Observations from today's ride (aerodynamics and drafting)

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

User avatar
fa63
Posts: 2533
Joined: Mon Feb 19, 2007 7:26 am
Location: Atlanta, GA, US

by fa63

This morning, I met up with a triathlete friend to join her during her interval session. We both have power meters, so I thought I would take a look at the data post-ride and see what I could gather. Here are the profiles from Strava; the top one is me (183 cm, 100 kg male), the bottom is my friend (163 cm, 55 kg female):

Image

I was in the front for the first ~3 km during warmup; she was ahead of me the rest of the time doing her intervals. The road is pretty much flat, essentially no wind. She was on a Cervelo P2 with aero wheels, aero helmet and a finely tuned position (she is pretty competitive in her age group, came in 29th out of several hundred at the nationals). I was on this, riding on the hoods.

Image

Using the speed-power data (and the Analytic Cycling website), I was able to deduce the following:

My frontal area, without the draft: ~0.743 m^2 (looking the the speed and power for the section where I was in the front)
My effective frontal area, with the draft: ~0.595 m^2
Her frontal area: ~0.465 m^2

For me, it is an approximately 20% reduction getting in her draft. Also, her frontal area is about 40% smaller than mine :) This is perhaps not surprising given her position on her tri bike (low in the front, flat back)

It took me 197W to average 33.9 km/h (21.2 mph) while being in the draft essentially the whole time. Without the draft, Analytic Cycling estimates it would have taken me 235W to achieve the same average. So I was saving about 16% by being in her draft. Value likely to be higher sucking the wheel of a less pint sized cyclist on a road bike and not a tri bike :)

Just thought I would share.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



bremerradkurier
Posts: 419
Joined: Mon Jun 13, 2016 4:18 pm

by bremerradkurier

Another interesting effect is very close drafting (4-6" IIRC) also lowers drag on the front rider.

peted76
Posts: 433
Joined: Mon Jun 23, 2014 10:30 pm

by peted76

Interesting results.. thanks for sharing.

alcatraz
Posts: 4064
Joined: Mon Aug 29, 2016 11:19 am

by alcatraz

What about clothes? Tight fit? Skin suit? Shoe covers?

/a

User avatar
fa63
Posts: 2533
Joined: Mon Feb 19, 2007 7:26 am
Location: Atlanta, GA, US

by fa63

She was wearing cycling shorts and a tight-fitting tri tank top. I was wearing a club fit jersey (tight, but not race tight) and Sportful bibs. No shoe covers for either one of us; that would be a little ridiculous for a training session [emoji3]

I plan to join her on the same course for the next couple of weeks, altering my equipment a bit each time. Just for fun really; don't want to turn this into a scientific experiment.

Alexandrumarian
Posts: 795
Joined: Mon Mar 28, 2016 6:34 pm
Location: Romania

by Alexandrumarian

For me it feels a little bit easier to pull hard if someone is behind even if not crazy close. I don't have a PM but when drafting at a casual, safe distance I see the bpm drop by a solid 10 points at 33-35kph and even 15 if I stay close (which I prefer not to). Never been in a large group or race, really curious how it feels when you are closely surrounded.

User avatar
fa63
Posts: 2533
Joined: Mon Feb 19, 2007 7:26 am
Location: Atlanta, GA, US

by fa63

There is a ride called the Airport ride in Atlanta, where 50+ show up regularly, which starts off pretty flat. When I am in the middle of the group along those sections, going 40-45 kph feels pretty effortless, almost like getting pulled along.

Then the hills arrive and I usually get dropped [emoji3]

benjaminm3
Posts: 48
Joined: Mon Jul 17, 2017 10:17 pm

by benjaminm3

I'm a little newer to cycling and so perhaps didn't know about power differences/weight for the same speed, but I found that aspect of your data interesting as well. (though I guess going straight frontal area/aero might be a bigger determinant in how much power you need to sustain speed) Thanks for sharing!

User avatar
fa63
Posts: 2533
Joined: Mon Feb 19, 2007 7:26 am
Location: Atlanta, GA, US

by fa63

benjaminm3 wrote:though I guess going straight frontal area/aero might be a bigger determinant in how much power you need to sustain speed


On the flats, absolutely.

bremerradkurier
Posts: 419
Joined: Mon Jun 13, 2016 4:18 pm

by bremerradkurier

fa63 wrote:
benjaminm3 wrote:though I guess going straight frontal area/aero might be a bigger determinant in how much power you need to sustain speed


On the flats, absolutely.


Everybody wants to be on a tandem's rear wheel on the flats.

youngs_modulus
Posts: 668
Joined: Wed Sep 20, 2006 1:03 am
Location: Portland, OR USA

by youngs_modulus

In fact, Benjamin, riding on a flat road at a steady speed, weight doesn't matter at all. Aero drag is nearly the only factor in the power required to sustain a respectable speed on the flats. Tire rolling resistance matters a little and bearing drag matters considerably less than that. But practically speaking, at any significant speed on the flats, aero is all that matters.

It's helpful to remember that the power required to overcome rolling resistance increases (roughly) linearly with speed, while the power required to overcome aero drag is proportional to the cube of speed. If aero drag and bearing/rolling resistance are equal at 10 mph, it doesn't take much of an increase in speed for aero drag to completely swamp the other terms.

Some interesting things fall out of this math. For instance, low rolling-resistance tires offer a bigger relative benefit when climbing than when sprinting or riding tempo on the flats. (Rolling resistance is proportionally a much bigger player at 7 mph than it is at 25-35 mph).

Also, many people assume (reasonably) that weight is a significant factor in sprinting since sprinting involves a lot of acceleration. And while F=MA is applicable here, people forget that the A is relatively tiny for bicycle racers, so the power ((force*distance)/time) used to accelerate from, say, 29 mph to 37 mph is pretty low compared to the power required to overcome aero drag at those speeds. Racers expecting to contest a sprint are often better off ditching their 1000-gram shallow-rimmed wheels in favor of 1450-gram aero wheels, which results in a significant increase in top speed (terminal velocity) at the cost of a nearly unmeasurable decrease in peak acceleration.

This is a really esoteric point in some circles, and many riders don't buy it. But the math is straightforward*, and since I plan to get back into racing with the hope of winning some sprints, I'm not going to argue about it too much with anyone, at least not locally.



* The math:
Kitted up, my bike and I together weigh about 84.5 kilos. Accelerating from 29 mph to 37 mph in 13.5 seconds (roughly 200 meters) works out to less than 0.3 meters per second per second, or a little under 0.03 g. These are tiny accelerations. To put this in perspective, the power required for me to accelerate 8 mph in 13.5 seconds requires 330 Watts. (The rest of my power, maybe another 750-900 watts, goes to aero drag). If I switch to a bike that's 2 kilos lighter (that's 4.4 pounds!) then the power required for that acceleration drops to 320 Watts. Meanwhile, deep aero wheels might add 0.45 kilos to the bike/rider system but save 30-50 watts at those sprinting speeds. As a back of the envelope calculation, the heavier aero wheels take ~1-3 extra Watts for acceleration but require ~40 watts less power to overcome aero drag. That's a net gain of ~37 watts; it's not even close.

This is why match sprinters ride usually disc wheels these days. It's also why when someone anounces that they can feel their new, 1-kilo-lighter bike "accelerates faster in a sprint" you know it's the placebo effect. None of us can tell when we're making an extra 5 Watts in a 1100-Watt sprint, and Watts are Watts, so the effect of sprinting on a lighter bike is also imperceptible. But when people think they can feel such tiny weight differences in a sprint, it's not because they're dumb or foolish. Human beings are wired to see causation everywhere. The instinct to connect correlation with causation is both what makes science hard and what makes science possible.

JackRussellRacing
Posts: 243
Joined: Thu Nov 03, 2016 4:32 pm
Location: USA

by JackRussellRacing

^ Outstanding post youngs_modulus. That's the kind of stuff I absolutely love on this forum. Mucho thanks!

benjaminm3
Posts: 48
Joined: Mon Jul 17, 2017 10:17 pm

by benjaminm3

youngs_modulus wrote:In fact, Benjamin, riding on a flat road at a steady speed, weight doesn't matter at all. Aero drag is nearly the only factor in the power required to sustain a respectable speed on the flats. Tire rolling resistance matters a little and bearing drag matters considerably less than that. But practically speaking, at any significant speed on the flats, aero is all that matters.

It's helpful to remember that the power required to overcome rolling resistance increases (roughly) linearly with speed, while the power required to overcome aero drag is proportional to the cube of speed. If aero drag and bearing/rolling resistance are equal at 10 mph, it doesn't take much of an increase in speed for aero drag to completely swamp the other terms.

Some interesting things fall out of this math. For instance, low rolling-resistance tires offer a bigger relative benefit when climbing than when sprinting or riding tempo on the flats. (Rolling resistance is proportionally a much bigger player at 7 mph than it is at 25-35 mph).

Also, many people assume (reasonably) that weight is a significant factor in sprinting since sprinting involves a lot of acceleration. And while F=MA is applicable here, people forget that the A is relatively tiny for bicycle racers, so the power ((force*distance)/time) used to accelerate from, say, 29 mph to 37 mph is pretty low compared to the power required to overcome aero drag at those speeds. Racers expecting to contest a sprint are often better off ditching their 1000-gram shallow-rimmed wheels in favor of 1450-gram aero wheels, which results in a significant increase in top speed (terminal velocity) at the cost of a nearly unmeasurable decrease in peak acceleration.

This is a really esoteric point in some circles, and many riders don't buy it. But the math is straightforward*, and since I plan to get back into racing with the hope of winning some sprints, I'm not going to argue about it too much with anyone, at least not locally.



* The math:
Kitted up, my bike and I together weigh about 84.5 kilos. Accelerating from 29 mph to 37 mph in 13.5 seconds (roughly 200 meters) works out to less than 0.3 meters per second per second, or a little under 0.03 g. These are tiny accelerations. To put this in perspective, the power required for me to accelerate 8 mph in 13.5 seconds requires 330 Watts. (The rest of my power, maybe another 750-900 watts, goes to aero drag). If I switch to a bike that's 2 kilos lighter (that's 4.4 pounds!) then the power required for that acceleration drops to 320 Watts. Meanwhile, deep aero wheels might add 0.45 kilos to the bike/rider system but save 30-50 watts at those sprinting speeds. As a back of the envelope calculation, the heavier aero wheels take ~1-3 extra Watts for acceleration but require ~40 watts less power to overcome aero drag. That's a net gain of ~37 watts; it's not even close.

This is why match sprinters ride usually disc wheels these days. It's also why when someone anounces that they can feel their new, 1-kilo-lighter bike "accelerates faster in a sprint" you know it's the placebo effect. None of us can tell when we're making an extra 5 Watts in a 1100-Watt sprint, and Watts are Watts, so the effect of sprinting on a lighter bike is also imperceptible. But when people think they can feel such tiny weight differences in a sprint, it's not because they're dumb or foolish. Human beings are wired to see causation everywhere. The instinct to connect correlation with causation is both what makes science hard and what makes science possible.


Thanks for sharing this. I knew the underlying physics but hadn't read it framed in this way before, and this is very helpful for thinking about what properties to optimize!

User avatar
Fixie82
Posts: 348
Joined: Mon Mar 18, 2013 3:45 am

by Fixie82

^ damn straight. That post is pure gold.

Follow on question regarding wheel depths. Is deeper better for wheels or is their diminishing returns? Is there a sweet spot at, for example, 60mm? Deep enough to get the Aero benefit but also maintain ease of handling? Or is it simply a matter of going as deep as you can go?

TobinHatesYou
Posts: 12456
Joined: Mon Jul 24, 2017 12:02 pm

by TobinHatesYou

Like all else, it depends on your goals and the ride conditions.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



Post Reply