Thanks for all the helpful discussion. And I guess this brings up a 'low level' issue that had not occurred to me before. I am guessing that most folks don't have a bike stand set up permanently so they just go in the garage to set up a new computer or GPS on their bike. This requires that you spin the rear wheel and cranks to 'wake up the sensor'.

So the question is 'how long does the sensor stay awake' - or how long to do have to figure out which buttons to push before you have to spin again?

The best answer is probably to spin them often while the pairing is being done. But it is a valid question, I would think.

From all that I can tell my Bontrager Trip 300 will 'stay awake' more than 10 minutes. I judge this by assuming that a reading of '0' means that it got data representing zero. A reading of a 'dash' means no data.

But there is an interesting characteristic here. When go out to my bike first thing in the morning, move nothing, and wake up the head, the head will try to connect. After 15 or so seconds it will go into ride mode showing no input (dashes for speed and cadence). If I wake up only one sensor (spin the cranks, not the wheel) both sensors appear to now be awake. I say that because I will get a zero reading for both speed and cadence.

I was not expecting this. But since the speed/cadence sensor is a single physical unit, it would appear to me that there is (most likely) only one sender that can send speed or cadence information (but not both simultaneously). If there are separate senders for speed and cadence the head must know that it is not connected to one of them, but for some reason chooses to interpret not connected as zero.

Or maybe there is a device profile for a dual device where a single connection can contain either or both speed/cadence info. Just curious if anyone knows. I do know that my Bontrager updates speed more often than cadence - or that is what it seems like. Maybe that is not exactly the case since you can go from very high to zero cadence in no time. I am hoping to avoid that experiment in the speed domain

dave

ps. Now that I think about it there is another option. There could be two senders but, in an integrated unit, waking up one wakes up the other.