Real World Aero Testing via Chung Method - Data Thread

[deek]

What do you mean by "wind-slope bias"?

[RyanH]

GSC10 Run 12: Cervelo R5ca* on 404s with 23mm Veloflex Criteriums
Recording Device: SRM PC8
Speed Sensor: GSC-10
Tire Pressure: 90F/100R
Beginning Temp: 65*F
Ending Temp: 69*F
Wind: 0 mph
Avg HR: 154 bpm
Avg Speed: 22.4 mph
Avg Power: 247 watts
Distance Covered: 15.662
Duration: 42:00
Estimated CdA: 0.3242
Quality Rating (out of 3): 2 (average)

Ride Data (Strava)
Google Sheets Aero Data

Notes: Rapha 2015 Pro Team Jersey

Test Setup:


Chung Method Aero Analysis



Atmospherics (beginning of session)


Comments:
Well, this was frustrating. Lesson learned, don't test on Bruin game days. Bunch of nit-wit drivers. I had the perfect run going, no wind, and then lap 4 some lady blocked all traffic arguing with parking. I had to hit my brakes. Power wise, I was running these laps a little hot to test my guesstimated difference between the R5ca and Foil.

* This session's power data came from an SRM vs the other bikes are using the same Quarq.

[RyanH]

@WMW, I agree but I think there's some value in seeing the variability introduced by multiple days. If we try to control everything, we're back to wind tunnel testing. Wind slope bias, if I understand what you're saying, is a big problem, especially at the Rose Bowl. I had to abandon two runs so far due to as little as 1-3 mph N winds which were slowing the first half of the lap by 12 seconds (5%). I'm not too keen on driving over to a place and testing, that's mileage I'm missing out on and I'd be sitting in LA rush hour traffic.


Yesterday, I was comparing no wind data points and came up with this:
Bike Avg CdA W@40 W@45
Foil 0.31305 301.6 418.6
R5ca 0.32275 309.8 430.4

Today, I ran my lap average at 7w higher than normal (damn you car!) and I was hitting nearly identical lap times to the Foil's laps from yesterday (8:24's +/- 2s). I think it's finally time to break down the Foil. The Fuji will be back in rotation tomorrow. I got a -17 stem to allow me to match the R5ca so I'll test that and then put on the Enve SES Aero bars later to see how much of a difference they make. I'm thinking most of the difference will be down to 38cm ctc vs 35cm ctc at the hoods.

[WMW]

What do you mean by "wind-slope bias"?

A course where you are going uphill into a headwind more than downhill into a headwind. Or downhill into headwind more than uphill into a headwind.

For out and back you want a fairly symmetrical half-pipe and a straight road. Hills on each end at about the same elevation, where you turn around. A loop course should be fairly flat to avoid wind-slope bias. You don't want or need any hills at all for a loop. Of course if you test when it is completely calm you won't have any wind-slope bias on any course.

[WMW]

Well, this was frustrating. Lesson learned, don't test on Bruin game days. Bunch of nit-wit drivers. I had the perfect run going, no wind, and then lap 4 some lady blocked all traffic arguing with parking. I had to hit my brakes.

You need to test on a road with no traffic at all.

@WMW, I agree but I think there's some value in seeing the variability introduced by multiple days. If we try to control everything, we're back to wind tunnel testing. Wind slope bias, if I understand what you're saying, is a big problem, especially at the Rose Bowl. I had to abandon two runs so far due to as little as 1-3 mph N winds which were slowing the first half of the lap by 12 seconds (5%). I'm not too keen on driving over to a place and testing, that's mileage I'm missing out on and I'd be sitting in LA rush hour traffic.

What's wrong with wind tunnel testing?

There aren't many places where you can find a suitable loop. I use a half-pipe that is ~1km/lap on a quiet residential street. I bet there is something similar near where you live.

[RyanH]

Normally there's little to no traffic at 6:30am when I'm there. Bruins football season just started so I wasn't sure what to expect. Now I know. I think one of the neighborhoods above the Rosebowl would be ideal as there is little to no wind and elevation change is minimal so no wind-slope bias. Downtown, even at 5:40am has 5-10mph winds and I've been scratching my head as to what is suitable in LA that won't be heavy traffic in the morning.

[WMW]

Don't worry about there being a little wind, rather find a course that isn't affected by wind, and has no traffic.

If you want to compare A vs B with any precision, then you really need to do many repeats sequentially and control every variable you can.

[KarlC]

Correct, but in the case of the aero frames with integrated bars, your selection is a lot more limited in optimizing fit/aero position so you have to be really careful with what you choose and if it's going to provide the exact same position as before with the exact same ergonomics.

I'm forming the opinion that integrated bikes (Foil, Madone and Venge) are not as theoretically fast as the S5 in the real world. Fit options with the S5 are much greater and I can run ENVE ses 40cm bars to give me 35cm width at the hoods. I have a pair on order which I'll throw on the Fuji after I get some new baseline runs.

How are you measuring the 35cm width at the hoods ?

[RyanH]

I'm not. That's the stated width from ENVE. I won't have them for a few weeks.

Sent from my SM-G930V using Tapatalk

[caddis]

Don't worry about there being a little wind, rather find a course that isn't affected by wind, and has no traffic.

If you want to compare A vs B with any precision, then you really need to do many repeats sequentially and control every variable you can.

Agree with this, I am sorry to say but I think these results are affected by too many variables to be comparable.

Why using so long route with traffic?

Maybe empty parking place or similar "outdoor velodrome" running track etc...

And IMO you have to test different set ups same time to be able to separate the marginal differences..

[kgt]

RyanH tries to test in real conditions and that is his big plus in comparison to in wind tunnels, velodromes etc. At the same time he does do many repeats sequentially and he does control every variable he can. Very few would ever have the patience to do that.

It is expected of course that in real conditions (wind, slopes, other vehicles, uneven tarmac etc. etc.) the pure effect of aerodynamics becomes much more complicated if not immeasurable.

That is what the test proves so far IMO and that is absolutely normal.

[RyanH]

One thing before I start this: trying to find a flat 1km to 3km course with little to no car interference and not much elevation change in Los Angeles is like trying to find a unicorn. I found a 0.88 mile loop to do in South Pasadena in a neighborhood. Only one car drove by me during the 3 hours I was testing and since the entire course is shielded by trees, wind was zero.

I did two runs each at different times of both the 404s and RZRs on the Fuji as well as a run with the Cervelo and 404s to get a point of reference. I was going to do runs with some box section wheels but I forgot to do a rollout so couldn't.

I brought a lot of crap this morning:



Test setup #1: Fuji with 404s



Test Setup #2: Fuji with RZRs



Test Setup #3: Cervelo with 404s

One of the streets as an example of the tree protection from the wind:



Run #1: Fuji with 404s



Run #2: Fuji with RZRs



Run #3: Fuji with 404s



Run #4: Fuji wtih RZRs



Run#5: Cervelo with 404s




So, imo, a no wind day at the Rose Bowl is producing much better data. For the person above that criticized the length I choose (5km), the above is why. If you look at the elevation chart of the Rose Bowl runs they are almost all perfectly uniform.

[RyanH]

GSC10 Run 13: Fuji Altamira on 404s with 23mm Veloflex Criteriums
Recording Device: SRM PC8
Speed Sensor: GSC-10
Tire Pressure: 90F/100R
Beginning Temp: 64*F
Ending Temp: 68*F
Wind: 0 mph
Avg HR: 148 bpm
Avg Speed: 22.0 mph
Avg Power: 239 watts
Distance Covered: 15.635
Duration: 42:43
Estimated CdA: 0.3234
Quality Rating (out of 3): 3

Ride Data (Strava)
Google Sheets Aero Data

Notes: Rapha 2015 Pro Team Jersey

Test Setup:


Chung Method Aero Analysis


Atmospherics (beginning of session)


Comments:
Zero wind, very clean run. Maybe she's not as much of a heffer as expected.

[WMW]

It is expected of course that in real conditions (wind, slopes, other vehicles, uneven tarmac etc. etc.) the pure effect of aerodynamics becomes much more complicated if not immeasurable.

The only reason it is "immeasurable" is because the results are obfuscated by too many shifting variables that aren't measured and accounted for.

[WMW]

One thing before I start this: trying to find a flat 1km to 3km course with little to no car interference and not much elevation change in Los Angeles is like trying to find a unicorn. I found a 0.88 mile loop to do in South Pasadena in a neighborhood. Only one car drove by me during the 3 hours I was testing and since the entire course is shielded by trees, wind was zero.

Sounds like a good course, but you can also do out-back on a quiet road that is book-ended by hills. For most people this is easier to find.