Over the past few months I’ve been working with Melissa Dowell to make these lovely Narwhal Skewers and now (after a good deal of prototyping, experimentation, poking, and prodding) they’re finally out. The process that went into these is actually kind of fascinating first because of how difficult it was to hone in on the right way to design these in CAD and secondly because of how many different directions we explored before finally landing on the final design.
The whole project started with some sketches from Melissa. From there we chatted about ideas, industrial design considerations, and manufacturing constraints. I felt that it would be reasonable to fasten a twisted rod into a cast metal handle and we could core out the handle to provide both the fastenings for the rod and minimize weight and material.
Melissa sculpted an oversize polymer clay narwhal that we finessed and modified until it had the desired gesture and proportions. From there, we took it to Numidas Prasarn to get the thing scanned and digitized.
Now, comes the frustrating bit. Ideally there would be a simple process for merging the digitized sculpture with robust CAD data to come up with a well designed product that takes into account all the manufacturing and industrial design considerations without losing the organic quality of the original sculpt. I could not, for the life of me, find that toolchain.
We tried editing the scan in GeoMagic and then taking it to Zbrush for surface cleanup and detailing. The scan data simply played havoc with Zbrush and we scrapped that workflow. Additionally 3DsMax didn’t like the model and refused to boolean the scan data with the CAD for the internals. Although I was able to import the data as an STL to SolidWorks to use for reference, editing anything about the STL was absolutely out of the question. I ended up using the scan as reference data to build the model entirely out of surfaces and then scaling it after the fact. Then, I shelled the model and added the fastenings. Even though it’s infuriating to create organic shapes in SolidWorks, it seems to be worth the time to hack away at it rather than depend on programs that have unclear interfaces between hard geometric data and scans.
Even though the process was difficult and not at all parametric I was able to mash some CAD data with the scan in Blender, and we produced two prototypes using the resulting models. The first one was utterly huge (which is a constant problem when working with dimensions in isolation to any real world reference) but the second printed sandstone model captured what we were looking for: turning the tail into a pleasing loop you can stick your finger through. Clark Sopper over at Gotham Machine manufactured some custom stainless steel skewers for us and we were able to stab some food. Once all the dimensions and fastenings were worked out, I dove into the SolidWorks process described above to crank out the final design.
All in all I’m very happy with the results. The skewers look like a genuine product, ready to add a slice of narwhal delight to anyone’s life.Title photo by Brainwomb. Prototype photos by Melissa Dowell.