AeroPod, PowerPod and Newton

I have the opportunity to "play around" with an iAero in conjunction with a CinQo for a bit and I was curious as to how the iAero stores the power samples. As I understand it, the CinQo transmits at a 4hz frequency the average of the power completed in a pedal revolution, with the data value transmitted in each "packet" updated upon the completion of each revolution.

So...with the iAero set to record at a 1s sampling rate, what value is stored every second? Is it an average of the 4 data points transmitted during that second, or is it just a "downsample" of whatever value is being transmitted at the sample time?

Also, what happens if the sample time is increased to longer intervals?

Thanks!

Tom_Anhalt wrote:As I understand it, the CinQo transmits at a 4hz frequency the average of the power completed in a pedal revolution, with the data value transmitted in each "packet" updated upon the completion of each revolution.

Hi Tom... so are you saying that every 1/4 sec the CinQo transmits the average power for the last complete revolution? If so, it seems like the average of the 4 would make sense since that is pretty straightforward... but I've no idea what actually happens.

rruff wrote:

Tom_Anhalt wrote:As I understand it, the CinQo transmits at a 4hz frequency the average of the power completed in a pedal revolution, with the data value transmitted in each "packet" updated upon the completion of each revolution.

Hi Tom... so are you saying that every 1/4 sec the CinQo transmits the average power for the last complete revolution? If so, it seems like the average of the 4 would make sense since that is pretty straightforward... but I've no idea what actually happens.

Yes...as it was explained to me, the CinQo transmits 4 data "packets" per second continuously. The information in the data packet is updated upon the completion of a pedal cycle. So...if you are cycling at a cadence of 60 rpm, that means that the power value contained in that packet is updated once per second, or after every 4 packets are sent. The head unit would be seeing sets of 4 identical values in a row. At a cadence of 80 rpm, the each pedal revolution is completed in ~.75 seconds, so there would be sets of 3 identical values in a row. Make sense?

The reason I ask is that even though it would make sense to average the packets over the entire sample period since that would be the most accurate method, experience has shown that typically isn't done in power meter head units. For example, even though the PT when set at 1.26s recording is logging the average power over that fixed time period (regardless) of cadence, if you switch the recording rate to 2.52s, it actually only records every other data sample instead of averaging. That's called "downsampling" and it's basically throwing away information. The most notorious example of this is the implementation of the Polar S710/720 power module. In that case, the power module actually transmits power values to the watch at a rate of approx. every 2 secs which are weighted averages of previous integer pedal cycles...but then the watch only stores (at a minimum of 5 second intervals) whatever value happens to be in the display at the sampling time. That's all part of the reason that the Polar tends to "clip" very short term, high power efforts. It also makes the longer sample period options pretty useless for recording situations where power is varying widely.

So...basically, I understand how the data is being transmitted to the iAero, I'm just wondering how that info is being handled in the head unit. After all this could have an effect on how desirable or not it would be to use longer sampling periods.