gr5

However, even with this grinding of the black ABS feeder, you should still be able to print fine. Just put it back together and see if it starts working again. Set the spring tension to the MINIMUM. You want the screw loose so that the white arm is all the way up. I suspect you have too many retractions at 4 hours into the print. Look at the gcode with cura - are there tens of thousands of retractions? Maybe play with the expert retraction settings to reduce the quantity of retractions. The UM2 is pretty good at retractions. I did a print where there was over a kilometer of retractions. A KILOMETER! wow. On that print each piece of filament retracted 8 times. On average. Some pieces more than that. But some PLA is softer than other PLA and I suspect this newer red filament is a little softer than the blue filament I tried. So look carefully in expert settings at the 2 "minimum" settings for retraction and consider setting one of them to maybe 2mm (the one about gaps) and the other to maybe .1mm (the distance of filament one). If you set the second one to .1mm then the filament can not retract more than 45 times on the same piece (4.5mm / .1mm).

I suspect #2 and it's a serious problem because that black ABS can clog your nozzle slightly and degrade the extrusion ability at faster speeds. I recommend you move the filament on to the floor for now until UM has a new feeder: Having the filament come in at a more vertical angle can help this if the problem is mostly at the entrance to the feeder but if the problem is elsewhere then it won't help:

Please push all 3 axis away from their extremes and try it again. With power off you can push the bed up and down but do it from the REAR of the bed - near the back of the machine - don't push up and down that hard on the front edge of the bed. It definitely sounds like your X axis isn't hitting the limit switch but I'm confused about the Y axis - that should at least move to the rear.

All the fans are very quiet. There are 3 fans. The rear fan comes on first before even the light comes on. The side fans come on when printing but you can turn them on before printing if you tell it to print and jump to the TUNE menu. The fans should be very quiet. The fans are quieter than the x,y,e steppers which are about as loud as a quiet dishwasher. Then there is the Z axis which is very loud but infrequent. As dkaygee says, it's probably loose screws or maybe the sheet metal that holds the side fans is bent/rubbing or something else is wrong. I accidentally put a screwdriver into one fan and it was noisy for a few days until I just bent the little spot back again. Now they are whisper quiet once again. I also had one of my 4 screws loose and it wouldn't tighten (just was loose and rattling) so I just removed it and one fan is held in by only 3 screws and it's fine.

X/Y resolution is just as important as Z. The X/Y is limited by nozzle diameter which is .4mm so anything you print will have a .2mm radius of curvature to it. For example if you print a cube, the 4 vertical sides will have rounded edges with .2mm curvature (the other 8 edges will be 90 degrees perfect sharp). Printing thinner than .05mm starts to be a waste of time. You can only see the layers because of the nature of the way light reflects off a shiny surface. If you paint .1mm layers they disappear.

We really need to see the video. Please make it public! There are 3 possible things that make an axis go the wrong way. 1) Either a limit switch is stuck, 2) the DIR signal is stuck, or 3) the stepper motor connector is wired backwards. #3 seems unlikely because that would have been caught at the factory. #1 is most likely but you say you tested it. X *and* Y axis not homing? This is very strange. How could both break during shipping at the same time? Need to see the video.

Yes! What those 2 guys said! One support for pointy bottom in center, and use lots of brim - maybe 10 passes of brim. Make sure you use some glue stick on your glass bed and make sure you heat to 40C to 60C. I wouldn't go over 60C for this part as you want it to stick really well. Make sure the bottom most layer is squished well into the glass. If not it might fall over during printing. But if you use glue, bed temp 40c to 60c and squish the first layer, and brim - then this thing will not fall over even if your cat attacks it.

Nope. Never. Not yet anyway and I've hit it harder than I should have. don't try to remove until temp is down to 50C as at 60C the bottom might still be above glass temp and it might warp instead of come free. At 50C, if you are impatient, then go for it. But if you wait until 20C it often just pops off on it's own or with very little force. This is because glass and PLA have a different expansion coefficient. My favorite tool to remove parts is a putty knife that I sharpened with a file. Once you get a little bit under the part, it comes up easy. Getting that first millimeter corner is the hardest part.

No idea about the time thing - the latest version of Cura is getting very accurate about the time, but the time that the printer shows is not as good.

A quick google search shows that printedsolid.com reselles colorfabb in usa but they don't seem to have that color. I'm done for now but try googling colorfabb - maybe they have more resellers in the US.

I think it's Colorfabb for two reasons: The filament spools are identical (were identical they switched to clear ones) and they are in the Netherlands. If you look at the filament spools on the websites of all the sellers of filament they are almost all different - no two places seem to use the same design.

I agree with Ian - probably just off by a tiny bit in the levelling. Also don't put the glue on super thick - a very thin layer is fine. I'm usually lazy and just put it on directly but I'm going to print a 20 hour print or something I wet a napkin and spread the glue stick glue around very thin. There's nothing special about 50C heated bed. Any temp from around 40C to 70C and it will stick equally well. However for large prints, more than 100mm across, 70C will help keep it from warping and lifting at the corners. But anything hotter than 70C (e.g. 75C is too hot) and the plastic doesn't cool fast enough and you get what *looks* like warping in the bottom 5mm (but is actually caused by the filament not being solid and the rubber-bandiness of PLA filament when still liquid).

Ian did you go to the mini makerfaire in hanover germany last December? I would think that would be a good place to meet and Sander might even join you there. You should have gone to the big one in Eindhoven last weekend. Sander was probably there.

I printed nylon PA6 on the UM2 and it worked fine. I used blue tape on glass. My notes say I printed at 260C and the bed was at 50C. I figure a warm bed will let the nylon flow into the cracks of the blue tape better. I probably cleaned the blue tape with alcohol as I almost always do but that isn't mentioned in my notes. I had to relevel with the blue tape of course. One of the objects was a christmas ornament so it was kind of a large sphere compared to the tiny point of the sphere at the bottom and I was nervous about it getting knocked over but it was fine.

I'd rather a hug than a kiss, but thanks. I don't really understand any of those fix horrible settings. Daid (cura author) has explained them several times but he thinks of polygons in a given slice yet we think of 3d objects. Anyway, that bracelet has an outer and an inner wall which is normally good for 3d printing but your requested shell thickness is .4mm yet the wall was .39mm in places and .41mm in places and it makes for a mess when printing it out - the .39mm walls were being ignored as too narrow for a .4mm nozzle. So we tell Cura that it is a solid part and that double wall is a fluke and ignore the inner wall, The "combine everything" helps us make it into one solid part - a bracelet without a place to put your hand through. Then we tell cura not to print: top, bottom, infill. That way it is now back to a hollow bracelet. As far as why you need to check 3 of the 4 checkboxes in fix horrible - I really don't know. I just started trying all 11 combinations at random until it worked (One might think there are 15 combinations but Daid told me that checking both "A" and "B" does the same thing as checking only one of them (I forget which one)).

If you really must fix the STL then this procedure might work (although it might leave you with holes in the print). I suspect it will work for you although you might need to drastically increase the light source: removing internal nodes/faces http://meshlabstuff.blogspot.com/2009/04/how-to-remove-internal-faces-with.html But really - just check the 4 "fix horrible" settings (except for "B") as that will only take a few seconds versus an hour installing and playing with meshlab.

Well I hate it when people make me repeat myself:

Very nice. I see your UM2 is printing in the background. And the filament is on the floor! Nice! That's the secret for a happy UM2!

The wiki is editable by anyone (you just need to register) and Dimensioneer knows this so he edited it himself a while back.

Oh - and if you change color or manufacturer of PLA or change your printing temp or speed or fan, then all these adjustments change a tiny bit. Again. Or if you change to ABS.

I strongly recommend you don't mess with steps/mm. This thing is calibrated to be spot on and if you mess with that then the software endstops will be messed up. For example if you think it is printing 3% small due to PLA shrinkage and you increase steps/mm by 3% then if you ever print something 225 or 235mm wide it will hit the clips or the wall and your part will get messed up. There are many factors that change the size of the part to not be what you expected. PLA shrinkage is one of the smallest effects. There are many others. For example if the bed is too hot the lower 3mm or so will kind of shrink in. due to a stretching affect that has nothing to do with cooling/density and more to do with elasticity (think rubber band) of a strand of hot pla. This same problem exists for holes in parts - they tend to be too small due to 3 factors. The 3% shrinking of PLA, the stretchy effect (the strongest effect) and the fact that CAD converts a circle into a series of lines (like a hexagon) but all cad programs *inscribe* the polygon so that it is smaller than a circle (never larger). This is fixed by doing say 40 sides instead of say 10 sides but the bigger issue is the stretchy effect anyway. No, the best solution, and what most people do, is you print parts twice. The first time you print it as planned. Then you measure all the critical tolerances and adjust by the error back in cad so if a cube is .1mm too narrow on the length then make it .1mm longer. If a hole diameter is .7mm too small then enlarge the whole by .7mm. After a while though you get really good at guessing the errors on the first try. This is nothing compared to CAD for "real" plastic parts made with injection molds where you have to also muck around with angles (after they pop out of the mold all angles change!).

lol. That's me. A moderator. I looked at it a little closer and didn't notice anything unusual. Anyway - be aware that if there are only 1 or 2 strings per part, and if you are printing multiple parts then this is a completely different issue. It's code that not many people use so it doesn't get quite as much attention - it's the between-print retraction stuff. Someone should fix this with a plugin or something.

Well you can lie and say that the nozzle is .3mm and Cura will act just like it is a .3mm nozzle and underextrude appropriately and put the lines closer together and such. It might, kinda work. I don't recommend it.

The black feeder on the back of your machine is made of ABS plastic. If you look at the entrance hole carefully you will see it is getting ground up (UM has some fixes for this planned). This means tiny bits of black ABS is flowing up through your bowden along with the filament and gets embedded into your prints, or worse, gets stuck in the nozzle as ABS flows better at 250C - not so good at 210C. If it is stuck in the nozzle for hours, the cooking converts the ABS into some other chemical that is like solid glue. Conisder keeping your filament on the floor to reduce this abrasion until UM comes out with a new feeder design:

Everything JB says is great except the 105% flow rate. On a UM Original that's good advice but on the UM2 that is more likely to cause "skips". If the pressure gets too high in the nozzle the feeder current is low enough that it is designed to skip backwards a bit. The problem is you then get zero extrusion for maybe 20mm of travel which leaves an ugly spot. Of course 5% of .3mm is only .15mm or about 1/8 the thickness of paper so getting leveling perfect is very important here and flow can be a way of correcting bad leveling (but only on the first layer - after first layer you definitely want flow at 100%).