gr5

Well when you switch the printer to direct IP that disables DHCP. DHCP is how devices ask the local router for an IP address so if it's static there shouldn't be a need for DHCP except maybe to get a DNS address possibly. So I would turn off the static IP address and ask the IT people to check it one more time. I strongly suspect the problem is hardware. I think that random address is what it switches to when it doesn't detect any network at all. So you tried a different cable but I'd try a different port in the building. Just to see if that helps. Maybe swap with a port used by a device that you know works fine.

I would try marketplace first but Artfabs method is a great workaround. I believe the tools say which range of versions they work with and ahoeben might not have set the minimum version all the way back to 4.0.0 because it's a pain to test in every version. Just a guess. So that might be why Artfabs had to do a manual install.

@wittsend - did you try rebooting? If so did you try the factory reset? Erin is very good at this - she has fixed hundreds of S5 printers. Note that after a factory reset I'm pretty sure you need to do XY calibration again. Probably lift switch calibration as well. But if you are only using one core you can skip XY calibration and lift switch calibration gets easier each time. I can do it in less than a minute now.

@Syl2014 - I don't think it was the firmware, lol. I think it was the reboot.

10mm/sec is crazy slow. Try at least 20mm/sec although that won't help. Also if you prefer dimensional accuracy over how nice the part looks then disable acceleration and jerk control in cura. You will get an increase in ringing which causes "ugly" ripples near sharp angle changes such as text or edges but will improve dimensional accuracy for example when using a caliper to measure things. Are you measuring the whole cube or only the face because corners tend to bulge out. So I'm wondering if the Y issue is at the corner or across the face? Please answer this question. So many times people don't answer critical questions and then 5 days and 30 posts later I get the answer and it changes everything. So I think the most likely issue is that your pulleys are wobbly - meaning the hole through one of the 11 pulleys (I think there are 11?) is not centered. This means the belt speeds up and slows down with each rotation. The problem could be with the pulley on the stepper motor or one of the 4 on the long belts. I could be wrong but if you stick with this hypothesis you can test it. If you print a cube that is exactly the same dimensions that is one pulley circumference then the problem should go away. By the way I'm thinking possibly the problem could be with the X axis and if you didn't have this issue then both axes would be 20.5. Or it could be on the Y axis. Another possibility is play/backlash. Push on the nozzle in both directions for the X axis and again in the Y axis. Push with less force needed to actually move the steppers. Typically play will make the part smaller in the axis that has play. If you don't know what this word means, read about it on wikipedia. So you might have X axis play making that axis shorter. And some other effect is making both axes larger in equal amounts.

So the fans are strange: 24V is sent to the two side fans but the fans are in series. Meaning each gets half the voltage (if they are identical loads). So they are supposed to be 12V fans but if one has a lower load then one is getting more voltage than the other. Which could in theory destroy one fan. Over cooking a fan usually results in an open but sometimes a short - and if you get a short then suddenly the other fan has 24V across it. This arrangement is not standard and not to spec for 12V fans. I doubt any fan has a specification saying "you can put two of these in series and double the voltage" yet the UM2 does it. Anyway my point is not just any fan will work in that configuration. These fans tend to have computers in them these days (I know - crazy, right?). The computer may detect overvoltage and shut down in some way or something. More likely you bought 24V fans but you need 12V fans? Or you bought 5V fans and you are lucky they haven't burst into flames yet.

It appears you are in the USA. I would contact support@fbrc8.com. They assembled your printer and if you give them your serial number and explain the problem they can tell you if it is still under warranty. Even if it isn't they will surely give you some hints as to what is wrong as we are not familiar with this particular issue (probably rare?). Usually someone here has a good idea as to the problem but apparently not this time.

Ah! I don't know. There are comments among the gcode at the top of the file that say how much filament will be used in total. But I'm not sure the printer displays remaining filament? does it? I forget. There are also comments on each layer estimating how much time to print the rest of the print. So those numbers will probably help the printer show accurate times when it gets past layer 100. In theory it should use about the same amount of filament either way. You can mess with that comment at the start of the file that says how much filament it will use such that it comes out perfect at the end but I don't know how that works exactly.

I haven't used the "insert filament" option on my UM2 printers in years. I always slide it in manually. That feature is only needed for the original "non plus" feeders. Always cut the filament to a point. Two cuts should do it. Not at 45 degrees but much steeper angle. That will help the filament find the hole in the teflon.

Does it slice if you don't pass it any configuration file? @ahoeben might be able to spot an obvious issue.

So you are talking about the 2 gears in the UM2+ feeder? I believe 0.5mm of play is too much. Not sure. @fbrc8-erin - what do you think? Does he need new feeder gears? (his profile says he has a UM2+ ext). As far as 2800 hours and 2400 meters of filamnt - it's common for those numbers to be similar - that's a lot but we've seen printers hit 10,000 meters. You might want to change the belts now. Definitely change the teflon part if you haven't done that (the teflon part in the head - the white part. Called sometimes an "isolator" I think?) That should be changed every 500-1000 hours.

If you don't understand my post above 100% do not do the splice!!!!!!!!!!!1 Just read it a few times until you understand or ask questions.

Okay so you need to read and learn about the G92 command. So you can reset one or multiple axes. You will have to reset the E axis to a specific value. So... In the gcode for these prints the E value will just keep getting larger. For example E1000 is when you have printed 1 meter (1000mm) of filament. The next E value in the file might be: G1 X100.13 Y75.33 E1000.2 (this says move X, Y and E axis to these new positions) That would mean it will be extruding 0.2mm of linear filament on the next "line" of filament printed. This is called "absolute position" on the E axis. So the 2 gcode files you are merging will be off by E and if you just start the next file on aprticular layer one might end around E500 and the other start at E470. The problem is it will back rotate the extruder by 30mm. Or if it's say 500mm you will run the filament backwards right out of the extruder. Or it might *overextrude* if the values are off in the other direction. So you hadd a single G92 command in between at the splice of the two files. If one is ending at E100 and the other one starts at E200.734 you can just add a G92 E200.734 <-- this tells the printer that the extruder is not where it thinks it is but is actually at location 200.734mm. In between the end of the first gcode file and the start of the next. Even better - use the E value that was the most recent just before the splice on the new file - so if this line of code gets thrown away just before the splice on the new file: G1 X100.33 Y134.3 E75.3 You can delete that line and all above before you add to the end but insert the G92 like this G92 E75.3

When you say "a good cleaning" hopefully you mean "several cold pulls" which you can do from the menu.

So I fixed this using an exacto knife that I had heated to a high temperature with a flame. The problem is the knife is very thin and doesn't store much heat so I had to heat it a few times (I think only twice? maybe 3 times) but I also had to remove the knife before it got stuck. And you have to just guess if it's still warm enough or time to remove. Also you have to be careful not to damage the white plastic head. It all sounds difficult but it took all of 30 seconds.

1) That is only for delta printers. Do not check "origin at center". If you don't know what a delta printer is then look that up using google images. 2) So UM2 has "software limits" for the "high side" of each axis X,Y,Z. So there is only one limit switch for each axis. The position can only be changed by changing the firmware. There are several positions in the firmware: purge position, the 3 leveling points, width of bed (software limit), nominal Z height and so on. So you can either 2A fix those and rebuild the firmware or 2B move the limit switch more to the left. Or possibly 2C, tinkerMarlin might let you set those. I'm 90% sure it doesn't. @tinkergnome? 2A) so I think you need to edit the firmware. I hope you know how to use good search tools. I don't know where those values are stored. Also for sure those values are also stored in the eeprom so even when you fix the defaults you will have to do a FACTORY RESET function on the UM2 control panel before it will switch to the new values. I suspect at least. Anyway first step is to find where those are stored. Probably in an ".h" file like Configuration.h or similar (Configuration_adv.h??). The best source code to start with for sure is tinkerGnome's version of Marlin (not Ultimakers): https://github.com/TinkerGnome/Ultimaker2Marlin how to build UM2 marlin

http://gr5.org/unbricking/

Nice. Hopefully your part won't be weak because you didn't fill in those gaps. I think it will probably be fine. I didn't realize your part had such thin walls. If I had then I might have figured this out. I was talking about there being extra green lines. Not yellow lines. Yellow is basically infill. Red is outermost line, green are inner walls (there can be many but typically 2) and yellow is anything inside that (infill, skin, gap filling). You still might want to play around with the "wall" value and make it exactly half the width of those curved walls. If you know what that value is. Otherwise you may have gaps in the walls and the part may end up weaker than if you allow it to do "fill gaps".

what? So far the answer is no. If you slice for 0.4 core with line width set to 0.4mm and you put in a 0.8 core it will not speed up anything - the AA 0.8 core will be printing 0.4 lines. But then there is more to your sentence - which doesn't seem to change the answer but I'm not sure I understand you. So I think still "no". The AA 0.8 core tends to print 4X faster because of the cura settings - because you are going to do 0.8mm line width (instead of 0.4 or so) and because you are going to print doubly thicker layer heights. Twice as wide, twice as thick, gives you 4X speedup. The AA 0.8 can print thicker layer heights for any given "print speed" partly because it has much larger hole in it so the friction is lower getting that volume of plastic through the nozzle hole. But in your case you are trying to print the lower half 0.4 and upper 0.8 and cura won't let you change the layer height part way through. It lets you change the line width, but not layer height. I could explain why but it won't help you. If you are only printing one part you should "bite the bullet" and just print the whole thing with a 0.4. If you are printing 100 parts you can fudge things. You can slice for each nozzle and splice two jobs together. With tricks to fix the jump in E value at the splice. But it's a little tricky and not worth the trouble if you can print this thing in under 4 days.

It's usually about 5 minutes maximum between when it at least shows a change in progress on the screen. If you power cycle it should continue trying where it left off. That works about 70% of the time. Or power cycling every 10 minutes up to 3 times. If that doesn't fix things and your printer is still under warranty call your reseller asap. Even if it's out of warranty they will give you the official unbricking procedure which involves getting under the printer where there are dangerous voltages so there are special instructions in a pdf they can send you. If THAT doesn't work then it's time to go to this website: gr5.org/unbricking/

Fix this by making wall width an exact multiple of line width (probably called "nozzle width" in other slicers). Do this for both slicers so you get apples to apples.

I think it has to do with wall widths. In cura look at these parameters: wall width wall line count (calculated) line width wall line width outer wall line width inner wall line width So for example if your wall width is 1mm and your line width is 0.39mm cura does this dumb thing where it does 3 passes! WTF? 3 passes to achieve a 1mm wall. If your line width is 0.4mm then cura does it in 2 passes. In the first case with 3 passese it underextrudes by a LOT on the inner most wall. In the latter case it overextrudes quite a bit on the inner wall. I think you have a case where cura is doing more passes per wall to get the same wall width than the other slicer.

I don't understand your point about extrusion volume. The extrusion volume is calculated based on each "line" that is printed by multiplying layer height by line width by line length. You do have an issue as you can change many things for just one area (volume really) of the model but not layer height. The AA 0.8 can indeed print a little bit faster with the wider lines. About twice as "fast" because if your walls are 0.8mm wide then you are printing 2 passes on the shell with the 0.4 and one pass with the 0.8. But if your wall width is 1.2mm then you save nothing. The big savings with a 0.8 - half of that comes from thicker layers which you can't really do.

how is 1.2 meters of filament an additional 36 meters? I don't understand your point.

Wow - that "gcode analyser" is very cool. Never saw that before. Did you put the correct values in for acceleration and jerk and such? I think you can get those values possibly out of the menu system on the creality? Is it accurate for your printer? I find that these time estimates can be off by 2X because you may have the wrong value somewhere. Anyway the slower print has twice as many retractions - that is the key difference I think. Did you play with "combing"? combing will reduce retractions and save time. I guess I would concentrate on the retractions. There are many settings that affect retractions.