RChung

Well, it's really an empirical issue of the size of the effect you're trying to measure, the variability in your air speed, and the change in CdA with yaw. Even if the wind were dead consistent, unless you rode your laps at exactly the same speeds in just the same spots, the amount of time you'd spend at each non-zero yaw angle would differ. That means the VE profile would differ; if the wind were slight (meaning your air speed changes were small) and the change in position or equipment were large, you might be able to get away with it. If not, you might not be able to tell. That's what I mean by "an empirical issue." The good thing about VE is that small things like that are discernible so you can try it and if it doesn't work you'll be able to see that the VE profiles don't (won't) match up. After a while you'll get the hang of when the wind is going to be bigger than the effect you're trying to measure. You might be able to determine differences between two frames with a little wind but you'd probably need absolutely calm conditions to determine differences between two front brakes.

Me too.

RChung

Well, you can go out front, above, or to the side. At "usual" TT speeds you need to be roughly as far out as the front of the front wheel and above it. Alternatively, you can be out to the side or up above the highest point of your head about 50 cm.

waterrockets 

Well, I guess that shows how long it's been since I've hit the power group. Chung on a Stick is pretty much what I was talking about, but didn't know about it.

Nate552

Agreed with those spots, but in front seems like the worst spot. I would prefer to keep the air on front of me as clean as possible.

Racer Ex 

On a less CdA note, One of the benefits of the IAB over bent arms is that you (well, me anyway), can sustain the position much longer, and you're holding far fewer muscles in tension so there's less muscular 02 demand.

I'd speculate that testing would show variability in which position might be more effective for each individual based on their build. It seems that taller, lankier riders do better with their arms apart (see Mini-Phinney's pursuit position) while smaller riders do better "scrunched" (IAB).

My own IAB shot.

rruff

Another option for that scenario, is to map the elevation of your course using either an iBike, or VE on a calm day, then calculate Virtual Wind. It would be more complicated and noisy, but you will gain some insights into how drag changes with yaw.

IMO you'd have to be really wealthy to jump into the wind tunnel before spending a few hours in field testing. From what I've seen, VE is nearly as good as most tunnels if you do it right... and if you have a PM it only costs time.

spinwax

WR, Ex and others doing the coast down testing. How many runs in each position/component change/bike adj/etc do you feel are satisfactory to eliminate so many outside variables?

Sorry if it was posted and I missed it.

waterrockets 

Yep, that's a really good point on comfort, and matches my experience. I think for a 1-2 minute explosive effort, I'll stick with deep drops, and for repeated or longer efforts, IAB.

The short/tall thing is perplexing. I don't anywhere near as compact as you, even though I've got a similar relative hip/shoulder height. I'm 1cm lower than this now, and feeling completely comfortable with it, but it's not really much lower.

waterrockets 

Based on my first couple tries at it, I'd say 10-15 runs would start to show repeatability if you wanted to test things like a helmet.

RChung

I've done this on two courses that I use. It's not quite as nice as having actual wind measurements for every test but having a solid baseline for comparison is pretty handy.

kleinboogie

Thanks for the video of the wind tunnel and smoke. Sure shows how a disc wheel and keeping your helmet tail down can improve airflow. And how other things really jack it up.

RChung

Depends on your protocol, the venue you're using, and how you're collecting data. The main coastdown measures are max speed (aka terminal velocity), rollout distance, and moment-by-moment speed. The smaller the change you're trying to detect and the lower the speed that you're testing at, the more precise you need the speed recording to be. So if you're doing coastdowns on a relatively shallow slope (or on a flat surface) you're going to need really precise speed recording in order to spot a difference. If you're coasting down a long steep hill, you might be able to get away with terminal velocity. I haven't had much luck with rollout distance -- small unevennesses in the road near the end of rollout make the rollout distances too discrete.

Racer Ex 

On my test course I usually do three runs per set up. It's a very well protected short and steep hill, and I pick my day and time to minimize variables. I wear my full aero gear, use my TT wheels, and am pretty diligent about keeping my head and knee position the same. I've got multiple sets at exactly the same speed to .2 km/hr. I don't pee or drink during the tests, and generally try to wrap things up in less than 2 hours to keep the temperature change to a minimum. The key here is stability. And as noted by Mr. Chung, more speed is good in these test, my ballpark is 52-54 km/hr.

As far as using roll out distance, I gave that up years ago. I think rolling resistance issue start coming into play in this equation, and certainly wind is going to affect things. there might be a really effective protocol, but my tiny brain can't come up with it.

spinwax

Thanks guys. Ex, Waterrockets was using the term "coast down" in his original post, so I just didn't want to cause confusion. Just curious to see what protocol you guys were using.

Interesting stuff.

Racer Ex 

It's my new cause. I'm old. I gotta have something to complain about.

SpongeDad

How short and steep? Just curious as to the point at which you think you're hitting terminal velocity. The longer hills around me aren't straight with consistent slope and the straighter ones aren't long.

Nate552

When I did terminal velocity testing I used a hill that was 9-11 % and about half a mile long. I would hit ~45 mph during tests. Oh, and Ex recommended to me to change my PT CPU to record kph for more accurate speeds.

ridethecliche

Food at the nursing home?

rruff

Old guys don't get to complain (about anything) if their FTP is 325W+...

Racer Ex 

Then Mr. Chung noticed that SRM does .2 km/hr increments (not .1 km/hr). If the PT does .1 that's the bomb.

My hill is around 11% and around 100 yards. I roll the start so I hit TV a bit over 1/2 way down.

Racer Ex 

I seem to be choosing curmudgeon over bike racer this year.

Nate552

PT does .1 increments. Some samples from a test file in kph
47.1 47.1 47.1 47.2 47.2 47.2 47.4 47.7 47.8

johnybutts

I'm going to go ahead and bump this because I'm planning some of these terminal velocity tests in the very near future.

RChung

It's like deja vu all over again.

tomato coupe

Thanks for the nice write up, however, you violated one of the cardinal rules of scientific or technical writing -- you used an acronym (IAB) without defining it. (Internal Affairs Bureau?) I'm sure its meaning is obvious to a lot of readers but, for me, the value of your write up is greatly diminished because I don't know what you're referring to. Cheers.