donmilne

If you hear the tocking noise as the feeder motor jumps back, then the motor is encountering too much torque, fighting too much resistance. You need to work out why. 30mm/s at 0.2 sounds ok, conservative in fact. You don't mention filament. What filament are you printing with, where did you get it, and at what temperature are you printing? How much printing had you done and at what temperatures, prior to the last two weeks?

I think you'll have to take the teflon spacer out and push the plug out with a screwdriver. The gold filament uses some kind of metal filler perhaps? Possibly the filament itself is a good thermal conductor, which might explain why it melted so far up.

Welcome aboard. It seems to be a standard initiation rite that everyone has to take the feeder apart on their first day, and then discover that two hands isn't enough to put it back together. :smile: In my case I held the feeder halves together with tape, freeing up one hand to hold the stepper, while the other held the feeder, a screw and a screwdriver... I suggest that you make |Robert| 's feeder high on your list of priorities. No tricks needed to attach that, and it's a much better design (easier to tweak for tension, easier to keep clean). https://www.youmagine.com/designs/alternative-um2-feeder-version-two

I don't understand either of the last two suggestions. If the fans are causing a temp drop (which they surely must) then shouldn't the heater controller function compensate? At 210 it surely isn't already 100% on?

Hmm. Read on Macduff!

I've also found a UK supplier for both rod and sheet :- http://www.par-group.co.uk/engineering-plastics/Vespel.aspx They also offer a non-Dupont equivalent for the sheet. No mention of pricing or ability to order online, though the sheet page does say "cut pieces available".

I was on a tablet yesterday which is why I didn't post a direct link (copy and paste is hard). Here it is now :- http://umforum.ultimaker.com/index.php?/topic/7338-ultimaker-2-collet-locks/

Use the Atomic method to clean out the nozzle (see online visual guide). Also, if you look back in my post history you should find where I posted an OpenSCAD script for an exact copy of the UM2 bowden clip. You do need it - you don't want the bowden tube moving. Edit: ah, problem solved while I was composing. The blockage was a bit more basic than I had in mind, but well done for finding the problem.

This could be nothing to worry about. Sometimes the filament gets snagged on the entrance to the Teflon spacer and won't go all the way in. Then of course the motor can push as hard as it likes but the plastic isn't melted, and it ain't goin nowhere. I try to put a point on my filament rather than square cut it. If that's not the whole story then tweaking the temperature up by 10C might help too.

I assume we're talking about underextrusion - though I can't see it clearly in that picture. The two possibilities that spring to mind are (a) deformation of the teflon spacer, (b ) getting towards the latter end of a reel (tighter coil == more drag in bowden tube).

I assume you've read the humungous thread that discusses improvements to the UM2 feeder? ... which generally involve junking it and replacing it with Robert's feeder design.

Oops, sorry, I thought "commenting system" meant the forum.

Ooh, bitching time! :-) To the above I'll add that I find the quote system here completely unusable when I'm on a tablet, it seems to add hidden markup that messes with the formatting. It's also obviously very annoying that images have to be attached in five steps, and zip files can't be uploaded at all. I don't know why this site doesn't use something proven, i.e. phpBB.

Those layer thicknesses are huge - underextrusion would be expected. Do you actually mean 0.1mm and 0.15mm?

Don't get me wrong: we all make a mess of our first prints, but I'd have thought they would get it right before going on air - and I found it amusing to think of the conversations that must have taken place right after the camera was switched off! :smile: (I'm assuming they tried to print way too fast in order to get something for the camera).

Perhaps you would be better off adding support manually.

I should say that I havent tried 240C with PLA myself, I was coming up with theories because Simmon asked for some. I'm more curious on why Simmon is running it that hot with PLA. Given what we now know about the Teflon part, ISTM this is risking early deformation of the part, so I'm interested in why Simmon is doing it.

Are you still running with the (excessively IMHO) high temperature of 230? That could make the PLA too runny and give you this effect. You can't extrude a liquid!

Yes, I know about the variable results with sdcards from my own SPI bit banging code, which is why all the cards I use myself are by SanDisk (though even they moved their fab lines to China recently, which has made their timing much less predictable IMO). For clients we only recommend SanDisk and Kingston. As I said above, I'm quite sure the (SanDisk) 8GB card I was using is a good'un, so my guess is that the firmware just doesn't like 8GB cards. Have you not seen the same thing?

Ordered from where? Do you have a link for the part you ordered?

I just caught the end of a BBC tech report, some guy talking about 3D printing. I stopped and watched because he was demonstrating an Ultimaker 2, seemed like nice exposure for my chosen printer. "Just create your design, copy it to sdcard and print it out. Simple as that", says our enthusiastic young presenter. Skit ends with camera briefly glimpsing at 1cm hockey puck like shape currently being printed... underextruding like crazy and in the process of lifting off the bed. Simple as that eh? :-)

Isn't the temperature probe a standard component? E.g. I see hundreds of PT100 temperature probes on RS, though it would be nice to have a less generic part number.

You maybe need to let the filament cool a little more before pulling it out. Perhaps let it cool another 10C. At the moment it's stretching life toffee instead of pulling out cleanly. Do you feel any obstruction when using the atomic method?

Should shmood! Better IMHO to establish a baseline: find out what your printer can really do right now. Then consider if that's good enough and what might be done if not.

I'm not sure how that improves on the existing feeder? Higher torque + single drive wheel == grinding.