: a ticket has been issued (and a response already received, nothing meaningful yet).

My experience with support is they issue a spare part and get it over with.
Whilst I am genuinely curious what could be causing this.

In the meanwhile I have actually managed the reproduce the issue, firming my belief it is a bug in the firmware. If that is indeed the case other people should realistically have come across the issue as well.
Hence me opening a topic and seeing of others like to pitch in with their experience.

Anyways, in case anyone comes across this post and is curious about the issue I tried to discuss:

I have an AMS 2 Pro and AMS HT connected to the H2D.

And have (today) been able to reproduce this problem 3x in a row, with the same results:

When I set the AMS HT first at 60°C & then AMS 2 Pro at 45°C it works fine, as expected.

But when I set the AMS 2 Pro at 45°C & then AMS HT set to dry at 60°C, the AMS 2 Pro goes higher than the set temperature (>60°C).

Awaiting for support to reply again to my latest findings.

16

You can have your opinion, but a lot of times the suggestion of contacting support has been given to users posting problems on this forum. It is very much a construcitive way to help.

1 Like
17

Sorry I didn’t mean to offend, I was just trying to tell him everyone knows what you meant.

I don’t know what is wrong with your unit, make a repro and submit a ticket, and then escalate it if they don’t seem to be taking you seriously. They should want to know about any external power supplies plugged in or if you are powering it from the printer only since that is a question I would ask.

Personally as an experiment I would delete the AMS’s, turn the printer off, unplug them, turn the printer on, plug the “last” one in first, then plug the old last one into the the one just connected. I realize you might not know which one is “last”, that is an addressing thing established by the order you plugged them in initially, not their physical position on the cable chain, but that’s what I would do since it looks like some kind of addressing error in the firmware.

18

Update: got a reply on my ticket.

With an apology for it taking a few days to reply, which was caused by discussion within the team and verifying/replicatig the problem I had reported themselves.
They have concluded and confirmed it is indeed a firmware bug and they are planning to resolve it in a future update.

Until then the advice is to either use 1 AMS at a time (for drying I mean) or have them run the same set temperature.

1 Like
19

I had a creality space pi that acted like this. It wound up being that the wrong PTC heaters were installed in it. I live in a 120v region and the PTC heaters in my space pi were those meant for 220v regions. The temp would heat up very slow and overshoot by a lot. Sounds very familiar to what you’re experiencing. Swapped out the PTC heaters for the correct 110v versions and all was good.

Maybe your ams ht has the wrong PTC heater installed in it?

20

Except in the case of possibly dangerous issues or situations. Not piling on but wanting to clarify.

Forums are great for passing on knowledge about minor issues or ways to do things better, but are terrible for most things having to do with safety because so many may not appreciate the dangers or have actual technical expertise to recognize or advise on safety-related issues.

Accepting advice on a forum for safety-related things is a great way to get in trouble. Not trying to slight anyone on safety advice but you never know if someone actually knows what they are talking about. There’s wide ranges of responses from well though out and informed to flip and totally incorrect.

With something going into “thermal runaway”, most examples of that result in fire or dangers of fire. As others pointed out, that is a dangerous situation had you been using the terminology correctly. The best advice anyone could give with something that could result in property damage and even loss of life is to advise contacting support and unplugging the machine, and keep an eye on it until it is cool.

The term used was incorrect for the situation, apparently, but the responsible approach by those who replied is to accept that there are extreme temperature issues of some kind that could be dangerous. It really was the correct advice.

And ultimately it was something nobody here could have really helped with since it’s apparently a firmware issue of some kind. Nobody here would know that.

You got the best advice here anyone could give. Faulting people for not advising you differently — and incorrectly — is not the best way to deal with a potential safety issue.

1 Like
21

It wasn’t meant as me disagreeing with the advice itself. Or as not appreciating anyone replying with said advice. What I meant was, that it was obvious for me to contact support about it. I would not come to the forum hoping someone would offer help solving it and I would not need support because of it.

I was looking for people with similar experiences, whilst contacting support in the mean time, since to me it seemed like a bug making it likely that I was not the only one experiencing this.

If others reported the same, perhaps they could tell me what the reply from support was. If it was a know issue or not and how to go about it.

But it seems that my misuse/interpretation of the term “thermal runaway” has put this topic of on a bad start which it unfortunately doesn’t seem the recover from.