Reverse Engineering a Part for 3D Printing!

One of the joys of buying a (very) used car is repairing all of the small issues that reveal themselves to you during the first months of its use (like the rear view mirror deciding to dive into the footwell in the middle of the motorway).

When I lifted the parcel shelf for the first time, the right side disengaged, and ended up in the boot. Hmm. Upon closer inspection, I noticed that it was missing the plastic pin that engages with the car interior.

I took the part out to inspect, and decided, this might be a good opportunity for a 3D printing reverse engineering project!

Try to be as precise as possible, especially for key measurements (diameters, thickness of parts that will mate etc). Produce a sketch with all of the key dimensions.

For this step, you're going to need a bit of 3D CAD experience (for this part in particular, not loads!). There are loads of good packages out there with tutorials.

For parts of this type, a parametric, mechanical CAD package such as Solidworks of Autodesk Fusion will work best. I used FreeCAD, because I'm a Linux user. Either way, the approach should be the same. For this part, there is a prism at the core, with two extrusions.

I created a prism in the middle, put a sketch on each opposing face, and extruded them to create the elements of the part. I then added the clip at the bottom as a separate extrusion.

After adding a few chamfers and holes, I was good to go! I've added the FreeCAD file to this instructable in case anyone is curious, or it's useful in any way.

Before you can print the part, you need to export it to an stl format. This should be pretty straight forward for the vast majority of CAD packages. The stl is effectively a mesh, which is a bunch of polygons glued together. To print, it needs to be closed – exporting a solid part from CAD should give you a closed stl.

Armed with the stl, you can either print at home, or use a service. I used 3D hubs. For the uninitiated, it's basically a network of people with printers who accept 3D printing jobs.

At this stage it's important to think about how the part will be used. I foresaw some bending along the Z-axis (on the pin), so I made a point to tell the printer not to layer (https://www.3dhubs.com/knowledge-base/how-does-part-orientation-affect-3d-print) the print along the Z axis. The printer managed to convince me to print along the Z axis, with a thicker skin (https://3dprinting.com/tips-tricks/how-to-choose-an-infill-for-your-3d-prints/).

You also need to think about the material. I went for the cheapest and easiest – PLA.

I had half expected the part to fail, and fail it did. Both the clip and the pin popped off. I also realised (doh!) that I should have had the mirror part of the original done, because it was intended to slot into the opposite end of the parcel shelf. I guess it was a good thing I had it done in the cheapest possible material after all!

I revised my design so that the clip was another milimetre in thickness, and the mirror of the original. Then I tried again with another printer.

This time I insisted that

the layering not be along the Z axis, and that the material not be PLA. I printed the part with layering at a 45 degree angle to the XY plane, and in PETG, which is stronger than PLA.

Finally success! I don't know how many cycles this part will last, but I will leave a comment when it fails with an estimate.