Z axis issue of UM2

[ltc 1]

I found my UM2 got some problems. Here is prints comparison of the file coming along with UM2.

File "IMAG0678.jpg"

File "IMAG0677.jpg"

When I print a bigger model, the deviation of Z axis is even worse. These are the result of prints from a working well UM and the faltering UM2.

File "IMAG0655.jpg"

File "IMAG0650.jpg"

 

You may see the print on UM is straightly built up, but the one printed by UM2 apparently leans to one side. Additionally, I found the distance between extruder and buildplate does not fit the height after leveling adjustment. I guess it may be caused by Z axis problem, and I need to know how to fix it. Any advise?

 

[illuminarti 18]

I doubt that it's anything to do with the z-axis. Most likely you have a loose pulley, or a belt or pulley that is rubbing on the frame.

Start by gripping the the four thick axis rods, and shaking them side to side. See if any move. If so, you'll need to tighten the pulleys while pushing them tight against the side of the machine.

Next, with the power off, try moving the head by hand, and see if you feel any resistance or rubbing, or hear any strange noises on one axis or the other. Sometimes the motor pulleys can be positioned slightly too far away from the motor body, and rub on frame.

Next, take a look at the two short belts as you print - do they show any signs of twisting or rubbing? If so that needs to be fixed by repositioning the pulleys (most likely at the motor end).

 

[ltc 1]

I think axis rods and belts looks good. Could you find any trouble in the video?

File "VIDEO0038.mp4"

File "VIDEO0040.mp4"

 

[illuminarti 18]

Nothing really obvious... although the movement of the head sounded like something was grating a bit, but I'm not sure if that's just the sound quality.

One other thing to check - look at the thin cross rods that run through the head. Are they properly seated into the black blocks on each end? Make sure that the x axis rod (left-to-right) isn't catching on the side wall of the printer. It can get pretty close sometimes... but it can be hard to see - maybe try slipping a folded piece of paper between the rod and the side panel as you move it back and forth. When it triggers the end stop at the back corner, it should cover about half to three-quarters of the silver lever on the end switch.

BTW, your glass plate is horribly gunked up with glue. Apply just a very light coat, and then spread it thoroughly with a wet (not damp) paper towel. Make sure the glue dissolves thoroughly, and then when it heats up and dries out you'll be left with the barest haze of glue on the glass.

 

[ltc 1]

Thank you for the glass and glue advise. But sometimes the glue doesn't stick well, and the PLA print gets warp as below. So I have to add more glue.

File "977193524340.jpg"

I try to check the X axis rod. Is it in regular status?

File "VIDEO0042.mp4"

I also find the gap at right side between rod and wall is bigger than the left side (I check in the video). Is it the normal and default setting?

 

 

 

[illuminarti 18]

Yes, it's normal for the rod to be closer to the left wall. It has to activate the endstop - but not catch on the wall. On the right side, it just has to fully engage in the sliding block.

It looks ok in the video.

 

[illuminarti 18]

Judging from your photos, it looks like the prints are leaning towards the front of the printer? If so, then the issue is with the y-axis - pay special attention to the motor, pulleys, and short belt in the back left corner. You can remove the motor cover by undoing the single screw that holds it on. It's located on the left side panel, a little to the right of the four screws with washers that hold the motor on.

 

[gr5 2,172]

It looks like you Y axis is slipping a little bit from layer to layer. So your part is moving towards the front of the machine and then sometimes towards the back of the machine.

When you have a sudden move it is usually caused by a loose pulley. But you have gradual - small move from layer to layer. This is almost always caused by the short belt rubbing near the motor.

You need to remove the back left cover to expose the Y stepper motor and belt. I would remove the motor and maybe add some washers or something to get the motor farther from the side of the printer - or maybe move the pulley on the motor closer to the motor - the gap should be much less than a mm, but not touching. Maybe 1/2 or 1/4 of one mm.

When you are printing - that belt from the motor up to the top shaft - if it is rubbing the wall you might be able to hear it or you might see it twist each time it changes direction - that would be more evidence that the belt is rubbing.

 

[ltc 1]

Thank you for the advise. I found the screws of Y-axis are loose, and they have been tightened this afternoon. Please check how it works now.

File "VIDEO0061_01.mp4"

I try to print a model, but there are problems. The surface of this model is supposed to be smooth. But you may see it is a little strange from the middle to the top.

File "VIDEO0053.mp4"

I also found the material feeder sometimes make weird sound.

File "VIDEO0052_01.mp4"

 

[gr5 2,172]

I'm glad you fixed the slipping Y axis.

The non-smooth surface is caused by "underextrusion" or simply "not enough plastic". This is usually caused by printing too cold or too fast. If you are printing < 240C then increase the temp to 240C might fix your problem.

If you still have underextrusion then slow down your print by 2X or make layer height thinner by 2X. For example go from 50mm/sec to 25mm/sec. Or go from .2mm layers to .1mm layers. Either of these changes will reduce the volume of PLA going through the tiny nozzle.

If you watch as well as listen to the feeder you can see it skips backwards. The feeder is designed to do this on purpose if the pressure gets too high. If it didn't skip backwards it would grind the filament to dust instead which is much worse because then your print fails and you have to take things apart.

This skipping backwards is what is most likely causing your underextrusion and again the fix is to print hotter and/or slower. There are hundreds of posts and experiments regarding this problem here - you may have a somewhat faulty printer (at least not as good as others) or it may be your filament is more viscous at your printing temps - so really the first thing to try is hotter temperature.

umforum.ultimaker.com/index.php?/topic/3976-almost-always-missing-layers-underextruding

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

 

[ltc 1]

I found another issue about Y-axis. Please watch the rod movement in the video. The rod looks unstable. How to calibrate

it? Can I get a replacement part from UM company? Or exchange a completely new printer?

File "VIDEO0064_01.mp4"

When I make UM2 stop printing in half way, it sometimes auto retract the filament out of extruder. Sometimes the extruder just go back to home position without filament retraction. Why does it happen?

 

 

[illuminarti 18]

Regarding the retraction, there was a bug in the firmware that was fixed a couple of

months ago. Make sure you have the latest Cura, and then use it to update the firmware.

[illuminarti 18]

I'm on my phone at the moment, so it's a bit hard to see the video - but yes, it looks like the axis rod is slightly bent. Check that it's fully seated at the back (nothing sticking out of the machine). I'd open a ticket with UM support also (support.ultimaker.com)

[gr5 2,172]

Yes, rod is bent. It *might* be enough to introduce very small errors. It might be worth fixing. It is unrelated to the problems you are having so far. DHL shipping is very very rough with the boxes. Absolutely perfect ultimakers often arrived after being sat on by elephants or something. I recommend you ask for *only* a new rod. It is very easy to replace - take a few pictures, then remove the set screws, slide out the old rode, slide in the new rod, tighten the screws.

 

[ltc 1]

Thank you for the suggestion of firmware updating. The issue of stop printing with material retraction is resolved.

In term of skipping backwards, I wonder if the sound is normal. There are 2 different kinds of sound when the feeder gear goes backwards in the video. I know the second sound is regular retraction (or the skip backwards sound?), but the first one sounds strange.

File "VIDEO0071_01.mp4"

Here you can see a part of the material has been ground and damaged by the feeder gear.

File "VIDEO0068.mp4"

I have raise the temp up to 250. That is how it works about 10 minutes after printing star. The filament cannot be added layer by layer.

File "VIDEO0074.mp4"

So I guess it is supposed to be Z axis issue. How to calibrate the Z-axis position?

 

[ltc 1]

@gr5

I think I found the key point of the skipping backwards issue. It is caused by the unsteady gap between extruder and build plate. The extruder sometimes runs too far or close to the build plate. When it is too close, the material feeder will keep skipping backwards because of the pressure. This situation will also lead to the material get ground. You may find the gap deviation in the video.

File "VIDEO0088_01.mp4"

I usually need to repeat the leveling calibration many times, and then the build plate can regularly work . It is annoying. Do you have any solution?

@illuminarti

After updating the latest firmware, I found the PLA material must be worked at over 240C. Otherwise, it will get underextrusion. Furthermore, the print looks not good as before. You can see two similar prints with different qualities.

File "IMAG0741.jpg"

File "IMAG0743.jpg"

 

They are printed at the same Cura setting, but the first one has much smoother surface. How to make it back to the better status?

 

[gr5 2,172]

the material feeder will keep skipping backwards because of the pressure

This is common but only a problem on the first layer.

I just watched video88 above. What is going on? It is printing much too high off the build plate. Something is wrong. Maybe where the motor connects to the Z screw is loose? Something seems very wrong with your Z movement. Also maybe the Z limit switch moves?

Does the glass ever crash into the nozzle?

For people who don't want to watch the whole video, he goes through calibration and it looks pretty good and then he starts a print and it goes to the wrong position - several mm off the bed when it starts printing. Also he sets the nozzle to 0C so it prints "cold" but that's besides the point. This is a test of the Z and the test FAILS.

Something is loose LTC - either the z limit switch or the Z screw to the motor or something inside the bed where the Z screw passes through.

 

[gr5 2,172]

I watched the video again. It's not hitting the Z home switch. You can hear it trying for maybe 5 seconds to keep going down but the Z axis is stuck. Maybe there is something under your bed getting jammed? Put the UM on it's side - you can see the limit switch on the bottom. Tell UM to "lower bed". Make sure the Z switch gets hit - you should hear it click.

Even better, you can turn off power and push down on the bed (well sideways will your UM is on it's side) and see the screw stick through the bottom of the machine and touch the Z switch. Hear it click. Make sure it clicks. If it does click then probably the other end of the wire isn't plugged into the PCB board - you can remove the cover of the PCB board with removing only 2 screws.

 

[ltc 1]

Thank you for your patience and the help.

I think the bed sinks down a little bit after it touches the Z switch. I don't know if the sinking gap makes the deviation.

File "VIDEO0092.mp4"

 

[gr5 2,172]

Uncover the electronics board underneath. The larger panel. Only 2 screws to remove. The wire may have fallen out. The connection is in this photo where it says "z-stop" mostly hidden behind a connector in the foreground but next to x-stop and y-stop.

 

[gr5 2,172]

Trace those (black I think) wires to the switch. If you have a multimeter, remove that connector and measure the resistance at that connector and when you push the switch the resistance should go from infinite to around 0 ohms (certainly less than 10 ohms).