Has anyone done a Computrainer VS Powertap watt comparison?
 

We just had an indoor TT on computrainers and was wondering if anyone had comparative data between a computrainer and a Powertap.

In comparing a borrowed Powertap VS my RealAxiom trainer show RA's wattage consistantly 60 watts less than on a powertap regardless of wattage. RA's 100 = PT's 160, RA's 400 :rolleyes: = PT's 460, etc.

So has anyone used a PT on a Computrainer?

Yes and my PTPro was about -5w compared to the computrainer. I use a Tacx Imagic trainer at home and had to calibrate it as it read about -40w below my PTP.

Nutz, I was hoping the Computrainer was -20 watts or so. But thanks anyway ;)

The thing about PT and any other power meter is that they will vary from unit to unit. I don't know that it would be much more than -+5 to 10w though.

I did a vo2max and blood lactate test back in september with my bike on a computrainer. I also had
my powertap SL on at the same time. I recall the two being within about 5W.

hmmm thats interesting.....have you ever calibrated it against a blood LT test?

and my questioned answered..........:D

I have never seen any unit-to-unit variation between Powertaps once each head is properly zeroed. I've checked the accuracy of several systems myself and know several others who've done the same. We all had the same experience. All PowerTaps were identical and accurate to the number of significant figures in the torque display (this can be verified by searching through the Wattage list). The only exceptions were with defective torque tubes, but in those cases the reported torque or power was so far off, that it was immediately evident that there was a problem. So while units may vary, the differnce is less than what is being reported and certainly below the 2% accuract limit.

Well I have. My Pro and SL differ about +/-7w.

How do you know what the true value is? Do you have some NIST traceable way of delivering power to the crank?

Are you sure you are not confusing accuracy with precision?

http://en.wikipedia.org/wiki/Accuracy

You don't need a NIST traceable power standard. All you need is a scale with the desired level of accuracy. With it, you can measure a known weight and then hang it from a pedal while holding the rear wheel stationary. From that, you can verify the accuracy of the torque measurement. Power is just derived from torque so if torque is accurate, power will be too.

You say the two units differ by 7 W, but don't give a power at which you see this difference. If it is a constant offset, I have to ask if you're zeroing the units correctly. I know that when I switch from one wheel to another, there will be a constant power offset unless I rezero. If the difference in power isn't constant but is a percentage of total power, have you tried the above procedure to check the accuracy of the torque for the two units?

It was a constant difference and yes I was zeroing the unit correctly. My PTPro was a 2004 (rebuilt in June of '06) unit and my SL is an 06 unit. The test was done by using my Tacx Imagic in a back to back test.

I did on the same day B. My average on the CT was 300watts exactly. My PT said average of 284 and NP of 289. There was one transmission drop for about 20 seconds, but that wouldn't make too much difference.

You beat me by one place, bastard! Good job though, very good effort from you.

I'll look up the data later tonight. I recall Max Testa saying that at certain wattages, the computrainer is not so accurate. I'll copy the details from my notes when I find them.

Thanks for the numbers. Nice job by you as well!