gr5

I see the problem. Clearly it stopped extruding. There. Solved. Oh. but why? Not sure. Is this filament brittle? Brittle filament can break in the bowden and then when the crack reaches the end of the bowden it can get stuck. If the rear fan stops spinning you can get clogs which keeps the filament from moving (is it moving? I can't tell in the video it's either not spinning at all or spinning too fast to be able to see - touch it - on the rear of the head while it is warmed up - it should start spinning as soon as the print head is at 40C or hotter). What else - well the filament may have grinded first and then stopped extruding - although it's hard to tell which happened first. You could have dust on your filament that travels through the bowden slowly and gets stuck in the nozzle - this is obvious because when you push filament through it won't come out anymore. Your filament may be broken, or bent, or too large for the bowden (you need to use 2.85 to 2.9mm filament - 3.0mm filament is too large - measure with micrometer). Your feeder may be defective. What happens if you push on the filament hard from the rear when this happens? Like with enough force to lift the printer off the table. About 10 pounds force. Next time please don't make me watch a 5 minute video - tell me where the problem starts in the video or edit it down.

It's because you followed my suggestion and layed it flatter :( You can reduce those "crop circles" by lowering your layer height but to reduce it more tilt the part back up the way it was. You will get these "crop circles" on the top of any sphere you print - the bigger the radius of the sphere the more obvious. So on a friendly octopus not so bad. On this HUGE sphere it's pretty obvious. It's just how layering works. This is why some people want a slicer that makes thinner and thinner slices as you get closer to the top of a sphere. Variable layer height slicing.

go to machine settings in cura. One of the settings places 0,0 in the corner. The other method puts 0,0 in the center. Could that be related?

The UM2 power supply is rated to 221W so definitely make sure it can put out at least 250 Watts (also known as VA so at least 250VA). But that is not the most important specification. Most of these UPS are rated to a certain amount of "watt hours". You need 221Watts X 0.25 hours or 55 Watt hours. In reality the printer doesn't use all 221W. The heated bed alone uses about 100W max and the rest probably uses on average about 75W. So if you are only heating the bed to 60C the bed is probably only on half the time and your printer can probably last almost 25 minutes with a 55watt hour UPS. I would recommend 50W hours or more. This is the most important specification so of course it is the hardest to find in the specifications. It's like not mentioning how much gasoline a lawnmore or car can hold. Usually instead they will tell you "max minutes at full load". So for example this thing: CP1000AVRLCD will put out 600 Watts for 1 minute which when mulitplied together gives you 600 watt minutes or 10 watt hours. Another way to look at it is that since the UM2 takes 100 to 200 watts it would last 3 to 6 minutes (600/100=6 minutes and 600/200 gets you 3 minutes). Everytime you drain the battery completely you probably lower it's capacity. So the first time maybe 6 minutes but the tenth time maybe only 5 minutes.

I'm pretty sure you could have rotated that part so it was more flat and needed less support.

Nice meter. Better than anything I have. Well the average is more important what was that? The power supply will deliver double rated power briefly but not continuously.

I think something is wrong with your model. Open it in cura again and look at it in layer view. Is it skipping a cm here and there and just not printing at all? Either that or your extruder is getting clogged for an hour at a time and nothing comes out and then starts working again.

How much volume is this? I've printed 18 cubic mm/second at 240C without any noticable problem and with 25W heater - although like I said before I have this error disabled in firmware.

Clamp meter? An inductive meter? Inductive meters only work on 60Hz and sine waves. It will probably read a bit low because it's less than 60Hz but it will read extra high because it's a square wave. I think the square wave wins - it can't be 6.5 amps anyway that's crazy. Just measure the resistance when power is off - that won't change much when it's hot. Wattage = V ^2/R or current=V/R

just set fans to 50% to get this part done. If you aren't in a rush I recommend you get a 35W heater. If you don't want to wait for that you have several options: 1) The UM comes with a 25W heater which is plenty but yours might be lower as there is a large variation so you might only have a 20W heater and you are pushing it too far. But I also see you have an olsson block and this error is very common because your Olsson block is probably touching your fan shroud. So you can try to fix that - for example raise the block up higher by tightening that round nut (no disassembly necessary but heat the nozzle to 170C first in case there is plastic in the threads). Or you could put polyimide tape between the block and the fan shroud. Or you could remove the fan shroud. or lower it slightly by loosening the four screws and pushing down and tightening. or you could just run your fans at 50% instead of 100% (that will definitely help avoid this error!). 2) You could get rid of this error in the firmware. It's a safety thing in case your temp sensor falls out or your heater falls out but I have this disabled in my firmware and have printed for years with no problems. 3) Like I said, you can just get a 35W heater. More about the feature and the firmware and where to get older firmware: You get an error if the heater can't move a certain amount in a certain time while driving full power (when it is close to goal temp it typically runs well below full power): Firmware Version 14.09 - does not have the feature 14.12 oct 16, 2014 - feature introduced. 20C in 20 Seconds 14.12.1 dec 15, 2014 - from 20C to 10C (still in 20 seconds) 15.01 jan 14, 2015 - from 20 secs to 30 secs (now 10C in 30 seconds) So for example you could download cura version 14.09 and update your firmware using that firmware version. Cura is unique in that you can have dozens of versions all installed at the same time - each version is treated by windows as a completely independent , unrelated program.

The circuitry for the heated bed and for each of the two nozzle heater circuits is identical. I looked at the actual traces inside and on the board and the mosfet switches are also identical. The heated bed is I believe about 100W or about 4 amps. So I don't think you should go over 4 amps. But you are talking about <4 amps so the electronics will be fine - it's just the power supply that you need to increase. Also if you don't use the heated bed and the nozzle at the same time it should also be fine. So for example if you are only heating the bed to 60C then I think the supply might be okay as is as long as you don't heat the nozzle until the bed is completely warmed up. Also note that there is a firmware setting where you can limit power to the nozzle. For example you can say 50% power (128) is max power to ever apply to the nozzle. This will be pulsed but the power supply can go FAR over it's normal limit during the pulses and will be fine as long as total power draw doesn't exceed it's spec for a longer duration e.g. 1 second. @coen? Is that correct? heated bed about 4 amps?

I think he is printing an architectural thing like a roman temple where he wants the columns 100% infill but the rest of the print only 20% infill. Is that what you want? If so you can do that in cura 2.x.x. The feature is called "merge" or something like that where you split the model up into multple parts (columns versus the rest) and then give the different parts different settings and "merge" them into one print. Keep in mind that 100% infill is really not much stronger than 24% infill as, like bone, the strength tends to come from the outer shell. If I mis understood the question please include a drawing - maybe draw on paper and photograph with your phone.

I recommend you just try it - it says without pva to use the model with "pins". I trust Labern's suggestions. But maybe go even slower - 25mm/sec.

This is a different issue than the Ultimaker filament issue but is just as interesting. I'll mention in the future not to let the silica pack touch the filament.

Actually your two shells seem to be touching nicely - it's the infill that isn't quite reaching the walls. I guess I would say you have a small amount of underextrusion - maybe 20% which is actually pretty minor. You could slow down your print by about 20% and maybe increase flow 10% and you probably should do these things. However your problem looks more like backlash. Backlash (aka play - if you aren't sure read about it's definition on wikipeia) can cause this. Backlash is rare on UM2go but it can happen. It can be caused by 1) a loose belt - particularly feel the tension on the belt that goes to the stepper. Push the head around when the power is off and compare the 3 (three! not two) X belts versus the Y belts. One of those may be loose. 2) Belts too tight or other friction - if they are extremely tight they can cause extra friction - again push head around with power off and see if one axis has very heavy friction. 3) Head loose or falling apart - try pushing 4 directions (e.g. east, west, north, south) on the nozzle. Is it loose? Does it move without the head moving? Could be a loose bearing. All of these are easily fixed.

Flow rate drop? No. Speed? Maybe. 20mm/sec * .2 * .05 (layer height) is 0.2 cubic mm/sec. At 210C (PLA, not PET) the .4 can do about 2.5 cubic mm/sec with .4mm nozzle. But this nozzle hole is 1/4 the area so I would expect it can print about 0.6 cubic mm/sec. So your speed seems okay. But I've never printed PET and maybe it's much more viscous at 210C than PLA is but I suspect they are similar.

Did you read above where it was figured out that it only gets brittle where it is straightened? Try taking a meter of filament and straighten half of it on a table with tape and wait 8 hours. Does the curled part stay flexible and the straight part become brittle?

Most likely you leveled with head too far from glass. Squishing the filament more will get you better adhesion. Also make sure the glass is clean. I clean the glass about once per month - just remove it and put in sink with soap and water, dry and re-install. Adding a little glue stick to the glass and spreading with wet tissue and letting that dry will help parts stick even more. But most important is to squish filament into the glass. Try turning the 3 leveling screws and equal amount - try 1/2 rotation CCW to get glass closer to nozzle. Blue filament is closest to ideal although the yellow example will stick like hell.

Dual posted. Closing this thread. Please answer this question here: https://ultimaker.com/en/community/21981-strong-light-weight-filament-and-design-help-for-designing-rc-airplane?page=1&sort=#reply-152792

I have a few thoughts. First of 100 pounds is - wow - a very heavy plane. Where are you going to store and build such a thing? I guess a garage because you will probably need a big door. Anyway, stay away from CF (carbon fill) filaments as they are no stronger and no more stiff (and stiff is bad for filaments usually). Nylon is amazingly strong due to it's flexibility but not my favorite material for quadcopters because it's a bit too flexible. Still it's amazing - you can drive a car over pretty much anything made with nylon and it will flatten but bounce back to it's original shape. Really I would stick with PLA for the strength as it's hard to beat. But I would mix it with CF rods that you can buy from places that sell RC kits. Those rods are really amazing and you can integrate them into your designs. For example when making the fuselage have many rods running the length of your fuselage such that they somehow attach or latch onto the PLA fuselage. You can use the CF rods for cross bracing as well. Really it's hard to beat balsa wood (coated with soemthing to add strength) when it comes to light weight and strong. As far as "injection mold" smooth - you could use PLA/PHA and completely submerge in acetone. There's a lot of discussion including helpful videos in this topic: https://ultimaker.com/en/community/10412-acetone-finishing-on-pla Or you can go the ABS route which is easier to process with acetone but ABS you will probably find has horrible layer adhesion unless you enclose your printer and turn the fan almost completely off. It takes a while to learn a new material. ABS is excellent for aircraft - it's a tiny bit more flexible than PLA (but same strength) and that tiny increase in flexibility means it is much much less brittle (think crashes).

These parts are wobbling while printing. There is no simple solution but there are several solutions. 1) You could put better support on these such that they don't wobble so much. The built in support of cura 14.X and 15.X could be better. Try cura 2.X (the latest) support features - they really improved the options on those and you can have the supports follow further up the ribs. 2) You could lay the ribs flat. They will then print much better. But you will get a larger area where support touches the rib that will look - less pretty - but at least the prints won't fail. This is probably what I would do. 3) Design your own supports. If you don't know how to use CAD software this might be an opportunity to learn one of them.

Okay - now I'm thinking you have one of those spools of PLA from a bad batch from the summer of 2015. Even if you just got your spool this week it may have been in a warehouse for a while. The problem with that filament is that when you straighten it out (like leading up to the feeder) for several hours (like over night) it gets very brittle and cracks. The cracks slowly move up the bowden when you next print and when they get to the end of the bowden often get hung up. The fix is to either send the PLA back to the manufacturer or NEVER leave the filament in the printer when it is not printing (just pull it out and let it auto coil around the spool until the next print). More details here: https://ultimaker.com/en/community/17102-why-did-the-filament-break#reply-120629

Cura can take image files such as a png file (don't recommend jpg as the compression adds lots of noize) and turn that black and white image into a height map such that black lines are tall. This is one way to make cookie cutters. You can switch on brim as well. But it won't let you add supports like you did for that tomato. But I supposed you could do that by adding a gray line to your drawing - then the gray line will be printed at a shorter height (shorter wall).

toms_tk has been silent, lol and yet people still post here. Toms - it's okay if you don't own an UM. You can still post stuff to the forums.

Yes - that's another good thing to try - but hit the switch twice quickly (click click) because the homing procedure heads to the switch fast the first time and then moves away about 1mm and then towards it a second time and expects the switch to to "on-off-on" in quick succession.