3D Printer Reproduction Continues

The model's not perfect. Find the missing fasteners...
...details, details.
Z-drive on the lead screw, just before adding the belt:
Y-guides: alignment is important, and a minor challenge.

In a previous post, we talked about planning, gathering necessary hardware, and 3D printing the RP-ready parts for the RepRap Mendel.  Well, that’s only a fraction of the process.

Assembly is coming along slowly but surely (during summer, things move at a different pace).  We’ve only had a small amount of time to devote to the printer, trying to fit it in between project work and other responsibilities.  However, now’s a good time to pause and give an update.

The physical assembly is nearly done.  We’ve been using the SolidWorks model as our guide.  This is not a perfect model (perhaps this has been corrected, but we have yet to find a perfectly clean version).  As they say: “beggars can’t be choosers.”  Just the fact that such a detailed model is available is testimony to the openness and generosity of the community.  But following the model is only so useful.  For example, as soon as you open the general assembly files, you notice missing pieces of hardware (not to mention the floating parts).  Thankfully, it’s easy to figure out which parts are missing.  The images at left show a typical condition: the ends of those rods are going to need washers and nuts.

The most useful aspect of following the 3D model is that the assemblies are grouped hierarchically.  This is probably standard practice in industrial design, but the logic built into the hierarchy is indispensable.  This is not a mere messy grouping of parts; each assembly has its own logic, based on the hierarchy.  After building one or two of the individual assemblies (the Z-drive or the Y-guides, for example), you begin to perceive the design logic behind the RepRap: get close, assemble, adjust, repeat.

We found it most useful to build each of the main assemblies first, rather than putting together the entire rig as you go.  Some parts need to be placed on threaded rod and secured before other parts; this was something that we discovered through trial and error…and no small amount of consulting forums and double-checking the model.  Though it’s slightly more work, it’s best to build each assembly first in its entirety, even if this means that small parts of the assembly will need to be disassembled later to add a belt or position the assembly on the frame.  This method allowed us to understand how each part interacts with the entire system (remember: we’re not industrial engineers).

We’re at the point where we’re beginning to work on the electronic portion of the project, wiring up the stepper motors and adding the extruder.  If we have learned anything important so far, it is this: the RepRap is not a kit.  The errors and omissions in the “instructions” (if you can call them that) are better thought of as opportunities to learn.  The overriding ethos of RepRap is striking a balance between community engagement – you have to ask others for help – and self-sufficiency – if you can’t improvise, you’re in trouble.

In our next post on this topic, we’ll talk about the wiring and electronic components.  Meanwhile, back to (architectural) work…



  • Roberto Carlos

    This is very impressive! I have yet to see one of these things in person, but I am aware of their capabilities. Great blog! I will keep following the progress of your second 3D printer. Best of luck!


    • dbelcher

      Thanks Carlos. The RepRap Mendel’s build performance will surely be underwhelming, when compared to the Objet Alaris 30 (which isn’t even the top-of-the-line model). However, the RepRap is only about 1/50th the cost of the Alaris, sooooo… Also, it’s just fun to build one yourself.

  • Arun

    Very interesting…and great seeing it person the other day. Please let us know if we can help with the “controls” aspects – with SNAP PAC controller & I/O.

    • dbelcher


      Thanks! We might ask for your help, considering your level of expertise.