Problems with feeder motor skipping back and causing under extrusion

[rotorhead 1]

Hi,

I have recently got my UM2 and have so far only printed a hand-full of small things. I have black and white PLA and ABS and the silver OEM PLA. I know a reasonable amount about 3D printing but am certainly not an expert and this is the first time I have owned my own machine.

The problem I have had is with my first large ish part. (12 hr build) I am trying to print at 50mm/s 0.4mm nozzle and 0.2mm layer in black PLA from printme3d.com. All my prints up until this part were fine, even using the same material and speeds but the first attempt at this larger part was terrible with gaps all over the place. I noticed that the feeder motor was skipping quite a lot so I have messed about with cleaning the nozzle, changing temps and speeds and pressure on the feeder motor and I made it slightly better but still not good enough. Eventually I realised that the filament going IN to the feeder was causing the problem. It was simply too hard to pull it through. With the filament just loosely coiled up and being picked by the feeder the problems seem to be almost completely sorted (I noticed just a few motor skips and the odd gap in between outer wall layers).

The PLA is on a 1kg roll and I have about 2/3 of it on the roll now.

I have measured the diameter of the filament and set the UM accordingly.

I have thoroughly cleaned the nozzle and bowden tube and once I established that the problem is the filament going in, I was able to get the speed back up to 50mm/s with a temp of 240 on 0.2mm layers with hardly any step-back on the motor.

The part I am printing has very little need for retraction, 1.6mm walls and 30% fill.

Oh and I increased the spring pressure on the feeder. The white markers were at the top, they are now in the middle - this seemed to be the only thing that helped the print apart from loosening the filament on the roll, no grinding yet but I'm not sure if I was correct to do this!

I have tried running the filament around the little black pin on the back, bottom right and it seems to make little difference if it's round that or not. The PLA seems very reluctant to straighten out. Even if I pull some off the roll to help the feeder, it tends to wrap back up again so every 5 mins I had to tend to the spool for 12 hours to get the part done!

I am just wondering what the best fix for this is? I have seen some mention online of using a different feeder, some people have said that increasing the power to the motor would be a good idea and I have also seen some alternative spool holders like ones with 4 metal rollers. I can't be the only one with this problem but then of course I could have completely misdiagnosed!

I have to say that I was very reluctant to increase the temp to 240degC for PLA but I found lots of justification for this online and did not experience any stringing, hairs or bubbling.

Sorry for the long post, tried to get as much useful info in there as poss!

[Labern 774]

increasing the feeder current can make the motor hot. As the motor gets hot it can heat up the drive wheel which can deform the filament.

What size/tolerance is the filament?

It sounds like your issue is worse then normal, have you tried printing with the grey UM filament?

A lot of people print This and This to help resolve some of these issues.

[rotorhead 1]

The filament is sold as 3mm but is actually 2.91mm. I have set the Ultimaker to this although I have tried it on the default 2.85mm and had the same problem. I have not tried printing the same part with the silver OEM material as it's quite a large piece. Perhaps I should?

Yes I spotted that feeder and was thinking I should try it. I like the idea of the low friction spoolholder too. Both seem like a good idea but like you say, my problem seems worse than normal. I see what you mean about increasing current to the feeder.

Presumably the feeder motor should ideally never skip back when things are working properly?

and presumably since the UM2 comes shipped with the tension indicator right at the top, that's where is should really be?

[Labern 774]

Yes the feeder shouldn't skip back unless there is to much pressure built up in the nozzle or a snag of the filament.

Sometimes the printers need some adjusting even from new. They just print a test sample and ship it off. as long as you have good impression like shown Here

Is the design your own or Someone else's. You can check in with cura 15.04 under xray view to see if there are problems in the model.

I'm not to familiar with the material you are using. It could be good to try the UM stuff, even if it for something else. Different brands can print very differently, even different colors.

How much of the ABS have you used?

Have you done atomic pulls?

You can try loosing off the 4 thumb screws on top of the head a bit. not so its loose but just holding.

A lot of people just put the roll on the floor on its side and let it uncoil on its own.

[rotorhead 1]

That's great info, thanks. It's all the various solutions I have come across on the internet all rolled into one post to be honest but it makes me more confident to start tinkering with settings now knowing that it is not normal that they come shipped ready set up.

My filament did look more like the fist image but I have increased the pressure now. I will pull it back out and have a look at it now and make sure it's somewhere between the 2.

I've not had much ABS through this machine at all but I did do lots of cold pulls afterwards and ended up getting lovely clean plugs. I'm fairly sure it's not the drawing, it's a very simple piece that I drew up and I've never had an issue before.

I think I will try everything you said over the course of the next few prints and see which one means I don't need to feed the filament by hand!

Thanks for all the help.

[rotorhead 1]

Thanks for all the assistance with this, I printed this and this and things seem to be going ok now. During all the experiments, it took me an embarrassing amount of time to realised that high feeder tension was actually more likely to cause the motor to skip - deforming the filament and simply making the motor work harder!

Tension is probably now lower than it was with the stock feeder, Z axis motor is on standard current, the low friction spool holder is brilliant even with a 1kg roll on there. I don't need to use the guide at the bottom of the rear, it just goes directly into the feeder. I have also replaced the head with an Olsson block and the head that comes with that.

So far the prints seem to be much better. I just need to experiment with temperature now, as well as try and resolve a few other minor problems (which will likely end up elsewhere on the forum!)

Thanks for the assistance I will return to this thread and post any more relevant info.