Nautilus 608 revised connector by brasshopper

Last crawled date: 3 years ago

I thought that the Nautilus gear spinners were great, math in hand (logarithmic spiral, and what should be a Fibonacci sequence for the shell partition spacing although maybe it isn't) and they printed well. But when I handed one to my wife she put it down after a second - the hardware I used stuck out of the thing and irritated her hands - it was sharp everywhere, it seemed. She suggested I redesign the connector and so I did. I do most of my design work, such as it is, in openSCAD. This was no exception. I built disks thick enough to allow the hardware to be completely countersunk. I made everything parametric. I have uploaded the stl files - but there is one extra - it seemed that two gears just were not enough - so I included stl files for a bar that will connect four gears. You can use the scad file to build on that if you want to, I have sort of wondered if a hexagon would close properly. Once I got the two bars as I wanted them it was easy to add more since it was just a matter of calling routines I had already written.
To give credit where credit is due, the nautilus shell objects are completely taken from the object I remixed from, without change. The only change I made was to the connectors. The shells have always printed well, and on my printer, the largest one needs a little heat to let me insert the bearing. If the bearing does not fit fairly easily, do not try to force it. The bearing retaining wall is thin and it will crack. Use a little heat from a hot air gun or even your heated bed, cover the object with a metal or glass cup and heat it until it is just sligltly plastic (but too much heat will cause the object to be deformed). You can do that if you don't have anything else, heat just until you are able to insert the bearing. It should be firm but not forced.
In the photo, you can see a double and a single bar in PLA and the yellow single bar is printed in PETG. I have yet to try one in ABS, but the tolerances are critical so I worry about shrinkage.
Edit: OK, I couldn't help but thinking that many of the other mixes of the original Nautilus used split bars so that you could more easily see the gears mesh. And it was not hard to create such a bar in openSCAD. In fact, I made a square bar and the same code worked for an oval bar (probably not strictly elliptical, although it is regular.
I have included a sample set of STLs for the styles. All of the styles can be generated by the new openSCAD I downloaded and you can probably even generate a four position bar if the values for the space between the slit bars is not too high.