Some of the wheels are friction fits in which case the low-friction areas are the holes where the shaft goes through the front and back plates. With brass framed clocks with steel shafts, the hole always wears in the direction of the force vector. I would imagine that a thicker shaft in our case (using PLA) would wear less. Steve do you have any quantitative data on frictional losses? Thanks, Andre Routh
I have done a few gear friction tests by chaining together a chain of gears and hanging a small weight on one end to compare relative friction between different gear tooth styles. Most come out nearly the same needing between 10 and 11 grams to start moving. It makes sense that involute and cycloid gears are both highly optimized. Much of the differences can be explained by different preparation when drilling the arbor holes to fit. A few tooth styles were significantly higher due to surface roughness from stringing. I mainly focus on designing gear tooth profiles that will minimize retractions to reduce stringing as much as possible.
As far as friction or wear with the 2mm arbors, I can only guess. The PLA will wear out long before the arbor. A larger arbor should last longer. My oldest clock has been running continuously for over 5 years. The escapement and gears look great. The arbors have stainless steel bushings that were used in my very first clock. There is no wear in the bushings, but I don't have a 5-year test case with simple PLA holes. I do have a few more recent clocks that ran for 6 months with no visible wear. I expect at least 5-10 years of life, if not more. If there is any wear in the frame arbor holes, I would drill out the holes and add stainless steel bushings. This should possibly add an additional 10-20 years of life.
Some of the wheels are friction fits in which case the low-friction areas are the holes where the shaft goes through the front and back plates. With brass framed clocks with steel shafts, the hole always wears in the direction of the force vector. I would imagine that a thicker shaft in our case (using PLA) would wear less. Steve do you have any quantitative data on frictional losses? Thanks, Andre Routh