nick-foley

Just tested the Rhino command with a 18,000 face mesh - the results are pretty good. It was a http://www.thingiverse.com/thing:157592 I had previously designed in T-splines, and then converted to a mesh file. Rhino was able to bring the mesh back to a nurbs surface very cleanly and without any noticeable loss. Original mesh: https://www.dropbox.com/s/qpglri1txy0fgfy/Urban%20Puukko%20v1.7%20-%20Base%20Mesh%20-%2018000%20faces.stl Converted Nurbs: https://www.dropbox.com/s/3ryacd44g0fzmde/Urban%20Puukko%20v1.7%20-%20Converted%20Mesh.stp

Rhino has a Mesh to Nurbs command, but it is slow and I've never tested it with complex geometry. There are plenty of ways to do this though - they just require some work. One workflow people use T-splines for, for example, is to pull and stretch a t-splines object to the important vertices of a mesh file.

Whoaa! One time I did see something glowing near the heater coil of my UBIS, but it stopped shortly after and my hotend has continued to function for months after. Maybe some of the heater wires are not fully insulated there?

Good advice. I'd never really used Rhino for displacement maps before, and though it is probably workable, it is definitely slow. All prior experience was in Alias, which has more control but is also probably a lot more than you're looking for in a piece of software...

Couldn't load an .AMF file so I applied your texture as a displacement map to a rectangle in Rhino. First step though was to open the BMP in photoshop and eliminate all color data, and renormalize your texture so that it was pure grayscale and had a more sensible range for applying as a displacement map. I also upscaled it. You probably should look for some higher quality image files if you want cleaner tiles when you print them. In Rhino's displacement map tool, you just need to increase the quality settings from where it starts and things look decent. Attached are the results: https://www.dropbox.com/s/9k43z7o8cgvw6kf/RoofTiles.stl https://www.dropbox.com/s/cdzntfv3586wipc/roofslateBig.jpg

Can you share the base model and the file you're using as the displacement map?

Also, you're underextruding, which is probably not causing the problem but definitely making it look worse.

Shouldn't be a problem, though probably worth noting that unless you find a consumer grade printer with a heated enclosure, any 3D printer that can handle ABS and has a heated bed is going to give you fairly similar performance regarding warping. UM2 will just give better quality and usability.

Curious statement, interested to hear your (...lack?) of reasoning behind it. PLA is a bioplastic, PET is one of the most highly recyclable polymers available, ABS is decent for recycling, and any thermoplastic is inherently very efficient to recycle compared to natural products (and probably some alloyed metals). Add in the relatively low waste of additive vs. subtractive manufacturing and I'm not sure where your sentiment gets any merit. If you're worried about extruding plastic causing environmentally harmful fumes, I think it's fairly likely that the delivery truck your printer arrived in produced more of those fumes on the single trip to your door than your printer will produce in a lifetime... if it does produce any at all.

I wouldn't be put off by the cost of injection molding - we build production parts all the time, and with the right partner and part design, you could easily get an injection mold for sub $5K. Same with DFM - it's a tricky process, but with a few hours of light reading and a few competent hours in CAD, you can get 90% of the way there - and then your injection molding factory will help you get the rest of the way.

Leo, my first guess would be that all of your prints are overextruding. You should try lowering flow settings until you see slight underextrusion, and back off from that just slightly. I get rough surface finishes like that when I'm overextruding slightly. About filament quality - I was doing some tests today of cheapo Amazon filament (Reprapper and Prototype Supply) vs. expensive filament (Colorfabb and Printbl) and I've got to say: the Amazon stuff is probably on par, maybe better. I think that they add ABS to the filament to make it cheaper, which does good things for surface finish. It just doesn't smell as nice when printing...

In your position I'd definitely keep an eye on the Merlin then. I haven't tried it, and though I am genuinely 100% convinced that pre-assembled/fully tested (ie UBIS) hotends being the correct choice for ultimate reliability, from a technical perspective the Merlin seems very compelling.

Scrap the printhead completely. Go with the UBIS hotend and the printhead I designed, or the Merlin hotend and the printhead foehnsturm designed. The stock one is OK, but it is guaranteed to fail sooner or later. I'm closing in on a year or so with 3 UM1's (one with well over 2000 printing hours) and switching out the stock hotend to an UBIS is the best upgrade I've ever made. Also, seriously, get your set screws tight on your pulleys.

Not speaking french, but I think this is a great project. A smart friend once told me that the secret to long-term success with a product is to always undercut yourself - it is essential to capture the low end of the market with a version of your own product. Ultimaker has a compelling product for the high end of the market - to stay relevant, they need to make a compelling low end product. I think a good starting point is a by using 3D printed parts for the printhead, XY blocks, and extruder. I have posted a few to thingiverse which might be a good starting point, others have as well.

Definitely seems like one of the many situations where it is better to ask for forgiveness than permission.

Anyone know of a filastruder/filabot/etc in the NYC area? I'd love to try one out.

Research "displacement" maps, not bump maps - a displacement map actually modifies the geometry, while a bump map only causes the appearance of modified geometry in rendering programs. Other than displacement maps being more computationally intensive, they are otherwise the same thing. You can do this in Rhino, using the "ApplyDisplacement" command.

Bet a wire is getting loose around your thermocouple board. Or some sort of EM interference along the TC wires.

My guess is that you have your settings such that infill speed is faster than perimeter speed? That would cause such a problem if you were approaching the volume per second limit at your given nozzle temp.

Any faberdashery suppliers in the US these days?

Merlin

It looks like the heel just barely touched. Though I agree, the ability to make objects float in Cura would be very useful.

I don't think there's much magic to printing extremely strong PLA parts... and contrary to frequently posted opinions, I believe PLA parts are generally much stronger than ABS parts (at normal temps) due to the much weaker interlayer adhesion of printed ABS. My "strong" parts usually have 3mm walls and 20+ infill, but don't expect infill to do much... with most parts you'll probably get the most efficient strength just by increasing your walls. Just make sure you're printing hot enough that your parts don't break along your build lines. Using a larger nozzle, or simply telling your machine you're using a larger nozzle in Cura, can do a lot to make big, strong prints happen quickly. It also makes infil more meaningful, since 0.6mm beads, for example, bond much better to themselves and others when making infil. These parts I printed are quite strong: http://www.thingiverse.com/thing:278290 So are these: http://www.thingiverse.com/make:71656 Both describe their settings.

I think you could improve the top teeth just by dropping the nozzle temp a little for the entire print. 0.06mm (or maybe 0.07 or 0.08) will show a slight improvement... beyond that I think you're way beyond diminishing returns. 40 a week sounds easy enough... I have a UM1, but there have been stretches where I ran consecutive 20+hour prints every day for weeks.

Yes.