Tom A. wrote:
An oldie, but a goody:
http://www.academia.edu/...mech_1998_14_276-291
Wind tunnel and "on road" derived power requirements match. 'Nuff said.
jackmott wrote:
Also not a PhD, but I have heard one say you have something wrong, but I don't know what, and at first I was thinking the same as you. I think maybe though what is wrong with your statement about error is that it assumes that the measured power is 100% accurate
Also if we are looking at the error of this plot to evaluate the effectiveness of these equations, we also have to consider that the wind tunnel measured CdA used as input into those equations is not 100% accurate either.
No idea where 2.7 watts comes from though.
The question here, really, is "how well do Wind tunnel and "on road" derived power requirements match".
If a guy with a PhD says I have something wrong - well, all I can do is be perfectly transparent and offer up the data I'm working off of. Other than that, not sure what else I can do... :shrug:
The journal article clearly states on page 284 (first paragraph) that the standard error of _measurement_ is 2.7W. I replicate this value with my data set (close enough!) but have to make an assumption about measurement instrument reliability. Reporting standard error of _measurement_ is an odd choice - in the context of how this plot has been presented it is confusing at best and misleading at worst. One of the junior authors that put his name on the article still seems to be confused.
The standard error that excel spits out is closer to 17 watts than it is 2.7 watts, for sure.
=================
Kraig Willett
http://www.biketechreview.com - check out our reduced report pricing
=================
Last edited by:
BikeTechReview: Sep 8, 14 19:36