Almost always missing layers / underextruding

[chrisp 10]

Oh, on another note, I noticed in your pic above that the wires from the side fans are rubbing up against the back fan that cools your nozzle. That alone can stop the fan from running, which might be another issue. This has happened to me.

 

[illuminarti 18]

Nicolinux - did you try doing a factory reset, to make sure your extruder current settings are ok, as I mentioned a page or two back?

 

[Nicolinux 288]

I hate to beat a dead horse, but can you check the gap between the teflon and the milled fitting again?

When my prints were failing everytime, the only change I made was to the space between the teflon and milled fitting. (Posted in this thread somewhere) Since then my issues of under extrusion are gone. I don't understand why, other than the transfer of heat in the form of conduction is very accurate where the heater, nozzle and milled fitting all sit together.

Or maybe I just got lucky and failing mechanical parts are now working? Very Unlikely

Everytime I use the Orange PLA from the Ultimaker store I got a clog almost instantly. I did buy some cheap white PLA off of ebay which is printing better than all of the colors I've tried so far(go figure).

I hope you find the solution!

 

Ok, the gap ist 2.2mm wide. I have found a nice workflow to 0 play with the bowden tube (head). Lower the gap between the teflon piece and the milled fitting, then insert the bowden tube and secure it. Then unscrew the milled fitting so it moves up against the bowden tube a bit. This alone has eliminated any play there.

Regarding orange PLA from Ultimaker. This prints beautifully on my printer. Way better than silver-grey or gold.

 

[Nicolinux 288]

Oh, on another note, I noticed in your pic above that the wires from the side fans are rubbing up against the back fan that cools your nozzle. That alone can stop the fan from running, which might be another issue. This has happened to me.

 

Good catch, thanks. It wasn't an issue until today. Now the cable decided to rub against the fan and make noise. Good timing

 

[Nicolinux 288]

Nicolinux - did you try doing a factory reset, to make sure your extruder current settings are ok, as I mentioned a page or two back?

 

Of course I did. Was the first thing I tried (because it was the easiest one...). No change though...

 

[chrisp 10]

Sander said in a previous post about the gap being between 1-2mm I think. Maybe it too much of a gap. Mine is at 1.5mm. I know this sounds ridiculous, but it made the difference for me. I was at 1mm before moving it to 1.5mm. As for the Bowden tube, it can be moved down after you adjust the gap again.

[Nicolinux 288]

[...]

However, if the filament was tightly coiled and badly oriented so as to add resistance to the feed process, then I definitely did see underextrusion. I'm pretty sure, also, that in all the cases I saw under-extrusion it was because the extruder stepper was skipping steps and slipping back.

 

Illuminarti, I have read the entire thread again and your first post with the extrusion test reminded me of something. In your tests with Ultimaker PLA, did you also run a test where the spool was on the default spool holder? For some reason my PLA is very stiff and even if I straighten it by hand I couldn't get it so far so it won't curl any more. I couldn't achieve a straight...ish line like the one you have here:

http://umforum.ultimaker.com/index.php?/gallery/image/2815-straightened-filament/

So I wonder, maybe my extruder is allright, but I just had bad luck and always tried filament that was too stiff. On the other hand, before I was aware of the underextrusion problem, I printed from the default spool holder without problems. But then again, maybe I was always under 4mm^3/s so I didn't notice it.

However, would you mind to run another test with Ultimaker PLA from the default spool holder? Would be nice to see if and when it fails. I could reproduce the test with the same filament here and we would have a direct comparison.

I have the folowing filament colors (all from Ultimaker):

 

• White

• Pearl-White

• Green

• Orange

• Red

• Silver-Gray

• Silver-Metallic

• Transparent

• Bronze (which I thought was gold but now see that they sent me the wrong color :shock: )

Or if anybody else would like to print with one of these colors, let me know and I will run the test too (the no-retraction version).

 

[illuminarti 18]

Sure, I know I have UM Bronze... will need to see which other UM colors I have. I'll run the tests later this evening.

 

[Nicolinux 288]

Sander said in a previous post about the gap being between 1-2mm I think. Maybe it too much of a gap. Mine is at 1.5mm. I know this sounds ridiculous, but it made the difference for me. I was at 1mm before moving it to 1.5mm. As for the Bowden tube, it can be moved down after you adjust the gap again.

 

Well I am at the point where I'd try anything, randomly hoping for success. You know, like in Monkey Island (or other old school adventure games) where you got stuck on a puzzle and started to combine everything with everything

I'll try it.

 

[illuminarti 18]

Sorry, Nicolinux - printed fine on my printer in both red and bronze UM filaments - both mounted on the standard spool holder. Red was a brand new spool opened for the test, bronze spool had just had a few meters used on it.

 

[Nicolinux 288]

Thanks. And no problem. This proves your printer is still very good.

I have a nagging suspicion that I will try asap. Maybe the temperature sensor fails during a print and the temperature gets so low that the extruder has a really hard time to keep pushing filament. This would explain a lot. I will print your test at higher temperatures. Let's see what happens.

 

[Nicolinux 288]

Ok, printed at 250° and it failed pretty late:

But I am not sure what I learn from this. I mean, more heat, more viscosity and it gets easier for the extruder to push filament. Printing at 230° still fails earling proving that this extruder of mine is a lazy b****.

One thing about terminology. I used the word "slipping" when I mean that the extruder produces a clicking sound and the knurled bolt snaps back letting filament back. This is oposed to filament grinding which does not produce any specific sounds and where the knurled bolt just turns and turns until the filament is grund down and won't push further through the bowden tube.

Is this everyone else's understanding?

 

[Nicolinux 288]

Ok, but what could I try next? Power sander - where are you?

 

[illuminarti 18]

Is this everyone else's understanding?

 

Yes.

Do you have any way to validate the temperature of the hot end - like an separate thermocouple thermometer you could attach to the heater block? Would be good to validate the temperature you're getting.

Also, have you tried hooking the printer up over USB to Cura, Repetier Host or some other tool that can capture temperature graphs? I'm not sure whether the UM2 even outputs temperatures over USB when requested, but that might be useful too - to monitor the temperature over time while you are printing, and see how it behaves...

 

[illuminarti 18]

Also, when you had the hot end apart, did you check that filament would slide through the teflon part without much resistance?

 

[gr5 2,167]

I used the word "slipping" when I mean that the extruder produces a clicking sound and the knurled bolt snaps back letting filament back.

 

Yes this is normal. I've never ground filament on my UM2. It slips back. The stepper loses steps. If you increase the current too high then the extruder grinds the filament which trust me is much worse!

 

[SandervG 1,521]

Well, those are a few hours well spend.

I read the entire 13 pages and wrote down some of my thoughts.

I am sure most of them were already thought of, but I didn't really read any reason for it not to be true, or at least a hint towards a solution.

What I think is of a great influence is the natural curl of the filament.

I think the extruder has sufficient power to pull the filament in (but it is far from being 'overpowered')

but it is struggling pushing it through the bowden tube.

Simon also said it needs to make an 180° turn, and when it is inserted the other way around that can cause quite some friction inside the bowden tube.

Have you tried inserting a small spray of WD40 in the bowden tube?

From my experience this has never influenced the print, in a way that the oil may be visible on your print.

For this same reason I have doubts with lazy susan, at least when it is lying behind your UM on the table, it will be inserted with an 90° angle. (first it is horizontal, and it needs to go vertical).

I also thought perhaps you could try faberdasherry filament. It doesn't come on a reel, so it has less of a curl.

Besides that, I know the quality is good.

With the temperatures, I double checked and R&D told me that the 190-260 filaments can actually go up to that temp when you are printing fast (150mm/s fast). And I believe Stefan that is also what you experienced? Better results with a significantly higher temp?

Stefan, what is your current status of the set screw in your teethed gear?

And did you try to push some fresh filament (that didn't go through the feeder yet) through the teflon when you had it out? You should have 0 (!!) friction.

Looking forward to hear about your findings.

I am now 100% up to date about this topic and will be more involved

 

[Nicolinux 288]

Yes.

Do you have any way to validate the temperature of the hot end - like an separate thermocouple thermometer you could attach to the heater block? Would be good to validate the temperature you're getting.

 

Not at home. But I have friends (with benefits)

I'll see if I can get a temperature probe or something alike.

 

Also, have you tried hooking the printer up over USB to Cura, Repetier Host or some other tool that can capture temperature graphs? I'm not sure whether the UM2 even outputs temperatures over USB when requested, but that might be useful too - to monitor the temperature over time while you are printing, and see how it behaves...

 

That's a very good idea. I will try it. Although if the temperature sensor is broken, it will report wrong values (or values that look normal). But combining this with an external temperature probe will provide ultimate conclusion.

 

Also, when you had the hot end apart, did you check that filament would slide through the teflon part without much resistance?

 

No I didn't do it conciously, I just check it for deformations and filament leftovers. I will take a look next time.

Got used to the idea that I'll have to take apart the hotend again before I get to the bottom of this problem.

 

[Nicolinux 288]

Yes this is normal. I've never ground filament on my UM2. It slips back. The stepper loses steps. If you increase the current too high then the extruder grinds the filament which trust me is much worse!

 

Yes I experienced it first hand. Filement was ground down, and stuck in the bowden tube. Also had to take the extruder apart to remove the mess and another piece of filament that got in there. Therefore I prefer skipping/slipping to grinding

 

[Nicolinux 288]

Sander, thanks for joining in. This issue is really troublesome for me because I can't print anything big or fast(er). Yesterday an easy print which was at 4mm^3/s filament volume failed due to skipping. And there are a few other users with the same problem.

 

Well, those are a few hours well spend.

I read the entire 13 pages and wrote down some of my thoughts.

I am sure most of them were already thought of, but I didn't really read any reason for it not to be true, or at least a hint towards a solution.

What I think is of a great influence is the natural curl of the filament.

I think the extruder has sufficient power to pull the filament in (but it is far from being 'overpowered')

but it is struggling pushing it through the bowden tube.

Simon also said it needs to make an 180° turn, and when it is inserted the other way around that can cause quite some friction inside the bowden tube.

Have you tried inserting a small spray of WD40 in the bowden tube?

 

You are right. It looks like the problem lies with the curvature of filament. I tried almost all orientations I can think of. The only one which produced good results was cut filemtent, straightened by hand and hold in hand the whole time. Clearly not the solution I am aiming for. Besides, Simon (and others) are able to use the stock filament spool holder without problems. And they also used Ultimaker filament (and I guess no WD40 spray).

Sadly I don't have faberdasherry filament here, just tons of Ultimaker filament And again, I am sure the filament is perfectly fine, and not the reason for my problems.

 

With the temperatures, I double checked and R&D told me that the 190-260 filaments can actually go up to that temp when you are printing fast (150mm/s fast). And I believe Stefan that is also what you experienced? Better results with a significantly higher temp?

 

Yes that's true. Printed the extrusion test (no retraction) yesterday at 250° and it failed only at the end. This was definitelly an improvement (link).

 

Stefan, what is your current status of the set screw in your teethed gear?

 

I tightened the set screw as much as I could. It should be alright. But I just got an idea. I will draw a line on the top of the knurled bolt. The line will go over the bolt and the rod from the motor. Then after printing, if I see that this line is not uniform anymore, this must mean the knurled bold slipped around the rod.

 

And did you try to push some fresh filament (that didn't go through the feeder yet) through the teflon when you had it out? You should have 0 (!!) friction.

 

Sadly I didn't. But I will do now.

One more info. The gap between the teflon part and the metal fitting is at 2mm now. I had it at 1.2, 1.5, 2.2 and 2.5 before.

So here is my todo list (in no particular order)

 

• mark the knurled bolt and motor rod to see if it slips during printing
  
• connect the UM2 to see if I get temperature readings in other tools
  
• provision a temperature probe and compare the temperature of the nozzle
  
• take the head apart and check if filament passes through the teflon coupler freely
  
• put the filament on the stock spool holder
  

 

[SandervG 1,521]

Sander, thanks for joining in. This issue is really troublesome for me because I can't print anything big or fast(er). Yesterday an easy print which was at 4mm^3/s filament volume failed due to skipping. And there are a few other users with the same problem.

You are right. It looks like the problem lies with the curvature of filament. I tried almost all orientations I can think of. The only one which produced good results was cut filemtent, straightened by hand and hold in hand the whole time. Clearly not the solution I am aiming for. Besides, Simon (and others) are able to use the stock filament spool holder without problems. And they also used Ultimaker filament (and I guess no WD40 spray).

Sadly I don't have faberdasherry filament here, just tons of Ultimaker filament And again, I am sure the filament is perfectly fine, and not the reason for my problems.

 

True, but if you could test it and it turned out to be the solution we then know what is going on and focus on how to solve it.

As of right now we are still guessing for what the problem is.

Hopefully your todo list will shine some more light on it.

(I don't think it is temperature related).

 

[gr5 2,167]

To me the problem could still be many things grouped into 3 categories, problems with extruder, problems with nozzle, other friction (bowden tube, kinking, spool).

I'm leaning towards problems with nozzle (such as burnt, black PLA coating the inside), but I haven't ruled out the extruder yet. You could do a definitive test to rule out the extruder. Put a short length of PLA in the extruder and with the servo powered up pull on the PLA until the servo slips or the PLA slips. Pull by using weights. When done measure the weights. If you do this test I will duplicate it on my machine.

You could even repeat the test with the filament 90% through the bowden if you want to see if that makes any difference (I doubt it).

 

[alanoo 0]

As a reference and as a complete 3d printer newbie (even if I have read the entire 13 pages, just for my personal knowledge) : Third print ever on my UM2 received this week

(UM PLA blue, 230°C, 50°C bed, filament on the stock reel holder)

Almost perfect... just so anyone lurking here could see these printers could definitely be good almost out of the box.

 

[IRobertI 520]

Finally got around to doing this test myself. And well.. it confirmed what I've been feeling all along, that something is off with my printer.

Ultimaker red @ 235C on standard reel holder (but "fluffed" to make sure there were no snags). Started hearing/seeing the first signs half way through the "4"-mark.

Guess it's time for me to do a proper tear down of the head and check the teflon piece as well this time.

 

[braddock 1]

Hey Sander, I had considered lubing the tube with inox (similar to WD40) but was worried about a few things... burn off smell, whether or not it would affect layer adhesion etc, but also, if it would casue the feeder to have trouble gripping the material, as innevitably, the lube will end up transfering back to the feeder also.