Feed Issue???

[Hammbone]

Ok... So I have had this happen to me a few times now. I have tried to search the forum but must not be using the correct wording to find any similar posts. Originally I thought this was because of clogged nozzle. Now I am thinking not so much.

I started a print last night before bed but woke up to filament all balled up near my extruder. It was a 9 hour print but probably on successfully printed an hour of it before the problem happened. I checked the nozzle by bringing it up to temp and it feeds like butter so I know it isn't a clogged nozzle. That being said, I am at a loss of what may be causing this problem. It appears that the filament stops feeding down the tube and starts to feed to the outside. This occurred on the same print before this one too but near the very end of the print. I thought perhaps my bowden tube wasn't pushed in far enough on the extruder end, so before this last print I pushed it as far as I could to where it was almost touching the cogged gear and still ran into the same problem.

I am printing with PetG and there are no signs that the extruder has shredded the filament like one would see if the nozzle clogged and the gear kept chewing on the same spot.

I am running a geared extruder. Could it be that my retraction speeds are too high?

Thoughts?

[img]http://i42.photobucket.com/albums/e303/CrashHammbone/20160818_082635_zpswxpukrzi.jpg[/img]


[img]http://i42.photobucket.com/albums/e303/CrashHammbone/20160818_082638_zpsj5mha0hc.jpg[/img]

[Nylocke]

Are you using Glacian's design for the EZflex? I've noticed it has an issue where the hobbed gear isn't hugged as tight by extruder body as it should and the idler bearing sticks out a bit too far too, so theres a lot of space that filament can slip out of. I've had this problem with Ninjaflex, but not with ABS/PLA. I haven't tried PETG. Is the teflon tube all the way up to the drive gear and do you have it cut to slide up into the gap between the drive gear and idler? If not that might help a little.

[Qdeathstar]

Hmm, I don't know.. Seems like a layer adhesion/temperature issue to me

[Qdeathstar]

It got high enough off the bed that the layer previously layed wasn't as warm and the new layer didn't stick.

I'd try to increase print temp/slow down print speed.

[Hammbone]

Are you using Glacian's design for the EZflex? I've noticed it has an issue where the hobbed gear isn't hugged as tight by extruder body as it should and the idler bearing sticks out a bit too far too, so theres a lot of space that filament can slip out of. I've had this problem with Ninjaflex, but not with ABS/PLA. I haven't tried PETG. Is the teflon tube all the way up to the drive gear and do you have it cut to slide up into the gap between the drive gear and idler? If not that might help a little.

You are correct Nylocke. It is Glacian's design. Do you have/know of a design that works better that may remedy this problem. I have had this problem with ABS since my switch to this design as well but it only happened after an actual nozzle clog.

Part of me thinks the gaping you mentioned and heavy retraction may have caused the filament to get out of alignment. Plus PetG isn't as hard as ABS or PLA so there is probably more play in it to contribute to the gaping problem. I had recently increased my retraction speed to 45 mm/s when I switched to PetG so I started a print with my old retraction of 30 mm/s to see if it may help. I hope to return home from work to a successful print. The tube is not quite touching the drive gear. I did cut it per your advise just before I started the print.

Hmm, I don't know.. Seems like a layer adhesion/temperature issue to me

It got high enough off the bed that the layer previously layed wasn't as warm and the new layer didn't stick.

I'd try to increase print temp/slow down print speed.

Thanks for the suggestion but you are seeing raw filament that has not passed through the hot end. It is coming out of the top at the extruder. So it isn't a layer height issue. Sorry for the bad camera angle. I can see how one would believe it to be coming out of the nozzle instead.

[Nylocke]

You could try using my original design strapped to the adapter plate glacian designed. The hobbed gears are the same diameter abouts, though you might have to shave some material off so the set screw doesn't interfere with the extruder body. I haven't tried this before but things should align and it might work better since I designed mine so the body would hug the gear. I would stick with the idler glacian designed though, its a bit more reinforced.

[Qdeathstar]

Oh.... Damn.. At least you didn't waste filament

[Hammbone]

You could try using my original design strapped to the adapter plate glacian designed. The hobbed gears are the same diameter abouts, though you might have to shave some material off so the set screw doesn't interfere with the extruder body. I haven't tried this before but things should align and it might work better since I designed mine so the body would hug the gear. I would stick with the idler glacian designed though, its a bit more reinforced.

Ok. I will see if I can get it to cooperate enough to print your parts and combine them with Glacian's. I came home to another failed print. It made it further than it did in the first pictures. The lack of consistency is troublesome. lol

Oh.... Damn.. At least you didn't waste filament

Definitely a plus but it is a pain to re-spool all curled up like it is. lol

[Hammbone]

Nylocke, your plate printed like a champ. I did have to dremel out a little bit more space for the cogged gear and make room for the set screw as you mentioned. It hugs the gear perfectly, sitting perfectly flush against it. I am trying to print the hilt again, so fingers crossed it is successful this go around. I'm tired of wasting filament lol.

[Nylocke]

Hopefully it works. I checked it last night to see if it would fit myself (had a spare of both sitting around).

[Hammbone]

I can certainly see how yours is an improvement, especially with soft filaments. The filament pretty much has no choice but to go into the bowden tube. lol. Judging by the tightness in the curl of wasted filament I had, I believe it was curling under the cogged gear as it continued to extrude then just threaded itself off and kept going. That being said, I'm not sure how this couldn't fix the problem.

Now the only thing I need to do is rework Glacians adapter plate to allow me to use the manual feed gear that attaches to the extruder to make it easier to feed filament. Especially with how tight of a fit your design is. If I posted the remix of your design and Glacians design together, do you think anyone would be interested in it?

By the way... I REALLY appreciate all of your help with this matter! And for all the other insight you have posted in the forum.

[Hammbone]

Ok... So I stopped by at lunch to check the print and found it had failed again. But it provided a little more insight. I no longer believe this to be a feed issue. Thanks to your extruder design it didn't start coming out the side and just chewed through the filament instead. This confirmed it is not actually a feed problem while improving my feeding for the future. lol.

My read out indicated the hot end was still set to the desired temperature and the reading from the thermosister was within 2 degrees of the setting. I cancelled the print and tried to take a reading of the hot end before it cooled too much. My thermocoupler indicated the hotend was about 20 degrees colder than it should have been reading compared to previous readings I have taken at the same temp setting. This would explain why it stopped extruding being out of the melting temp and all, plus why when I brought it up to temp later it didn't reveal a clog. I think my thermosister may be going bad. I have never had one go bad. Is this something that could happen if it was going bad?

[Nylocke]

Normally they don't go bad unless you overt-temp them. Try taking a temp reading from inside the hotend (stick it down into it and make sure its in the nozzle area) when its up to temp. Its not uncommon for them to run a little low. What temp are you printing at?

[Hammbone]

Ok. I will see if I can get it to feed into my hot end to see what it reads. I am printing the PetG at 246 by meter measurements taken where the heater block meets the nozzle. But that means my hot end is set at 265. So if I have been measuring incorrectly and my temp is actually closer to 265 then that could be the problem. It could be heat soaking into the CZ causing a clog during retraction.

The weird thing is, I printed this same print on the same settings and completed 7/8 of the print flawlessly. Now it struggles to get past an hour or so on the same print. I can bring it up to temp and extrude filament with ease which made me think maybe the thermistor is shorting out during the print.

On the flip side, it printed your design without any problems though it is only an hour print and doesn't have as much retraction as my other print. My retraction is set at 40 mm/s and a distance of 1.6 mm. I have a transition of 2mm on a Prometheus Hotend.

[Xenocrates]

Have you ever applied thermal paste to the heat break/heat sink interface? Is the fan still running properly (And mostly at speed)?

That being said, someone recently had a similar problem with the thermistor reading low. Are you sure it has good thermal contact (if it's an E3D like it looks to be, perhaps a ceramic based thermal paste in the thermistor hole would help)?

[Hammbone]

I am running a Prometheus V2. It didn't call for any thermal paste on assembly so I haven't used any. I will check to see if I have any laying around. I will also make sure it is making good contact with the heater block. I am thinking of switching to a threaded thermistor as well. Fan is still running properly.

[Hammbone]

Just an update on my temperatures.

I attempted to measure my temps as Nylocke suggested however, I was unable to get my thermocoupler far enough into the nozzle. That being the case the Prometheus V2 has a hole just under the thermistor hole. Figured this to be a good place to measure temp and my thermocouple fit without any problems.

Here is a temp chart reflecting my numbers...

Thermistor reading Thermocoupler Measurement
270 260
265 256
260 250
255 245
250 241
245 238
240 234

I plan to chart the rest of the temps but for PetG this should be sufficient. These definitely seem more accurate than my prior measurements. I wish I could get measurements from inside the nozzle though for the most accurate measurements.

Would I be correct in thinking that the nozzle could be clogging from too much heat soak near the CZ because I am printing too hot?

[Xenocrates]

If the thermistor is reading higher than the thermocouple, I would think it's the other way around, it's clogging because it's too cold.

[Hammbone]

If the thermistor is reading higher than the thermocouple, I would think it's the other way around, it's clogging because it's too cold.

Even though I am adjusting my slice temp to match the desired measured temp? So measured by my previous method, I thought I was heating to 246 but by my new measurement I was actually heating to 256, possibly higher because of my thermocoupler placement. My PetG calls for 230-250 so I would be over heating at 256. I had a problem in the past with ABS clogging my CZ because of overheating and it retracting into the CZ. So I thought it may be a similar case.

[Xenocrates]

You posted a table with the thermistor and thermocouple readings which seemed to show that the thermistor was reading higher than the thermocouple, which would mean that say you wanted 245 as a desired temperature, you would have to set 255 rather than 245 directly, as the printer reads the thermistor.

So if the table is the other way around, and the first column is thermocouple measures, then yes, you are correct. I'm going to suggest that either way, you do try adding some thermal paste to the coldzone/nozzle interface, and potentially play with your geometry to reduce heat soak.

I would aim to get the thermistor to agree with the thermocouple, personally. That gives you a way to check it periodically for further drift and make corrections, as well as having a standard to check against other people with. But that's up to you. Playing with print temps, since this apparently is a quickly recurring issue should be pretty easy. Drop down to the lowest temp you can manually extrude at, and increase it until you have results you like.