When pedalling a Bosch e-bike when its motor/electricity is off, how many watts are lost compared to non-ebike drivetrain?

If it differs, let's assume a Bosch Performance CX and 250 watts input power at 90 rpm.

  • 8
    Remember a downvote by itself doesn't help improve the question. Ebikes are here to stay, and they're a great way to lower some of the barriers-to-entry for cycling, like "its too far" or "there's a steep hill" or "I'm too out of shape" If it is worth downvoting, it's worth commenting to help improve the question or the answer.
    – Criggie
    Jun 21, 2021 at 0:37

3 Answers 3


In theory - none because the motor has a freewheeling clutch that is completely disconnected when it is not providing assistance. That is when you're travelling over 25 km/h (location-dependent) or have not engaged the assistance.

However you're dragging around a heavier frame, with a motor and wiring, and maybe the battery pack too. All this extra mass will take energy to get to speed, and have an increased cross section to affect aero.

Also the tyres will be selected to suit the total weight of bike+rider, so they will not be as light/supple as a non-ebike and will have more loss.

  • Also, you will note that in reality there is nothing such as a true "freewheeling clutch"... There are notable differences here between different brands of ebikes when it comes to how much resistance the drive train gives when the motor is turned off.
    – fgysin
    Jun 24, 2021 at 8:17

The answer by Criggie excellently explains that the motor doesn't add any pedaling resistance. I find the claim true especially on by Bosch Active Line Plus system -- I have never observed any pedaling resistance above 25 km/h.

However, there's a bit of extra weight. Around 5 kilograms to be precise.

If riding at 23 km/h average speed on varying terrain including hills, my power simulator shows that a 15kg bike plus 70kg rider requires 129.6 watts. Add 5 kilograms of weight and the required power increases to 134.3 watts. So about 4.7 watts more power needed. I'd say this is so significant that you don't want to use an e-bike without assist.

I know this isn't 250 watt input power but I find sustained 250 watts unrealistic except for athletes that surely won't use an e-bike.

  • 6
    "I'd say this is so significant that you don't want to use an e-bike without assist." Can you elaborate on this, and explain it in a way that's easier to understand for those of us who have never ridden an e-bike? 129.6 to 134.3 is a 3.6% increase, which is very small. Does it feel much harder to ride an e-bike with no assist than a regular bike?
    – Szabolcs
    Jun 21, 2021 at 11:15
  • 2
    @Szabolcs has a point; one measure to gauge the loss would be to calculate the speed increase could the additional power be used for thrust instead of friction. This simulator seems to give me 15.68 mph or 25.2 km/h for 100W, and 15.85 mph or 25.5 km/h with 103W. Jun 21, 2021 at 22:30
  • The significance (or not) of extra weight and resulting 4.7W increase will largely depend on the rider. For a small untrained female with a 90W FTP it will be very noticeable, for a large trained male with a 400W FTP not really noticeable at all. Personally I would say it's insignificant, but I could understand people wanting to argue otherwise
    – Andy P
    Jun 22, 2021 at 11:14
  • @AndyP short of athletes who wouldn't be using an e-bike anyway, It's probably more realistic to assume both the 90W and 400W rider would produce the same power they always did and not (or barely) notice the fraction of a km/h reduction on their average speed. That's well within the fluctuations you can expect from unfavourable weather or traffic anyway. Jun 22, 2021 at 12:38
  • Also depends a great deal on environment: living somewhere with inclines makes weight a lot more noticeable (though granted, these are the environments where the benefits to an ebike are largest). Jun 22, 2021 at 21:41

I think the two answers are not strictly correct as the 2nd generation Bosch motors use a gear ratio between the pedal input and sprocket output that is something like 2.5*; hence the small sprocket on these systems. There is inherent loss within these systems but to measure it in the real world you would want to use a calibrated pedal wattmeter (or crank) and a rear hub powermeter like the Powertap to see what has been lost in translation.

It may be that the newer system with the large sprocket is directly connected to the pedals and the motor is fully disconnected, but not everyone rides this. Also, we can say that the losses are "minimal" but they still exist in any transmission component.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.