gr5

Not humidity. Could be heat - what is your ambient air temp? If over 22C then that could be related. Did you check #2 and #3 above? Is it brittle? Did you count the retractions? The gcode file is easy to read - it has comments for the start of each layer and G10 and G11 are the retraction and unretractions. What is the exact diameter of your filament (accurate to .05mm) do you have calipers? If it's greater than 3.000mm then that's a problem.

You are the first person to have this issue that I know of. And I think I've read every post on this forum for the last few years. Bad wire to the Z stepper motor? Oh - the Z stepper driver can possibly overheat. Are you in a hot climate? is it much hotter than 21C where you are? I wouldn't think Belgium is that hot. Usually when the Z stepper overheats it only shuts off for 1 second and then starts back up no problem. There is a "fix" where you lower the current to the Z stepper. This fix is in the latest firmware. I don't think this is the problem though. I would hook a USB cable between a computer and your printer and install pronterface (it's free): http://koti.kapsi.fi/~kliment/printrun/ Connect to the printer with that and command the Z to home and then go to position zero: G0 Z20 G0 Z10 G0 Z0 G0 Z1 G0 Z2 G0 Z2.2 The above commands tell the Z to move to various positions. Start with 30 to be safe (not sure if the head hits the glass around Z=20 or around Z=0. Probably Z=0. See if you can get the problem to duplicate. See if it's just sticky and starts moving again - or if the Z is completely electrically silent - when servo is off you should be able to push gently down on the bed and it should slide easily. I'm guessing the Z stepper is fine (but not certain!) and there's something wrong with the Z screw - it may need cleaning. Or maybe the two vertical rods and bearings are bad and jam up but you said it felt fine so that seems unlikely even though this is a very common problem. Z screw issues are also a common problem but it doesn't stop for 3 layers usually - usually the Z stops moving only 1 layer or moves half a layer height instead of the whole thing. You really need to isolate this problem into mechanical or electrical.

Yes. That filament was popular back then and shipped with every printer because it matched the Ultimaker color scheme (note similar color on this web page here).

It's a custom tool. There's only one or two on the planet. It's steel I think and machined to incredible tolerances.

I'm 90% sure if you slow to 30mm/sec it will make it look much better - certainly if you were printing PLA that would fix what I'm seeing here. However if that's not enough, XT is not as easy to print as PLA so I don't know exactly what the problem is - I would start by printing cooler - you can mess with this in the TUNE menu while it's printing to see if temp makes a difference - I would try printing half speed in the problem areas and keep lowering the temp by 5C at a time and mark the test part with a sharpie at the layer height where you change temp and/or speed and keep notes in a log (you think you can memorize it until you are at the 10th temperature change). I strongly recommend you print things like this in PLA. It will be stronger and better looking than in XT. XT is a great material but it took me about 100 prints to get really good at PLA. It took me another 30 to get marginally decent with higher glass temp materials like XT, ABS, Nylon. These higher glass temp materials are MUCH trickier but it can be done.

With a better feeder (meduza 2X or bondtech) and the RACE nozzles I can print quite a bit faster with a 0.8mm nozzle.

I think this is the same object up close:

They were all solid cubes and I was in the TUNE menu non stop increasing and decreasing speed to see how fast I could go such that the infill was still touching. Even if I found the cube it would be garbage because when I was at 20mm^3/s it was underextruding which was about half the time (half the time underextrucing) because I was constantly changing the speed. For "real" prints I usually try to keep things at half max speed or slower. Oh wait - I have video... but the object is too blurry I think to see much - this was with my old phone a few years ago - I guess it had low quality video (now everything is HD):

Cura 14 is still available. Lots of people created configuration files for Cura 14. There aren't so many yet for Cura 2 as the official released version as of today (august 2016) is too slow for most people (there's a new unreleased version that is amazingly fast - when that comes out people will suddenly start using Cura 2). As people (you?) create machine configurations for Cura 2 they will get slowly included in the release. The Ultimaker employees working on Cura are currently trying to implement lots of new features and fix bugs and it's not in their purview to create profiles for competing machines but they are quite happy to include them in future releases if someone creates one. I know a few profiles for other machines have been created already for Cura 2.X.

I put jerk in quotes because it's not the right word but it's the word that Marlin uses in the code and on the settings in the menu system. So when Marlin or Ultimaker printers talk about "jerk" they are talking about an infinite acceleration - an instant change in velocity at a vertex (there's a lot of vertexes in a print - one for each gcode approximately). The same word: "jerk" is used unfortunately in the firmware for: Marlin, Repetier, Sprinter, Redeem, and more. The jerk setting on UM by default is 20mm/sec. So if you are going into a 90 degree corner at 100mm/sec it will slow down to actually 14.14mm/sec at the corner (it's complicated).

90% of "3mm" filament is actually 2.85 or 2.9 +/- .05. But there are a few (very cheap chinese) suppliers that actually ship true 3mm filament which you can not use on an Ultimaker. Please post this kind of question in it's own topic as it is unrelated to 1.75mm.

There are lots of "tiny gaps" issues that are completely unrelated such as Z axis slipping slightly or underextrusion and slicing issues so please post a photo.

good question. I'm not sure of the answer - this is a new slicer (cura 2.x) and I have never seen the orange before - not sure what that is. You can increase infill overlap from default of 15% to 50% or 100% to see if it fixes this but I've never had to do that. Maybe you have some play in your print head - e.g. belts too loose or too tight but unlikely with UM2plus. If you have gaps I would try printing at half speed and I think ti will fix your problems (just set to 50% in TUNE menu while printing).

Is this UMO? Best fix for droop is to lower print temp by 10C and get a second head fan on the right side.

neotko is an extreme expert at UMO - more so than me. So read his advice carefully. Some other ideas though more specific to your question - if you do retractions that are too long the heat moves upwards with each retraction. And future retractions then pull the heat higher up each time. If it gets above the white teflon part it's a problem. Also make sure your bowden is all the way in so there isn't much room for filament to expand above the teflon. What is your retraction distance? I use 4.5mm on UMO but my bowden is nice and tight (no play). If you do get a new teflon, get it from 3dsolex as theirs is a newer material that is more durable. Consider maybe putting a small fan blowing on the upper block area to keep the teflon cool. UM did exactly that in the UM2 (the rear fan).

A few things can cause the filament to stop coming out of the nozzle and grind at the feeder. The problem is distinguishing if it happens first at the feeder or first at the nozzle. 1) Excessive extruder heat. When weather/air temperature is warmer (above 20C) the combination of air temp and heat from the extruder motor moving to the filament can heat PLA enough so it gets soft in the feeder and it gets squished flat and eventually fails. This can be obvious if you know what to look for - right after a failure there is 5cm of flattened (half round) pla in the bowden just above the feeder. 2) Filament brittle. A year ago there was a bad batch of Ultimaker PLA. Some people still have it - the symptom is if you bend the raw filament it breaks much too easily and breaks in the bowden. It gets stuck temporary at the end of the bowden and so the feeder is grinding, grinding until it's all dust at the feeder. This is easy to deal with so if this is the problem I can point you to more info. 3) too many retractions. Does your print have lots of retractions in the layer that it failed? Open the gcode file and delete everything except one layer (somewhere near the problem layer) and count the G11 codes - easiest way is to "replace all" G11 with something else like GXXX just to get the count/qty. If you are doing 10 retractions for every spot on the filament this can grind it up too much and occasionally cause problems. "grep" is also a great too to count G11 in your gcode file. 4) Printing too cold/too fast. I doubt it's this as you are only printing .06 thick layers.

There was. All new blocks are fixed now so I'm probably wasting my time. Some of the blocks had a very very tiny ridge that kept the nozzle from screwing the last 0.05mm or so. Carl at 3dsolex has a tool to fix them - it takes about a second per block to fix the ridge.

I don't understand why the homing procedure would move it right past this spot but when printing it gets stuck. Is this a UM2? If it's a UMO or UM2 you should be able to push the bed up and down by lifting and pushing at the back of the bed near the z-screw. It takes quite a bit of force - about the amount needed to lift the machine. Once it starts moving it doesn't need as much force. Pushing down is much easier. See if you feel something wrong. Maybe the screw isn't fully attached to the Z stepper? I've heard of prints not moving for maybe 1 or 2 layers in Z but they usually start moving again after 2 layers of moving less than nominal.

I think the lines are rough because the part it's about to print doesn't have smooth edges.

I have replaced about 10 blocks and the damn temp sensor is VERY HARD to get out. But I always seem to manage. I use a sewing needle to pry it up but first I spray WD40 into the racks around it (makes a mess - put newspaper or rag on the glass to protect the glass from oil) then I heat it to 180C and hold block with pliers (so I don't burn fingers). Once it reaches 180C I cut power (if heater or temp sensor slips out while heating the heater will quickly head for 1000C and glow like a light bulb!). Then I hold block with pliers so I don't burn fingers and pry out the temp sensor with the needle until just enough sticks out that I can grab the metal part of the sensor with needle-nose pliers and twist like a screw back and forth while sliding it out.

So homing fails? Or homing works but plate stops moving at a certain height and you get a massive ball of filament on the "final" layer? or it only gets stuck a few layers and then starts moving again? Photo would help.

I'm experimenting. 6Nm is quite a bit of force! I'm now making weaker versions. My current tool lets you pop out one torque and pop in another version. Anyway Carl is fixing all the blocks such that 0.5Nm should be plenty. Hopefully (fingers crossed).

Here it is. 0.4Nm for pla/pha. Well mine is 0.8Nm even after using it hundreds of times. I printed it using Ultimaker orange pla. I have a newer version of the tool that is much flatter. Still tweaking it. It can go to higher torques. I have a 2Nm version and a 6Nm version for the block V3. The 6Nm version breaks with the 1/4 inch drive so it has a longer hex hole instead which delivers 6Nm with no problems. I need to refine this.

You can get rid of those surface lines by unchecking "combing". This will turn retraction on for those moves but will slow down the print slightly as you will be retracting on internal moves.

Can you just add a few drops of cyanoacrylate glue for now (aka superglue)? Even though there is a slight gap on the top layer it might still be extremely strong. That sucks. what's different about your test piece? Location on bed? Support? Are you sure they were both at 190C? Usually when I do a test piece I run it at several settings changing things like temperature on each layer and taking pictures and changing speed and so on. I'm wondering if it does this on every layer. I guess now you need to make the test piece fail. I would try high temps and fast speeds. Maybe. I don't know. The nice thing is you can change most of the crucial parameters from the TUNE menu while it's printing so each layer can be different settings (temperature, speed, flow) to see their affect quickly (within a minute and sometimes within seconds). I always have paper and pen ready to take notes live so I can do maybe 10 tests in 10 minutes.