In order to analyze data from multiple power meters you first have to record the data, which isn’t straight forward because most cycling computers will allow you to connect to only one sensor of each kind.
Possible methods I tried so far:
– Use multiple headunits. Uncomfortable because you need to charge and operate all those headunits, and unefficient because you need to export and synchronize the data by hand. Including left-right balance and other advanced metrics can be cumbersome: the only workflow I found so far is to first import .fit into Golden Cheetah, then export to Golden Cheetah’s JSON format, then convert from JSON to CSV.
Not that I’d put too much importance into left-right balance for which some are quick to emphasize that there is no scientific evidence for the benefit of using balance in training, but, as soon as you start comparing left-right independent power meters like pedal-based meters or the Pioneer pedaling monitor, and see some deviation, you need to dig into left and right power data.
– Use North Pole Engineerings WASP unit and their iOS app. Good if it works and if there are also many other sensors you’d like to record (like multiple Moxys): you just get one large file with synchronized data. Unluckily, the iOS app sometimes swaps data columns (which you then need to restore by hand, figuring out how they swapped), sometimes crashes, seems to get all those power spikes reported from the P1-Edge combination, just much more often. If these are all just issues with their iOS app, writing your own app might be the way to go; just didn’t have enough time so far to do so.
– Use one of the multi-user PC apps like PeriPedal (or PerfPro) that are targeted to spinning classes/gyms. It’s not a portable solution and you get separate files, but the data is at least synchronized. PeriPedal does not support advanced metrics yet.
– Use IpWatts on an Android phone. The app is free, so if you already own an Android phone with built-in ANT+ support, it won’t cost you anything, and there is only one item you need to have charged and ready. Records power and balance data from up to 4 power meters. It’s actually made of two separate apps: IpWatts and IpSensorMan, which captures the ANT+ data and can relay that data to compatible apps, making it possible to run multiple apps using ANT+ in parallel. I tried it a few times so far and it seems reliable.
Interestingly, it let’s you distinguish between no data and 0 W by having a separate data column for a “data/no data” flag. Percentage data is I believe received separately from power and may therefore drop separately; the app’s description says it simply uses the last value in case of bad reception but my data shows that there are also cases where it records as 127 (for “no percentage data available”, the value used in case of a downstream power sensor like KICKR or a power hub) or 0. So, this app seems to (partially) address the issue of most head units of mixing up no data and 0 W data. On the negative side, in order to benefit from this, one needs to implement data analysis differently from the usual workflow.
Other minor incoveniences I found so far is that operation isn’t totally smooth because of the need to run those two apps separately, connecting to all sensors can take some time (or I just haven’t found out yet how to do it correctly), visual appearance isn’t sophisticated (but heck it’s a free app), and you need to fish the recorded data from the Android file system and copy/send by hand.
Two possible future improvements I’d actually be willing to pay for:
– Cadence is now recorded from only one sensor. Accuracy of upstream power sensors (btw that’s I believe a term Alex Simmons’ created and includes pedal-based, crank-based and spider-based power meters) depends heavily on correct cadence, so, for comparing such power meters, you’d want to record their cadence data separately.
– I might be wrong, but the app display seems to refresh more often than at 1Hz, so, it might be that IpSensorMan captures and relays all the power meter data that usually arrives at about 4Hz. So, it might not be too difficult to extend the app to actually record all that data instead of averaging and reducing to 1Hz, which would really make it a stand-out feature.
Ipwatts sample data with corresponding Garmin Edge data: https://www.dropbox.com/s/47ugy9aqvofbsue/20150925%20IpWatts%20Edge%20comparison.xlsx?dl=0