Please help me set up Repetier, although I remain hopeful that you'll write that tutorial and post it up here for people other than me (Jonimus and BrandonW, for starters, but also the rest of the internet). My filament is coming very soon, and I want cases that look as good as possible, so settings like those seem like what I need. Thanks in advance.

Edit: With geekboy's very patient help and Jonimus' occasional comments, I got Repetier and Repetier-Host installed on my printer and computer respectively. Once I finally got some rudimentary settings in place, I printed out a tiny 20%-scaled model of my TI-84 Plus calculator case. I hope I'll be exploring some more complex models with Repetier soon, but I'm already very impressed with the quality of this. The blob are unfortunate, but that seems to be unavoidable with this printer.


Edit #2: since I split the thread off from this one, be aware that I'm working on a 3D-printed TI-84 plus case for both fun and future Cemetech contest prizes/general swag.

So, question(s).

Will this have the classic Cemetech logo (so the lightning bolt is visible) added to the outer case? Will this come in Cemetech Red (or equivalent?) Will you be doing older calc cases, or even the Nspire? (My Nspire has no case :< )

Otherwise, looks like fun

I haven't decided how to put the logo on yet. I almost would like to try to put the current logo on, as long as I can also approximate the lightning bolt.
I have $50 of ABS filament arriving today or tomorrow, and half of it is a beautiful red. I'm going to start with the TI-84 family-style case, then go on to the TI-84 Plus CE case. I'll probably also do the TI-Nspire case if I have time, though. And I remain hopeful that I'll eventually have the time and inclination to model the actual case of the calculators (ie, not the slidecase), so that I can transfer the electronics from a real calculator into a custom red Cemetech case.

It is!

Kerm, I think you can have the logo as an svg and import that into something like blender to save it as an stl. Then Solidworks can turn that into an extruded cut into the back of the case. So, the text and the lightning bolt would be two different files so they can be cut in at two different depths.

Hmm, that might be tricky, as I unfortunately was dumb enough to create our logo as a (very high-resolution) raster image. I still have the font for the text, so I can use that directly in SolidWorks, and for the lightning bolt I guess I could trace it in a vector graphics program. Any suggestions for that? Can InkScape do that?

Inkscape can do that.

In Solidworks, (to get the letter outline effect) You can use the font twice and once, big with the letters right up against each other and the other small, with the letters separated using the separation input. Then cut those in at different lengths and then the lightning bolt on top maybe using similar layering. This will be so awesome and that tiny case you printed is so cute!

I ended up succeeding in cutting the text and lightning bolt out of the model, if you take a look at the case topic. Anyway, I have general printer problems that I think I need to resolve. I tried setting the layer height to 0.1mm for this 25%-scaled calculator slidecase print, and that was problematic, but the 20%-scaled calculator slidecase I printed on Monday had similar problems. Specifically, the bottom few layers are way, way too thin, and the extruder feeder slips against the filament for the first few layers. This should be six 0.1mm layers, and it ended up being so thin that at the bottom it is transparent. I think that the printer is misjudging the starting height of the bed. I already tried to level it, and the corners are within 0.2mm of each other, and I also went through the auto-levelling process and stored the new bed offset. What can I do to fix this?

Neat! Those are looking sharp! (Even though they are rounded. )

EDIT: As for your question, I'm afraid I can't help you there.

Thank you. I found this bit of information that may help, but I'm still looking:

Edit: I adjusted the z-offset by 0.2mm in Slic3r, and it seems a bit better now. Ive been reading everything I can to learn more about Slic3r, Repetier-Host, Repetier, and the Da Vinci 1.0, and I'd like to think I'm making progress, but there are still a few things that are stymieing me. The latest issues are largely related to supports. I'd like (if possible) exactly the type of supports that XYZware creates: they're thin, don't use much material, and most importantly, are easy to remove. The supports that Slic3r has been generating on its own with my parameters are very thick, and almost impossible to take off. They also make the text on the bottom of the case completely illegible. Any ideas how I can configure Slic3r to make those sort of supports? Thanks in advance.

Can I get a screenshot of the support settings pane? You probably just need to tweak things in there.

I currently have:
Code:


I think I need to go ahead and do this step in Repetier's installation instructions, which I ignored because I had spent a total of 7 hours leveling my bed over the preceding two days:

Reading all this trouble shooting, makes me glad I don't have a 3D printer. Why do you guys have to level it so often, seems a little counter productive. Keep up the documentation and work, when I finally get one I'll have some excellent trial and error posts to read.

Yeah, levelling is an unexpectedly (at least for me) painful part of the process. You can get around levelling by printing a raft under your piece, which is a solid, level plastic base that you can later peel off your print. However, using a raft means you waste more time and plastic making your prints.

I'd value my time more than I'd value the filament I'd save. Besides, it looks like you used more filament printing than if you had just printed said raft.

Perhaps you should consider getting an extruder to help recycle failed prints?

One thing that is still bothering me is the Z-probe should be able to compensate for slightly out of level beds using the axis mapping. Of course the less it has to compensate the better but it shouldn't be this bad.

Also I believe there should be an offset somewhere that adjusts machine Zero to extruder Zero but I haven't seen where that would come from other than the Autoleveling being run, but it would still need to know the distance from the Z probe tip the the extruder tip.

I spent time on Saturday to level my bed perfectly: as far as the probe can tell, my corners are within 0.01mm of each other, which is well below the probe's precision. Nevertheless, the head still appears to be mushed into the bed, and with a first layer height of 150% (ie, 0.3mm on a nominal 0.2mm layer height run), the extruder ends up dragging a huge ball of ABS in front of it by the time it finishes the first layer. I found that the configuration.h for the Repetier firmware for the Da Vinci 1.0 specifies a probe offset of 0.28mm, so I'll try changing the z-offset in Slic3r from 0.1mm to 0.28mm and see if that helps. I'm surprised that since it has that information though, it wouldn't already be applying that offset. Jonimus has requested the data from the M205 command, which I'll post here shortly.

That offset should be set in the firmware and the Z-probing should adjust its compensation matrix with that offset taken into account. Hopefully the M205 dump will give me some insight hopefully on what is going on.

Here's what I get from running M205:
Code:


Edit: Setting the Z-offset to 0.28mm from 0.1mm seems to have not made things better, and possibly even made them worse. I then printed six 10mm x 10mm x 2mm "cubes", offset from 0.0 to 0.5mm off my bed. The one with an offset of 0.3mm appears to have printed the best, so I'm going to try again with my z-offset set to 0.3+0.28 = 0.58mm.

Edit #2: I tried again with 0.58mm, and it actually came out really well. I'm going to go up another 0.02mm to 0.60mm tomorrow and try printing out a whole case, probably at a 0.2mm layer height for now since I'm not sure it will come out well. Slic3r does such weird supports that I don't know if I should bother trying them, but I guess I should try. The images below show:
(1) The result after adjusting to an offset of 0.28mm (the extruder hit that huge burnt lump of scraped-up ABS and got offset by over a centimeter;
(2) Six flattened cubes offset between 0.0mm and 0.5mm. Bottom-left is 0.0mm; moving right increases 0.2mm, and moving up increases 0.1mm. The top-right is nominally 0.5mm above the bed. Subjectively, the 0.3mm offset seemed the best, so I set the z-offset to 0.28mm+0.3mm and tried again with 0.58mm
(3) A nice 20mm x 20mm x 2mm rectangular solid printed with that 0.58mm offset.
Click any image to enlarge it.