Almost always missing layers / underextruding

[woofy 12]

@Nicolinux,

I share your frustration on this. My own um2 is a little better than yours but not much. I think maybe now is the time to get it back to Ultimaker. They can compare every aspect of your machine against a good one and get to the bottom of this.

@Ultimaker

My own suspicion is that there is no one thing causing this, but the cumulative effect of many issues.

Maybe the motor may be a bit weaker.

Motor current may be slightly low because of component tolerances.

Filament grating on the extruder casing.

Bowden tube too narrow of too much friction.

Maybe an updated hot end design is required.

In the end though, whatever is needed to fix this is likely to be new parts that need to come from Ultimaker themselves. With increasing numbers of people on these forums complaining of extruder skipping, hopefully Ultimaker will be able to divert more resources to a solution.

 

[hacklordsniper 1]

I hope they will come to some solution before shipping my printer. I even plan to ask them to hold my order little bit more until they solve this problem. It would be better if they keep us informed about progress

 

[Nicolinux 288]

Ok, I've disassembled, cleaned and assembled the head and extruder. Extrusion test fails right after 4mm^3/s now.

The issue where filament won't pass the teflon piece remains. Even if I blunt the tip it won't pass easily. The nozzle was not clogged, the dent in the "inside nozzle" shape must have been an air bubble. I didn't replace the set screw for the knurled bolt because they sent me the wrong screws.

@Ultimaker: People, really. I am very grateful for your generosity, but you should check your packing/shipping workflow. I am just one single customer and received already a few wrong parts. This can't be profitable for you if it happens often...

So, that's it for me now. I've had enough. Don't want to turn this into a hate thread though. I still believe the UM2 rocks, and when I print very slowly, quality is still astounding. I'll contact support and hope they'll accept the printer for repairs or replacement.

EDIT: Uploaded images.

 

[woofy 12]

So, that's it for me now. I've had enough. Don't want to turn this into a hate thread though. I still believe the UM2 rocks, and when I print very slowly, quality is still astounding. I'll contact support and hope they'll accept the printer for repairs or replacement.

 

I agree, UM2 print quality is top notch. :cool:

Just want it to be faster.

@Nicolinux, Please keep us posted on this.

 

[hacklordsniper 1]

@Ultimaker: People, really. I am very grateful for your generosity, but you should check your packing/shipping workflow. I am just one single customer and received already a few wrong parts. This can't be profitable for you if it happens often...

 

Thats not nice, and not to mention the time your printer does not work while waiting replacement. But however if you are in a hurry you could buy a allen set screw in local hardware store (i can buy them in a small village type town here), so in Germany you should really have more choice.

Im glad you decided to ship the printer back, i hope your printer will be fixed and we will know what is the problem then with it. Let us know how it goes.

 

[mr.-waldorf 0]

Hi Guys,

More bad news, after 2 days without any changes, the underextrusion is back. Can't go over 6mm3/s and changing the flow % doesn't make any difference

other day test with 100% Flow (failed at 4mm3/s)

 

other day test with 90% Flow (passed at 10mm3/s)

 

Today test with 100% Flow (failed at 6mm3/s)

 

Today test with 90% Flow (failed at 6mm3/s)

 

There wasn't no diferences between tests, even the room temperature was the same...

 

[Nicolinux 288]

Hi,

here's another thing I noticed (just for the record). When changing filament, the tip looks alright, but it drags a long strand through the bowden that ends in a blob. I had this before but now, since I switched the teflon coupler, it happens every time. The blob gets stuck inside the extruder because it can't pass the knurled bolt on its way out. So I always have to fiddle with at toothpick to get it out. Worst case I have to disassamble the extruder.

EDIT: Forgot the image:

 

[Dim3nsioneer 557]

Hi,

here's another thing I noticed (just for the record). When changing filament, the tip looks alright, but it drags a long strand through the bowden that ends in a blob. I had this before but now, since I switched the teflon coupler, it happens every time. The blob gets stuck inside the extruder because it can't pass the knurled bolt on its way out. So I always have to fiddle with at toothpick to get it out. Worst case I have to disassamble the extruder.

EDIT: Forgot the image:

 

I had a similar effect once or twice on my UM1. It happens less if you change it at lower temperature, 180°C worked quite well for me while I got this stranding above 200°C.

 

[schafe 3]

For those of you who have not been following my C-130 topic I offer the following gem:

Cura said the print would take 28 hours. I started in Cura with a print speed of 60mm/sec and .1mm height. The flow rate was 100% and temp was 250C. 60 times .1 times .4 equals 2.4 mm3/sec as programmed in Cura. The night before I had a print come out perfect with those exact settings. The only thing that changed was the filament radius on the spool. Early on in the print I heard the extruder skipping so I lowered the print speed to 80% or 1.92mm3/sec. Still heard the skipping so I went to 70% or 1.68mm3/sec and went to bed. Woke up to this. My filament was on the stock spool holder. I'm changing that tonight.

I've been watching the olympics almost every night - except for figure skating. I'm inspired by the scoring that goes on. I propose a failed print scoring system based on the blarp. Things like print time, print speed, and if the print is a total failure will be considered. I give this print 9/10 blarps. Took over 30 hours and had a really low total extrusion speed. I'll still use it and will fill in the gaps with balsa filler.

 

[yellowshark 153]

250c is way too hot. Most PLA filament's have a quoted upper range of 220c.I would print .1 at 60 m/s at round 210C,maybe 205c. I printed .2 at 60m/s at 215 this evening.

 

[mr.-waldorf 0]

250c is way too hot. Most PLA filament's have a quoted upper range of 220c.I would print .1 at 60 m/s at round 210C,maybe 205c. I printed .2 at 60m/s at 215 this evening.

 

I bet that you have a UM1 instead of a UM2, right?

 

[schafe 3]

Just had another thought. Has anyone looked into mounting the extruder directly on top of the X/Y axis thing, whatever it is called? It seems possible to get it there without much effort. That way you would just shove the filament straight into the heater. The axles seem strong enough for the weight of the stepper motor.

 

[mr.-waldorf 0]

Just had another thought. Has anyone looked into mounting the extruder directly on top of the X/Y axis thing, whatever it is called? It seems possible to get it there without much effort. That way you would just shove the filament straight into the heater. The axles seem strong enough for the weight of the stepper motor.

 

I considered the same as you for a while, should work way better, unfortunately I think that due to motor sizes and adding some kind of material feeder It will be very difficult to install all this in the head and keep all the clearances and maybe weights. Maybe a very powerfull and tiny stepper? Not sure if available. I like the concept it seems to me the easier way to feed the filament, unfortunately this is a bit more complicate. Filament has a extrusion limit inside a 0.4mm nozzle so if we push to hard it will strip the filament also. The way to go here I think its measuring the filament motion, this could improve a lot the underextrusion problems. The motor "jump steps" are like safety steam valves when they are overpressured they release the pressure out of the system, so if we control the pressure the release valve should stay always close unless there was a malfuction. Its needed some control motion sensor, some changes to the board and some Daid hard work.

There is a lot of ways to improve the printer and this underextrusion problems but it must be Ultimaker to lead the way, develop new solutions and apply them fast. We all payed a lot of money for a printer that doesn't do what Ultimaker announces. I suspicious that they are working hard on this but would be great to have more information about the solutions and implementation deadlines.

 

[schafe 3]

After a booth review, my print above gets a perfect score of 10/10 Blarps. When I removed it from the build plate the layers had terrible adhesion and the whole thing was flimsy. The "fuzzy" final layers came completely apart.

 

[gr5 2,167]

Something is wrong with your printer Schafe. I suggest to start with a thorough nozzle cleaning. Also I strongly recommend you never go above 240C. You can cause the PLA to gunk up your nozzle at that kind of temp. You might want to take the nozzle completely off and "burn it clean" in gas flame of some sort. Try not to get it *too* hot - just hot enough to burn everything to ash but not hot enough to melt the metal.

 

[yellowshark 153]

Correct, no UM2, waiting for the dual extruder. Fair point that different printers may need adjustments to settings but, IMHO, if you have to run at 250c to print at 60 m/s then there is something wrong with your printer; just has to be.

 

[schafe 3]

Yep, I did the heat up the nozzle, stuff in filament, cool to 90 then pull out gunk trick and got quite a bit of brown burned PLA from the print mentioned above. Did that trick 4 times. On the 4th time it still was pulling out some brown PLA. I guess 30+ hours at 250C was just too much. My son was excited to get one of his prints going so I started a new one a few hours ago and left for work. 245C, 60mm/sec, 100% extrusion and .1mm layer height. Thats 2.4 mm3/sec which I know is very slow. When I left everything looked fine.

I should mention the PLA from the above print was purchased from Makershed. It definetely behaves differently than the PLA that came with my UM2.

Sorry for so much belly aching. I just had no idea that nozzle cleaning and maintenance would be such a big issue.

Those of you who have experience with the original Ultimaker, was the nozzle such a problem with that machine?

 

[gr5 2,167]

I'm starting to think there are several causes to underextrusion that can be "erratic":

1) Filament tangling, filamant tight diameter coming off reel. Because the feeder is so much lower on the back of the UM2, the angle that the filament enters the feeder is at a strong angle. Especially when you are getting low on PLA.

2) bent filament. I found that a typical bended filament as it comes off the spool has trouble going through the white teflon part in the print head. If the filament is straight, it is so loose you can spin the teflon part around the filament but get a 10" diameter bend and it takes quite a bit of force (maybe 1 or 2 kg) to get through this part alone.

 

[gr5 2,167]

Those of you who have experience with the original Ultimaker, was the nozzle such a problem with that machine?

 

I went about a year until I cleaned my nozzle on the UM Original the first time. But it was NASTY. There was something in there - some kind of dust that did not melt or burn even in a gas flame. Of course the nozzle comes of the UM1 in less than a minute which is nice.

I think you were unlucky - just don't go over 240C in the furture for PLA. Temps over 240C are meant for nylon and ABS. Even this print you just did at 245C is hotter than I like to print PLA.

I went a few months on my UM2 before my first nozzle clog. It also was easy to clean - I used the method where you insert filament from above and let it cool to 90C and pull it all out. Eventually you get most of the clog out.

 

[IRobertI 520]

One thing to keep in mind though. Even though the head is at 2xxC doesn't mean that the PLA reaches that temp. Depending on how fast you print the PLA might not even come close to that temperature, especially on the UM2 with its much smaller heater block. That said, it's probably not a good idea to print over 240ish degrees like gr5 says since a print is rarely at a constant speed with minimum layer times kicking in and whatnot.

 

[schafe 3]

My print above turned out pretty good. I used the Makershed filament. Stayed with the above settings except I lowered the extrusion to 95% because I heard the clicking. I give my recent print 2/10 Blarps. Just a few underextruded holes.

 

[schafe 3]

I'll try lower temp next time.

 

[yellowshark 153]

Tangled and bent filament…

Yup since I have been using Faberdashery filament I have got use to the filament laying on the floor and flowing freely. Now when I use a reel I lay that on the floor and pretty much unfurl as much filament from the reel as I need for the print.

This applies more for a half empty spool, but I do it anyway, I then pull a finger and thumb along the filament and basically stretch out the filament removing the curls (substantially) laid into it from being in the reel. So there are no tangles and it also stops nasty bends being in the filament.

 

[gr5 2,167]

Okay - I did a new test! I got a second nozzle for the UM2 when I was at NYC 3dprintshow. I had Joergen drill it out to .7mm diameter (thank you Joergen) and printed as fast as possible without extruder skipping. The result?

18mm^3/sec at 230C

More details here:

http://umforum.ultimaker.com/index.php?/topic/4127-um2-extrusion-rates-revisited/?p=38869

But the point is that the tiny heat chamber is not the limiting factor. It was a compelling hypothesis that the UM2 nozzle/heat chamber was too small, but it has now been disproved.

 

[mr.-waldorf 0]

Hi George,

I was planning do the same and this post saved me some € on a new nozzle. This is very important data! Thanks

 

I'm starting to think there are several causes to underextrusion that can be "erratic":

1) Filament tangling, filamant tight diameter coming off reel. Because the feeder is so much lower on the back of the UM2, the angle that the filament enters the feeder is at a strong angle. Especially when you are getting low on PLA.

2) bent filament. I found that a typical bended filament as it comes off the spool has trouble going through the white teflon part in the print head. If the filament is straight, it is so loose you can spin the teflon part around the filament but get a 10" diameter bend and it takes quite a bit of force (maybe 1 or 2 kg) to get through this part alone.

 

This is also very important info. Since my last extrusion tests I decided to change the spool for a new one to see the diferences and I was able to achieve the 10mm3/s again. son of a b****

I will make a few more tests but right know it seems that my underextrusion problems happens on a half empty spool... ARHHH!