Calibrating it with a friend's powertap is a good idea. I've seen others here remark that the KK equation is within +/- 3% or so, but not a bad idea to confirm if possible. I'm not so worried; speed here should be constant after the unit is warm, unlike the road. So higher speed means higher power.
As for the large jumps in power from one speed to another, that's the way power works. It isn't linear. If you look at the KK equation, you'll see that it has exponents. These are a little steep at higher powers, though.
The general equation for air resistance is: 0.5 * air density* drag coefficient* cross-sectional area* (velocity squared). That velocity-squared term is the one that'll change the most on the road.
Basically when you're training to get faster, you have to increase your power by a lot to go just a bit faster. To increase your speed by, say 5% (20 to 21 mph), you have to increase power by about 10%...to get to 23 mph (15% increase), it's about a third.
200 17.6 225 18.6 250 19.6 275 20.4 300 21.3 325 22 350 22.7 375 23.4 400 24.1 425 24.7 450 25.3 475 25.8 500 26.4 525 26.9 550 27.4
I used Excel and the KK equation to make a little power chart, to give me a rough idea of what I'm doing (Watts vs. mph). This also illustrates how steep the resistance gets at higher levels. I couldn't find the numbers that KK uses for the rider weight, etc (i know the slope is 0.1), but analytic cycling puts 500 W at about 30 mph. They do seem to diverge. Not that you'll probably be spending that much time at 500 W, but the power needed here is probably higher than on the road.
Last edited by tadawdy; 02-23-10 at 06:54 PM.