JohnInOttawa

That's an interesting innovation. I was just starting to wonder what I should do with my UMO. This might answer that question. I tried to figure out how the support arms are configured. I get that there is yarn strung between the holders on either side, but can't quite see how that is incorporated. I presume the print is paused, the yarn strung flat, then the print resumes and the yarn is just buried in the filament, but when I see the pictures, it looks like the yarn is under tension downward. is the whole thing strung like a raquet, then lowered to create that tension before printing resumes? Thanks! John

Same here, colorfabb with a 0.6 (hardcore). no issues with clogging, and I went to 0.1 layers. As an aside, I found this material printed with almost invisible print lines, but as I was going to apply finish, I sanded it anyway, and sanding worked as expected for a softwood like pine. I did have to pay attention to infill pattern, wall thickness and print temperature, I was printing cylindrical items with a vertical hole for a lamp wiring shaft, and my first attempt had a weak layer band that fractured almost in a crystalline pattern around the narrowest portion of the print. Cura layer view was helpful in seeing what was going on, I had managed for a few layers in the fracture zone to create just wall layers that did not divide evenly by the line width and didn't really give enough time for the layer to cool. This area did declare itself visually with a darker band in the material, looked like wood that had been burned by a router bit. By increasing that zone just a little the band disappeared and a test to destruction more than met my strength requirements. I hope this helps! John

Let's try this approach: 1) please identify which slicer you are using - if it's Cura, then you're in the right place. If it's not, and this is not an Ultimaker printer (looks like the Tevo is a prusa style) then you might be frustrated at suggestions here that don't make sense for you. 2) I would try a different file of the same height but using less material so you can get past the shift point sooner - a vertical bar 2.5cm x 2.5 cm or whatever would be stable - just to see if this issue tracks with height, or time. If you get the same issue at the same height, again, I would think mechanical aspect. 3) Assuming you use Cura for a moment, I would sink the model in Cura to 25 or so layers below the problem area and print just that part of the model, to see if the shift occurs at the same place in the model regardless of how many layers have been printed to that point. If the shift happens anyway. then you can start to zero in on other things and you won't have to wait so long to see the problem and test a potential solution. I did for a time have a prusa style printer, which also experienced layer shifting at a specific height. In that case, it was indeed dirt on one of the Z screws that caused an asymmetry in the gantry, when I put a bubble level on the table it was level but the gantry no longer was. Not saying it's your case, but when things like this take place, it's among the easiest things to fix. John

Wow.

Not a printer I am familiar with, however if the problem is always at exactly the same height, regardless of how long it took to get there (i.e. narrow, low infill jobs get the same problem at the same height as large, high infill ones), then I would look at a mechanical problem instead of Gcode. Mechanical problems are not all broken things, it can be as simple as dirt in the Z axis leadscrew that causes more friction and missed steps. This is pretty common, especially in printers where the bulk of jobs stop at about the same spot, so debris gets pushed to the same level and congeals with the lubricant. Since you indicate that this all started when you recently decided to print bigger things, I'd start there, remember that most leadscrews are multi-threaded, so you may have three groove lines to inspect and clean per leadscrew. John

This topic raises another aspect - I see more folks referencing IR temperature when assessing component health. While IR is an excellent method for non-contact temperature reading, the ease of use of current instruments belies the difficulty in getting a true value. We use calibrated IR handheld 'thermometers' and imagers in research settings and I've used them to assess stepper temperatures as well, on my UMO, during a long print. Depending on where the sensor is aimed, the surface texture and IR reflectivity and nearby heat sources, a handheld, single-point reporting IR thermometer can under, or over-report temperature. Over reporting tends to happen on surfaces with pits and valleys that create temperature wells, whereas under reporting is a function of surface material, specularity of the surface and surrounding sources. Error can make up more than 90% of the value that reaches the sensor. An imager is similarly affected, but tends to have adjustments for focus, emissivity, background temperature and transmissivity of the airmass (which can be a factor in a hot space like an enclosed printer) and reports thousands of points, so it's easier to see what is going on and correct for it. The other thing to keep in mind with IR reading is that it's solely a function of surface radiation. If there is any form of airflow over the surface or between the surface and the sensor, (like print fans tend to produce), it's going to divide the reading from reality. I've been meaning to post some shots of reading effects on steppers and heated beds. I was hoping to also get some images of a print in progress to show relative layer temperatures, but I will need some form of time travel to get the rest of my stuff done as it is.... John

This looks to me like coasting - you're not extruding in the gaps but you'll lay down oozing filament. Try it again with no coasting and see.... John

I have a 0.3 and 0.25, on a UMO and a UM3 so not the same thing. But maybe this is related. Can you share what adjustment you made to feed rate? In my UMO, I had to almost cut it in half for the smaller nozzle. John

Thank you. I asked if the fellow would provide some traceability on the printer, that seems to be the end of the discussion. Maybe the community would consider forming a list of missing and suspect machines. Probably wasn't needed before, but as you fellows slowly take over the world, there is obviously a secondary market starting..... Thoughts? -John

Hello everyone. Strange question of the day. A fellow has reached out to me, knowing that I have an Ultimaker 3, asking if I would like to buy his, new in the box, everything included, never been used, Ultimaker 3 for a too good to be true price. No receipt available, but he tells me it has never been registered for a warranty. Hmmmmmmm. Now I don't want to go around accusing anyone, but neither do I want to buy stolen goods, if indeed that is what is going on here. I know that there are blacklists for some other hardware, is there some way for me to check the serial number of this unit and see if it's legitimate or 'other'? Thanks in advance! John

Thank you. Your comment about slowing down in corners resonates - these are small teeth and corners are exactly where things are getting messy. I'll try to explore ways to change how that path is laid down. If it was CNC I would know how to change that toolpath, but I don't have that skill or knowledge (yet) here. Fortunately, there are some smart people to ask here ? For a dummy tower, is that simply a matter of creating a separate STL and importing it to Cura, or can I do that within Cura itself? Thanks again!

Well said. Looking forward to the next chapter of fractal land ? J

Well, that discussion would occupy an afternoon and more than a few cups. While I make my living from math based stuff, it would be a cold and barren pursuit if not for art. At least for me, if not for those things that remind us of beauty, I'd soon forget to look for it in my struggle to get everything done. The other night I was passing an isolated thunderstorm over northern Canada. Alone in the dark, the cell was lit continuously from within by lightning, the intensity of the storm created an ionized glow immediately overhead, while still further aloft was a green, dancing curtain of northern lights. All the way from the ground to space appeared to be lit up. Our aircraft and systems are an engineering marvel. Could not hold a candle to that vista. Art wins, at least IMO. Cheers J

I'm always taken aback when I see creation in progress. The revisualization (if that is a word) of an abandoned fuselage into something that could be the centrepiece of a sci fi show. Pretty neet. I look at that same original photo and think about how salvageable that build could be, what would come of corrosion or stresses on parts that were only supposed to be in their current state for a few hours, instead of the months or years it has been. Which is why I make parts with my Ultimaker, while you make art! Thanks for sharing. J

That looks like AN225 number two taken in the Ukraine..... John

This thread is related to my nylon printing 'adventure'. The part of the print I am struggling with is an upward facing gear-toothed circle. It looks like the print is suffering from severe overextrusion and the nozzle is dragging material out to the perimeter. Replaying the layer view in Cura, I am starting to wonder if the line width required for these gear teeth is narrower than I have set. I'm using a 0.4 nozzle currently, have adaptive layers in use per a suggestion here and that has improved things somewhat. Base layer height is set at .05 and width at 0.4. When I look at how it tries to print that, I can imagine why there is excess filament. When I increase line width to 0.45, it looks better in Cura. I imagine it will slightly underextrude and that might offset whatever is going on. Of course, if underextrusion is what I am doing, I would rather not do that do the rest of my print. Can I set up a line width change to start at a specific layer? Thanks once again... John

Appreciate this! I will isolate that part of the print and try it again on Thursday. I had not considered adaptive layers and will adjust retraction, will post the results back here. I had also slowed down my print speed, maybe too much. I'd welcome advice on what others have found to be optimum. John

Sorry I didn't add it earlier, was soaking out the PVA. Post is updated with the print pic. Thanks again John

First part is done, new day, new design, new question! I'm printing a part now that has a circular 'dog', upward facing teeth around the perimeter. Cura showed no issues, but, while the rest of the print turned out fine, the toothed ring turned into a nondescript raised roughened, mostly melted, area. I'm pretty sure I am missing an essential technique for managing the small, rectangular pyramids that represent each tooth. I ran a 0.4 AA core, Ultimaker black nylon, 0.1 layer height, 250C and no fans. Thanks in advance! John

Sooooo Good news and 'learning'. Glue stick (with good advice applied, thinned out with a damp tissue), worked perfectly. Flat, no warping, came off without taking any glass afterward. 'Learning:. Ironing. I had slowed down the print speed to ensure the quality was where I needed it. Worked great. But then I decided to tick 'ironing' at similary low speeds, on a piece with lots of corners. From the already pretty good finish, Ironing produced three distinct textures. At the start of the corners, surface quality was slightly better than non ironed. Mid length lines were amazing, just about zero discernable lines. But then we got to the corners on the end, and, well, it just created a melted blob, effectively destroying what needed to be a flat surface. No biggie, this wasn't intended to be the final print anyway, just checking dimensions, but note to self. Easy on the ironing with nylon. (BTW this effect worked perfectly on the PLA version of this print). Thanks again for the advice! John

Thanks gents! I bought a glue stick and will try out the slurry. Thought of doing it this morning, but thought it might take some time for the first run. Of course, with the printer tied up for the rest of the day now, I guess I have time.... Much appreciated. John

dumb question of the day - no longer have the original glue stick. Which ones use PVA? As an aside, I must be the only woodworker I know who no longer has any PVA glue in the shop. Everything has additives that I probably don't want on my build plate. If there was an anti-matter version of perfection, I would achieve it.... Thanks John

I understand you have kept your filaments in bags with dessicants. That might not be enough. How long ago was the PVA first used? To my eye, it does have the look of humidity. I use a filament dryer, dried my PVA completely two weeks ago and sealed it with fresh dessicant. I could hear it popping as it printed today. One area of a support failed the same way as print #1. Just a thought..... John

That sounds like a good plan. Thanks! John

Good morning. I just finished a job with Colorfabb Woodfill on my UM3. I ran it with a hardcore 0.6, results were better than expected, good adhesion, excellent surface finish, sanded to smooth and accepted stain without any unexpected blotches. Now, my prints were simple, no overhangs. But so far, so good. John