Real World Aero Testing via Chung Method - Data Thread

[KWalker]

Do you think this is more valuable for race results than doing rides that might have a stronger training benefit? Or do you get to do this in addition to that?

[KWalker]

Do you think this is more valuable for race results than doing rides that might have a stronger training benefit? Or do you get to do this in addition to that?

And the above poster is correct- your specific Fuji with your position and dimensions suggests that it is as aero of a setup for you to ride compared to the others tested, but cannot be traced directly to the frame itself. Your position and interaction with the frame likely has more of an impact.

And no offense, but the cda numbers you're posting up really are not that low for a road bike, quite on the higher side of what I've seen testing myself and others on various road and TT setups so I really can't imagine that the frame is your limiter considering it's impact on the system. There are even some (then cat3s and 4s) I helped with Chung testing a while back who had some pretty unpredictably impressive figures that more than compensated for their non-aero frames (both oddly on Cannondales) and are likely what helped them obtain success in events you wouldn't think they were built for.

I'd wager some mobility work and perhaps a better bike fit (even if some name-droppable person has fit you) would be much more useful. I'd be more curious as to the effects of those changes then the frames actually. But if the testing is fun, keep nerding out on it because this should be fun.

[RyanH]

My biggest issue is getting motivated to ride early in the morning. Training in the afternoon is hit or miss since I can be zapped by work and zero desire to push myself on the bike. Through fall and winter, I'll begin the same training plan as last year which will overlap with the testing. I think Chung method model should be flexible enough to accommodate non sprinting intervals.

Kwalker, curious what CdA numbers you have seen with people riding on the hoods? I thought the low 0.3's was a fairly good number considering non optimized TT positions begin around 0.27. Specialized did their WIN tunnel testing on positions and the faux aero was 0.27 and drops were 0.32 so me posting up 0.31 to 0.32 on the hoods, not sure how you're coming to the conclusion I'm giving up a lot on position. I am going to test a longer stem and see what the effects are as well as see if I can cope with the extra length.

I do agree with others that so far the results are not very comparable as my position is changing between bikes. Even slight changes are showing notable differences. That's the big take away so far. It's a learning process.

[kgt]

The outcome of the test so far is pretty clear IMO: a more aero frame is not translated as a faster one. The result may be rather disappointing for some people but that's it.

[cyclespeed]

The outcome of the test so far is pretty clear IMO: a more aero frame is not translated as a faster one. The result may be rather disappointing for some people but that's it.

Or at least that the variables are so tiny, that an elbow a bit to the left, a small gust of wind, a passing car, etc. is enough to disguise them.

[RyanH]

Yes and no, the CdA differences can be averaged over many runs to hopefully minimize the noise.

The differences aren't that nuanced, the lap times of the Scott were I think 5 to 6 seconds faster per 8.5 minute lap than the Cervelo. Those lap times tended to be fairly consistent.

[cyclespeed]

Am I right in saying that most of your testing covers over 20 miles? At speeds of approx 21 / 22 mph? (35km/h).

The drag force at 43km/h is more than 50% more than at your speeds. If you could ride at this speed, even for shorter distances, it would flag up aero differences much easier.

Now riding at 43km/h is a tall order for any rider, so unrealistic for extended testing, but basically what I'm saying is, is it possible to make your tests shorter, but faster?

[justkeepedaling]

The outcome of the test so far is pretty clear IMO: a more aero frame is not translated as a faster one. The result may be rather disappointing for some people but that's it.

All else equal it is. When you throw in different groups, seatpost, stem, drops, cranks, and even position, no one can tell what the heck is going on. I guarantee that that doesn't stand as the scientific method for any journal

[KWalker]

My biggest issue is getting motivated to ride early in the morning. Training in the afternoon is hit or miss since I can be zapped by work and zero desire to push myself on the bike. Through fall and winter, I'll begin the same training plan as last year which will overlap with the testing. I think Chung method model should be flexible enough to accommodate non sprinting intervals.

Kwalker, curious what CdA numbers you have seen with people riding on the hoods? I thought the low 0.3's was a fairly good number considering non optimized TT positions begin around 0.27. Specialized did their WIN tunnel testing on positions and the faux aero was 0.27 and drops were 0.32 so me posting up 0.31 to 0.32 on the hoods, not sure how you're coming to the conclusion I'm giving up a lot on position. I am going to test a longer stem and see what the effects are as well as see if I can cope with the extra length.

I do agree with others that so far the results are not very comparable as my position is changing between bikes. Even slight changes are showing notable differences. That's the big take away so far. It's a learning process.

I have seen quite a few people with road positions around .25-.3. Mine was .26 last time I tested, the .25 might have been due to some testing protocol error, but that same rider is .24 on his TT rig! It might be fairly good for dudes on the interwebs who tend to document their testing publicly. .3 was for a rider who clearly had a high/ugly position. Many of the low numbers were final- we adjusted the bike fit for power and efficiency. In most cases the seat height was slightly lower than predicted by just fitting and riding on a trainer, reach was a tad longer and bars almost always a bit higher. There is a ton of room for road bike improvements through fit, but that alone is going to result in MUCH higher reductions than this frame testing stuff.

From what I've seen, allowing more forward pelvic rotation usually aligns the spine and drops the shoulders and head typically. Optimizing spinal extension and pelvic rotation can dramatically reduce rider height and usually a better bar/hood position can further allow the rider to drop their head and bring their elbows in closer.

Considering you said you were dropped at Winters on the climb, I'd look internally for more motivation to train since you could hide in the pack with whatever aero and sprint/attack near the end, which is how almost every cat 3 race goes anyways.

[RyanH]

@KWalker, sorry I'm having a hard time believing that. Attached is a screen shot of Chris @ Specialized in the wind tunnel in the drops position which netted him a CdA of around 0.32. He looks pretty aero to me. Would you care to share some data to substantiate that?

If I remember, I'll have someone take a picture of me riding on the weekend but I'm 5'9 on a small frame (S/M in Fuji, 51 in Cervelo and 52 in Scott). I normally run a slammed 110mm -17 stem but tested a 125mm -17 on the Fuji today. Bars are 38cm CTC at the hoods. Not exactly a cafe cruiser position, but hey, if there's room to optimize my position and get it down to 0.25 to 0.27 on the hoods then I'm all ears. At 0.25 with a FTP of 300w, I could hold, theoretically, 26.7 mph on the hoods!

[justkeepedaling]

That is a really relaxed drop position. I'm about four inches lower in head height with almost 90 degree bend in elbow in the drops

[RyanH]

@cyclespeed, when covering a fixed distance, aero differences are more pronounced the slower you go (time wise).

In example:
40K@200w
Speed @ CdA of 0.32: 34.072402
Time: 70.438 minutes
Speed @ CdA of 0.31: 34.412557
Time: 69.742 minutes
Time Difference: 0.696 minutes

40K@400w
Speed @ CdA of 0.32: 43.864276
Time: 54.714 minutes
Speed @ CdA of 0.31: 44.312922
Time: 54.16 minutes
Time Difference: 0.554 minutes

So, I think it's a generally incorrect statement to say that aero benefits kick in at X speed. If you're racing an Iron Man for the first time and expecting to average 15-16 mph on the cycling leg, a very aero setup is going to net you a lot more time saving (in absolute terms) than the top contenders. I'm too tired to calculate the percentage differences, which may be the inverse. But, moral of the story is that aero benefits everyone regardless of speed traveled.

[RyanH]

That is a really relaxed drop position. I'm about four inches lower in head height with almost 90 degree bend in elbow in the drops

I don't know how big Chris Yu is so it's missing some context but the point was that that is what 0.32 looks like, not necessarily that he's in the most aero position for the drops. My data points are around 0.32 and I'm on the **hoods** yet it is being suggested that my position can be improved by up to 0.7 CdA. I'd love for that to be true, but I'd also like to see some data to support that people are getting usable hood positions (meaning, they can still ride comfortably in the drops, for say, an entire crit) in the 0.25 to 0.28 range and weigh >135 lbs.

[RyanH]

GSC10 Run 15: Fuji Altamira (125mm stem) on 404s with 23mm Veloflex Criteriums
Recording Device: SRM PC8
Speed Sensor: GSC-10
Tire Pressure: 90F/100R
Beginning Temp: 66*F
Ending Temp: 68*F
Wind: 1-3 mph N (a lot of wind)
Avg HR: 153 bpm
Avg Speed: 22.4 mph
Avg Power: 243 watts
Distance Covered: 15.661
Duration: 42:00
Estimated CdA: 0.3104
Quality Rating (out of 3): 1 (below average)

Ride Data (Strava)
Google Sheets Aero Data - ***All Test Runs are in this Spreadsheet***

Notes: Rapha 2015 Pro Team Jersey

Test Setup:
Bottle front, fred can with tire in the rear cage.

Chung Method Aero Analysis


Comments:
Tried a 1.5cm longer stem. Wind was going this entire run, which is pretty easy to identify in the graph. First half of lap one had headwind, so plus VE, tailwind on the back, so you see VE dip, same with the second half of lap 2. Lap 4 and 5 look like they had headwinds on the first half but not as much tailwind on the backside. This is a discard run.

I may be eating my words above because even with challenging wind, which should have resulted in a high CdA estimate, this is one of the fastest runs I've done. If this holds true, interesting what an extra 1.5cm can do (queue the jokes now).

General On-Going Notes
For those just joining or missed a few comments here and there, I'm going to add some pertinent information that I've gathered, and put it in this section.

GPS vs GSC-10. The first 15 runs had to be discarded since they used GPS. GPS was causing too much variability in the runs.
Wind. A north based wind on this course results in a headwind on the climbing section and tail wind on the descending portion. At a constant 3mph, that can increase lap times by up to 4%, which is huge. On the flip side, a south based wind (tailwind) only has a 1% affect. Going forward, any lap with an observed North bound wind value of more than 1/4 flag will be considered a discard run. I'll post the results, but keep in mind it should be excluded since the effect is rather large and there's too much uncertainty to make an adjustment.
Position. This is still under development but it looks like small changes in position can have drastic impacts on my overall CdA. If a hood angle is causing me to be slightly less relaxed on the bike (how bent my arms are) then that can completely erase any aero benefit of a frame.

[KWalker]

Chris' position doesn't look all that great honestly. He's pretty "tall" and his head is substantially above his back with quite a bit of spinal flexion. Fairly standard road position. Looks like his seat is preventing anterior pelvic rotation as well.

What I have left on my existing laptop and hard-drives is simply a summary of estimated results from runs I believe. Would have to check. The first time I ever tested my position was around .3-.31 and I was under the impression that things looked fine, however, my seat was too high and too far forward and causing some hunching like in Chris' picture above. Lowering the seat and dropping it back a bit reduced my shoulder height by almost 4 inches! Unfortunately when I moved cross country a few boxes were lost and stolen, one of which contained all of my hard drives with this data, my academic research/grad data, and a number of other items. I should have something in email when I have a chance. I have not performed such consulting since as its pretty damn boring. Even then considering it was a service people paid for, I'd ask their permission to share their data just as I wouldn't share their power files, blood lactate tests, skinfold tests, etc.

I don't want to derail your thread further, but looking back a few pages at your power numbers, riding hrs etc. I am amazed that you would have trouble in a race such as Winters.

The quickest link I can find on my work computer is simply from my grams and isn't at full gas/relaxed. As power increases, pelvic roation increases and the back straightens out and flattens while the elbows bend. Also using a terribly un-aero helmet for my back shape. Normally the orange panels on the sides of the jersey are parallel to the ground: https://www.instagram.com/p/4mV1y7SCjh/ ... karstegram. Capable of holding that position and actually more elbow bend for entire 60min crits, however, my hoods position is better. A terrible reference, but