How to handle AMS feeding errors mid print?

Hi,

maybe iḿ just dumb and missing something out but how do you guys handle such situations?

When for example got a tangle, the printer stops in error…but when i try to “de-tangle” the spool i cant

because for this i have to unload the filament. Otherwise the feeder from the AMS starts pulling like crazy when i try to get the filament out. But in an error paused print i CAN´T unload the filament over the menu! It simply don´t let me do that….i have to cancel the print to get this done…seems dumb :frowning:

Am i missing here something?

Greetz

Let me ask the most obvious question: Could it be that your filament had already been unloaded without you noticing? Some errors cause the printer to unload the filament on its own, so you neither have to nor can select it.

Unload means that the filament is pulled back far enough that it is no longer in the nozzle. Was the filament still in the nozzle? It doesn’t rewind all the way back - just from the nozzle and extruder gear.


If it has already unloaded and is no longer in the nozzle, you can manually rewind the spool by hand. Depending on which AMS you have, you’ll need to push the mechanism so that the filament can move freely. Please check the wiki for more information.

Above post by retrosharky is correct.

Further information, filament tangles do not really happen with reliable manufacturers. If you have a lot of problems with this, review your filament handling procedures and/or choice of filament supplier.

Well this happens not all the time. In about 800h print time since i own the H2C i maybe got it 3-4 times. But i wanted to clarify it.

If it was still in the nozzle i can’t tell but i DID pull the lever on the AMS pro and HT (it happened on both) and still its was pulling instantly when i tried to fiddle it out. I was frightened to damage the feeder drive so didn’t fight it and gave up the print instead.

No you don’t…

He has space to move it, when i got these errors the tangled part is already direct over the feeder inlet so you can’t move it. Only way would be maybe to unroll the spool if possible enough to take it out…

But the initial question was if i getting something wrong about the unload function and answers suggest that you can’t do it as long as the printer thinks filament is “in use” which seems like a bug for me.

I could be wrong but it seems to me that in order to save your print, cut the filament, resume, let the filament runout, while you untangle spool. When the printer runs dry, it would pause and ask for filament which you would then reload and resume.

I am having this same issue. I have two posts on this right now. What I am finding is that the filament already used on current or previous jobs already loaded into the AMS is cut square by the print head cutter. Then when called back to be used, it gets stuck at the print head, or the track switch (H2c). For me, I have noticed this error on both my left and right nozzle as I ran tests to see.

Yes, the AMS needs to be able at any time to retract filament all the way back up and out and if the only sensor in the system is at the print head itself, that is an issue. We should be able to automatically retract fully at any point.

The solution I have come up with is pull the filament out manually, put an angled cut in the filament and reload.

I think it’s intentional that you have to pull it out manually, as explained in the Bambu Wiki. If an error occurs, users are expected to think about it and work out what has gone wrong.

Having a “rewind” button could cause significant damage. Imagine if the filament got tangled, wasn’t cut, or was still stuck in the nozzle, and the AMS was pulling on it like crazy. That can only go wrong and would probably even break the motors.

Even though it’s a hassle, it’s the safest way to deal with a problem. Otherwise, I’d expect to see tons of AMS units that are broken because people just keep pressing “rewind” like crazy. Then we get comments like: “I’ve already pressed it ten times, and now my AMS has burned out.”

Larry_Hair has perfectly summarized how to deal with the problem.


Beside: As mentioned earlier, tangled filament is almost always the result of user error - specifically, failing to maintain the proper tension on the filament while loading it. It is virtually unheard of for tangled filaments to be shipped, because that is technically almost impossible.

That’s why MakerWorld is also stuffed with filament clips.

This topic is as old as 3D printing itself, but of course, not everyone realizes that you always have to maintain the tension on the filament - even when loading and unloading it, as well as when storing it.


Strictly speaking, it’s not really a tangle but more of a “slip-through.” If the tension isn’t maintained, individual loops will slip out of place. There result is a tangle - the cause a slip-through.

Yes Larry_Hair’s answer seems to be the best workaround. As said earlier, i only wanted to know if i’m missing something in the functions.

But i must object the point with “always” user fault. I mostly use Sunlu and out of around 30+ Spools i burned i got these 3-4 were it was inside a fresh spool! I also use big bearing balls inside the spools to prevent them from jumping when they get empty.

To me, pulling the filament out manually is what is going to break the AMS. I just came home and started the 3rd print. Did the exact same thing I described, the blunt filamnet made it to the printer and would not load.

After getting the error, i pulled it back out of the printer, back out of the track switch and back out of the HT. The manual pressure exerted even while depressing the tension release inside the HT is concerning.

I cut the filament at an angle and it went right through. In one of my threads, one of the responses wanted to see the PFE tubes. I am starting to think the 3d printed 4 in 1s and the 3d printed speed couplers and all the non-bambu products may be adding to the issues. I will be pulling all that off and running some tests but watching the filament load, it does seem the extruder itself is where it is stopping.