Severe under extrusion stops mid print consistently

[ericwilsonart 0]

I have tried about everything I can think of and have read many threads and tried troubleshooting all the suggested things via all the suggested methods.

What is happening consistently is the print is failing over and over. I have had this problem on both of my machines but more on one than the other. My guess at this point is that the temperature of the head is dropping or fluctuating. My reason for theorizing this is that the print color on a good print is almost consistently the same while all the prints that are failing have color variations and gradations and when I manually turn up the temperature the color changes.

What have I done so far?

1. Disassembled the head assembly and cleaned the brass nozzle out thoroughly till nothing other than brass was visible on the inside.

2. Disassembled the bowden tube and manually inspected the resistance of the depressed PLA. Result negative, no resistance was found.

3. Disassembled the feeder cleaned out debris. I doubt this had any effect on extrusion.

4. Tested the feeder with weights (20lbs) and found that of the said 22lbs limit the material slipped back at about 15 lbs. Either way I highly doubt this was the problem.

5. When printing and close to ultimate extrusion failure, the feeder stepper motor starts to 'tock' and slip back suggesting that there is too much load on the hot end. While the print is failing sometimes turning the temperature way up for a brief period of time, maybe a few minutes then turning it back down seems to correct for the problem for a while anyway then eventually maybe hours later the print will fail gradually by under extrusion to no extrusion. I have tried to increase the material temperature after extrusion has stopped and nothing comes out at that point.

6. It is not a blocked/clogged nozzle. If I abort the print and manually heat up the nozzle I can easily feed material through by hand with a thick stream extruding straight down creating a nice snake coil on the build tray.

What were my settings?

Quick print profile settings with ONLY the infill adjusted.

i.e.

Quality

Layer height (mm) .06

Shell thickness (mm) 0.8

Retration Enabled

Fill

Bottom/Top thickness (mm) 0.6

Fill Density % 7

Speed and temperature

Print speed (mm/s) 50

I have tried to print small to large parts at these settings and with the default infil of 5%-20% all of which have failed. In the middle of all these failed prints I have had two successful prints with even color and absolution no extrusion problems. This is probably the most frustrating part, not knowing why it randomly without changing any variables will print successfully 20% of the time.

I'm out of ideas at this point.

Here are some images:

A successful print next to a failed print at the same settings.

http://imgur.com/LAu2mz1

Here are a number of prints that failed. keep in mind that

there was one in between all these that was successful.

http://imgur.com/7Pai0Vr

Here are a number of larger prints that all failed in the same way.

http://imgur.com/05ByB6H

Thanks in advance for your help on this. I hope there is a reasonably easy solution.

 

 

[gr5 1,071]

The one setting you dont mention is temperature. What temperature do you print at normally?

Switching from ABS back to PLA is usually a bad thing to do. The ABS seems to be all gone but there are bits in the feeder and the tube and in parts of the head and bad things can happen. Unless you need these to be left in a car on a hot summer day I would stick with PLA only as it is much easier to print and you get better looking parts. Or print ABS only. But try not to switch. Or possibly have only one printer for ABS and one for PLA.

This might be part of your problem but maybe not. If this is the problem then the only suggestion is to stick with PLA for the next 30 prints and see if it eventually gets better.

Other possibilities...

- The 3rd fan on the test head - if it's not turning that would explain all your symptoms (failures after an hour or more). Does it spin? Is it actually blowing air? It should start spinning as soon as you turn on the machine - even before the lights.

- Filament melting - pla only - the glass temp for PLA is around 60C. If the feeder motor gets hot enough the knurled sleeve gets too hot and the PLA starts to soften at the feeder and you get flattened PLA coming out of the feeder which then jams in the bowden.

- Tangled filament or filament coming into the feeder at an angle. Put the filament on the floor! Seriously. Just try it.

- Too hot and too slow. .06 layer and 50mm/sec is nice and slow. So keep the temp low. 190C. Certainly no hotter than 200C.

You are suspicious of your temp. You can go to the tune menu and see if it bounces around. If so then there is a firmware update that speeds up the PID cycle and improves temp stability for some users in the latest Cura. Or you can try using high temp heat paste to get better thermal conductivity on the temp probe and heater.

You can test the temp of the test head. This video explains how with just some filament:

 

 

[ericwilsonart 0]

It was failing at 210º so I bumped it up to 220º I will do the test seen here in the video and see what kind of result I get. So are you saying the lower the resolution the lower the temperature should be also? On a side note...Today I spoke with someone that owns a 3D printing company and he said the formula for minimal layer thickness should be 80% less of your nozzle diameter. So if I multiply the .4 mm nozzle by .2 that gives me a value of .08 or .08mm layer thickness. Please check my math haha. If that is true than .06 mm or .02 as I have done in the past is outside the range of safely printable tolerances.

ABS in nozzle? No I cleaned that thing out for hours with pipe cleaners acetone and plenty of fire Like I said I could see the color of brass only inside, it was like a new part. I will stick to PLA either way. It is much nicer for bigger parts like you say.

Regarding the fan. I will have to check that next time it fails.

Not sure I understand the part when you said "- Filament melting - pla only - the glass temp for PLA is around 60C"

Filament is not tangled. Plus I have changed out the roles many times.

Thank you. I will report when I have some findings.

 

 

[DidierKlein 660]

A lot of people print in 0.06mm layers i think that the UM2 does a good job with going lower (but in my own experience going lower will not be worth the time spent printing).

Can you tell us the version of the firmware you're using? Some printers have temperature fluctuation that can cause problems like yours (but i must say, apart from the color difference most of the prints you show look pretty good).

This temperature difference can be seen during printing when you go on tune -> temperature

 

[gr5 1,071]

I think that .06 is pushing it and you start to get quality problems - especially with overhangs and bridging. the thin threads of PLA break too easily. So in some areas quality is slightly better and in some areas quality might be quite a bit worse. But this is not likely your current problem.

 

Not sure I understand the part when you said "- Filament melting - pla only - the glass temp for PLA is around 60C"

 

Boil some water in a glass in a microwave and then remove from the microwave. Take a piece of PLA and stick it into the water (it's fine if the water has cooled to 70C by now). Leave it in there for at least 10 seconds. Remove it and bend it and squish it and play with it then blow on it for 2 seconds and repeat until suddenly it is solid again. PLA has a melting point of around 160C and a "glass temp" of around 60C. When you are above 60C or so it starts to act like clay instead of a normal solid. It doesn't act like a liquid until around 160C. But it is very soft at 60C. So if the stepper motor for the feeder is at 60C and it is conducting heat through the shaft and the knureled sleeve directly to your filament it can soften and squish the filament which then gets stuck in the bowden just above the feeder. Or it just doesn't move forward at all (it grinds the filament to pudding).

Regarding ABS. Tiny pieces (ABS dust) can be in the feeder knurled wheel and get transported slowly up the bowden with PLA. Eventually the pieces reach the nozzle. Tiny pieces can be in the bowden. Or in the feeder. Usually a little bit of this is not a problem. It's very common to have issues when switching between ABS and PLA and I just don't recommend doing this more often than one in 100 prints as you may want to throw the machine out the window on days when you switch.

 

[gr5 1,071]

I'm not sure why you would bump it up to 220C. I recommended you lower the temp to 190C. You are printing at an extremely slow pace and there's no need to leave hot PLA in the nozzle which can turn to gunk if you leave it too long at a hot temperature. Also because you are printing so slowly (.06mm layer) the heat can migrate up the head and soften the PLA up above the area where it is okay and cause clogs in the upper part of the print head.

However if you speed up to .1mm layer then 210 or 220 might be appropriate.

Did you mention if you tested the 3rd fan? Make sure it blows through the head, not suck. When you first turn on printer (not during printing).

 

[gr5 1,071]

Did you do the nozzle test in the video?

 

[ericwilsonart 0]

Thanks for getting back to me.

Didier Klein you asked

Q. "Can you tell us the version of the firmware you're using? Some printers have temperature fluctuation that can cause problems like yours (but i must say, apart from the color difference most of the prints you show look pretty good)."

A. There are no signs of fluctuation of temperature. I have kept an eye on this.

Q. Can you tell us the version of the firmware you're using?

A. Cura is up-to-date but that is not the same as updating the firmware on the machine right? I will look into.

gr5 you asked

Q. Did you mention if you tested the 3rd fan? Make sure it blows through the head, not suck. When you first turn on printer (not during printing).

A. Yes the fan seems to be fully functional at all times even during extrusion failure.

 

I'm not sure why you would bump it up to 220C. I recommended you lower the temp to 190C. You are printing at an extremely slow pace and there's no need to leave hot PLA in the nozzle which can turn to gunk if you leave it too long at a hot temperature. Also because you are printing so slowly (.06mm layer) the heat can migrate up the head and soften the PLA up above the area where it is okay and cause clogs in the upper part of the print head.

However if you speed up to .1mm layer then 210 or 220 might be appropriate.

Did you mention if you tested the 3rd fan? Make sure it blows through the head, not suck. When you first turn on printer (not during printing).

 

This makes sense I will try and print from the beginning at 190º C. If I turn it down to 190º mid print or when extrusion has slowed way down, extrusion stops all together. According to your theory of heat traveling up the PLA and making it too soft/wide to pass through the teflon spacer would mean that once the print has failed due to overheating the PLA @ 210º-220º C lowering it to 190º C and not seeing any signs of extrusion is because the damage is already done. This is interesting and seems like it would be the culprit to me. I will have to see how it comes out from the get go, but my concern is that at that heat from the beginning the walls will not touch. (See the link) http://support.3dverkstan.se/article/23-a-visual-ultimaker-troubleshooting-guide#wallspace

I will let you know though.

ABS dust should be gone I have blown out the bowden tube 3 times and disassembled and blown out the feeder box assembly 2 times so I imagine that is not the case but maybe....

 

[ericwilsonart 0]

Another question regarding the speed. This might be a general question that is well known already but; if the profile settings within Cura says "Print speed (mm/s) 50" but when it is printing it says under "tune>speed>100%" on the machine.

Does that mean that the speed setting is actually 100% i.e. 100 mm/s or that it is 100% of the input value set in Cura? I can't imagine that would be the case that the setting would be relative as that would lead to a lot of confusion, but that it would literally always mean 1%=1 mm/s all the time. If this is the case I would need to manually match the speed to what the profile value in Cura is calling for? If that is true that needs to be fixed in my opinion

 

[titanoid 3]

or that it is 100% of the input value set in Cura?

 

This.

 

[DidierKlein 660]

The speed setting of the print is set by Cura, so 50mm/s at 100% = 50mm/s!

Firmware is the version installed on the printer, you can see this in the maintenance menu i believe.

Cura is the software for slicing but is used also to update the firmware on the printer.

Maybe you can test your extrusion with the cylinder test also:

http://umforum.ultimaker.com/index.php?/topic/4586-can-your-um2-printer-achieve-10mm3s-test-it-here

You have to print at 230°c for this test, this is to see the limits of your printer.

Also did you change anything regarding the retraction settings (on the printer)?

 

[francfalco 37]

Hi original poster,

Did you ever solve your problem?

I fear that after 3 months of perfect, no problem printing, i seem to be having the exact same problem as you?

My setup is UM2, PLA, Cura slicing - nothing fancy.

Nozzle seems fine and like you, only starts to 'motor skip' well into the print.

I'm stumped?

Any updates please let me know

Cheers

 

[DidierKlein 660]

Hi FrancFalco,

Did you try to clean the nozzle with the atomic method?

What is the brand of filament? Are you sure it's 2.85mm ?

Did you check if your teflon is deformed?

Many things can cause this, you could give us more information (printing parameters), and maybe a pic or two

 

[francfalco 37]

Hi moderator,

Further onfo fyi...

Nozzle was cleaned using the atomic method.

I have used that method before with complete success, but this issue doesnt seem to be a nozzle clog. Filament loads correctly and the print/ purge start of the print is good.

Filament brand is Faberdashery, a brand i use exclusively and has never given any problems. Size and quality is alway consistent

I slice with Cura, usually standard settings.

Havent checked the teflon collar, that does seem to be a possible area of issue. However, I would be surprised to find any wear/ distortion in that, the printer is only 3 months old and i only print PLA at 205c max. But it may be the cause.

I havent checked if the firmware is latest, could that be an issue?

I havent visually checked the 3rd fan is working, although it does sound as though it is?

As i mentioned, this is an issue that has appeared very recently almost from one print to the next. Models that my printer was printing perfectly, now always airprint.

The print starts fine, but well into the print the under extrusion starts and always ends in the extruder blocked.

With my limited experience of FFF printers, it does appear to be heat creap affecting the filament, but i cant think what would cause such a sudden major change?

I have a support ticket with the company i purchased the printer from, iMakrUK.

I will try to resolve this myself, otherwise, hopefully those guys can fix the issue.

Please let me know your thoughts

Regards

Franc falco

[DidierKlein 660]

Faberdashery is awesome filament so we can rule that out.

I'm guessing it doesn't get tangled up either?

The Teflon should not be deformed if you print exclusively at 205°c, it can happen but like you i would doubt that.

Have you tried printing a bit hotter?

Does it always happen or some prints go well?

Fire alarm here i have to go :mrgreen:

 

[DidierKlein 660]

False alarm

Something to add:

 

 

The one setting you dont mention is temperature. What temperature do you print at normally?

Switching from ABS back to PLA is usually a bad thing to do. The ABS seems to be all gone but there are bits in the feeder and the tube and in parts of the head and bad things can happen. Unless you need these to be left in a car on a hot summer day I would stick with PLA only as it is much easier to print and you get better looking parts. Or print ABS only. But try not to switch. Or possibly have only one printer for ABS and one for PLA.

This might be part of your problem but maybe not. If this is the problem then the only suggestion is to stick with PLA for the next 30 prints and see if it eventually gets better.

Other possibilities...

- The 3rd fan on the test head - if it's not turning that would explain all your symptoms (failures after an hour or more). Does it spin? Is it actually blowing air? It should start spinning as soon as you turn on the machine - even before the lights.

- Filament melting - pla only - the glass temp for PLA is around 60C. If the feeder motor gets hot enough the knurled sleeve gets too hot and the PLA starts to soften at the feeder and you get flattened PLA coming out of the feeder which then jams in the bowden.

- Tangled filament or filament coming into the feeder at an angle. Put the filament on the floor! Seriously. Just try it.

- Too hot and too slow. .06 layer and 50mm/sec is nice and slow. So keep the temp low. 190C. Certainly no hotter than 200C.

You are suspicious of your temp. You can go to the tune menu and see if it bounces around. If so then there is a firmware update that speeds up the PID cycle and improves temp stability for some users in the latest Cura. Or you can try using high temp heat paste to get better thermal conductivity on the temp probe and heater.

You can test the temp of the test head. This video explains how with just some filament:

 

 

 

You can also test your extrusion here:

http://umforum.ultimaker.com/index.php?/topic/4586-can-your-um2-printer-achieve-10mm3s-test-it-here

 

[francfalco 37]

Yep, faberdashery is great and no i check it doesnt get tangled.

Tried hotter, cooler, no difference.

Is there a video/ thread to show how to remove the teflon collar?

It happens on all prints, apart from very small one. My 20mm test cube seems to print okay.

[DidierKlein 660]

 

Obviously you don't need to do all that is done after 4min30s (or so)

 

[gr5 1,071]

If you take the white teflon isolator out, you might want to keep going and put the nozzle in flame to burn out anything in there.

When you have it apart, take some badly curved PLA and slide it through the teflon isolator. Test for friction.

You might want to drill out the isolator slightly.

When you put it all back together, try to lower the nozzle as far as possible (by spinning that nut with the holes in it) such that the spring pushing against the isolator is not pushing too hard. But there needs to be a little bit of gap so that the teflon is definitely touching the brass.

 

[francfalco 37]

All very strange or maybe not!

So i performed yet another 'Atomic method' nozzle cleaning last night, (i'd done this a couple of times before without success) but maybe the difference with this one was that it produced a perfectly formed 'clean cone' result. I also slid a guitar E string wire up the nozzle from the bottom for good measure.

I also updated the firmware and went back to Cura 14.07

Printed without problem! No under extrusion, no skipping motor, all was back to normal.

I'm still not totally happy with the slicing of Cura, surface blobs and Z scaring. But i'll take those anyway to not being able to print!

Hopefully it won't let me down tonight... fingers crossed!

Cheers

 

[DidierKlein 660]

Interesting, either the good atomic pull out was the reason or the firmware update.

If it's still working nicely and have some time to try updating back to the firmware you had when you were experiencing this, to see if it's firmware related?

 

[francfalco 37]

If it carries on printing nicely, i'm not going to touch a thing!!

 

[DidierKlein 660]

:mrgreen: :mrgreen: Lol I understand