Need a bit more oomph! for the hills?

Non-circular chainrings have been available in cycling since the 1890’s. More recently, Shimano’s Biopace disaster has spoiled the market for oval chainrings. The Harmonic (1994) has been re-launched in 2004 under the brand name O.symetric with some important successes in professional cycling. In 2005, the Q-Ring (Rotor) entered the cycling scene. However, non-circular chainwheels have not yet conquered the cycling world. There are many reasons for this: the conservative world of cycling, the suffocating market domination of an important manufacturer (and sponsor) of circular chainrings, the difficult bio-dynamics not understood by the users and last but not least, it is not easy to measure and to prove the advantages of non-circular versus circular. Any reasonable non-circular chainwheel has about 50% chance of being better than the circular shape. The only question is: what is the optimum shape and how large can the difference be? The objective of this paper is to compare different chainring designs. Relying on a mathematical model a biomechanical comparison is made between circular and non-circular chainrings. The results of the study indicate clearly that (Criterion 1) for equal crank power for both, circular and non-circular, the peak joint power loads can be influenced favourably by using non-circular designs. For equal joint moments (Criterion 2) for both, circular and non-circular designs, the model calculates differences in total crank power and differences in peak joint power loads. The results of both criteria are mostly concurrent. The analysis also indicates that shape as well as ovality, but also orientation of the crank relative to the chainring are important parameters for optimum design. It is found that some non-circular shapes are clearly better than other designs. The mathematical model can also be used as a tool for design optimization. Besides the commercial available non-circular chainrings, also some ‘academic’ non-circular profiles are investigated.

See the paper on this here:
http://www.noncircularchainring.be/