Daid

Some more details about the NetFabb marlin combination can be found here: viewtopic.php?f=9&t=479 The E14 is not hex. The "E" stands for "Extruder", so it's 14 steps per "unit" of extrusion. Not sure what a "unit" represents in NetFabb, but E14 is needed. Upgrading to marlin should be quite painless if you use: http://daid.eu/~daid/marlin_build/ Set: "M109 settings, Wait Time:" to 3 seconds Also, if you are still printing with ReplicatorG you need a baudrate of 57600 instead of the Marlin default of 250000 And you can always install the stock firmware from within ReplicatorG.

1) 10.000 a year is about 30 a day (7 days a week). Depending on your printing speed/quality you need a few hours per 6x6x6cm model. Using a single printer won't cut it. Also, if you produce 10.000 of the same items 3D printing might not be the best solution. 2) Smooth depends on your model, your calibration, and a few other factors. But I haven't seen perfect smooth yet with my printer. Sanding might work, but melting the outside with a acetone or something also works. 3) 25-30 euro per kg. If you use PLA. However, for medical supplies, you might need medical grade materials, which I don't even know if they exists. 4) Depends on your thread size. http://www.thingiverse.com/thing:12728 this works. But for "normal" thread, people usually cut in the thread with taps.

"Exterior only" support is broken... because, it should have been "Exterior Only" with an upper case O. Stupid bug that I made. (If you change "newui/mainWindow.py" yourself, it should work) Thanks, all 3 flip actions indeed look broken in the slicing. I'll have a look at it. The object is movable by adjusting the machine center in the machine config tab. (The confusion has been noted ) A beta like this is perfect to get all these bugs out. I'm not adding features right now, and just fixing the bugs. So we can have a bug free release pretty soon.

To give a very simple and clear answer to your main question. No. The Ultimaker will not match those prints in quality. It's a pretty awesome machine, but those first two prints would really push the limits of it. And the 3th one, it won't look as, sandy as those circles. This octocat: http://daid.eu/~daid/IMG_20120107_214334.jpeg is still one of my favorite prints. It's detail is very good, but you still see the layers, and it has a few flaws. As for stringing, stringing looks like this: http://daid.eu/~daid/IMG_20110927_000944.jpeg , it's quite simple to cut away the strings afterwards. You can get better stringing results then this print, compared to recent prints the strings in this one are pretty bad. But I still get strings. However, with Skeinforge "Comb" feature, the 3th object should be able to print without strings.

The jumper is indeed there to power the Arduino from the 19V. This is useful for printing from SD cards.

ABS needs a higher temperature then PLA. But remember, around 300C your PEEK will start to meld. Which is kinda very extremely bad for your machine.

That error is... well, not sure where it comes from. But it still works after the error. You can fix that error by installing .NET runtime 1.1 (which includes the DLL) EDIT: Identified the offending file. It's "/python/Lib/OpenGL/DLLS/gle32.dll", it's safe to remove the file, as SkeinPyPy doesn't actually do anything with it. The other error, might have something to do with the model you are slicing. I've seen the same thing happen once. Running the stl file trough http://cloud.netfabb.com/ might help. But I'm also interrested in the model and setting you are using to cause the error, maybe I can fix it.

PLA won't survive in the dishwasher as I understood it. You might want to get some ABS for that. Printing from within NetFabb isn't stable yet, so printing from RepG or PrintRun is recommended. Skeinforge has support settings, hidden somewhere. But you might want to upgrade to Marlin and SkeinPyPy: https://github.com/daid/SkeinPyPy/wiki (Beta3 has an upgrade to Marlin buildin). Even if you slice with NetFabb, it's better to print with PrintRun and Marlin. Because it produces better results. I wish I could say "Support works in SkeinPyPy!" but Beta3 has a bug which makes support only work if you pick "Everywhere", the other 2 options don't work. And you usually want exterior support. You do need to adjust the start code of NetFabb a bit if you use Marlin, to insert an "M92 E14" in the beginning.

What does the log say? Most likely your laptop is lacking SSE2, which should be noted in the log message.

Tried the newer RC2 marlin with PID? http://daid.eu/~daid/marlin_build/

I've just uploaded Beta3 of SkeinPyPy (With new Userinterface). It's fresh. It's hot. It's downloadable: https://github.com/daid/SkeinPyPy/downloads It's loved by the Ultimaker team (even though they have Macs, and the Mac version is not ready yet) It's less buggy then the Beta2! It's got X-Ray vision! It's packaged with a newer PyPy version, so no more "!?!Something went wrong during slicing" It's assisting you in verifying some basic machine functions (like heater and end switches) It's just... good, and you should download it. Now for the long, and not so boring story. I was at Ultimaker HQ today, thanks for the invite Erik! I enjoyed it! We talked about a lot of things, but we mainly talked about the software, and how bad ReplicatorG is, and how much a replacement would be so much better. And the replacement was there, it was SkeinPyPy. And nothing is set in stone. But it's possible that after the next ReplicatorG release from UM, SkeinPyPy might be the next thing they'll be working on, software wise. They like the simpler UI, and the First Run Wizard. But they also suggested an even simpler UI (select profile -> select model -> print). Which I also like as "beginner" UI. We did a lot of testing, and fixed quite a few tiny bugs. Renamed the "DepthComplexity" to "XRAY!", as it looked a lot like X-Ray vision. And we ran into a ton of problems with the MacOS version. We fixed a great deal of them, but we could use someone with more MacOS development experience. Especially packing software correctly was hard to figure out (not solved yet) Now, this Beta3 is not perfect yet. There are still a few UI things we didn't like. Like it should be possible to switch between GCode/Model view, especially after restarting SkeinPyPy. There where also ideas about showing what each setting does with pictures. It's also missing a GCode sender, so it's still packaged with PrintRun. There are many ideas, but I figured to put this release out there so we can get more feedback. It's also better then the Beta2 because of all the bugfixes. And, I got their new unreleased, half announced new pieces of hardware, thanks for letting me test that! No, and I won't be telling what it is. It's a secret And I've seen work on a printer head without PEEK. It was a fun day, to hear all kinds of problems they face as a small startup, which tries to deliver all over the world. To see a random unannounced customer show up at the doorstep to pickup his missing parts, just because he was in the neighborhood. And too see how busy they are with support. Thanks again for a fun and productive day, and thanks for the lunch UM Team! (And start doing those lunches on a fixed time so you don't skip out of them )

Schematics should be posted soon. (Says Erik) The jumper is for powering the board from 19V. That the fan gets powered from USB is a side effect (bug?), but you can safely remove the jumper to get the old situation.

Did you check if the dimension plugin was enabled? (The GCode file should have E values behind the G1 commands)

Also, http://www.orbi-tech.de sells PLA-90, which is suppose to shrink less, so should give less warping. (Never tested it myself)

That's a bug in PyPy (doesn't always happen, depends on the model). If you rename the PyPy folder it will use Python to slice, which will be as slow as Skeinforge, but you will get the new interface. Beta3 will solve this issue, and hopefully will be out later today.

Your circle doesn't look very round to me, maybe you have a belt tension problem? That also causes the infill to connect less to the outer walls. Else, you can increase the perimeter infill overlap value. But that's not really recommended. The thin walls is always a problem. With SkeinPyPy it's a bit easier to get that right if you select the right wall thickness (so the 2 walls connect) but there is no perfect solution.

The idea is that you have the smaller X/Y rods exactly parallel with the thicker X/Y rods in the frame. There is a little room for skew there, and you don't want that as it puts extra stress on your machine and makes your printed models skewed.

Ok, that sounds odd indeed. The 1.5.4 PCB doesn't have a jumper there. I might speak with a few UM team members tomorrow, if I don't forget I'll ask about it.

Most likely the 1.5.5 and 1.5.6 updates are just minor corrections to the layout. Schematic wise they are most likely the same as the 1.5.4 version. Where are the 1.5.5 and 1.5.6 versions? No idea. What are you trying to understand? The schematics are pretty straight forward.

That's not the schematic you are looking at, it's the layout. The power and ground are connected to the power/ground planes. If it wasn't connected it would show a "rats nest" line. (yellow line) I'm not sure how/if you can show the power/ground planes, but they are there. Maybe it's a hidden polygon instead of a layer?

What part is not coherent between the schematics of 1.5.4 and the 1.5.6 board? I surely hope they did not change the pinout.

Are you using Marlin? Are you belts properly tensioned? Temperature can be lower, which might help. 100% infill will require perfect calibration. Try 90% infill so if you have a bit of over extrusion it will flow out during the infill and not during the outer edge. Most likely you have a pressure buildup which starts to cause extra extrusion around 1cm of height. This fan shroud does wonders for small objects: http://www.thingiverse.com/thing:10501 I printed these: http://daid.eu/~daid/IMG_20110921 ... small.jpeg and they look perfect. But they where at 20% infill.

Small? From the first angle yes, from the front angle you can see it's a lot larger then you think. It could easily hide an NEMA17 I'm all for the "plugging the nozzle with a pin" idea, as this should work even better then retraction (also works at high temperatures), but it must be very hard to build...

Wait? The fan cable is suppose to go trough a hole? I have mine just "flying". Look at this photo at the top: http://daid.eu/~daid/IMG_20111023 ... small.jpeg No problems with it so far. For the fan shroud, if tape doesn't work. Try to see if you can place the fan on the bed so it blows over the print, with that you could be able to print a fan shroud. The stock fan shroud disassembled itself during my 3th print. I'm not sure why they aren't doing something about it. Must be the most replaced part of the whole machine.

You use the 3B part to put between the slider blocks and the edge of the frame, this will give you the same distance at both ends of the X/Y rods. It's just a simple measuring tool in this case.