This is the 14th post in a series detailing my efforts to construct a RepRap Mendel, an open source 3D printer.

Introduction:

The RepRap Mendel is a complex machine, that is quite challenging to build and calibrate. However, there’s nothing quite like the finished product, and the satisfaction of having completed one of the early generation rapid-prototyping devices makes the challenge and effort worthwhile. In total, the RepRap Mendel Build took 3 months and 14 blog posts, which are listed below for completeness:

1. Part 1: Prologue
2. Part 2: Awwww Nuts
3. Part 3: XYZ
4. Part 4: Toil and Trouble
5. Part 5: Extrudapalooza
6. Part 6: Electronics Testing
7. Part 7: I Love it When A Build Comes Together
8. Part 8: Calibration
9. Part 9: Fix It!
10. Part 10: Calibration (Part 2)
11. Part 11: Build Platform Blues
12. Part 12: Fix It! (Redux)
13. Part 13: Victory!
14. Part 14: Lessons Learned

Takeaways:

Over the summer I constructed a Makerbot Cupcake, which took 3 days to assemble. By contrast, the RepRap took 3 months. Why the significant difference? Listed below are the important lessons I learned in building my device:

1. Lesson 1: Mendel, or Prusa Mendel?

   Many months ago, I was asked by Marc which model of Mendel and which extruder model he should print for me. At the time, the Prusa Mendel was a fun novelty, but didn’t have the appearance of official RepRap support. Not wanting to miss out on development and upgrades, I requested the parts for a standard Mendel. While comparable in printing ability, the Mendel is considerably more difficult to assemble than the Prusa relative. If you’re considering building a 3D printer, I HIGHLY recommend considering the Prusa model as an alternative to the standard Mendel.

2. Lesson 2: Do your research!

   There are numerous ways and suppliers that can be used for obtaining your RepRap Mendel, as detailed in the Mendel Buyer’s Guide. When I was purchasing non-plastic parts for my Mendel, I did a cursory search of suppliers and choose Mendel-Parts as my supplier. This proved to be a mistake, and a costly one at that, as I dealt with insufficiently designed parts, shipping delays, and a lack of customer contact or tech support of any kind. In hindsight, I should have checked the #reprap IRC channel and requested suggestions for a supplier; had I done so, my build time could have potentially been considerably shorter.

3. Lesson 3: Choose your materials wisely:

   Mendel-Parts shipped my Mendel kit with 1.75mm PLA filament. Unfortunately, they also sent my kit with a hotend set only compatible with 3mm filament. (I would ask for them to send the correct filament, but they lacked customer contact of any time, so such requests were not possible). Over the summer, I constructed one of the last batch of Makerbot Cupcake CNC printers for my family members, and, as a side effect, had a spare roll of ABS plastic I could use for test prints. While a good, strong material, ABS is notoriously difficult to work with, and a good portion of my work getting my machine to print involved tweaking it in such a way that ABS would stick to my build platform. There have been numerous suggestions in the RepRap community to make use of PLA plastic, as it is deemed easier to print, melts at a lower temperature than ABS, and doesn’t require a heated printbed to print practically. Additionally, PLA is plant-based, instead of petroleum-based, so it is more sustainable and environmentally friendly.

4. Lesson 4: First Extrusion Test:

   One of the major aspects of your printer design is the extruder/hotend. I assembled my system and then tested my extruder, which was a mistake. Furthermore, my first extruder test was “live” — I had plastic loaded in the extruder barrel for the test. Unexpectedly, the PID control system in the Sprinter firmware was extremely poorly calibrated, and as a result dialing in 230 degrees for the extruder temperature resulted in severe overshoot to 260 degrees. At that temperature, the plastic carbonized and hardened, destroying that hotend and necessitating a replacement (a several week proposition).

5. Lesson 5: Beware the opto-joust!

   At the time of this writing, RepRap firmware, and RepRap electronic systems, rely on endstops to “home” the extruder (establish an origin position for the device). A common current approach to implementing this is to use thin pieces of metal. These metals are inserted into small gaps in optical endstops, triggering them and halting the axis. It is fairly easy, especially during early testing, to inadvertently have a flag skip a gate. If this happens, kill power IMMEDIATELY. Don’t think, just do it. (That goes double with the Z-axis: have your hand over your power switch or power plug, and get ready to pull/flip).

Next Stop: Everywhere!

The RepRap may be built and ready to print, but there is still much to do. From here lies the task of device calibration and tweaking. As good as contemporary 3D printers are, they are still lacking in many basic needs, and aren’t yet ready for the general population. Until the day comes that 3D printers are as simple and easy to use as a paper printer, there will be a need to develop and improve these devices, and I intend to be a part of this process. In short, this is only the beginning, and expect to hear much more about the RepRap and its uses in the months and years ahead.