nick-foley

Let us know how it goes!

v2 seems to be working well. Support on thin contact points is more manageable for sure. It may be trying too hard to find small overhangs and support them, though - on my latest print, it is building small support structures on the sides of what is essentially a perfect cyclinder. STL here: https://www.dropbox.com/s/fk1iw3f7p3retwl/Spring%20Crown%20Mount%20v0.5.STL Not much of a problem, but worth noting.

Yikes. That is a process. We have yet to get a clog with this hotend that wasn't solved by refreshing the filament, elevating the temp, and cranking the gear a few rotations by hand. Maybe we're just getting lucky though. Anyway, as an additional data point for this hotend, we seem to be getting better performance regarding retraction and clogging when printing in ABS instead of PLA. Several prints in with Ultimachine Natural ABS @ 280°, 3mm of 30mm/s retraction, getting pretty clean results and no clogging problems.

Double check the integrity of your thermocouple wires, particularly where they connect to the printhead PCB and where that PCB connects to the cable assembly. Also check that the transition from the metal housing of the thermocouple to the wire mesh sleeve of the thermocouple looks good. I had a similar problem a few weeks ago and one of those connections was starting to come loose.

AAaahhhhhgghh balls... Got one of the Pololu drivers installed properly, but fried another in the process. I followed the Pololu instructions, putting the stepper into full step mode (all jumpers removed) and tuning the pot until the test pad voltage was 0.65Vish, but it still wasn't working properly; the movement was very jagged even back in 1/16th step mode. I then started making (...admittedly arbitrary) adjustments to the pot and something fried. I'm not exactly sure what happened - I may have set the current limit way too high, or I may have just accidentally shorted something while ham-fistedly metering the test pad. Anyway, if other people ever need to do this, here's the pragmatic, low risk set of instructions for installing your pololu driver that I used successfully on the second driver I purchased: 1) Turn the pot fully counterclockwise 2) Install it in the machine... in the correct orientation, you idiot! 3) Turn on the machine and connect it to Cura 4) Jog the machine through the print interface. Listen to the sound the motor makes. It will sound wimpy and quiet. 5) Make a small (1-2°) clockwise turn of the pot and jog that axis some more. Listen to the motor. It will sound less wimpy. 6) Repeat step 5 a few times. 7) Eventually, it will sound less wimpy, and more angry. Dial it back until both the axis you are replacing and your stock driver axis sound equal amounts angry/wimpy. 8) Print yourself some victory. Now I just need to replace the second stock driver that I broke when I was trying to debug the first broken driver... Still just one stepper driver away from a fixed machine...

Perfect, thank you.

The Pololu drivers did not come with heatsinks. Supposedly, they (the black versions) are rated up to 1.2A without a heatsink, which is also supposedly around where they should be tuned for an Ultimaker... so I might be ok. The potentiometer seems to be of a different resistance value though, I think 20K instead of 10K. My plan was to just turn the pots until they matched resistance values, but since neither the pots nor the wiring is the same, I need to meter the current. Can someone explain where I need to meter current when tuning the potentiometer?

I would attempt to loosen the back panel of the z-stage and reorient the brass nut so that it has more wiggle.

I have a "fit" 3D model I made when designing my printhead that I can share. It seems to be fairly accurate, but it is external details only. What are you using it for?

Is it possible that the plywood of the back panel is warped and is rubbing on the back of the Z stage when it gets that low? It should be fairly simple to debug the source of this problem - remove the lead screw from the assembly and move the Z-stage up and down by hand. See if the resistance changes. Does the resistance noticeably change when you are turning the lead screw by hand?

Using the stock extruder drive. May look into other options. I've been thinking about reaming the hotend... maybe .2 or .3 mm. I've reamed a stock hotend to accept 1/8" nominal filament for doing experimental plastic printing, and the results have been great - both with larger filaments, and with the standard 2.85mm filament as well. It seems like it will be harder than ream this hotend though, because the stainless steel tube will be much less readily drilled than the brass, and I believe the wall thickness of the thermal break is even less than the stock hotend. However, I'd like to try it. Sanjay... any thoughts?

Printing with some clear PETG now @ 265°, no fan. Sticks well to an MDF print bed. Clarity isn't perfect, but it looks good overall. Even at this temp, though, layer adhesion isn't great.

Just as a follow up to my earlier clogging problems with this hotend - it is now clear that the problem is less directly related to retraction , and more directly related to minor filament deformations from the extruder drive, clogging once they make it to the hotend. Because of the tight tolerances in the bore of the hotend, it is very easy to get plugs or extrusion slowdowns - which then cause more grinding at the extruder drive, which cause more clogs... etc. The big takeaway has been that as soon as any clogging, grinding, or slowdown in extrusion is noticed, we now pull the filament and make sure to remove with a fine file any deformation that might be present. Also, we use some very high extruder drive spring force, which seems to reduce filament grinding opportunities. Overall, despite these minor problems, we are getting pretty great reliability from this hotend and are using modest retraction settings successfully to get very high quality prints.

I have been interested in printing in some amorphous metal glasses, and I might be able to try this soon. I think this may be the correct path for FDM'ing metals.

I would also make 100% sure that there isn't any aluminum swarf stuck in the bottom of the threaded pulley holes. No matter how hard you tighten the screws, if there is material down there, the pulleys will slide.

Seems like a bunch of extra work for limited reward and a ton of unnecessary weight. If I were going to go for another material I think it would be CNC'd ABS, but the wood is probably better from a performance perspective. That being said, I don't think even laser cutting your own wood panels is a good use of time/resources. I own a laser cutter, and when it came down to it, I wasn't able to source HQ 6mm plywood in the US for a price that made the laserless kit a cost savings... without even beginning to put a value on my time. I think the only way a significant improvement is going to happen to the UM frame is through large volume production of stamped metal panels - and that is only an interesting option when the rate of design modifications and improvements has slowed down significantly.

Sooo..... I appear to have just fried two stepper drivers and the Extruder 1 connector of my PCB. Apparently they kill themselves if you put the driver PCB in backwards? (...for the Extruder driver, backwards = the same orientation as the other steppers. Brilliant! Should be a keyed connector there, or an orientation indicator screened onto the PCB...). Annnnnyway... It seems like I should be able to climb out of this hole I've got myself in by changing the firmware to use Extruder 2 as the primary extruder for the machine when I receive my two pololu replacement drivers. Can anyone comment on this? Have I possibly fried my board entirely?

I would get a spare hotend kit. Even if you don't break your stock one in the early days, you'll appreciate having a fresh one to swap in there at some point in the future.

I think the copolymer mixture that happens as the two filaments blend in the nozzle is probably why it is sticking so well. Dual extruders probably wouldn't produce that sort of mixing, now that you mention it. Still, it will probably be similar to creating a micro-heated bed whenever you print the first few layers as an adhesion layer.

Great to know. Do you add heatsinks to the pololu boards?

While attempting to debug my machine which is repeatedly pausing printing and then resuming when "disable steppers" is selected through the Ulticontroller, I noticed that the potentiometer on one of my stepper drivers had been damaged, and as I tried to put it back together it broke apart completely. I now have a few questions... - Is there a US source for UM stepper drivers? The shipping from Europe is just too much for such a small piece. - Is it going to be possible to replace the broken pot with a 2.5K resistor (the measured resistance of my other pots) and bring the board back to life until a replacement comes? - Is there a P/N for the pot so I could replace it completely? - Is there an available schematic or layout for the driver PCB so I can see what I'm doing? - What do I need to know about using a Pololu driver instead of an UM driver? Anything? I ordered 2 as a short term solution in case I can't repair this one. Any help would be appreciated!

I had to come back to rave further about how amazing this new support material is. Just finished another long print with substantial support and the entire support piece came off with a quick twist, leaving behind a great looking surface underneath. This one software upgrade is a pretty transformative improvement regarding what 3D printers are capable of.

Loose pulleys, probably the hard to access ones on the short belts in the back of the machine.

I just tested this, and it seems to be calibrated correctly - putting the thermocouple in a pot of boiling water read 100°C. It is producing a somewhat shiny surface, just less glossy than PLA. Layer bonding seems to be acceptable initially, but thin parts break along the bond lines, easier than PLA parts would.

As a quick follow up, I did some ABS printing on my mostly stock machine on Friday. They were small-ish parts, but the bed adhesion wasn't a problem when printing directly on the plexi platform. My problem was that the Ultimaker Black ABS I have has weak layer bonding even when printed at 270° with no fan. The parts looked great though. That matte finish of ABS really gives an appearance of finer surface quality compared to glossy PLA. I'll probably search out some natural ABS and see if the layer adhesion improves. Can anyone recommend the best brand and color combo for good (high strength and high surface quality) results in ABS?