phil-t

Nice to know about the upgrade kit! I've already done the Olsson block / 35W Heater and new temp sensor, then printed an alternate feeder/fan duct, but having the official upgrade kit would be definitely something I'd like to fit!

Hi Izzy, I quite like the 'O' mount at the back, that way if you did add the original mount holes, when you undid those, you could still swing the fan out the way, or just unscrew the rear head bolt to completely remove it. To be honest, I only print PLA with a 60 degree bed temp, so it's more a fit and forget I'm after, but some slotted mount holes would allow me to move it up/down and get the height perfect! I have had a couple of shutdowns due to the excess air, so am just messing with the fan speeds and duct heights to find the perfect settings!

Hi Izzy! I've just printed a set of the fan ducts, they seem really good, and certainly a massive amount more air than stock! My only comment on a quick print are that Instead of the 'clip' mechanism, I'd love it if the inner face was elongated and two holes put in which allowed it to screw into the existing shround mounting holes so it was strongly mechanically mounted. The reason for this is that it 'rattles' slightly and has a bit of play that means it often drops slightly on one side. The fan mount is genius by the way, it really holds those fans in very well and I don't worry at all that one might fall out! I only print PLA, so am hoping with a 60 degree bed temp that it will last a while.. And I've found pretty much any more than 50% fan speed can start cooling the head a little too much! I have a suspicion I'll be able to get very nice overhangs with this setup!

Some good ideas already! For me, I'll echo some points already discussed, plus my own. 1. Improved filament guide, there are a few more enclosed filament guides that don't allow the filament to pop off like the original one (e.g. https://www.youmagine.com/designs/ultimaker-2-filament-guide) 2. Bearing'd spool holder with conical retaining nut (e.g. https://www.youmagine.com/designs/low-friction-um2-spoolholder) 3. I would love a flex3drive feeder or similar to print more materials easily. Or improved feeder that has a door mechanism that makes changing filament less prone to straggly bits getting stuck in it. 4. Improved hotend with no PTFE coupler issues, I'm still scared to even go near ABS and know it's a case of when not if my current coupler starts having issues 5. four post main bed mechanism, maybe with 2 drive screws etc, but something to make it 'solid', the cantilever design works quite well, but definitely one area that causes some precision issues 6. Ethernet connection so we can have a web-interface to control/monitor prints remotely 7. Built in camera, and you can monitor it from the web-interface. 8. Feed detection and other simple detection mechanisms added to help detect and stop bad-prints for unattended use, and also stop catastrophic issues if the head gets impeded. 9. Dual extruder's - one for de-solvable supports That's about it! I think the main usability issue for me would be cured with an ethernet interface, I contemplated using my Pi 2 with OctoPi, but I want full UltGCode, so perhaps some collaboration with OctoPi or similar would help? And points 1 and 2 are in response to me getting work and friends to buy UM2's recently, and it's the same 'essential' first few improvements they've all ended up printing, so you do think it'd be great if UM improved those simple aspects.

I can't remember the last time I had to calibrate the bed (Ultimaker 2), maybe 30 or so prints ago and am still happy with the calibration. I use a 200 micron first layer, printing direct on glass, and get a very even/glossy finish.. I am also a bit dubious of using any auto-levelling that constantly requires the adjustment of the Z-Axis, it would most certainly have an impact on print quality as opposed to the normal mechanical calibration. Saying all that, I'm not against innovation, so I see no harm in asking..

I would also say that 95% plus seem perfect at the moment, although the last 10 prints have been perfect, no finishing required. I started out with a very hit and miss success rate, the improvements have been partly sorted out small issues with the printer, part educating myself on material flow rates, and partly educating myself on what does/doesn't print very well.. I do design my own stuff when I can, and my success rate with that is very high.. some stuff from Thingiverse/Youmagine can be a bit weird, not that it doesn't print, but more it needs more finishing that I like.

Hopefully final update! Removing the USB cable from the UM2 has seemingly completely solved the problem! I've been printing (touchwood) for nearly 2 weeks and not a single hiccup!!! I thought it best to update the thread, since if anyone else gets this issue, they will at least known one possible cause!

Yeah, it's a 5V DC basic fan, nothing fancy!

Whilst I've been sorting out my UM2 issues, the fan was very loud indeed, so I bought a 25x25x10mm Sunon Maglev fan and used m2.5 x 12mm cap head screws and used a dremel to take some excess bulk on the outside of the fan to make room for the hotend feed wires.. I will say it's much quieter, but not silent, although I am only 2 feet away from the printer, its very tolerable.. it stated 5m3 flow rate, and it certainly is pushing a nice amount of air through! http://www.ebay.de/itm/Lufter-5V-0-43W-25x25x10mm-5m-h-16dBA-Sunon-MC25100V2-A99-/231085071506?pt=Geh%C3%A4use_L%C3%BCfter&hash=item35cdbe5c92 Mind you I've yet to print anything, so I'll see if anything is affected!

Oops, My bad, I didn't read your post and failed to spot you where referencing your i3!

Hi, The first thing to do is obviously a standard Bed Calibration using the Maintenance->Build Plate menu option, but I then print something with a nice large first layer, and manually adjust the screws as it's going around (always adjusting the screw when the head is furthest away, and doing it by trial and error, I can get small objects to put a perfect solid polished finish on the bottom layer at 0.1, however I've recently just discovered my glass plate is actually bowed so I can't do larger objects with a 0.1 layer just yet! There is no 'option' to change, the build plate calibration built in pretty much gets you with 0.2mm or so following the instructions and using a thin sheet of paper as directed. e,g, this is the bottom face of a small halloween badge I did for one of the kids.

Thanks again for all the great advice.. I found a countersink bit in work that I borrowed, I removed the levelling screws, and the heated bed cable retaining clamp and use the countersink in a screwdriver handle (it has a hex shank) and did all the screw holes by hand.. It took 2 hours to do the glass retaining screws and bed levelling screws, but all is nice and flush now!! Just for a laugh, I have indeed turned the glass over so the bow is topmost and am doing some test prints and I'll keep an eye on it whilst UM support get back to me! At least I'm OK for small jobs in the centre of the glass! you never know, the heat/head pressure on the first row might magically bend it back over time!! I'll phone around some more places tomorrow to see if I can get any cheap deals on glass, it's the rounded corners (needed for clearance with the front clips) and polished edge (to stop any cuts) that seem to add that bit more cost, but I'll see what I can find!

I've wound up the spring tension (which is effectively pulling that screw head down) quite high, and it's clear the countersunk hole diameter is a shade too small, or the screw heads are a shade too large! I work next door to a company who sell fixings amongst other stuff, I'll see if they have some M3 x 20mm countersunk bolts in to try, if not I have found a countersink bit in work, which I could try. I think I'm correct in saying I need to address the proud screw heads before thinking about new glass, since that is more than likely the reason it has bowed in the middle?

Thanks, I know support are snowed under reading other threads, so I usually ask here first since you lot are very knowledgeable! But as it seems like the answer to my issue is going to cost a bit of money I've raised a support ticket to see what can be done.

After getting to grips with the UM2 (thanks to this forum), I've just found a bit of an issue. I noticed that larger prints had a really uneven first layer, I've been experimenting with 0.3/0.2/0.1 bottom layers and levelling the bed. After a while I've noticed that while I can level the bed using the manual tweak of the screws to just dial in a perfect perimeter thickness, I'd get odd patches of infill on the first layer that looked like the head is miles away from the surface. On inspection, I've noticed that A. The levelling screws are slightly proud of the aluminium heated plate, for example, the best one is: So I guess a bit of additional countersinking is in order, and the aim would be to be slightly under the surface? B. The glass is warped! - I was shocked to see it's actually so obviously bowed by a mm or more, for example, if I take a ruler, and take a picture of the glass with the ruler on it, then turn the glass over and take another image (to rule out any issues with the ruler not being straight), I get: I've already phoned some local glass places, but they all want £30-£50 for a one off ground edge toughened glass which shocked me, then I wondered why I was just faced with spending £10 for a decent countersink bit, and up to £50 on glass which may well have sagged in the middle due to being supported on the proud bed levelling screw heads, so would anyone think this is a case of pleading with UM support to see if they can help out?

Just to chime in and say this affected me the other day, I aborted in the middle of some long in-fill lines, and the head instantly starts cooling while the buffer is being emptied, resulting in a blocked nozzle.. Obviously you can work around it to some extent by aborting when it's working on smaller moves, but I think it'd be nice if the firmware could make this fool proof!

I'm using this thread to document the issue, so will keep posting anything I find. One oddity I just had was it stopped again, but exactly when my PC came out of sleep! It could be USB related, either the PC suddenly powering the port or some other interaction, I forgot I actually had it hooked up via USB and will be relieved if that's all it turns out to be! I'll keep printing with no USB connected and see how it goes over the next few prints! Just to be clear, it's not resetting the board due to the serial port opening (I've since done that, using KiTTY to talk to the UM2), it was potentially causing the X/Y to just stop mid-print, even though when you abort the print, it can move the head 'home'..

Just an update, It printed OK for a few days, no stops/pauses, but today it's started playing up again! I was trying to do a test print of Emmet's Gear Bearing to tweak settings for good dimensional work, and after a few aborted attempts due to incorrect settings, I started a print which just stopped after 20 minutes. The machine is sat there, it thinks it's printing, showing the time left and I can go through the usual tune menu's etc. I could abort the print, at which point the bed and X/Y home, so clearly it can move the head!.. If attempted the print immediately after this, the bed warms, the head heats up, it does the pre-extrude and then stops, no X/Y.. I aborted and tried again, same thing. I had to power cycle the machine, then it started printing, but after 1 hour it has stopped mid-print again. I'm going to try another SD Card and see how it goes, but I'm wondering if there is anything I should look at?

From my own brief experience, I've found the Ultimaker PLA doesn't flow so well at 190-200C on my UM2 and I was getting the same issues. I think some of it is learning your own printer and whilst doing a few solid infill prints, I took the opportunity to play with extruder temperature and material flow rates with speed, I've found for my machine upping the temp of the extruder closer to 230 seems to make for neater infill lines, and a material flow of 110% seemed to also improve the infill consistency, with that done, I've been happier with 0.2mm layer prints now. There is definitely a necessity to 'understand' your printer and each material, I do small tests while printing more uniform parts using the tune menu to see the effect and try to arrive at a happy set of values.

Thanks, I'll bear that in mind! I think in this particular case, the file was intact since I later printed it OK from the same SD card that had not left the UM2, I'll keep an eye on it, I've set up an IP Camera so I can monitor it during the day, I only work 1.5 miles away and can pop home any time should I spot it stopping again!

That is good advice, the closer the Z Axis guides, the less chance of 'bounce'! I'm also just getting to grips with understanding the printer, and the support on the forums really helps!

Being honest, no I didn't (My card reader does have an activity led that I always wait to show no activity for a few seconds before removing), but I never removed the sd card at all between it stopping and eventually restarting the print, and its happily completed that same print after resetting the um2. I've since removed the card to add the next file to it and have used the 'safely removed the device; in windows just in case though.. I presume the issue of not ejecting would be an incomplete file/corrupt file system in some way?

This is a bit of a weird one, but I started a print earlier, and after about 20 mins I came back to hear no real noises (other than the fan), when I looked in to the build area, I saw that X/Y had just stopped, and the head was sat in the same place, extruding away merrily.. The display was indicating it was mid-print with the progress and time remaining still showing, and I could abort the print, at which point the bed/head all moved back as normal after aborting. Here's the pics.. After aborting, I tried to restart the print (after removing the old item off the bed) and it moved the head to the correct place, moved the bed up, but nothing extruded and again X/Y wasn't working, it just stayed at the front left of the build area. So I power cycled the printer quickly, and I assumed the nozzle may have blocked due to sitting around extruding into itself, and did a full material cycle, atomic clean and reloaded the material. At this point I happened to glance on my shelf and noticed the big blob of plastic that I found around the entire nozzle (under the guard) that I had to remove the day I bought the printer home (I bought it second hand and didn't spot it when checking it over), it was lodged above the nozzle guard enveloping the nozzle. Just to be safe, I factory reset (It's already got the latest firmware when I first launched CURA the other day), and then re-levelled the bed, and restarted the print which is so far doing OK. My question is really what could have caused the issue? I'm now a little nervous because I am starting to suspect the previous owner may have run into the same issue judging by the amount of green PLA melted around the entire nozzle block (now removed and cleaned), which I fear may have happened to me had I not spotted it within a few minutes. Just to re-cap, the symptoms where just no x/y movement at all but seemingly everything else working as expected, and it would home/move fine when I aborted. I will obviously leave it printing, and see what happens, but I just wonder if anyone has any advice or may have heard of this before?

Howdy! Is that Newnham-on-Severn? I used to commute through there most days until I moved closer to work!