Nicolinux

Does it really make sense to use this for a temperature test? I mean if you wait too long with changing temperature, you'd also get more volume per second because the test is designed to increase it over time. Ideally I think you would need an object with the same volume where you can test the temperature gradually.

@3Dmaker4U: If that's the case, then my former tests with flow rate up to 150% should have produced different results, but the strands were still not touching. Also looking at the underextrusion test where it fails really late (between 13-14mm^3/s) I'd say this printer doesn't have problems with it. @MartijnvG: Thanks for the detailed response. Yet I still think this issue is unique to my printer (or others as well who don't care much about surface quality). I did some more tests: Printed the standard 20mm cube at different speeds and feeder settings. Default print settings for all cubes: layer height: 0.2 shell: 0.4 bottom/top thickness: 0.6 infill: 0 travel speed: 150 combing: disabled location: center of heated bed There is no real difference between 40mm/s and 20mm/s. Also the temperature (230°) is pretty high for these low speeds. I've also changed the feeder in between (Bas sent me a new case with the metal grommet). For the "tight" tests, I've set the tension at the second notch. Here are two closeups of the filament that has been through the feeder: This one is with retraction involved (during a print). And this one is just through the feeder once during a print (I undid the bowden at the feeder and just let it move through). The filament didn't reach the hotend, so I guess here, less back-pressure is involved.

Nein bin ich nicht, aber das kann man ja ändern. Und dann gäbe es auch 3D-Druck-Projekte

Ein weiterer Vorschlag: https://www.cccmz.de/ Findet in Wiesbaden stadt (in der Nähe von Mainz und Frankfurt). Ich komme aus Wiesbaden und habe einen UM2.

Yes you are right - I meant layer height... Updated the post. thx

Just another remark. The bright spots (where the light shines through) are the spots where the strands do not touch at all. That's why the square looks so frazzled. So this is not a minor issue for me.

The belt climbing issue is resolved - I printed my own spacer and replaced the one that was wrongly used on the bottom right rod. I couldn't find other places where the belt behaves in an unexpected way. And regarding the first layer - that's pretty easy to explain. Since the first layer is wider than the rest (0.3mm) it just hides this problem by making the strands overly wide. On the two prints above I changed the first layer height to 0.2mm (still have not got to grips with Simplify3D - this was my first print with it). I think I will finally take out the rods today, check them if they are straight. I will also check the pulleys. Not keen to do it, but won't accept status quo either

I did, here are the results: It looks alright when it is on the bed, but a bit wonky when held like this (I rotated the image to match the printing orientation): I then printed another one with the infill at 90 degree (to get the other axis): The heat distribution is not even (it is worst on the top left corner). I have printed with other colors or course and at higher temperatures. The strands not touching problem is never gone, just very much less visible with higher temperatures. And it always is worst on the upper right corner of a print. That's why I think that it is something mechanical. I also printed the hollow pyramid but the stringing at 230° was not that pronounced compared to 210°. (210 left, 230 right)

I sure am. The underextrusion problems are gone with the new UM2. I just tried to make a connection to the actual strands not touching problem.

I think gr5 is also right. If you look on the vertical side walls, you'll see that entire layers are missing or receding inwards. This is a sign for underextrusion. My problem is that all lines/strands except for the first layer do not properly touch eachother so the surface is not "watertight". But your's looks like it is leaving out the outer shell completely.

This looks a bit like my issue with strands of filament not touching. Check out this thread: http://umforum.ultimaker.com/index.php?/topic/5375-top-layers-not-touching-um2/

I am not sure what this could be. Normally I'd say bed leveling, but the problem is not sonsistent and it recovers on the upper left. It looks like the nozzle scrapes over the already printed layer. Maybe the bed is warped. Could you take the glass plate out and see if it is completely flat? Also if you have other means to test the flatness of the heated bed - that'd be the next thing to try.

@Dim3nsioneer: Well that's why I ordered the better pulleys from user foehnsturm a few months ago when I still had the UM1. @Sander: Is there any chance to not have to take all rods out and all pulleys? Because that's a b**ch to do :/ Apart from that, I have other problems. Printed a thin walled object and the surface looks bad. It should be smooth and not riffled: Also had some underextrusion (printed at 210° and round 40 speed). I did run the underextrusion test again (at 230°). It looks fine and fails late, but here again - underextrusion and then it is fine again: @3Dmaker4U: I did measure it (because a layer was sticking out on a print). It was close to 0.2mm so this should be fine. ---------------------------------------------------------------------------------------- So I did a few more tests. Printed the standard cube and paused after every layer to take a photo: So there is always a pattern - a part of the layer is printed almost fine, and the other part is bad. I re-leveled the bed and made it to be a little closer than usual but it didn't make a difference: Then I printed at 230° and the problem is almost gone. That's not good though because I'd really like to print cooler to reduce stringing and all kinds of other effects: Next things to try: pulleys (don't want to but I can't rest if it does not work) firmware reset/update Here are a few notes about my workflow. I don't think I am doing anything special. Printing with Faberdashery white. I use Martin's loose filament spool holder and the standard UM2 spool holder: The printer is in my living room now and sits on a flat, low table:

As far as I know, they do. So you should be able to configure the webcam output via the OctoPrint interface. Check that you can access the webcam url directly - so that you see the stream. If this is working, then OctoPrint should pick it up too. Also check that OctoPrint has permission to write to the config file.

Good news, at least Gina (the OctoPrint developer) is aware of this issue. I hope she will tackle it soon.

You need to add something to the config.yaml file: Open ~/.octoprint/config.yaml (just create it if it doesn't exist yet) and add the following lines to it: webcam: stream: http://<your Raspi's IP>:8080/?action=stream snapshot: http://127.0.0.1:8080/?action=snapshot ffmpeg: /usr/bin/avconv Source: https://github.com/foosel/OctoPrint/wiki/Setup-on-a-Raspberry-Pi-running-Raspbian

Sorry to crash the party but I see some (very faint) underextrusion. Also there is some banding. Did you by chance print with Faberdashery white and which settings did you use? I am asking because I have a somewhat similar issue.

Pulleys on the steppers are fine. I had to tighten them a lot, but the problem persists.

Hey Markus, You forgot the following video

Ok, I spent most of today trying to fix this issue. 1. Axis Alignment I re-aligned the short rods which took ages because the pulleys on the long rods are hard to reach and tighten (the fine gentleman who assembled my UM2 didn't care to align the pulleys so the screws face in the same direction). While doing so I found out that he also messed up assembly and fitted the wrong plastic spacer on the x-axis (front right when facing printer). This lead to a problem where the belt sractched the corner of the pulley. Here is a video of the issue (the belt bends weirdly when the slider block approaches the pulley: I printed a new one and it works fine (hopefully the friction won't produce that much heat as for the spacer (PLA) to soften)... 2. Possible Underextrusion After aligning, I printed a test cube and the pattern of the non touching lines changed. They are evenly spaced now: Printed another one and set flow (gradually) up to 150% Then I remembered what you guys said about possible underextrusion. I used a Taulman 618 sample to perform a through nozzle cold pull. (read about nylon to be very useful for that here is the - http://bukobot.com/nozzle-cleaning). And indeed, there was some gunk in there. The tip of the Taulman filament looked like this: Then it was clean for a couple of pulls: But then all of the sudden (I barely could pull the filament out), it looked like this: Continued with cold pull and finally concluded with PLA cold pull: Conclusion The strands do not touch (it seems to be worse on the top right half). I still think it is something mechanical. When a print starts, after the preparation and initial extrusion, when the head moves into position - the sound it makes is a bit unusual. It sounds strange - like it would resonate through the entire frame and amplify the movement of the head. I don't remember this sound characteristic from the other two UM2's. I am past anger and disappointment now and start to not care any more. If this continues I'll put the printer on ebay and be done with it for good. I'll try one more thing (check the pulleys on the steppers for the x/y axis). Since most pulleys were not tightened properly when I got the printer, maybe those on the steppers have been overlooked too.

Hey Markus, Just a little feedback. The fact that changing filament is not easy, is a deal breaker for me because I do change filament on every print (almost).

I wonder why no one is interested in the elapsed time. Since print time estimates are way off, it would be very nice to actually know how long a print took (just in case if you print another one and need to decide if you want to invest the same amount of time again).

Das Ultimaker Rot ist auch ziemlich... rot. Hier ist ein Beispiel: http://www.thingiverse.com/make:72748

Hi, I was just about to start an indepth analysis of what I could be doing wrong and write a lenghtly post about my workflow when a thought struck me. The problem is mechanical and very sucky because I am not going to take entire UM2 apart to fix it. The strands are not touching because the small rods (where the head moves onto) are not parallel to the X/Y rods. I printed an axis align tool and checked it on all axes. When I latch it onto the 8mm rod and slide the head so a small rod fits the outmost notch (6mm) and then move the align tool up - the small rod doesn't fit into the notch because it is farther away than at the bottom. Unbelieveable but I am really that guy with a huge amount of bad luck when it comes to 3D printers :mad:

That's not so easy to measure. But it is around 4mm. The line and spacing pattern changes from bottom right to top left. On the bottom right two lines are touching followed by a gap. On the top left the single lines are evenly spaced (and still not touching).