One of the most significant obstacles I can see in testing the relative aerodynamic performance between the two frames is to make sure your body position is constant when doing trials for each frame. Small differences in how you are positioned on the two frames could be more significant then the differences in the frames. If you are simply wanting the relative difference between the two frames including the effects on how your body fits on the bike then that would be much easier.
Here is my suggested routine for testing the relative difference including the effects on your body position from the difference in frames.
find a 1/2 to 1 mile straight stretch of road with 100m acceleration zones.
pick a set of wheels and power measurement of choice.
go to planned stretch of road and warm up the tires for 10 min or so.
pick a acceleration power and steady state riding power.
accelerate through your acceleration zone and at the boundary of the beginning of the steady riding zone stop pedaling momentarily and then continue pedaling at the planned steady state power. When you reach the end of your steady state zone let off the power abruptly. perform a U-turn at the end of the acceleration zone at this end of your course start the process over. Complete two loops of your course.
then switch to your other bike and repeat the process,
note use the same wheels and tires from the other bike by swapping them out.
do this procedure several times in a day and on more than one occasion with also taking not of the weather conditions that day.
If you do more than one on the same day take a little break in between each pair of trials. Gather 10+ trials of each frame.
Don't forgot to always zero your power meter befor each trial
For data analysis
for each pair of trials you did for the frames do the following.
take the time average power and velocity of each trial after trimming out the segments of steady state riding.
take the ratio of the time averaged velocity to time average power for each frame so you have the mph/watt for each frame
take the ratio of the (mph/watt) ratio's for the two frames for the trial pairs.
do this for each pair of trials, for example you start with frame A do a trial then do a trial with frame B that is one pair.
Then average all the ratios to get a more certain result. Take the standard deviation of the mean to get and idea of the uncertainty in the average.
How to interpret this.
If you take the ratio of the (mph/watt) ratio for frame A to B then frame A will be performing better if the ratio is >1 if the ratio is < 1 A is not performing as well as frame B. The average of all these ratios gives a more certain result.
Last Edit: 1 year, 9 months ago by howardjd. Reason: grammar
I think however you choose to get your "response" metric you'll have to use a rule of thumb in order to determin approximate number of trials necessary to be able to have the experimental power to reliably detect the amount you'd like to detect (below is for p<0.05):
there will be some work just to understand what the stdDev of your method is...then, you'll have to state what difference in frames you'd like to be able to reliably detect.
In the tunnel I've used historically, I'd suggest doing this set of 12 runs:
beta =[0 10];
drop = [0 -2 -4];
reach = [0 8];
...for each of the frames. this ought to give you the statistical power to detect a CxA difference of ~0.0025 with a p value of around 0.05 (about 80% of the time). more importantly, though, it has a false positive rate of 0%, and a reversal rate of about 0.1 %. This set of runs would take about 90-100 minutes in a tunnel and on top of that, you'll get real good heuristics for reach and drop (and insight into any interactions).
If you want to field test this in a velodrome, I'd suggest doing this matrix ~6 times depending on how good the repeats look (you can't do the beta runs in the velodrome, so you'll need to double up the runs to compensate) in order to give equivalent stat power. Oh dear, that sounds like a piece of work! Based on the field testing I've done, that test would take a whole heckuvalot longer than 2 hours.
Using an Ibike does not improve field testing std dev in my experience.
Last Edit: 1 year, 8 months ago by kraig. Reason: tidy'd up a few numbers.