gr5

Can you post just one picture that shows this? And zoom way in on the picture and recrop please so that it's obvious. What *I* see on these pictures is that the shape of the tip of the nozzle is kind of pushing the plastic around like a plow. Not a lack of material.

For UMO, yes. But UM2 needs custom nozzles. Make sure to get the 3mm filament version of the filastruder nozzles and not the 1.75mm version.

By the way I just checked and hairspray, wood glue, and glue stick all contain PVA (and lots of other ingredients).

@fheineman please update your user settings to show what kind of printer(s) you have. For ABS on glass, 110C bed temp is recommended. It takes 10 minutes so you might want to get that started before you load your part onto the SD card. Make sure you use the "brim" option. At least 6 passes. 10 might be better. Make sure first layer is squished down. Please provide a picture so we can see your brim. The glue stick is great for PLA and okay for ABS but I find hairspray is better for ABS. I could be wrong maybe they are about the same. If you *do* use glue stick you want it very very thin - it should dry to invisible. So use a wet napkin to spread it thin and let dry for a minute (will dry fast at 110C! better to spread water when still only below 70C). For hairspray use aquanet unscented. Spray on tissue and then spread onto glass thin layer. It's good for laundry stains also - spray on stain on clothing before washing.

How hot does that part of the wire get? I would tape the heater and temp sensor together with kapton tape (can handle something like 600C) and then turn it up to 250C and see if that brown spot is a problem. If so then yes, you might as well cut it all open and solder a new wire in. But use lead solder (hard to find now) which can handle up to 300C. Typical electric circuit solder melts around 200C I believe. Or maybe 1 inch up the cable is below 200C?

First of all front/back lean would be the Y axis. So your X axis is fine you only have to worry about Y. Two things can cause that lean. One is short belt rubbing on wood frame. The other is loose pulley screws. To test for rubbing, move the head front and back and look in the rear left corner at the Y stepper and watch the short belt. Does it twist different directions when you move the head? If so it is rubbing the wood. Add some washers for spacers under the 4 standoffs for the motor mount. Or move the pulley closer to the stepper (loosen set screw, slide so almost touching - maybe .5mm - then tighten the hell out of set screw). The UMO also should have come with a spare set of "silver" set screws. The ones that comes with the pulleys are black. The silver ones are better (pointier). I'm still using my black ones. I just tightened the hell out of them. So tight that the hex wrench is twisting. Make sure you get all 6 (six) pulleys on the Y axis. Most especially the 2 on the short belt (motor and above).

The corner cutoff isn't to prevent breakage - it's so that it doesn't squirt plastic out - then squish it all over the nozzle tip making a mess. When I built my first heated bed I drilled a big hole in the bed to let the plastic through that spot at 0,0. The priming/leakage ends up going through the hole to the area underneath the bed. No harm to the forthcoming print.

Sorry I look at things with a microscope so to speak and only look at a tiny problem at a time. Nicolinux printed your .5mm wide trace part and saw the same issue - "big" air gaps between traces. CLEARLY SEEN. I mean the gaps are maybe 30-50% as wide as the traces. It looks more like a screen than a flat top. However in all your orange/red cubes above they don't seem to have these air gaps. I don't see the same problem. Instead I see melted plastic pushed around by the nozzle. It all looks normal and typical quality to me. I didn't see *any* bad cubes in the red and orange recent posts. If you consider these bad then I think it's a different problem. Even in your cutaways the infill looks about the same to me. I don't know why you think one is better than the other. So I'm confused about that. The only clear problem I see is in a single print that nicolinux was able to repeat and with the .5mm shell width. Or maybe I'm just too impatient to look carefully? Instead of posting about 20 cube pictures how about just 2. A "good one" and a "bad one" with complete settings (ini file) for each and what % speed you printed at. I feel like you did that and I claim it is probably the .5mm issue. Maybe you should post a single picture of a cube with (.4 or .8) walls and .4 nozzle and .2mm height at 230C and 40mm/sec that has the same issue and as bad as Nicolinux photo. Also you could link to the gcode.

By the way - most of that mess is when I tried to remove the relay to replace it with a new one I bought. But in the end I gave up and just shorted it.

Sounds right. Here is my repair. Doing what I did is safe as if the relay *does* close it won't hurt anything. You could test this by holding a wire across these 2 points while someone else tries to move a servo with the menu or something.

It's true you can print on cold glass. It helps to heat it to about 90F with hairdryer first or print extra hot first layer (250C?). I recommend thin layer of any PVA glue as well. You can get ordinary (but thick) glass at any glass shop. They will cut to any size and grind the corners. I recommend you cut off the corner where homing occurs so that the nozzle does not touch the glass during homing just to make it less messy. It should cost you about $10 to $20 for a custom cut glass with ground edges so it isn't sharp. Attach with binder clips to existing bed. I recommend that the glass is smaller than existing screw heads so there is no bump. Don't forget to do rough leveling with power off before printing or you might break the glass on the very first homing sequence.

I still think the problem is related to trying to print .5mm wide lines with a .4mm wide nozzle. It should be possible but maybe some nozzles have more trouble than others? By the way if you ask for .7mm wide shell it will make two .35mm passes. If you ask for .9mm wide shell it will make two .45mm passes. For a .4mm nozzle I believe it will do traces .3 to .6mm wide on a single pass (75% to 150%). So .7mm is easier to extrude than .5mm. You should really stick to multiples of nozzle width if you can (0.4 0.8 1.2 1.6).

Okay - I think I found the problem. I looked at the gcode. I didn't like what I saw. cura is doing everything right. You have shell thickness set to .5mm. This should almost always be a multiple of your nozzle width (.4mm) which is a major part of the problem. Changing this to .8mm (or .4mm ) should fix your cube. What is happening is it is slicing each layer for a .5mm trace which means you are sort of overextruding by 25%. That's not easy. The amount of extra energy needed to pump out that extra 25% of plastic is certainly more than 25% extra - I don't know if it's 2X or 1.5X but it's more than 1.25X. Because it has to squeeze that filament into a tiny area between the nozzle shoulder and the part. Now Cura doesn't just make the *shell* width .5mm but also all the infill. Both the sparse infill and the top/bottom infill. Everything is .5mm traces! This is just how Cura works. I don't know why. If you change your nozzle width to .5mm it will change NOTHING in this slicing result. But don't do that because if you later change your shell to .8mm and nozzle is .5mm things will be even stranger. Just - always make your shell width a multiple of the nozzle width. Until you are doing strange things like printing tiny text. But for a cube? Just follow that rule of thumb. I checked the top layer traces and they are extruding exactly the right amount for .5mm wide and .2mm tall traces. I checked the spacing between traces and they are .5mm apart. So Cura did things as expected. Now how likely should this underextrude on a good UM? Well you are printing .5 X .2 X 50mm/sec which is 5 cubic mm per second. That's pushing it - the absolute limit my machine can do is around 6 cubic mm at 220C. Thats where you expect some underextrusion. I recommend printing half that speed at 220C (3 cubic mm) so 30mm/sec. But because you are asking for an extra 25% overextrusion to fill those .5mm wide traces you might want to go slightly slower - maybe 20mm/sec. Now at 230C you should be able to go much faster. About 7 mm^3/sec so if you decide to print at half that rate that's around 35mm/sec printing speed. Again because you are trying to push .5mm wide traces through a .4mm nozzle you might want to go even a little slower. Instead - I strongly recommend you change your wall width to .8mm (or .4mm) and then you can print at "normal" speeds. I am getting all my "max speeds" from this chart here which is designed for .4mm shell and .2mm layer thickness (dark blue line): http://umforum.ultimaker.com/index.php?/topic/4127-um2-extrusion-rates-revisited/ So for example at 230C you can see the max speed is around 90mm/sec which is 90 X .2 X .4 or 7.2 mm^3/sec. Again I recommend printing at half that (about 3.6 mm^3/sec at 230C) for a good margin of error. You can see why I like to print hot when I print fast. I usually don't go over 240C though. If you leave PLA at 240C for too long it can turn to nozzle-clogging-gunk so you don't want to set it to 240C and walk away for 30 minutes. Anyway this is all fine as long as you print nice and slow.

Is it leaning like the 4th picture down on the left side? http://support.3dverkstan.se/article/23-a-visual-ultimaker-troubleshooting-guide

"leaning downwards" usually means leaning left, right, forwards or backwards. These are caused by either a loose X belt or a loose Y pulley - usually the on on the motor which is the hardest one to get to. But really you should post a picture of a "leaning downwards" print. To post a print do "gallery" in the top left of this page, then click "upload" button. Later make a new post and click "my media" next to smile icon.

If you end up needing a new PCB you have to contact support here: support.ultimaker.com Personally I would just jumper out the relay. I'm an electronics engineer and so it's easy for me.

I don't think you get the red lights until you try to turn the heat on. So it does seem like the same problem but it might not be. The best test is to measure each of the pins of K1. When it is supposed to be on (always) the coil pins should have at least 2 volts across it. Probably 3 or 5 volts. Then it should conduct the 24V from somewhere to somewhere else. You really need to know how to use a multimeter and read schematics to be sure this is the exact issue. If it is then it's okay to "jump out" the relay as it isn't used yet. The idea was to hook up safety equipment to it (fire alarm? kill switch?) but that never happened. Yet. If you know how to read schematics it is 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

Do you have a picture? Can you post it? That would be very helpful for future people with the same issue.

I will look at the .2mm layer gcodes tonight.

I have found measuring nozzle diameter is very difficult. I did it by inserting a needle (with cone shaped tip as all needles have) until it stops moving inwards, then marking the needle with a marker or tape. Then removing the needle and measuring it's diameter where the needle stopped. But it shouldn't matter much. You can print .4mm wide lines with a .35 nozzle just fine. Cura extrudes the correct amount for .4mm and puts the stripes .4mm apart. The actual nozzle diameter matters but it can be off by 25% no problem.

Parts don't stick so well to fresh tape so you can go that route (don't clean the tape with isopropyl alcohol) but then your parts might not stick well enough. So I recommend you just tear the blue tape. Whatever it takes. Use a putty knife. Then to remove the tape off the bottom, soak it in a plate of isopropyl alcohol for about 5 minutes. The tape then come right off with no effort. The thing that sucks about this is then you need to repair the tape before every print. I don't know what to say about that.

I'm not familiar with the first run wizard since if I ever ran it then it was probably a year ago. There is a relay, K1 that can be "sticky". If the relay doesn't close then everything has power except for the following: 4 stepper motors, nozzle heat, bed heat. I had a problem with this relay. If I turned the machine off for an hour and turned it back on the relay would work. If the machine was warm and I turned it off and on again it would *not* work. My fix was to give it a tap. The larger cover underneath the UM2 is easily removed by removing only 2 screws. I recommend you do this just to do the same test I did. The fix is to get a new white PCB but it's important to make sure you definitely have this problem. Here is a video demonstrating the temporary fix: Remember - this is only one possibility. There are other possible problems.

I just tested and it took a 7 pound weight to lower the table (at first nearly double that but then starting 20mm down it only needed 7 pounds to start it back up. On lifting 7 pounds lifting (estimate with 7 pounds in one hand, lifting bed with the other) was enough to lift it also. 7 pounds is 3kg.

Another interesting not - more firmware changes - possibly related to this issue. This is from Simon aka illuminarti: So I'm thinking the PID update rate for the nozzle was quadrupled so that the "bang-bang" changes to the heat bed no longer mess up the temperature at the nozzle which I assume caused the problems in the pictures above. Note that even if you got 14.09.1 you might not have gotten the changes because Cura only does a "factory reset" if the version changes a significant amount. So 14.09 to 14.09.1 does *not* load the new settings without a manual "factory reset" but 14.09 to 14.11 *does*.

I'll send you an email from my personal email. If you reply then I'll consider your email valid and move you to the members group.