IRobertI

It's always a little bit difficult to tell on a recording but I'm suspecting belts even though the noise is a bit more "sharp" then what I would normally associate with a tired belt. But it could well be the recording skewing the sound a bit as well. Often it's the short belt that is the most noisy so one quick thing you could test is to loosen the screws that holds the motor in place and lift it up (to try to disengage the belt from the pulley) and then move the print head manually and see if the noise goes away/reduces. You might have to remove the motor completely if the belt is stubborn and still rides too much on the pulley to test properly. The long belts can also be a bit cranky but not as easy to test for.

Big shoes, big beard and a thundering voice. One of my favourite, albeit a bit silly, memories is still when I was going to meet up with Sander and some other people a few years ago and I wasn't quite sure where I was going. Far in the distance I could hear Sander's booming laughter and knew I was on the right track 😄 And a big welcome to Mariska, I'm sure you'll do great even without a beard 😉

Glad to hear it helped 🙂 Say hi to UM-support for me 😛

Might be a little bit of LocTite in the threads that is causing you problems. In my experience switching screws around has always worked when one of them was being stubborn. Inspect the threads of the screws for damage or debris and clean them up and see if that helps. Heating the screw up a bit before inserting it might work to soften leftover LocTite a bit. Note that I have never tried that myself, just a thought I just had. And obviously you'd need to be very careful not to melt the plastic parts of the head. So maybe skip that altogether 😄 If all else fails your local Ultimaker reseller(s) can help you get new screws should that be necessary.

That is correct 🙂 There's a longer one, the one I remove at 17 seconds in the vid, and then the shorter one that remains in the head that goes through the white vertical PCB that remains in the head in the video. That's the bearing that usually has issues.

You would need to take the head apart and put it around the outside of the bearing to take up the extra space. Here's a video I recorded for a customer on how to take the head apart. This was to show the lifting mechanism so I don't take out the bearing in the video, but you will get close enough here to get it out. If needed you can also remove the print head board that the bearing goes through. It's held in place with one screw from the back. https://www.dropbox.com/s/q356qgu1tsfqtfv/head_disassembly.mp4?dl=0 I would start with just a single wrap of something like painters tape, it doesn't take much.

I would check the bearings in the head. If you grab the head and try to twist it around the rods that go through the head you will find that for the X rod it will be quite firm and not really wiggle much at all. On the Y rod you have a much smaller bearing in the head and it's normal to have a wee bit of wiggle there. But if the bearing is in bad shape you might find that you have a bit too much wiggle. One sign of a bearing that isn't doing so well is a buildup of a lot of black... "goo" on the rod. It could also be that the "slot" for the bearing isn't holding it in place as well as it should. A very thin piece of tape around the bearing might take care of that.

Google translate: Bleef je de hele tijd in het tune-menu tijdens het checken, of heb je het gesloten en weer geopend? Als ik het me goed herinner is er een rare bug die er altijd is geweest waarbij de temperatuur van de nozzle niet begint te stijgen als je in het tune-menu bent tijdens het opstarten. Het kan ook zijn dat de verwarming van het mondstuk kapot is. Ga eens naar Maintenance -> Advanced -> Heat up nozzle en kijk of het mondstuk opwarmt als je de temperatuur daar verhoogt.

Google translate: De temperatuur van het mondstuk begint te stijgen wanneer de temperatuur van de bouwplaat in de buurt komt van het doel

Yes, I think that would be a good next step. But I would try gluing it first, if that works for a print that usually fails we can pretty much conclude that the arm was the problem. If the glue is good, it will probably keep working just fine like that and you wont need to buy a new feeder.

Yeah, I'd say the feeder arm is the culprit here. What you can try to do is take it apart (gently, the little tabs are easy to break, so go slow and careful) and glue it back together again with some epoxy or super glue. That's a cheap way to test if you need to buy new parts or not. Or just keep using it if the glued together arm works fine.

Does the arm of the feeder look to be in good condition and not splitting apart? Especially at the end where the ball bearing is. If you grab the part that moves the filament forward and pull/push on it, does it have a lot of movement? On old feeders the gear on the back of the feeder sometimes moves out of position so that there's a lot of movement back and forth which can sometimes cause issues. On newer feeders the gear is riveted in place and can't be removed, on the initial feeders you can just pull the gear off.

E-step calibration is not something that is generally needed or done on an Ultimaker 3, something else is causing your problems. If you describe the problem in detail, ideally with pictures as well, we can start figuring out what's going on.

You can use anything of roughly the same thickness. I've used folded up pieces of paper in the past.

You have both an UM3 and an S5, you could just take a look 😉 Go here: http://IP-OF-PRINTER/analytics and at the bottom you can download a CSV-file with all the stats. Example (these are not the only columns of data)

Not without some sort of 3rd party solution like the Mosaic Palette or similar. The Material Station is only compatible with the S5. The UM3 doesn't have the needed flow sensors in the feeders for starters.

Please post a few pictures of this, or even better, a video if you can.

Are you sure it's not a piece of leaked plastic that formed itself around the sensor? I think there's a bit of internal "sleeving" inside the sensor IIRC which can be what you're seeing as well, but I don't remember it creeping out that far. Either way, if it's reporting sane temperatures I don't see a problem. Keep an eye on things the first time you fire things up again. The heater should go further in than that. You should be able to see the tip of it through the little slot at the front of the block. You likely have some plastic in there that is preventing you from inserting it fully. Do you have a heatgun? That's the best way to heat the plastic inside there up enough for you to get the heater all the way in. An alternative is to briefly turn the heater on while it is inserted half way into the block. If the sensor isn't in the block, or if things aren't making proper contact the heater will quite quickly get extremely hot since it can't dump all the heat into the block and it has no temperature feedback to know when to turn off. So turn it on for say 5-10 seconds at a time to heat things up.

Have you tried going directly to the stream? That would show you if the camera is still alive and there's something else going on preventing you from seeing the output. http://IP_OF_PRINTER:8080/?action=stream

Glad to hear it 🙂

It's not difficult to do. You need a micro sd card, the firmware image and software that makes it bootable once inserted in the printer. There's a PDF with instructions for it (which I wont post openly, I prefer that customers get in contact with their resellers before doing that).

You were too fast 🙂 I edited my post above to list the three scenarios I think it could be.

No, not the screws in the frame. When you turn the printer over onto its side you will see that there's a plastic cover that covers up the entire bottom side of the printer. That cover is attached with 8 screws and is the one you need to remove. The cover has a speaker attached to it which is connected to the mainboard underneath the printer. The cable is plenty long but good to know about so you don't yank on the cable by accident. I'd say there's three possibilities: - The display has died for some reason (pretty uncommon). Very easy to replace if so. - The cable is loose - Something is wonky with the firmware. This could likely be fixed with a firmware recovery. Your reseller can help you with information on this. Of course it sucks to have to fix a brand new printer, but as I said, it might be less hassle than packing things up again and having to send it back.

To start with I'd check the cable for the screen, just in case it's something silly like the cable having gotten disconnected in shipping. That would be a lot less hassle than doing a return if that's all it is. Unplug it, put the printer on its side and remove the screws on the underside. Remove the cover (mind the speaker wire) and check the cable, it'll be obvious which one once you're under there. Unplug and replug both ends to make sure it's well connected.

No, not without hacking the firmware. Which for obvious reasons isn't really recommended. Personally I think it's a pretty small price to pay unless you're printing lots of 10 minute prints.