eldrick

E-steps calibrate the number of stepper-motor steps per unit (mm) of filament movement. If you have the E-steps or Flow set OK, then retraction shouldn't change. UM defaults retraction to 4.50mm, but many of us use a bit higher value, like maybe 5.50-5.75mm.

I also have this problem - noise just like that during the X-axis moves. I'm also seeing two other symptoms: - The thick fore-and-aft rod along the right side shows signs of wear - when I wipe it before lubricating, I get black stuff in the oil, which doesn't show up on the other rods. It is plainly wearing more rapidly than the others. - Considerable belt-wear dust on top of the X- and Y-motors. This has been visible since the machine was new. Any ideas on what might be misaligned and how to correct it?

That Mk 8 pulley image is impressive - I just ordered one too. Looks like a lot less of a filament-grinder than the Ultimaker version, and should reduce friction in the Bowden tube. What E-step value do you use with it?

Recycling plastic works if and only if you have a source of lots of known-identical material, with no contaminants, labels, etc. And recycling 3D printed plastic is an exercise in futility, unless you like mud-colored stuff.

So far, spare parts support to the US from UM has been limited, to say the least. I have had great service from illuminarti, and I'm hoping that he soon takes over US support for Ultimaker, as UM is doing a really lousy job of aftermarket support with parts: none offered for purchase online, overpriced when you can get their attention, slow response, and now outlandish shipping costs. If UM does not step up to improve their parts support, I'll be inclined to sell the machine, and certainly unable to recommend their printers.

I have been trying to get a spare brass printhead nozzle for my UM2, so I can drill it to .5mm for faster printing. I contacted Sales@, and was quoted a very high price for the tiny piece: Euro 69.50 = US$95, which is more than enough to buy Two entire J-head extruder heads with thermistor and heater - significantly more complex parts. However, not having another source, I elected to purchase it anyway, until I saw the incredibly high cost that they quote for shipping: Euro 25.96 = US$36. This is simply nuts - they could use International Post and mail this fingernail-sized piece of metal to the U.S. for less than 4 Euro. I view this as stunningly bad customer support, as I can buy similar small bits from England with postal shipping costing less than 5 Pounds. I conclude that Ultimaker simply has little interest in customer support outside of Europe, as their policies make it prohibitive to keep their machines running here in the U.S.

Just measure the thread-to-thread distance - that's the pitch.

Try printing it with fresh glue stick and the bed heater turned off. PLA will usually stick just fine on a cold glass bed, and there is no point in heating it at the bottom while you are trying to cool it just above the bed.

While searching for a camera for the same purpose, I came to the conclusion that ip-attached cameras all seem to be low-res and pricy, compared to USB-attached. Probably a bandwidth constraint. Logitech has several USB-attached cameras that can focus down to 7-10cm, and have wide-field or 1080P.

248C is 25-40 degrees too hot for the Ultimaker PLA sample, which is why you got melted-looking layers in the print shown. The sample robot file I got with mine sets the temp to 203C, which worked for me. Try it at 205-215C and set the fan speed to 100% for much better results.

I've given up on reproducing the nozzle, and now I'm trying to order a new one from Ultimaker, which is also proving to be difficult. I've received one response with a steep price (€ 69,59 excl. VAT and shipment) from sales@ultimaker, but have not been able to get Zenalda Furtado to tell me how to actually order or pay for it. Ultimaker still seems unwilling or unable to provide post-sales hardware bits. I'm hopeful that illuminarti can provide a nozzle here in the US.

Youmagine has the worst tagging and search facilities I've seen in years. It is stunningly difficult to find items that are related to, for example, the Ultimaker 2. Why on earth don't you implement the (free to use) Google search extension for the site, so people could find what they are looking for?

One thing that will prevent update is that if Pronterface/Printrun is active and connected to the UM2, you will be unable to update the firmware via Cura until you Disconnect from Printrun. ...says the voice of experience...

venkel, the designer, has uploaded several fan duct designs to youmagine. Here you go: https://www.youmagine.com/designs/fanduct-another-idea These are improvements to the point where a couple of us find that we turn the fans down from the previous setting. An ABS shroud also gets rid of a surprising amount of noisy resonance and fan noise vs. the OEM metal one. Difficult to find on youmagine, as youmagine tagging and search are remarkably primitive and ineffective. Why they don't use Google search on the site is beyond me...

Just checking, but you did change the nozzle size setting in Cura/Advanced or other slicer, right?

With a coil lying on the floor, you also pick up an amazing amount of nozzle-clogging debris, particularly in a room with a 3D printer... Really bad idea, IMHO.

Ultimaker uses it between nozzle, the metal collar, and the aluminum heatsink (the copperish stuff you can see smeared on the heatsink around the collar), and recommends that you use anti-seize with copper in it, to avoid multi-metal corrosion. This was from another post on the Forum somewhere - someone asked what it was, and whether it should be removed. Pretty much any hardware store anti-seize will have copper, and often aluminum, in a lubricating base. Usually available in a small tube of thin paste. To apply, use a toothpick, and wipe a very small amount around the female inside threads and in the holes, so that putting the parts together pushes the stuff in, instead of just wiping it off the outside of the male part or thread. You don't need much. Don't put any inside the nozzle barrel. A Ferrari-certified mechanic once told me that you could tell how good any auto mechanic is, by whether he always puts anti-seize compound on all the metal parts that fit together.

That's a great tip - I've got some Taulman nylon, and I'll give that a try next time. Thanks!

I've been printing a two-month-long project with white PLA. I've used the atomic method dozens of times, and recently got into the habit of doing it after every multi-hour print. I often got some brown stuff with a few flecks of black, and keep repeating cold pull until it comes out clean. I'm damned if I know where the black flecks come from, as I use only high-quality filament, keep it clean in sealed poly tubs, run it through a piece of poly foam as a wiper in front of the extruder, use Roberts extruder made of white PLA, and I can see nothing in the Bowden tube. Before this extreme cleaning, I had done the cold pull several times until clean, and the nozzle was still lined with black stuff when I removed it. The atomic method is better than nothing, but not by a lot, IMHO. Very frustrating.

Actually I just posted a comment about the .skp file, which is not remotely of engineering-drawing quality. You could try sending them the .stl for printing, but good luck finish-machining a nozzle from that. One tip on reassembly: use a little anti-seize compound in the holes for the heater core and the thermistor (as well as the aluminum heatsink). It will make them a lot easier to remove and insert without damage if/when you have to take them out again, and might even improve thermal conductivity a tad. I discovered when removing the thermistor that it slips in 1/8" further than it had been, because it was tight in the hole without the lubrication of the anti-seize.

Posting a Sketchup drawing of the machine is a long way from Open Source, as far as I am concerned. From the files supplied, there is no way one could really reproduce the machine. As example, I'd like to have some replacement nozzles with different diameter orifices machined. There is no way on earth that a machinist could work from the supplied .skp file to get the nozzle specs to work from, because if you extract the nozzle from the .skp file, you just get a collection of boxes with holes of partially-specified dimensions. Oh well, I personally don't see a lot of value in Open Source hardware anyway...

My guess is that the stuff was carbonized PLA. If you let PLA sit in the nozzle without extruding for even a few minutes at temps over 180C, it tends to turn into something like burnt sugar. It wasn't really that much work - maybe spent 20 minutes over and above removing/reinstalling the printhead and nozzle. While in the shower this morning, it occurred to me that if the drawings are available, I'd consider having a local high-grade machinist do a run of UM2 nozzles in several orifice sizes: maybe .35, .4, .5, and .65mm?

My UM2 has roughly 1000 hours of use now, printing both ABS and PLA. I've had some recent under-extrusion issues, even after cleaning the nozzle with acetone and applying all of the known "fixes". (Roberts extruder, increasing extruder motor current to 1500, better-fitting clips on the Bowden ends, etc.) So I took the dam thing apart again, and this time I resolved to do the job right. First, I soaked the brass nozzle in methyl chloride to remove any trace of PLA, then in acetone (Note: both are fire hazards and have the potential to poison living things). Then I swabbed out the inside with Q-Tips repeatedly with acetone, cleared the orifice with a needle, and inspected the nozzle again. I wasn't satisfied with the condition of the opening (slightly out of round), the flat on the outside of the tip (not much flat there), or the inside of the barrel (coated with black crap). So I raised the ante: - I got out the Brasso (metal polish with rouge in it) and used it with a half-dozen Q-Tips to polish the inside of the barrel until I could see bare metal again. I don't know what the black stuff was, but I believe that being smoother and fresh metal should considerably reduce friction and back-pressure in the nozzle. - Then once it was all reassembled and the bed re-leveled, I put a piece of 1500-grit sandpaper (very fine, used for polishing paint on car repairs) on the print bed, with the nozzle just touching it with light pressure. I manually moved the paper a few inches under the tip in circles. This left a very thin streak of metal on the sandpaper as evidence that it was lightly sanding the tip flatter. The result has been a noticeable improvement in extrusion and print quality. When extruding a few mm prior to a print, the plastic no longer curls up against the nozzle, and outside periphery layers are visually better-aligned and smoother. My UM2 is now pretty-much back to its original level of operation and print quality. Of course a new nozzle would have been a lot less work, but UM2 parts still don't seem to be available here in the U.S. (If a replacement nozzle had been available, I'd have bought one and drilled it out to .5mm, as .4mm is just too fussy and clog-prone for me.)

Or use Robert's extruder, and don't bother waiting for the "Change Material" routine. Because you can disengage the drive wheel, you can remove and feed filament thru the Bowden tube manually.

It would be strongly preferable if Cura had a Randomize Start Points feature for the outer layer, to avoid leaving visible scars in the print surface where layer starts are vertically aligned, as often happens now. This is particularly obnoxious when printing any cylindrical objects with Cura. The current (and apparently future) implementation gives up significant print quality in favor of trivial print-time reduction. Take a look at the Kisslicer/Slic3r/Simplify3D implementations - each allows you to randomize starting points, and/or request that start points be restricted to corners or concave areas, and Kisslicer even allows you to specify a range of angles on the print, where starting points can be "hidden" in unobtrusive parts of the piece. Until at least randomize is available, I cannot use Cura to manufacture either of the sport-fencing products that I manufacture and sell, which is disappointing.