Principles

Classifieds design principles for an efficient bicycle drivetrain.

The core of CLASSIFIEDs philosophy is to create the most efficient drivetrain solution combined with uncompromised shift performance and increased reliability

EFFICIENCY: IT’S THE WHOLE SYSTEM THAT COUNTS

The core of CLASSIFIEDs philosophy is to create the most efficient drivetrain solution combined with uncompromised shift performance and increased reliability. For the cyclist this means the following characteristics:

  • Having the right gear at hand for optimal cadence;
  • Having instantaneous availability of the right gear when a sudden chance is requested;
  • Having the highest crank-to-wheel-efficiency as possible.

The ultimate bicycle drivetrain should have all of these three characteristics.

On a modern bicycle drivetrains however there are known compromises. Below the major compromises made, either by a 1x or 2x system.

Characteristics1x2x
Optimal gear for cadence
Compromised by large steps between sprockets and/or limited ratio coverage
Not compromised as small steps between sprockets and large ratio coverage is possible
Instantaneous availability of the right gear
Compromised as no bail-out gear is available
Compromised by front derailleur shift performance
High crank-to-wheel-efficiency
Compromised by smaller chainrings and sprockets (10T, 9T)
Compromised by smaller diameter of second chainring


Regarding optimal crank-to-wheel-efficiency it is known that:

  • The chainline should be as straight as possible, to reduce chain friction losses and axial forces on the crank and hub bearings.

    In the figure below it is illustrated that unwanted axial forces are introduced when going to the outer sprockets. The straighter the chain line the better the efficiency of the system.
Classified fig01 wit copy
  • The chainring and cogs should be as large as possible, to reduce chain friction losses and radial forces on the crank and hub bearings and to reduce the polygon effect.

    In the figure below it is illustrated that with the same pedal force the chain force, and also the bearing forces on the crank and hub are 45% (1/0,686) higher with a small chainring (F2) compared to a big chainring (F1). The bigger the chainring the better the efficiency of the system.
Classified fig02 wit copy

Also on the cassette bigger sprockets run better. When having a 9T and/or 10T sprocket on your cassette you generally use smaller sprockets all the time, even when you are not using these extremes.

Classified has found a solution that creates an optimal blend between a 1x and 2x drivetrain system in which all the above compromises can be eliminated by introducing a special reduction gear under the cassette.

Characteristics1x Classified2x
Optimal gear for cadence
Not compromised as small steps between sprockets and large ratio coverage is possible
Not compromised as small steps between sprockets and large ratio coverage is possible
Instantaneous availability of the right gear
Not compromised by a super-fast powershifting hub reduction, replacing front derailleur and giving a bailout gear possibility
Compromised by front derailleur shift performance
High crank-to-wheel-efficiency
Not compromised by one large chainring on the crank, not smaller than 11T sprockets on the cassette and a very efficient hub reduction
Compromised by smaller diameter of second chainring

By using a reduction gear between the cassette and the wheel:

  • Compared to a 1x, the chain line will be improved as the reduction gear creates virtually 45% (1/0,686) larger sprockets and there will be less need to shift to the real larger sprockets on the outside of the cassette.
  • Compared to a 2x, the chain forces will be improved as the reduction gear creates virtually a small chainring. The fact that the 0.7 ratio of the hub still uses a large chainring up front means that we do not increase chain tension by 45% as with a small chainring. This results in lower chain, bottom bracket and hub bearing losses.

LET’S EXPLAIN THE EFFICIENCY OF THE 1XCLASSIFIED SYSTEM IN MORE DETAIL.

“THE MOST EFFICIENT GEARED HUB IN THE WORLD”

First we needed to address where the gearing should be located and which gear topology it should have. These two are obviously also intertwined constraints. 1x-drivetrains typically have a sleek look and aerodynamic benefits on which we didn’t want to compromise. The emergence of one-piece cassettes meant more space under the cassette. This provided the ideal place to house a two speed gearing system and create an overall super sleek package. The idea to develop the most efficient 1x drivetrain with a compact geared hub, with outstanding shift performance was born.

Regarding gear topology it is known that when used in the right configuration planetary gear sets have an extremely high power density and efficiency, when certain design constraints are met. As our design goal was to exploit the higher efficiency of larger chainrings, we want to use the planetary gear set to replace the small chain ring on the crank with a ‘virtual’ one.

This means three things:

  • With the 11t as the smallest sprocket on the cassette, the highest ratio of the gear set should be 1:1. In other words: the planetary gear set can be by-passed, resulting in a rear hub that works as any other rear hub;
  • By using the planetary gear set only to decrease the ratio, we were able to choose the most power dense and efficient gear layout possible. In this layout the outer ring gear is connected to the driver as its input and the carrier with its planet gears forms the output, connected to the hubshell. The sungear is locked to the hub’s axle. In transmission design it is known that this topology results in a higher overall gear efficiency than the individual gearmesh efficiencies, due to input forces being directly transferred, and relative rotations being low;
  • By only using the planetary gear set for a single gear (again, effectively replacing the front derailleur), we were able to design a hub that has no additional friction losses in 1:1 as the not-in-use gears are then locked together and rotate as one piece. This is not possible in multi-speed geared hubs.

A GEARED HUB DESIGN WITH MINIMIZING EFFICIENCY LOSSES IN MIND STILL MEANS EFFICIENCY LOSSES, RIGHT? TRUE.

With our selected planetary gear set topology a theoretical gear efficiency of higher than 99% can be achieved in the 0,7 ratio and 100% gear efficiency in the 1:1 as the gears are then locked and rotate as one piece.

As explained by using the reduction gear between the cassette and the wheel there are advantages on system level compared to a traditional 1x or 2x system:

  • Lower chain losses compared to 2x system by always using the big(ger) chainring on the crankset, as demonstrated by Friction Facts
  • Next, the radial forces from the chain acting on the hub’s bearings and crank bearings are lower when compared with using a smaller chainring up front to create a similar ratio. Thus, in the ‘low gear’ you have less (hub/crank) bearing losses than with a 2x system.
  • Additionally on system level compared with a 1x setup you can now use a straighter chain line when going for the larger sprockets, as the 0.7 ratio creates virtually larger sprocket sizes. This also contributes to a higher drivetrain system efficiency.
Classified fig03 wit copy

Improved chainline by creating virtually bigger sprockets with the CLASSIFIED hub.

CAN THE IMPROVED CHAINLINE COMPENSATE THE VERY SMALL LOSSES IN THE GEARING? YES.

At CLASSIFIED we developed specific measuring equipment to be able to measure the total drivetrain system efficiency from crank to wheel with 1W accuracy. As the overall system efficiency is the only thing that counts, we have measured the efficiency of the hub including the crank, chain, rear derailleur, wheel load and frame with its specific stiffness.

We built up a gravel bike with a Shimano GRX 1x groupset + CLASSIFIED hub with a 42T chainring and a 11-34 cassette.

During the measurement the output speed was kept constant by controlling the input torque on the crank which is measured. In this way it was measured how much input power is needed to ride a certain speed. By comparing the required input power in two different settings with the same output speed and similar overall ratio between crank and wheel the loss difference can be accurately measured.

CHAINRING-SPROCKET-RATIOHUB RATIOTOTAL RATIOCRANK SPEED (RPM)WHEEL SPEED (RPM)REQUIRED INPUT POWER (W)
42/28
1:1
1.5
110
165
46,2
42/19
0.7 ratio (0.686)
1.516
108,8
165
45,9
42/21
1:1
2
113
226
63,3
42/14
0.7 ratio (0.686)
2.058
109,8
226
63,8
42/17
1:1
2.471
109,3
270
78,4
42/12
0.7 ratio (0.686)
2.401
112,5
270
79,0

Measurement results

The above measurement shows that from the middle sprocket and upwards the total drivetrain efficiency is higher in the reduction gear than in the 1:1 having the same total ratio by changing the sprockets. So the better chainline and lower bearing forces improve the system efficiency and compensate the very low loss in the gearing.

In the above ratio comparison the crank and wheel speed are similar, so losses in crank and wheel are comparable. At higher input powers the drivetrain efficiency increases due to the no-load losses that create an offset in the torque loss. This is also explained by Ceramic Speed “Chain efficiency vs Rider output.” With higher loads the efficiency will be higher as well

Knowing the ideal drivetrain is the 1x set up with a big chainring in the front and with as few cassette sprockets as possible (to improve chainline and chain durability), there is still a need for at least 16 gears (based on an average 10% gear step and at least 400% gear spread).

Multiplying the number of cassette sprockets by two without impairing efficiency or shifting performance is the ultimate solution. This is what Classified has made: a drivetrain with 11 sprockets that functions like a 1x22, but with a bail-out gear always available, every time!

Roëll

Written by Roëll Van Druten