Metaluna

I'm thinking about getting a Wahoo Elemnt cycling GPS to share among 2-3 bicycles. Normally this would require getting separate speed and cadence sensors for each bike (or just doing without like I am now with my Garmin Oregon) however it appears that the Wahoo cadence sensor can be attached to shoes as well. This seems like a no-brainer way of saving $40 per bike (after the first one), since the sensor travels with you as long as you always wear the same shoes.

Is there any drawback to this approach? The only thing I can think of is that the sensor might give spurious readings if you hop off the bike and are walking around in range of your GPS or phone, maybe messing up your average cadence stats. I don't know how well the Wahoo handles this situation or if it's even an issue. Any opinions/experiences? How does the Elemnt handle multiple bike profiles? Maybe it will get confused if it sees the same sensor associated with more than one bike? Or is that a function of your tracking software (I use RideWithGPS and sometimes Strava, where I have to manually tell it which bike I'm on when I upload a track)

TimothyH

Are you talking about Wahoo RPM sensors?

Wahoo says that the RPM cadence sensor should be placed as close to the center of revolution, toward the bottom bracket if at all possible.

I'm not sure how this jives with placing it on the shoe but I'd love to hear your results.


-Tim-

Metaluna

Yes I was referring to the RPM sensor(s) that use accelerometers (I assume) rather than the Blue SC that I think is magnet-based.

The setup instructions I found on the Wahoo site don't seem to mention anything about placing the sensor close to the axis of rotation. They only say to put it on the inside of the NDS crank arm. The sensor does move in a different way when it's on a shoe vs a crank arm, so I wonder if that rule would apply in both cases. I think the force vectors the accelerometer sees would be quite different between the two options.

Instructions | Wahoo RPM Cadence Sensor

TimothyH

The difference in the way it moves on the crank vs shoe was my concern as well. This was in the instructions linked above...

I also found this...

It is clear that they want the sensor oriented a certain way.

A photo of how mine is mounted on my fixed gear bike is below. There simply isn't clearance between the chainstay and crank arm. It works fine.

Otherwise, all I can offer is generic comments about the sensors.

One reason I chose the RPM was because they do both Bluetooth and ANT+. My Polar V650 supports only Bluetooth and my Garmin Edge 820 only supports ANT+. Six sensors on three bikes paired with both the Polar and Garmin GPS and work well.


-Tim-

TimothyH

If I can find the clip for shoe mounting I might be willing to try it for you.


-Tim-

Stick69

I couldn't get it to work well on my shoe. The holder wouldn't fit on the Velcro straps only the ratchet straps. Apparently that wasn't quite the right angle.

Metaluna

Interesting. They may only be using a 1 or 2-axis accelerometer, which would be less sensitive to centripetal force in certain orientations. Either that or it just makes it easier for their filtering algorithm to "find" the rotation if you start out knowing the sensor can only be in a few possible orientations. I believe moving it closer to the axis of rotation would create a greater centripetal acceleration since the sensor has to be "pulled" into a tighter arc for a given angular velocity, further elevating that force component above the noise (also, as you say, it gives more chainstay clearance).

I imagine with shoe mounting, the sensor is going to wobble a lot more through its arc due to the foot being free to tilt in several directions, plus the radius is going to vary throughout the pedal stroke since the top of your foot is going to be a couple of inches farther away from the bottom bracket at the top of the stroke than it is at the bottom, so the acceleration changes. Anyway, I'm sure I'm way overthinking this. It probably works fine in most real world scenarios, but it does seem like shoe mounting in theory would be the less reliable of the two mounting methods. I should have the sensor in my hands this weekend (unfortunately the Elemnt itself is backordered ) so I might be able to play around with it a bit via their iPhone app.

wphamilton

For just cadence I don't see much downside to it. Technically they'd just calculate the magnitude of the acceleration which would look periodic with repeated maximum and minimums, and the frequency of that would be the cadence. Smooth it with a running average over several cycles and it would be accurate.