Teensylu boards often ship with LUFA CDC pre-installed from China. While this is nice we still need to upload a firmware to the board. Unfortunately this is not as easy as Pi and using a modded Arduino IDE is necessary. This is because the board has an AT90USB1286 chip which has an USB converted integrated and for some reason is not officially supported by Arduino. I found some instruction on the reprap wiki and a blog but they did not work for me. What I don’t like about the wiki page is that it describes the procedure of uploading with lots of additional tools which are totally not necessary. Rubbish I say. Edited it already but the changes are awaiting moderation or some kind of anti-anarchy-precaution-system-ish thing. To hell with it, I’ll post my findings here.
The very first prints on new 3d printers tend to be somewhat nearly perfect here. Which is strange because first prints on new printers should look ugly, messy, not proportional and overall malformed. What then after some time happens is even more strange. With tweaking and tuning they start to look ugly and all of the above stated.
So after a very nice 0.5mm thin wall print and the Groot bust I ran into alot of troubles with the Merlin Hotend and the Wallace printer itself.
After these first unmatched prints everything went downhill. I will list it in non-chronological order…
So finally my RepRap Wallace printer is ready to print. This printer is a mixture of the stock RepRap Wallace and sponnets Wallace++ M8 Edition. I choose a Gregs Wade Extruder Reloaded with an passive cooled Merlin hotend. This is my very first print:
And this is my second print
Really cool for a pretty early stage after completing the machine.
While building the Rostock Mini (Modification) I came to the point at which all the parts I needed were printed. And leaving my 3D printer resting for maybe weeks without a meaningful task was not an option I started to print another 3d printer.
As stable builds are boring and for the anxious and cautious kind of people I went for an experimental design: The Wallace !
Actually it is a remake of the printrbot design. I like the simple design over the Prusa i3 which needs frames made of wood or acrylic glass. This Wallace instead only needs some printed partes, hardware store threaded rods, smooth rods and some belts. While reading the reprap wiki and skipping through thingiverse I came upon the Wallace++ M8 edition. The name really got me in the first place. “M8 edition”. As the stock Wallace uses smaller Nema14 steppers and M6 rods (threaded+smooth) I decided to give it a little more extra steel. Cause the more steel the merrier.
So I printed the M8 edition and found that the motor ends were a bit to fragile in my opinion. Also the bearing retainers that hold the bearings in place were a bit to slaggy. So I went back to the original Wallace and replaced the parts I did not like from the M8 edition. This means my Wallace now consists of the following parts:
Additionally I found a really cool upgrade to the X and Y axis http://www.thingiverse.com/thing:37787. It replaces the timing belts by racks & pinions. Komb achieved really great improvments using this upgrade so I decided to give it a try on the Y axis at first.
Since my TwoUp has a limited build volume I had to reduce the racks tooth amount by to from 40 to 38. This could be easily done by just entering the new value into the OpenSCAD file. Love this!
I already bought some flexible couplings for the Z axis but lately I’ve read somewhere on a blog that flexible couplings are not good for Z axis as they are too unstable. Didn’t know that but the guy who said this seemed legit and an experienced 3d printer builder so I trust his words. Static couplings can be found by the dozen on thingiverse too.
A word on accuracy on fit
I realized that the bearings did not push fit into the X axis motor and idler parts of the Wallace++ M8 Edition. The author made the inner diameter of where the bearings go 0.5 mm smaller than the bearings so they can not move around. When I printed those I was not able to push the bearings in. I tried to use a vice but crushed the whole printed part. After some googleing I found out that I had to calibrate the thickness of the printed walls. There are two tutorials, both different in the process but with the same outcome. More of less. First there is the official slic3r tutorial (https://github.com/alexrj/Slic3r/wiki/Calibration) and second there is Ian Johnsons tutorial over at Solidoodletips – Wall thickness/Flow rate. Personally I like Ian Johnsons tutorial better because it is more “engineering-style”.
Though why calibrate the wall thickness ? It’s easy: If you don’t do this inward holes become too small and outward holes and wall become too big. Hence push fit designed parts won’t push-fit fit.
But there is one thing that is entailed after calibrating the wall thickness. I realized that when printing with a calibrated wall thickness the infills do not quite hit the perimeters. Means alot more needs to be tweaked inside slic3r to get a decent looking print. I havn’t got the time to deal with such details so I just deal with it.