Login required to started new threads

Login required to post replies

Aero field testing with Best Bike Split
Quote | Reply
So I did my first aero field test today and was rather surprised by the numbers.

I was testing a TYR sleeveless tri suit (the carbon zipback) vs a LG M2 sleeved Trisuit.

I did an out and back course that was roughly 30mins total since BBS recommends using a file that is at least 30mins long. I kept my watts at 180 which is a bit lower than my current IM pace. Since I had to go through a neighborhood first, I hit the lap button at a spot where I was up to speed and could make sure I hit it at the same spot on the way back.

Here are the results from those two laps (both 7.74mi:

Sleeveless Suit: Time- 22:37.1 Avg. Speed-20.5MPH Avg. Power-181 NP-183

Sleeved Suit: Time-21:44.1 Avg. Speed-21.4 Avg. Power-182 NP-185

So the power was slightly higher on the second run, but it was 53 seconds faster

I ran it through Best Boke Split’s Aero Analyzer and I got a CdA of .2677 for the first run and .2486 for the second.

If I’m interpreting that right, the sleeved suit is 19w faster.

Am I missing anything? Does this all sound about right?

-Nate
Triathlonpal.com
Flaer|Team Kiwami|Nuun Hydration|Honey Stinger
Twitter: @N8deck
Quote Reply
Re: Aero field testing with Best Bike Split [triathlonpal] [ In reply to ]
Quote | Reply
Why are you surprised? Sleeved suits will be faster, provided they fit well.
Quote Reply
Re: Aero field testing with Best Bike Split [EnderWiggan] [ In reply to ]
Quote | Reply
I guess I just didn’t realize it was that big a difference.

The sleeveless also fits tighter... but the sleeved suit is so much more comfortable... I’ll take comfort AND speed any day :)

I also just want to make sure my testing is on target since I have a few other things I’d like to test soon.

-Nate
Triathlonpal.com
Flaer|Team Kiwami|Nuun Hydration|Honey Stinger
Twitter: @N8deck
Quote Reply
Re: Aero field testing with Best Bike Split [EnderWiggan] [ In reply to ]
Quote | Reply
EnderWiggan wrote:
Why are you surprised? Sleeved suits will be faster, provided they fit well.

Not always.
Quote Reply
Re: Aero field testing with Best Bike Split [triathlonpal] [ In reply to ]
Quote | Reply
That delta is way too large IMO. I’d try to find a route that minimizes your application of the brakes, test on a very calm day, and do way more runs.
Quote Reply
Re: Aero field testing with Best Bike Split [GreenPlease] [ In reply to ]
Quote | Reply
Or just always wear a sleeved trisuit, forget testing that as more often than not a sleeved suit is faster so stop wasting your time and test position, or if that’s dialed maybe helmets?
Quote Reply
Re: Aero field testing with Best Bike Split [triathlonpal] [ In reply to ]
Quote | Reply
triathlonpal wrote:
If I’m interpreting that right, the sleeved suit is 19w faster. Am I missing anything? Does this all sound about right?

You have the basic idea, except you need a much shorter course (loop) and many more laps. And repeats where you switch back and forth, like 3 laps with A, 3 laps with B, repeat 3 times.

Then you will have enough data to tell something. Where I live a shift in the wind could have easily accounted for a greater difference than you measured.
Quote Reply
Re: Aero field testing with Best Bike Split [rruff] [ In reply to ]
Quote | Reply
rruff wrote:
triathlonpal wrote:
If I’m interpreting that right, the sleeved suit is 19w faster. Am I missing anything? Does this all sound about right?


You have the basic idea, except you need a much shorter course (loop) and many more laps. And repeats where you switch back and forth, like 3 laps with A, 3 laps with B, repeat 3 times.

Then you will have enough data to tell something. Where I live a shift in the wind could have easily accounted for a greater difference than you measured.


Agreed on this point. A very very small shift in wind can account for the difference you measured. To illustrate, one random data set I have from a competitive triathlete looking to KQ, is on what seemed to be a fairly calm day. Was doing out and backs on a flat stretch of road, with CdA being computed for each stretch.
Using standard GPS and elevation data:
Out #1: CdA = 0.249
Back #1: CdA = 0.279
Out #2: CdA = 0.249
Back #2: CdA = 0.277

Include aero sensor measurements (i.e., effect of the wind), we found:
Out #1: CdA = 0.260
Back #1: CdA = 0.262
Out #2: CdA = 0.260
Back #2: CdA = 0.260

Here is the crazy part - the wind effect is on the order of 1 [kph] for these CdA results. It doesn't take much to throw off aero testing results unless all variables are carefully controlled and you do multiple laps or multiple out-back scenarios.
I will admit, this particular data set turned out to be really nice, as we got very consistent results from using the aero sensor. It also emphasizes the importance of accurate wind information when testing (or waiting for that ideal test day with zero wind).



Chris Morton, PhD
Associate Professor, Mechanical Engineering
co-Founder and inventor of AeroLab Tech
For updates see Instagram
Quote Reply
Re: Aero field testing with Best Bike Split [AeroTech] [ In reply to ]
Quote | Reply
Yup, that calm day rarely is. Even indoor velodromes have airflow.

Developing aero, fit and other fun stuff at Red is Faster
Quote Reply
Re: Aero field testing with Best Bike Split [AeroTech] [ In reply to ]
Quote | Reply
AeroTech wrote:
rruff wrote:
triathlonpal wrote:
If I’m interpreting that right, the sleeved suit is 19w faster. Am I missing anything? Does this all sound about right?


You have the basic idea, except you need a much shorter course (loop) and many more laps. And repeats where you switch back and forth, like 3 laps with A, 3 laps with B, repeat 3 times.

Then you will have enough data to tell something. Where I live a shift in the wind could have easily accounted for a greater difference than you measured.


Agreed on this point. A very very small shift in wind can account for the difference you measured. To illustrate, one random data set I have from a competitive triathlete looking to KQ, is on what seemed to be a fairly calm day. Was doing out and backs on a flat stretch of road, with CdA being computed for each stretch.
Using standard GPS and elevation data:
Out #1: CdA = 0.249
Back #1: CdA = 0.279
Out #2: CdA = 0.249
Back #2: CdA = 0.277

Include aero sensor measurements (i.e., effect of the wind), we found:
Out #1: CdA = 0.260
Back #1: CdA = 0.262
Out #2: CdA = 0.260
Back #2: CdA = 0.260

Here is the crazy part - the wind effect is on the order of 1 [kph] for these CdA results. It doesn't take much to throw off aero testing results unless all variables are carefully controlled and you do multiple laps or multiple out-back scenarios.
I will admit, this particular data set turned out to be really nice, as we got very consistent results from using the aero sensor. It also emphasizes the importance of accurate wind information when testing (or waiting for that ideal test day with zero wind).

While not controlling for changes in environmental conditions is a problem, I'm not sure the example above is completely apples to apples with the OP.

Their CdA estimates were the result of a combined out and back run, not separated out and back runs.

Taking a simple average of your results, i.e. (Out CdA + Back CdA)/2, then
#1 = 0.264 vs your corrected values of 0.260 / 0.262
#2 = 0.263 vs your corrected values of 0.260 / 0.260

Of course we can't really do a simple average for this but you can see it's not quite as bad as it first seems, with just a +0.003 (1%) bias in each case.

But I agree, a better methodology that reduces environmental differences is a good idea.

http://www.cyclecoach.com
http://www.aerocoach.com.au
Quote Reply
Re: Aero field testing with Best Bike Split [AeroTech] [ In reply to ]
Quote | Reply
Wind is thine enemy.
Quote Reply
Re: Aero field testing with Best Bike Split [AlexS] [ In reply to ]
Quote | Reply
Same effect as illustrated by slides 36-37 here:

https://www.slideshare.net/...nd-tunnel-compressed
Quote Reply
Re: Aero field testing with Best Bike Split [AlexS] [ In reply to ]
Quote | Reply
AlexS wrote:
AeroTech wrote:
rruff wrote:
triathlonpal wrote:
If I’m interpreting that right, the sleeved suit is 19w faster. Am I missing anything? Does this all sound about right?


You have the basic idea, except you need a much shorter course (loop) and many more laps. And repeats where you switch back and forth, like 3 laps with A, 3 laps with B, repeat 3 times.

Then you will have enough data to tell something. Where I live a shift in the wind could have easily accounted for a greater difference than you measured.


Agreed on this point. A very very small shift in wind can account for the difference you measured. To illustrate, one random data set I have from a competitive triathlete looking to KQ, is on what seemed to be a fairly calm day. Was doing out and backs on a flat stretch of road, with CdA being computed for each stretch.
Using standard GPS and elevation data:
Out #1: CdA = 0.249
Back #1: CdA = 0.279
Out #2: CdA = 0.249
Back #2: CdA = 0.277

Include aero sensor measurements (i.e., effect of the wind), we found:
Out #1: CdA = 0.260
Back #1: CdA = 0.262
Out #2: CdA = 0.260
Back #2: CdA = 0.260

Here is the crazy part - the wind effect is on the order of 1 [kph] for these CdA results. It doesn't take much to throw off aero testing results unless all variables are carefully controlled and you do multiple laps or multiple out-back scenarios.
I will admit, this particular data set turned out to be really nice, as we got very consistent results from using the aero sensor. It also emphasizes the importance of accurate wind information when testing (or waiting for that ideal test day with zero wind).


While not controlling for changes in environmental conditions is a problem, I'm not sure the example above is completely apples to apples with the OP.

Their CdA estimates were the result of a combined out and back run, not separated out and back runs.

Taking a simple average of your results, i.e. (Out CdA + Back CdA)/2, then
#1 = 0.264 vs your corrected values of 0.260 / 0.262
#2 = 0.263 vs your corrected values of 0.260 / 0.260

Of course we can't really do a simple average for this but you can see it's not quite as bad as it first seems, with just a +0.003 (1%) bias in each case.

But I agree, a better methodology that reduces environmental differences is a good idea.

I agree, it is not apples and apples comparison here. Just illustrating what kind of quantitative change in wind can result in the same order of magnitude change in CdA (i.e., the idea of a shift in the wind playing some role in the variability between tests, even if they are out/back). Indeed, simple averaging doesn't quite work in this case due to the non-linear relationship between drag and relative velocity.

Chris Morton, PhD
Associate Professor, Mechanical Engineering
co-Founder and inventor of AeroLab Tech
For updates see Instagram
Quote Reply
Re: Aero field testing with Best Bike Split [AeroTech] [ In reply to ]
Quote | Reply
I think people should use both a direct measure power meter AND something like a wind method Powerpod so you can determine the wind speed vs ground speeds.
Quote Reply
Re: Aero field testing with Best Bike Split [triathlonpal] [ In reply to ]
Quote | Reply
when i use BBS aero tool, i use at least 5 exact same rides to average before settling on it. somewat easier for me with central park in the mornings being empty

808 > NYC > PDX > YVR
2024 Races: Taupo
Quote Reply
Re: Aero field testing with Best Bike Split [hadukla] [ In reply to ]
Quote | Reply
hadukla wrote:
when i use BBS aero tool, i use at least 5 exact same rides to average before settling on it. somewat easier for me with central park in the mornings being empty

Is that 5 laps of the same course on the same day? or 5 rides on different days?

-Nate
Triathlonpal.com
Flaer|Team Kiwami|Nuun Hydration|Honey Stinger
Twitter: @N8deck
Quote Reply
Re: Aero field testing with Best Bike Split [triathlonpal] [ In reply to ]
Quote | Reply
Different days since the aero analyzer is supposed to account for wind conditions via dark sky (and I could not hope to hold the same necessary watts for 5 different rides in one day!)

808 > NYC > PDX > YVR
2024 Races: Taupo
Quote Reply
Re: Aero field testing with Best Bike Split [triathlonpal] [ In reply to ]
Quote | Reply
As others have said on here, wind will definitely play a big roll in differences. We have a few updates working to help with that, especially on the out and back specific testing efforts. From the look of your data it could be that the second run had a bit more cross component and a slightly higher yaw that we don't handle as well in terms of forming a good 0 yaw baseline component.

Originally we designed it to use race data to get a good approximation for modeling, but there are a lot of enhancements we plan to add. Dark sky is typically pretty good but when we have done field testing we have an anemometer to do spot checks on the course. That being said BBS is currently kind of the good in the good, better, best scenario. We have started to test other CdA measurements options from 3D scans with CFD (STAC), additional field testing methods/protocols, aero sensors, and of course wind tunnel/velodrome testing to help refine our algorithms for general use as well as provide a comprehensive review of all the options available to get the best aero input into our models.

-----------------------------------------------------------------
Founder: BestBikeSplit
Amazonian
Quote Reply
Re: Aero field testing with Best Bike Split [Mrcooper] [ In reply to ]
Quote | Reply
I'm the one that sent you the STAC scanner. How's the testing going so far? I really liked what Andrew did for me over the winter with the virtual scanner. I ditched my old helmet and pulled in the elbows a bit. And we got my CdA down closer to .22. I've compared this to the aero analyzer on BBS and it's within the range of what STAC says. The big test will be this Sunday at Florida 70.3.
Quote Reply
Re: Aero field testing with Best Bike Split [rruff] [ In reply to ]
Quote | Reply
+1
Quote Reply
Re: Aero field testing with Best Bike Split [Mrcooper] [ In reply to ]
Quote | Reply
Would it be fair to say that the Aero Analyzer has its limits on training rides where you are braking a lot and dodging traffic, etc?

I have input a few of my long rides and the values just don't make much sense. The average speed of my training rides are probably 3-4 mph slower pushing the same power as I would in a race, due to all of those external variables.

I would think the model couldn't tell the difference between you tapping on the brakes to slow down slightly, versus you sitting up straight in the wind, both of which would have nearly the same effect.

Strava
Quote Reply
Re: Aero field testing with Best Bike Split [sch340] [ In reply to ]
Quote | Reply
We attempt to throw out data around obvious decelerations / accelerations and attempt to cluster on race / climb(or relaxed) positions, however in training rides where there is a lot of up and down and other places where you are focused and in position it is definitely noisy! We have though about adding a zoom recalc feature for specific sections similar to the power plan and time analysis zoom along with several other updates around weather with some new available services.

As for the STAC sensor I finally have the tests I want to do but they are pretty general fit/position/bike type as oppose to small changes. It is a really cool device and excited about the concept and product.

-----------------------------------------------------------------
Founder: BestBikeSplit
Amazonian
Quote Reply