Tuesday, August 12, 2014

Me, Myself and Mendel

The Beginning

 I've been spending the last year with Mendel, not the founder of the modern science of genetics Gregor Mendel but the RepRap printer.

The project started by some research on subject which led to building the RepRapPro Mendel because of the possibilities to easily extend it to Tricolour and ease of finding standard parts such as 8mm threaded rod. RAMPS 1.4, the chosen one, has established an image of reliable and stable modular RepRap electronics, at least in my perspective.  The parts  were ordered by means of the bill of materials, which I highly recommend to compose in such projects as 3D-printer. As much as possible of recycled materials were used such as cables from old computers and bearings from my beloved skateboard. After finding and acquiring everything the building process itself was quite straightforward. Measure - Fasten - Repeat...

Finding The Balance

 As I expected after spending multiple hours reading RepRap.org/forums and investigating the experiences of the community the calibration process turned out to be endless road of finding the balance. After slightly changing a parameter such as layer height you might end up with unstable printing because only to mention few properties such as printing speed and cooling don't match the change in the process brought by the change in the layer height. As a independent electro-mechanical system with a bunch of  imperfections due the fact that the 3D-printed parts are not as accurate as plastic parts gained from casting process the real bond between you and your printer sprouts  during the essential calibration process.
[bfr - bridge flow rate, v - velocity (mm/s), T = temperature]
My own path to the equanimity started by not well documented process with an objective of finding the correct amount of steps per unit, in this case steps/mm. This pretty intuitive procedure left me with 3D printer capable of moving an axis with an error lower than 0.01mm; which is the resolution of my caliper to be honest. This procedure also contained bed leveling and extruder calibration. I performed the calibration of the bed by simply iterating the height of the bed while moving the nozzle corner by corner. Extruder calibration is comparable to the calibration of the axes.

Next in line was the real torture of me and my printer, the infamous bridge. First I had to find the correct temperature which depends on the desired print speed and the filament. I started by printing 20mm x 20mm x 10mm wall and found out that the temperature shouldn't exceed 175°C for my speed settings and the PLA I possess. After successfully printing the wall I started to temper with the bridge flow rate and the bridging velocity in Slic3r - the best program to slice your solid geometry in my opinion. Disregarding my previous findings I started with 180°C and the result seem to support that. As my thirst for adventure was fulfilled I found out that that the best result was achieved with 173°C 40mm/s and 1.5 bridge flow rate. In fact I wasn't satisfied and I did let loose the restless part of my mind and experimented with even lower temperatures. I found out the temperature which left me with jammed extruder and raised it 3 degrees which left me with suitable bridge at 157°C, 1.0 bfr and print speed of 40mm/s.

Hereafter

Now I'm able to print with decent speed and accuracy.. The next upgrade will be completely rebuilding the frame as this design is slightly wobbly. I've already printed few spare parts for the rainy day and for while I'm going to print with this configuration.


Links:

Mendel
Thin wall box
Bridge