"Peace of mind isn’t at all superficial, really…It’s the whole thing. That which produces it is good maintenance; that which disturbs it is poor maintenance…If you don’t have [peace of mind] when you start and maintain it while you’re working you’re likely to build your personal problems right into the machine itself.‘’
- Zen and the Art of Motorcycle Maintenance
I needed to write my own notes for this problem so figured this is as good a place as any, might be helpful for someone else. It is fixed but not necessarily solved - meaning my prints are stable but not sure what fixed it. Looks like there’s a few posts around and a general consensus that this is a common problem where heat creep causes the filament to overheat in between the drive gears and extruder.
The issue: Prints kept failing mid-print, where the filament would stop extruding but the printer wouldn’t figure it out and so the printhead would continue working through the rest of the G code, just the motions without any extrusion. I am going to call this “heat blobbing”.
Specific conditions which caused this: Any and all prints, any and all filaments. Tended to occur with longer prints but happened with shorter ones too - I think it’s just a law of large numbers thing, and the underlying issue exists regardless of print length but presents more readily the longer the print time.
Background: I had a fairly upgraded X1C with AMS which was working great until around the 1000 hour mark. Upgrades included:
- Capricorn PTFE tubes in parts of system
- Glass riser with Wham Bam fire extinguisher
- Biqu modified print head cooling ducts
- E3D Obxidian nozzle
- Bento box
- Print head PTFE tube bracket thing
- Wham Bam PEI sheet
Around the 1000 hour / 6 month mark, the prints started failing a lot. Think it was two unrelated, but overlapping problems - heat blobbing, and then an unrelated PTFE tube wear out. The PTFE tube wear out was causing various “Unable to feed filament” issues. This was particularly problematic because the symptoms overlap with the heat blobbing (feed errors), and I thought it was one problem rather than two. I won’t go into the PTFE issues here, this is well covered. I eventually figured out the PTFE tubes should have been replaced 2-3 months ago and did this.
The problems abated, with no feed issues but remaining heat blobbing failures. This was also around 1-2 months of constant issues - as a result I decided I needed to go back to square one. I took off all upgrades, including the AMS and software side (I went back to the stock nozzle and filament profiles including stopping the “no poop” G Code). This removed all the “Filament feed” errors upstream of the drive gears, leaving only the heat blobbing and making it clearer that there were two errors here. I then focused on trying to fix this one. (this is also the relevance of the above quote, because I didn’t immediately figure this out lol, I spent a week or two despairing that solving the one issue I thought had figured out (and partially I did!) didn’t fix my printer, leading to the 5 stages of grief before doubling down)
So the main symptoms remaining:
- Filament feed failure mid-print, but printer doesn’t figure it out
- Extruder gears continue to turn downstream but won’t push filament. Can push filament upstream out of print head
- Clog in system, cannot manually feed filament after print failure
- Burnt blobs and other similar artefacts found on print or extruder up to failure point
- Filament blobbing or bowing out when removed from print head (I’ll attach pictures below)
The main distinguishing feature was the blobbing and bowing out and the nonstandard nature of the clog. I got pretty quick at disassembling everything as I had to do it 30+ times, and these consistencies emerged:
- The extruder was sometimes clogged, but usually not. I could disconnect it from the gears and extrude manually by pushing filament through with a no clogger tool. There was some clogging, but this was more down to my bad practices, and I’ve now taken up cleaning between each filament type change. I don’t think this was a root cause, more than causing heat build-up (explained further below)
- The drive gears were tough to move but not exactly “clogged” by which I mean that i) disassembling usually revealed very little if any debris, and ii) with some force, I could pull the filament out backwards. After getting the old filament out and with no extra cleaning, the drive gears could extrude with no problem
- The drive gears themselves looked very good to my eyes, and I swapped in multiple extruders including new Bambu Official ones, so these were isolated out
- There were cutter issues, it would usually get stuck or have trouble cutting. I swapped the blade (not for the first time) but the old blade seemed fine
- When pulled out, the old filament usually had a distinctive and significant blob at the cut end
It’s a bit of an overlap with timing but all this was happening as I was taking off the upgrades as I mentioned above. Swapping to the bone stock slicer profiles led to a big improvement, but I still had some large print failures around 4-5 hours in. I then took off the toolhead cover with the Biqu fan and started leaving the door open, and this has led to consistent printing - I’m 30+ hours in with no repeat of the error!
Diagnoses: I think, as has been said elsewhere, that this is being caused by overheating in the toolhead leading to heat soaking up in between the extruder and drive gears. This is an issue because the filament starts to melt, and gets bowed out into the small empty space between the two parts, where it gets stuck. Since this is downstream of the filament sensor at the top of the drive gears the printer doesn’t realize and keeps printing (and indeed continues to display green for the filament fill).
Causes: Excessive heat is the general problem, but I’m not sure exactly what causes it at the moment. The print head gets too hot, the heat soaks up, bows up the filament, and gets the whole thing stuck. Below are my current thoughts as to what exactly might be causing it, but I’m only ~30 hours into getting the printer working again, and don’t have much appetite for scientific testing at the moment so it’s all more hypotheses.
-
Ambient temperature - I live in NY and the temps have gone up by ~15c over the last 2-3 months. This has obviously contributed, and now my printer is consistently ~35c when not printing just because of residual heat soak from printing. There’s also the tangential issue of humidity doubling and tripling, meaning I have to be a lot more consistent about my humidity management, whereas I could be lazy about this with basically all filaments in the dry winter. Fixes: proper humidity management, AC and leaving door open with particularly sensitive filaments
-
Extruder heat, filament cleaning - I think being lazy about filament cleaning was contributing. I spent the first 1000 hours constantly swapping between filament types, from PLA silk to PA-CF, without much of any cleaning. This was a testament to the Bambu X1C which handled it like a champ, until it didn’t. The main contributor here is that the trash builds up in the extruder, causes inconsistent heating, and this travels upstream. **Fixes: proper cleaning diligence, cleaning filament between type changes
-
Biqu hotend fan ducts - I am definitely not trying to rag on this, it works well otherwise. I don’t know for sure, but I think it might have contributed to the issues. I think it’s known the stock hotend fan position means it draws up heat from the plate and is inefficient for cooling. I think the Biqu fan might be exacerbating this by increasing air flow to the hot end. This lowers temperatures at the extruded filament, but this temperature has to go somewhere and might go upstream somehow? The print I’m linking below doesn’t work with this mod unfortunately so I cannot test this directly until I make a modified version. UNTESTED AS OF YET
-
Toolhead cover - I installed the Biqu hotend fan ducts on the toolhead cover that came with the X1C. I was a bit of a noob and in the 50 odd times I’ve removed this, I’ve made a bit of a mess of the connectors. It is possible that the electronics have been messed up, or maybe the fan just needs clearing out. When I removed the Biqu ducts I did so by just swapping in a new toolhead cover. I am unsure if removing the Biqu or swapping the toolhead cover entirely was what made the improvement.
-
Slicer settings - I think retraction settings are the main contributor from the software side, though not a huge contributor in the grand scheme of things. I’ll try to download and upload my latest custom profile, but to be honest it suffered from “feature creep” and I’ve turned on a lot of beta OrcaSlicer settings which adjust retraction specifically. I think this abnormal retraction causes issues because the hot filament gets pulled out and draws heat up above the extruder
Conclusion - I think this is mainly a case of a lot of upgrades and modifications combining with bad maintenance to cause consistently abnormal heat conditions at the gap between the hotend and drive gears. The fix is to return to stock conditions, particularly paying attention to modifications that affect the heat profile such as nonstandard extruders, slicer profiles (especially those which affect retraction), and fan arrangements. Beyond this it’s about dialing in temperature settings and managing the internal temperature by small fixes such as leaving the door open. Potential specific testing for the future - I need to test this out properly with a thermometer in the future to see what the temperatures are in the chamber and specifically around the tool head. I think this heat build-up is likely temporary spikes, hence why it can go 4-5 hours without a failure, but it is consistent. Most likely it’s operating close to the edge of acceptable heat soak, and then some random factor pushes it up beyond.
There is also a potentially more general problem however, with the extruder fan engineering, where it’s position causes inefficient heat management in the printhead - once more extreme ambient conditions come in, this heat soaks the whole system. Engineering in an OEM heat shield so that it only draws in cold air from up higher above the print head is the best solution.
A smaller related issue is that the filament sensor is upstream of the drive gears, meaning filament faults in between the drive gear and extruder aren’t noticed. There might be a software fix for this which relies on the thermistor temperature variance combined with webcam AI feed monitoring to be able to stop prints when there is no extrusion. Haven’t looked into this just an idea.
I am by no means an expert, this is my first 3d printer. Please let me know your thoughts.
To-do:
-
Add back in modifications piecemeal to test what continue to be stable. This will likely mean adding a part in and running temperature towers, and then some type of repeatable LT test piece since as noted above it’s likely a random spike causing this
-
Install thermometers in various parts of chamber to more carefully monitor potential spikes. Including specifically: i) chamber by stock sensor, to test stock sensors accuracy, ii) hotend fan intake, iii) gap between drive gear and extruder, iv) extruder, to measure thermistor accuracy (might be underreading temperature)
-
Test fan shroud effect with thermometers, and then test versus Biqu and versus modified Biqu with fan shroud
-
Test old toolhead without Biqu or new one with Biqu to isolate which was the problem
Sources:
Biqu duct
