gr5

Raising temp to 240C just for the first layer helps quite a bit - try that also. You can then lower it again for the second layer. A setting? Well you can tell the printer to do .35mm shell. This will put the traces .35mm apart instead of .4mm apart. It will also extrude less plastic - it will basically act like you have a .35mm nozzle. Changing the nozzle size doesn't always affect the amount of material to extrude or the distance between fill lines - changing shell does. Not sure exactly why. Anyway this will certainly put the lines closer together but it will also give you very thin walls. Or you can do .35mm nozzle and .7mm walls (or any multiple of .35) and you will get thicker walls and still get lines close together in the infill. I don't recommend this. The problem is that leveling is incredibly sensitive to the gaps between lines. For example if your bottom layer height is .1mm (thickness of paper) and you do accuracy .03mm (that's hard to do!!!) then you will get a 30% error which is quite a big gap (or over extrusion) between lines on bottom layer. So the first thing to do is make the bottom layer .3mm thick (3X less sensitive to leveling). The second thing to do is as it is printing the bottom layer - if the lines are too far apart, turn the screws slightly until it's looking good. Then cancel the print and start over. You can alternatively adjust the "flow" parameter but this has a huge delay - it's about 10 to 20 lines in the firmware buffer so if you make a change you have to wait for 20 lines to go down or so before you see the result of your change.

Still this is extremely common cause. Mark the shaft and the pulley with a tiny dot with a permanent marker. If a limit switch signal is triggered you get the same sort of pattern where you get a sudden shift in one layer. How is the friction of your two axes? Push them around by hand with power off. Maybe you need a little light oil? Maybe belts are too tight?

Yes. Or raise the temperature. higher temperatures let the PLA flow more like water instead of toothpaste. What temp are you using - 240C is usually my upper limit with PLA. Higher than that gets a little dangerous that you will bake the PLA into a carbonized clog if you don't keep things flowing.

First of all - the larger diameter is a good thing! Not a bad. I wish I could get other filaments with larger diameter - I have a bit of trouble printing the last bit of filament because the radius of curvature is too tight and I get underextrusion. I only print the last 1/4 of the spool on my UM Original. So spend $10 and get one of these turntables aka lazy susan - it's what I use for unspooled filament and it works great. You put it on the floor and it actually rotates as the filament is sucked up into the feeder.

Yes. 50% feedrate in TUNE menu with 30mm/sec print speed results in 15mm/sec print speed. 40% would be 12mm/sec print speed. This is much slower than you need. 20mm/sec should be plenty slow. The idea is the temperature has more to do with the color so you can set the temp on the printer. Then if you print another robot in blue you can raise the temperature without changing anything in Cura. However you can look at "tweakAtZ" plugin which lets you set the temp exactly and can vary for different parts of the print. The stringing was mostly on the antennas - it's very small and easily removed with fingers. That robot looks great. You can improve the tops of the antennas by printing 2 robots but I think it's fine. A few more things: Travel speed should be as fast as possible - this will help *break* the strings. The default is 150mm/sec. You can probably go to 250mm/sec but 150 is a good start. Infill speed and inner shell speed are typically *faster* than the outer shell speed. This is because you don't care about the quality inside so much. But I recommend you leave these at the same speed as the shell speed. Don't slow down - don't speed up. If you are in a rush this is a good thing to increase the speed on (infill speed) or just try 0 infill as the robot prints fine with zero infill. Bottom layer speed should be nice and slow so you have that set good right now. 35mm/sec is plenty slow enough though for most thing. In fact I rarely go slower than 35mm/sec for anything. You can print this much faster and still get good quality.

The additives to PLA that give it color affect the melting point. Gold PLA is fine as far as I know. I've printed with UM Gold and looked over my notes and I had printed 2 things that I still have at 220C and they came out beautiful - they were not as tricky as the robot (not much overhangs). But be warned every time you switch colors there is a chance that you will have to adjust temperature. It's not about quality - it's about melting point and viscosity and glass temperature. Getting back to your robot - the main advice is what robert said - lower speed to 20 if you want the quality I got. And stay with 190C which is a good temperature for that slow slow speed. But if you print at 96mm/sec - you might need to increase temp a little to get that PLA to flow through that tiny nozzle hole. The pressures can be quite high in the nozzle and these reduce quality. nalalth's advice about shell is also critial - 1mm is bad choice for shell - I hadn't noticed that. Use .8mm (two passes) or 1.2mm (3 passes). But never 1mm except for extremely specialized things which you will hopefully never have to print. Actually I wish Cura would turn that yellow when you choose something not a multiple of the nozzle diameter. I rarely print at 20mm/sec. In fact I think that was the most recent time (about 9 months ago). But this was kind of a contest to get the best possible quality. I usually print at 35mm/sec for show pieces and 75mm/sec for things that no one will see. I've learned to be patient and not rush a print and letting the printer run for 36 hours no longer scares me (although most of my prints are under 4 hours).

Did you create a ticket at support.ultimaker.com? Refer to the ticket number in all communications. Daid (posted above) is an employee of UM but he's a software guy. UM's support is very friendly and helpful and gives out all kinds of free stuff but they are understaffed and slow right now. Hopefully that will be fixed. Your fastest solution is to try the three things I mentioned above - they might find the problem pretty quickly. If you don't know how to use a multimeter, maybe you have a friend/neighbor who does? UM started out selling a hobby kit - so only people who weren't afraid to take the thing apart would buy them. Now people are buying these printers (which are very easy to take apart!) who are afraid to do that. UM is slowly adjusting to this new type of client. So UM is partly relying on this forum and people like me to do the customer support.

Sudden shifts in only one axis are almost always fixed by tightening the set screws on the pulleys. There are 6 pulleys on the X axis (not 4) and the 2 on the short belt are the most likely that are slipping (the one on the stepper and the one on the other end of the short belt). The UM Original usually comes with a second set of spare set screws that are shiny (the originals are black). The shiny ones are better although most people are fine with the black ones but you have to tighten the hell out of them. If you don't believe me mark your pulleys and shafts with a permanent marker so you can see if the pulley slips. Less likely causes are high friction in the X axis or rubbing belts against the case but those usually cause gradual shifts - not sudden shifts like this.

Yeah - that would be the firmware. To do it in Cura wouldn't make sense as you need to know characteristics of PLA at various temps and length of bowden and all kinds of stuff. There's been some more talk recently of making the firmware compensate for the delay/pressure of plastic in the nozzle. I don't think anyone is working on it. There is an existing feature that does this in marlin but it's, well, not quite correct implementation. But it does help really crappy printers quite a bit. Not sure if it helps the UM2. Never turned on that feature myself.

By the way, printing at 150mm/sec .25mm layers with .4mm nozzle is (multiple the 3 numbers together) 15mm^3/sec which is way above what my UM2 can do and very few people have a UM2 that can achieve that throughput of plastic even at 250C. It is however possible with the right filament and a brand new nozzle, and brand new teflon isolator. You can see the results of printing at different speeds here and you can test your machine (after you unclog the nozzle) with the original test (that only goes up to 10^3mm - download it in the very first post): http://umforum.ultimaker.com/index.php?/topic/4586-can-your-um2-printer-achieve-10mm3s-test-it-here/ Or the tougher test that goes higher here (again - first posting): http://umforum.ultimaker.com/index.php?/topic/5436-a-tougher-extrusion-test-o/

260C will just make things worse as PLA at 250C and higher can slowly be baked into a difficult to remove gunk. I think it is caramelization. 180C is a safe temperature to leave the head for many minutes and warm enough that PLA is soft enough to remove and work with. It's common for the head to hit the part due to shrinkage on the top warping the edges inward and lifting the corners especially. I'm just really surprised that the z screw would move part way and then not the rest. And only on some models. I'm not surprised at all that the head hits the part and not surprised at all that it stops printing. So I would be damn sure Z really stops and I would experiment with it using pronterface and set the Z to various values and watch the bed go up and down and see if it has trouble at some point. pronterface is very easy to use: http://koti.kapsi.fi/~kliment/printrun/ Back to clogged head. There's lots of solutions. The easiest is the "atomic" method described in extremely long detail here (but is very easy to do): http://umforum.ultimaker.com/index.php?/topic/4118-blocked-nozzle/?p=33691 Altermatively combine that method with something that can fit inside the nozzle tip such as either hypodermic syringe or acupuncture needle. Or the most reliable method and the most work - take the head completely apart and put the nozzle into some gas flame and burn all the PLA and teflon (from black feeder) out of the head (but not so hot as to melt the brass). Then soak in acetone for an hour, then put it back together and perform the atomic method until it comes out clean. I've only had to do this once.

It could be pinched servo cables. It could be a bad PCB. It could be a bad power supply. If it were me I would: 1) Inspect all the servo cables carefully. There are 2 metal covers in the back two corners that hide the servos. Each one has a single screw to take it off (this entire printer is easy to disassemble). Take those off and take them off slowly and look at the cables at the bottom to see if maybe they are damaged (insulation bent or wire coming through insulation). Do the same thing with the larger cover on the bottom of the machine particularly pay attention to where cables go under the cover. The cover on the bottom comes off easily after removing 2 screws. 2) If that doesn't show anything obvious and the cables to all 4 servos look good (no possible short circuits) then I would leave the cover off the bottom PCB to increase cooling. 3) I would also hook up a volt meter to the 5V power supply (not 24V) somewhere on the PCB and do a print and monitor the voltage carefully. See if the voltage slowly drops to 3V. You can pick it up at TP16 and many other places. TP16 is on the "left" side of the PCB (with text around edges in normal orientation). Or you can look at the UM2 PCB schematics and board layout in this pdf here: ULTIMAKER 2 SCHEMATIC - click "raw": https://github.com/Ultimaker/Ultimaker2/blob/master/1091_Main_board_v2.1.1_%28x1%29/Main%20Board%20V2.1.1.pdf

Anyway you will never be happy if you don't experiment and find the limitations of your exact printer and then compare it to other's limitations to see if there is something wrong with your particular printer: http://umforum.ultimaker.com/index.php?/topic/4586-can-your-um2-printer-achieve-10mm3s-test-it-here/

Is that through a .4mm nozzle? I searched for 10 minutes with google and looked on cubify website and couldn't find the nozzle diameter. But you have more interest than I do - maybe you can figure it out? I assume the cube has a larger nozzle which offers less resistance and allows you to print much faster but the detail won't be as good (XY resolution). Everyone posts the Z resolution but it's harder to find the nozzle diameter.

Diamond age. Very nice filament but no longer sold in USA. But there are better filaments in UK (I'm jealous). But this is mostly irrelevant as I can get this kind of result with every standard PLA I've tried including many from Ultimaker. That particular Robot was before I had a UM2 (last fall) so it was done on UM Original but I've printed robots on UM2 just as good. Exact settings for the robot in the photo above are in post #19 here - also there are more photos zoomed in closer above that posting and other discussion about getting quality robots (other objects have other challenges so this is mostly UM robot related): http://umforum.ultimaker.com/index.php?/topic/3038-can-your-ultimakerultimaker2-print-such-quality/?p=28588

Does cura show the layers in layer view? The background slicing engine may have crashed - for example if you have too many millions of polygons.

Connection issues debugging steps to isolate your problem - but before reading this - is it possible you download UM2 firmware into a UM Original? Or UM Original firmware for the UM2? If so don't panic, you can fix it. usb debugging (post #7): http://umforum.ultimaker.com/index.php?/topic/5100-how-to-update-um2-firmware/?p=46156 also win 8.1 drivers: http://forum.arduino.cc/index.php?topic=205837.0

Do you have retraction turned on by the way? Without retraction one should expect stringing like the 3rd part in top picture.

2nd and 3rd from the right in your first picture look good - use those settings. You have some stringing on some of those parts in the top picture. That can be improved by lowering the temp, lowering the speed, and increasing retraction distance. E.g. 70mm/sec, 220C, 5.5mm are probably good to start with. In picture #2, the 4 parts in the background, the bottom layer looks like crap. It was printed much too high off the build plate. Re-level with paper. You don't need to touch the 3 screws, just do digital leveling with the dial and paper at the rear screw only. Paper should be grabbed a little bit by the print head. If you turn off power after leveling you will lose it due to a bug recently found. After power cycling it levels too high. You can fix *that* by turning all 3 screws such that the bed moves up by .2mm. And don't relevel. Until Daid comes out with a fix. But your levelling was off by much more than .2mm I think. Make sure the bottom layer is .3mm thick so that the leveling doesn't have to be "perfect" and you can have an error of .05mm and still get beautiful bottom layer. The two parts closest to the camera in the 3rd, final picture look pretty good but if you want better you appear to be getting some tangling on your filament spool. Move the spool to the floor to get further improvements and if you want it even better slow down the printing to 35mm/sec.

If you are patient you can make the Z scar almost non-existent by slowing it down to 20mm/sec because you don't get any oozing when you print that slow. Also it helps to crank up the Z acceleration to lower the time it sits there moving the Z. Daid wouldn't it be even better to move the head to an "infill" region of the part during the Z move? Most of my parts have no visible Z scar by the way (at 35mm/sec).

It is possible to get much better overhangs - this is my best UM robot:

Ignore that - you don't have a second extruder. Wow! Much too hot for PLA. Dangerously hot - a little dangerous - leave PLA at that temp without printing for a half hour and you will have to take your nozzle out and burn all the PLA out. Anyway - do bed at 50-60C for PLA (100C for ABS). 75C for PLA is a last resort after you can't get any of the other methods of getting the part to stick to the bed and you still get major warping (corners lifting) on large parts. 75C will give bad quality on the bottom 5mm or so where the air is too hot. It will pull inward kind of look like warping but it's just that the material is too soft on the corners when the next layer goes down. 75C won't affect more than 20mm off the bed. I vaguely remember that gold needs lower temp - maybe even 190C. But start at 210C or 220C. If you get any underextrusion (holes in the wall, or skipping feeder) then slow it down to 35mm/sec but 50mm/sec should be fine regarding overhangs. Also make sure your side fans are working. They need to be at 100% by the time you get to the first overhang. By default they come on at 5mm up. Make sure minimum and maximum fan speeds are both at 100. Many printers fan connections disconnected during shipping so this is a likely problem. Test that the 3rd fan at the rear also comes on (the 3rd fan comes on as soon as power is on - even before the lights). To test that side fans work do: PRINT, select any print, then immediatly go to TUNE and set fans to 100%. Turn off machine when done testing so that it doesn't start printing (will not start printing while you stay in TUNE menu).

You can get it even tighter if you loosen the 4 thumb screws 3 or 4 turns, do what you describe above, then tighten them back up. If the bowden keeps slipping even after that, you should take it out, cut off the last 2mm and put it all back together again.

Z dimensions are extremely accurate and should be within about .1mm. If you print in a heated chamber (recommended for ABS) then there will be shrinkage from the heated chamber temp to room temp. XY dimensions will shrink for PLA about .3%. Not sure about ABS but I think it's about double. I believe both materials have a similar density/temperature graph and it's pretty linear in solid, glass, and liquid states - there arent major sudden expansions or contractions at any of those transitions. So mostly you only have to worry about glass temp to room temp shrinkage which is much more extreme for ABS because glass temp is much higher. Anyway - this doesn't answer your question. The mechanics of the printer itself should give you around .1mm accuracy and better than that repeatability. The speed you print, the acceleration settings, cause over/under extrusion (repeatably) on corners such that you might get a bump out on a corner if you print "fast" because the bowden has some sprint to it and the filament keeps coming instead of slowing down on the corners. The most extreme innacuracies will be for "bores" or vertical holes. There are 3 factors that shrink these things. I always add a minimum of .5mm to all diameters. Sometimes even 1mm. The 3 factors are shrinkage, number of lines inscribed in the circle (fewer lines = smaller hole e.g. 4 lines gives you a square hole) and the rubber band effect when printing the hole that pulls the lower layer inwards as the current layer is laid down or the upper layer may even be printed more inwards - think tight rubber band being glued down into a circle with the glue not quite dry. This effect seems to be worse for PLA because it has a lower glass temp. In other words, PLA does most of it's shrinking while still liquid/glass. ABS does much more shrinking after it is a solid. Repeatability - if you don't change any filament, temperatures, fan speeds, printing speeds - should be at least .1mm, probably half that. So you can put chamfers and other compensating factors onto your parts and get a non-conforming part to print so much better and still print very fast.

It's fine to have cubic voids, but not double walls. Try robert's suggestion first, then look at the part in "xray" view. Any red areas are a problem. They are generated by passing a 3D line through the model and counting how many polygons it passes through. Even polygons are shades of blue and white. Odd polygons are shades of red. Red indicates an internal wall that doesn't have meaning or a hole in an intended wall. If you have no red zones then unchecking all the "fix horrible" boxes should result in a perfect part. The exception is if you have to vertexes very close together (less than maybe .2mm). When Cura slices a layer it interesects a plane with your polygons (which are unordered in an STL file). Then it has a bunch of lines. It tries to figure out which lines intersect and tries to create loops of lines and assigns those together to be a "surface" but sometimes it gets confused if 3 or more vertexes are very close together. Also STL files have an indication of the polygons which side is "inside" and which side is "outside". You can tell Cura to ignore that with one of the "fix horrible" settings.