Originally Posted by
Banchad
I've emailed the manufacturers of the bike I use(Matrix IC7). Their website claims the power meter in the bike is accurate to +/- 1% and never needs recalibration due to it using photocells which I believe is an extremely similar concept to the Ergomo Pro Powermeter that was floating around a few years ago. I'm taking this with a hell of a pinch of salt as the maximum I've ever seen on my Garmin Vector is 1400 ish whereas I'm seeing 2000+ on this static bike. I'd try to compare it on the other static bike they have in the gym but its out of commission for the moment.
This is where being able to download the data will come in handy. If you hit 2000w in one particular file, you will see some 1700, 1800, and 1900w values leading up to that and after that. Also, in other files you will see similarly high values.
My max wattage I've ever recorded was 2,170. I feel confident with that number being that I could regularly clock over 2,000w on any good day.
Originally Posted by
Banchad
I didn't know that about not needing to be accurate at all power ranges to claim a certain accuracy. Interesting.
I did know that about that about Garmin headunits only being able to sample at 1 a second. I'm writing my dissertation on power meters and have had to take this into consideration. Either way it'll be interesting to see what they come back with.
Originally Posted by
JimiMimni
Market is magical! They're claiming it's accurate to that percentage, but NEVER disclose what their calibration procedure is, or what ranges are tested, so it's completely plausible that (especially dynamic!) efforts can slip through the cracks. They aren't lying, they're just not offering full disclosure. A new professor savaged our calibration procedures for equipment in grad school because it was a completely static procedure.
If you're doing actual scientific research, beg, borrow, or steal an SRM. I believe their track-specific model has a 50hz sampling rate. At any rate, it is WELL above the typical ANT+ 1Hz sampling. I hope your dissertation is smooth sailing!
+1
Banchad, sampling rate is VERY important. A lot can happen between two 1-second samples
This is why SRM made the Scientific model which samples at 10Hz (10 samples per second) which is tops in the industry. It's faster than any system that uses Bluetooth can muster being that Bluetooth transmission is capped at 2hz (I think). Unfortunately I don't believe that the "Science Track" is that model. I don't know why they are using the name "Science" here. It's their normal Track power meter that measures at most 2hz (which is awesome and industry-leading, btw), but it's not the older Science model. This one is only $2,600. The Science model was much higher.
SRM PowerMeter SRM Science Track
That being said, they make all of the power meters by hand and will custom make you one. My track PM was custom.
If you don't have the time and or money for a custom power meter, I'd consider using published and accepted cycling power meter data from guys like Martin and build on their data sets.