My printer stops extruding randomly, what am I missing?

oliof

@infiniteloop stock ender3 hotend has ptfe line go all the way to the nozzle, and with a bad heatsink fan and high retracts this can happen (plus clogs between the inevitably appearing gap between nozzle and tube when the cheap coupler gives out).

Alsp your after market tube is transparent and likely not proper PTFE, so it may not have the right inner and outer diameter.

Dont waste money on Capricorn tubes, not worth it. Get a nice after market all metal hotend (NF Smart or Phaetus Dragonfly BMS are decent choices), a replacement coupler for the filament feeder (or a whole new feeder), and normal ptfe tube).

Check the heatsink fan for proper operation.

With an all metal hotend reduce retract to 1mm or less. If you keep the ptfe lined one, keep retract at 4mm or less.

infiniteloop

@oliof Not sure of what you want to tell me:

I have an all metal hotend, and it has its merits if you want to print ABS, Nylon etc. at high temperatures, but it’s overkill for PLA, not to mention the inherent problems with an all-metal design. But that’s OT.
You are right in that blue tubes may not be worth it, but my point was not to recommend crap ("not proper PTFE") instead. In the context of this thread, I just wanted to alert the OP of potential problems with a deformed tubing.
Thx for your recommendations on retraction settings. Due to the fact that I mainly use PETG, mine are different.

oliof

@infiniteloop i meant to mention OP, sorry.

deckingman

Here is another vote for not using Capricorn tubing. I tried it one of my experimental multi-input hot ends and it was easily pushed out of the coupling by pressure build up within the hot end, whereas "generic" non-branded PTFE tubing did not suffer the same way. I'm not sure if that is due to the Capricorn being exactly the correct OD whereas the generic stuff might be a bit oversized, or whether the Capricorn is just a bit more "slippery". Either way, "generic" PTFE tubing was gripped better by the coupling than the expensive Capricorn tubing.

I did discover a way to fit PTFE tubing into a heat sink such that it'll never move but it's a bit extreme. There is a product called "Technicoll X-935" which is a two part cyanoacrylate and primer specially designed for gluing PTFE/PP/PE etc. Apply the primer/activator to one surface and the cyanoacrylate the other. BUT, you have about one second to get the PTFE into the heat sink. If you are not very fast, then the liner will be stuck part way inside and the only way it comes out is by drilling it out. This is the voice of experience talking...........

re_tour

@oliof To be honest the tube came with the machine, I just dug it out of the box and changed the old one. I had my suspicions tho that it may be a crappier replacement, since the one which was fitted to the hotend looked different.

I bought a new coupler as @infiniteloop pointed out that it may be shot(which it absolutely was) and now I'm printing a calibration cube. I don't want to jynx it, but so far its seems to function normally again.

The cooling fan is a noctua 10x10 fan with a step down converter, which I don't think is an issue(but I've been wrong before sooo...)

I may buy a dragonfly if I have this issue again though since the coupling cost me 5€ and the dragonfly is around 40€ but would eliminate this kind of problem altogether.

Thank you for the tips!

re_tour

@infiniteloop Alas, failiure. Around layer 40 it yet again got stuck inside the tube and couldn't move further down the hotend. I've had enough with this hotend, as @oliof said I'll try the dragonfly and see if this keeps occuring with that hotend also.

Deep breaths.

infiniteloop

@re_tour

I'll try the dragonfly…

The dragonfly is certainly fine, hopefully it can eliminate your problem. With your current setup, there would just be two suspects left: nozzle too cold or hotend too hot. However, if you ran your test without the part cooling fan, we can exclude the nozzle from the equation. OTOH, the Noctua should well do its job to cool the hotend. My best guess is that the air flow might be less than optimal - from your photo, I can’t spot the details. With the dragonfly, that’s all obsolete, but maybe you keep an eye on this potential problem when you print your next fan duct. Good luck, and keep us informed of your progress. Hope to see some perfect benchy soon, not another cabriolet version

re_tour

@infiniteloop I will look into the issue a bit further during this week, I have ordered the dragonfly, but that will only arrive on the next week. Until then, who knows, maybe I find the offending factor and can eliminate it. I will be back with updates and hopefully with a complete benchy!

re_tour

@re_tour said in My printer stops extruding randomly, what am I missing?:

@infiniteloop I will look into the issue a bit further during this week, I have ordered the dragonfly, but that will only arrive on the next week. Until then, who knows, maybe I find the offending factor and can eliminate it. I will be back with updates and hopefully with a complete benchy!

After two failed attempts I let the hotend cool to see why the filament gets stuck and guess what! The ptfe tube has no molten pla in it. So whatever happens, it happens below the ptfe tube. Since the thermistor and the heater are working fine, I can only guess theres clogging...which is just not possible, I already changed the nozzle and saw no sign of residual pla in the entire hotend which could cause clogging.

If the issue will be the same with the dragonfly there may be some petrol and cigarette lighter involved in the following troubleshooting step.

infiniteloop

@re_tour

So whatever happens, it happens below the ptfe tube.
I already changed the nozzle and saw no sign of residual pla

Between these two parts, the heat break should do its miraculous work. But, as you state:

I can only guess there’s clogging...

Exactly. Usually, a hotend simply works. If it doesn’t, we can only hypothesise what goes wrong. Some time ago, I spent weeks figuring out that the chemistry of a certain PETG filament slowly built up unflamable debris in the hotend - I couldn’t even torch it (and believe me: I had enough petrol and cigarette lighters at hand). So: strange things happen.

That’s not to say your case is similar. You just encounter a similar mess. So, hopefully, you can resolve the issue with an exchange of the hotend. Anyway, the ultimate way is follow one hypothesis after the next, to take one step at a time. At the end, we know better.

deckingman

@infiniteloop said in My printer stops extruding randomly, what am I missing?:

............... Some time ago, I spent weeks figuring out that the chemistry of a certain PETG filament slowly built up unflamable debris in the hotend - I couldn’t even torch it (and believe me: I had enough petrol and cigarette lighters at hand). So: strange things happen..........

Someone much more knowledgeable than me about chemistry, once told me that if you over heat PET-G, it will "carbonise" and that the only way remove said carbonised filament is by drilling it out (or use explosives) . So perhaps lowering the temperature of your problematic filament might prevent a recurrence.

infiniteloop

@deckingman

So perhaps lowering the temperature of your problematic filament might prevent a recurrence.

Thx for the explanation: that's gold standard, deckingman style - appreciate it. Well, I tended to apply explosives: new nozzles and tubing, the filament is on my eternal blacklist.

oliof

@re_tour said in My printer stops extruding randomly, what am I missing?:

The cooling fan is a noctua 10x10 fan with a step down converter, which I don't think is an issue(but I've been wrong before sooo...)

I would not rely on a Noctua 40x10 for hotend cooling. A 40x20 may work if there is free airflow "behind" the hotend. Its quiet because it has almost no air pressure.

re_tour

@oliof That noctua has been there for almost half a year and there were no issues with it.

@infiniteloop So...today I got back to troubleshooting after leaving my printer alone for a few days so that my frustration would go away somewhat. I also picked up the phaetus dragonfly bms hotend today and put it in my printer.

I set it up and everything, bought a new generic ptfe tube (no fancy capricorn stuff) and after I put everything together I started the print...and yet again...the issue was the same...I couldn't believe my eyes...I was in utter disbelief...how, what, why...I have changed everything, there can't be any issues...I even pulled the ptfe tube out and put the filament in directly to check if there was any clogging, but there wasn't. And then...when I clicked on extrude on the duet surface, I heard the extruder stepper moving, but the gear didn't turn...then it hit me: I have upgraded the extruder part with a metal kit, which included a new gear for the extruder, but instead of the factory pressure fitted one I now had one with two miniature screws...which I didn't tighten strong enough, which (sometimes!!) led to the gear slipping...not always, just randomly.

So, turns out that I have been very negligent when putting the printer together the last time, it could've been one mistake or multiple failures at once which caused the print failures.

At long last, I can finally present a fully printed test cube! I really hope that this is where this story ends and that – at least for a while – I can print my creations in peace again.

9fdb6571-9ce1-451c-b6ee-dd4b04464da9-kép.png

Thank you @infiniteloop ,@oliof , @Exerqtor ,@jay_s_uk and @deckingman for the help and keeping the spirits up!

Exerqtor

@re_tour
Glad to see you finally got it sorted out! It's easy to loose track of where and what might be the issue if more than one mod/rebuild/new part is being introduced without testing between each step, I've done it myself more than once lol.

infiniteloop

@re_tour Nice little cube - reminds me a bit of the first iMac… sorry, last century

Sure I’m glad you found the culprit. In my eyes, the most valuable achievement after this tour de force is that you now have profound knowledge of how each and every component of the filament path works - you’ve really touched every single aspect. Congratulations, truly

re_tour

@infiniteloop I cheered too soon...

Now that I started printing a part that I made, I started having problems.

It seems like I've got some heat creep going on, however, I can't reproduce the issue manually. I already set like 0,5mm retraction, I have the same result.

I tried pushing the ptfe tube as far as I can down the hotend but it won't go any further...

infiniteloop

@re_tour

It seems like I've got some heat creep going on, however, I can't reproduce the issue manually.

A heat creep develops over time, sometimes, it needs an hour to show up. So you’re right: it’s hard to reproduce in tests.

Maybe it’s a good idea to take a break and to summarise your current configuration: I understand that you have a working extruder, a dragonfly hotend and fresh PTFE tubing? Now, what kind of filament did you use, at which speed, which temperature?

Assuming all of this is fine, we are left with a topic @oliof brought up: is the cooling of the hotend sufficient? Can you publish some images so that we get an idea of the air duct and the air flow before and behind the hotend?

re_tour

@infiniteloop Yet again, I must say that I did something really stupid, which I just came to realize....

So...the thing is, that to make my life easier during wiring, I put the part cooling fans and the hotend fans on the same step down module...which means that unless I manually turn the part cooling fans on, the hotend fan won't be on until like the second-third layer or so...which I guess is plenty enough time for a clogging to happen, but correct me if I'm wrong.

But anyway, yes. Extruder stepper motor is working, motor gear fixed in place, marked with a sharpie to know if its slipping. Ptfe tube check, hotend is good, coupling is good. I've tried different PLA-s, Gembird, 3D Jake eco PLA and filamentum. Hotend was set to 220°C with the filamentum one and 210 with the gembird and 3DJake.

infiniteloop

@re_tour

the hotend fan won't be on until like the second-third layer or so...which I guess is plenty enough time for a clogging to happen …

The hotend fan should be triggered at 45°C - you'll have to add the time until the hotend reaches printing temperature to the time for the two layers. Let's see what happens after you've rewired the fan …