gr5

Different question = different topic. Maybe you should start a new topic asking about glow in the dark filaments.

Not going. I'm in Boston. I usually go to the Makerfaire in NYC in September though.

Maybe like me, the beta testers had a different version of firmware on their UM2s (I have tinkerMarlin).

Cura 15.X used to have this wonderful feature where the last line in the file contained all the settings (meant to be read by a computer, not a human). This also allowed the great feature "import profile from gcode" which allowed one to alter all the cura settings to the settings used to make a particular gcode file. Cura 2.X and 3.X don't do this. They don't write all the settings to the gcode file and they don't read settings from a gcode file. I suspect this is what you were looking for. Instead there is a new feature called "save project". This is actually in some ways better than the old feature. It saves all the settings and also it saves the position and scaling of every part and it saves the contents of the STL files such that when you load the project all the parts are there on the bed in the same position/orientation and with the same settings as when you saved the project. Therefore EVERY time I create a gcode file I also save the project. That way I can go back later and see the settings. I'm not sure if this carries across versions of Cura (I think it mostly works). In other words if you save in cura 3.0 can you read the project in cura 3.1? I think maybe it mostly works. Not sure.

I (thegr5store) only ship to USA so use 3dsolex.com or another reseller in europe (3dsolex posts a list of resellers if you click on the earth/globe symbol on the 3dsolex page). Anyway you need not just a hardcore but you should get a sapphire nozzle to go with it. The ruby/sapphire nozzles are called "everlast". Don't buy a ruby nozzle meant for the UM2 - make sure you get one for the hardcore.

I have bought 2.2kg spools from printedSolid. Works great. I rarely put filament on the spool holder anyway. Usually I just put it on the table behind the printer or on the floor behind the printer. Spool shape is not important to me.

Don't search for "2.85 filament". Search for "3mm filament". That's what most people call it. printedSolid.com sells a great variety of filaments. meltink3d in florida has a nice line of filaments protopasta has a lot of amazing and interesting filaments amazon has some really cheap ones (e.g. esun) and lots more. I didn't mention matterHackers as kman mentioned them.

Sorry I didn't see this until now. The first thing to make sure - push harder on the clear part until you hear/feel it snap in. You don't have to do this with the UM cores but the hardcores need a little extra push. Are you printing with the silicone rubber installed in the print head? I was printing with a 0.8mm nozzle and 0.3mm layers and the silicone pad was missing and that got enough wind inside to make it so the core couldn't keep the nozzle hot (now that I think of it, that was a UM core, not a hardcore - still - the silicone rubber helps). It could just be a bad/intermittent temp sensor wire (or heater wire). If it's the temp sensor wire you will also see the temp bouncing around. If it's the heater wire it will cool a bit unexplicably and get heater errors. Heater error and it's *usually* the heater. Temp sensor error and it's always the temp sensor connection/wiring. If problems persist you should ask the seller for a new hardcore.

That metal plate you spoke of. That is meant to hold the cores consistently every time - within maybe 10 microns I'm guessing. Because that way once you calibrate the X/Y position on a pair of cores you don't have to ever calibrate them again because that metal plate is supposed to hold them in the same position every time. So it's kind of critical I think.

Damn! I suspect this will never get resolved. So as far as I can tell from a dozen or so reports, tinker 16.01 is fine but newer versions have some subtle bug that corrupts memory occasionally. It's very sublte and the right things all have to happen at the same time. Once memory is corrupted anything can happen. Power cycling fixes it. I was hoping you had a procedure that could somewhat reliably create the problem within 5 minutes. But instead one has to print for probably 10 hours and even then it's not reliable. So I have only two things for you to try: When you switched to tinker did you do a "factory reset"? If not I recommend it. There are a few dozen parameters that are stored in eeprom such as steps/mm and width of platform. These values are stored in a different place in each version of Marlin so not doing a factory reset can cause some suble issues. Please turn off geek mode while printing. See if that seems to fix it. It seems to have solved the issue for one person (maybe I think). Now realize you don't need tinker firmware to use the DDG. There are other ways to set steps/mm. For example you can set steps/mm with a single gcode and then save that with another. So you can install normal Marlin if you want. But it has to be a version of Marlin that rotates the extruder in the right direction at least which I think the UM2+ firmware does.

That works also but if you don't have good bonding the part will be weaker - but I suppose in your case that doesn't matter. You might want to overextrude by 10% and print slow and or hot - if you want it to be air tight.

Hairspray is excellent. The next of a few tricks you need to do is to squish th bottom layer more. Maybe. although that print I see there looks like the skirt is transparent so you probably squished it quite well. The next trick would be brim or rounded corners which you got on most corners but the top branch of this part not so much.

@StephanK - many people don't use change filament. I doubt Tinker uses it. So there could be a bug that slipped by. Or maybe this is a known bug that was fixed but this is very interesting to me because it could be related to a rarer bug. What version of tinker software did you use exactly (which printer type/variation). What does it show for the version on your display. Can you get it to repeat? If you start from powering on the first time, what would be the procedure for me to get this issue you see on the display and the weird homing behavior? I have um2go, um2, um2ext so I can try to repeat your test with the exact same printer type and the same firmware.

And ultigcode (the default) gives you full control of temperatures on the printer itself.

This usually happens when your part is wider than tall and then when it comes loose from the bed. At that point the nozzle is dragging around your print like a hockey puck and filament is still coming out and it has to go somewhere so it goes "up". The solution is to keep your part sticking to the glass. I could tell you how to get your part to stick to the glass but it's better to also know why the techniques work because then you can modify the techniques. So to become an instant expert you should really watch this entire video (sorry it's so long but it contains the results of many experiments and I tried to keep it short):

I understand what you want but being a triangle every layer will have a different thickness so this might be difficult. Certainly this is easy to do on the bottom layer. For one thing I would set infill to 0% and set wall thickness to 100mm such that it prints it solid with only concentric infill. This will help but there may be gaps. I would then play with line width a bit to try to get rid of most of the gaps - at least on the widest layer I would get it perfect.

Yes. Cura does multiple steps. One of the early steps is to "slice" which means cura takes each of the layers and represents a plane and intersects that plane with all the triangles in the STL file (an STL file is an unordered collection of triangles - there is no information in there that says which triangle goes with which). This is done *after* the rotation that you specified so the triangles are already rotated into their new postitions. At this point (for each slice) cura has a huge list of straight lines in 2 dimensional space (because we intersected with a plane. Those randomly ordered lines are then linked up into loops. The edges of those loops are considered the WALLS even if they are "holes" inside the part. At the same time cura tries to figure out which side of these loops are inside versus outside. The TOP/BOTTOM areas are figured out differently - obviously everything printed in the very first slice is "bottom" but there can be "bottom" portions half way up the print as well. TOP/BOTTOM is figured out in 3 dimensions. Walls are figured out in a single 2 dimensional slice.

ALWAYS look at the print in layer view before printing the part. I'm guessing the wall got too thin for a bit. You can either make "line width" smaller (and make wall thickness a multiple of wall line width). Or alternatively try checking "print thin walls". And ALWAYS look at the print in layer view before saving/printing. You would have noticed the problem without starting the print.

This is known bug in 3.2.1. 3 bugs combined to create this issue. There are many work arounds - one of which is to add 256 or more characters in comments at the bottom of the file. one is to downgrade. more details here just a few posts above your post. Cura 3.1 should be fine.

A photo would help. I'm wondering if you are asking for something completely different than kman and I are thinking you are asking about. Well let me ask you this - when you say "removing top/bottom" do you mean in CAD? Because if so I think the answer is no. But if you want to remove them in cura (but they exist in cad) then I think the answer is yes. If you just put up an infinitely thin wall in cad cura will slice this wall PERFECTLY. To within a billionth of an inch accuracy. It will print nothing which is what an infinitely thin wall is: nothing.

I"m a bit confused because the "Z hop when retracted" should have taken care of the problem. I thought. I guess it fixes most of them but occasionally it doesn't retract (if it retracts too much it can grind the filament). Another solution might be to increase retractions (always retract) and I know you got ground up filament and a failed print but you could lower the print speed by X2 to reduce the pressure and reduce how hard the feeder has to work. Then loosen the feeder all the way to the top position (weakest squeezing of filament).

Okay. Yeah. Check those wires.

I've never tried the auto level so I don't know if it does a good job. I've had my Um3 for over a year and never had any interest in even trying it. It probably sometimes does a perfect job and sometimes not perfect? I don't know. But I know if you mess up manual calibration it will indeed cause the behavior you describe. So I would try that.

I can delete this if you really want but I think this might be helpful to someone else? But if you are embarrassed I'm happy to delete the post. Just let me know.

You never answered my question. I don't want you replacing something that isn't broken.