Simulation
Classified PowerShift: Full Drivetrain Efficiency Analysis
Full drivetrain power-loss model for the Classified PowerShift system, combining chain articulation, bearing, and hub losses to quantify when the chainring-size advantages of the 1x configuration compensate for the hub's reduction-ratio efficiency penalty — validated against hub-level and complete drivetrain test rigs.
The Engineering Question
A previous study established the efficiency of the Classified PowerShift hub in isolation: in the 1:1 ratio it behaves like a conventional high-end rear hub; in the 0.686 reduction ratio, the active planetary stage introduces a measurable but modest efficiency drop — roughly 99.0–99.25% depending on load and speed. Hub efficiency, however, is only one component of the drivetrain-efficiency puzzle.
For a cyclist or engineer, the relevant question is what happens at the complete drivetrain level. The PowerShift hub does not only add a small loss in the reduction ratio — it also changes the operating point of the rest of the drivetrain. For the same total gear ratio, a Classified 1x configuration can use a larger front chainring than a conventional 2x setup in the small ring, which reduces chain tension, bottom bracket reaction forces, and chain articulation losses. These are exactly the mechanisms that dominate drivetrain efficiency in many cycling scenarios.
The Model
The full drivetrain power-loss model combines:
- chain articulation losses,
- roller-tooth contact losses,
- chainline offset losses,
- bottom bracket and rear hub bearing losses,
- and the previously validated PowerShift hub loss model.
The central question the model addresses is not whether the hub in reduction ratio is less efficient than in direct drive — of course it is — but whether the reduction in chain and bearing losses can compensate for the additional hub losses.
Results and Validation
The simulation answer is conditional. In some climbing or high-power scenarios, the additional hub loss in the 0.686 ratio is more than offset by lower chain tension, reduced bearing loads, and a more favourable chainline. In other scenarios — where the hub ratio is not needed or where a conventional drivetrain already operates near its optimal chainline and sprocket range — the efficiency advantage disappears. Drivetrain efficiency is not a single number; it is a function of load, speed, cadence, gear selection, and course profile.
Both model levels were validated experimentally. The hub model was compared against measured PowerShift losses on a dedicated hub-level efficiency test rig. The full drivetrain model was then validated on a second test rig capable of varying chainstay length and chainline offset. Parameter identification was used to fit friction coefficients in the chain and bearing loss models; the fitted values remained within physically expected ranges — a necessary condition for the model to qualify as an engineering tool rather than a curve fit.
The work demonstrates why drivetrain configuration in competitive cycling is increasingly a simulation problem. A 0.6% hub-level efficiency difference can produce a qualitatively different system-level conclusion once chain tension, chainline geometry, and cadence effects are propagated through the complete model.