Monthly Archives: November 2013

Rotor Power Meter – First Experiences

Rotor are a newcomer to the power meter scene. However the pedigree is fantastic – they make fantastic cranks, and the power Meter technology comes from a leading manufacturer of laboratory level power measuring equipment based in Italy. Alongside that, the price is reasonable and the accuracy is promised to be very good (<2%).

So, the order was placed in December 2012 and the waiting game began. Due to the demand it was not until June 2013 that the crank was ready for me. Due to other commitments I wasn’t really able to fit it until late July.

The cranks came well packaged in a nice glossy box as you’d expect:
Rotor Power Cranks in box

And they’re easy enough to put together:
Rotor Cranks assembled

After a few months’ testing some issues became apparent:

  • The cadence is only accurate at a very narrow range.
  • The signal strength is limited compared to other ANT+ devices and is prone to interference.
  • The power measurement is very consistent given a 3 second average, but instantaneous readings are jumpy and can be confusing.

I think the main cause of the inaccuracies is the cadence measurement. There are no magnets so it measures the peak-peak measurement and then estimates the measurement based on that.You can confirm this by pushing the pedals rhythmically whilst stationary and it will start reporting cadence and power measurements.

Due to this style of measurement the cadence is only predictable given two conditions:

  1. The riding surface is smooth, and
  2. The pedalling action is not smooth.

I found that the measurement was most reliable at high torque / low cadence conditions. And it was least reliable at high cadence / low torque conditions. Going over roads that contained segments or joints seemed to report jumps in cadence and power, though the 3 second average was pretty steady.

To address this and other issues, rotor released a software update, SW8 or SW08 depending who you talk to. I was very excited about this update but be warned: the update is done wirelessly! I ran the update and the crankset stopped responding at about 13%. After that it was completely unresponsive and had to be sent back to be reset.

After being returned it was responsive but was not returning any torque readings. I tried reinstalling the update, this time I taped the ant dongle to the crank, wrapped the whole lot in aluminium foil and it still failed… but was responsive this time. 5 update attempts later it succeeded, but still the same issue occured and it needed to be sent back. Oh well.

Rotor Power Meter wrapped in foil
Rotor Power Meter wrapped in foil

I have reflected on this a little. The update worked successfully after I closed all other applications and stopped all services, including a webcam and web server process that was consuming abut 50% CPU. I believe that the update process has no error or sync checking, so if a packet comes too early, too late, or is missed, the process fails.

So, if you are updating your cranks:

  • Have a fast computer ready
  • Ensure there is no interference
  • Close all other applications on your computer
  • Shake the cranks every now and then to keep them awake

And here we are, 3 weeks of waiting and no answer on when the cranksets will return. I can only hope it’s soon.