Road Bike, Cycling Forums banner
1 - 20 of 26 Posts

· Registered
Joined
·
735 Posts
Discussion Starter · #1 ·
Hello,

I think this is the best place to answer a simple question. If two people are riding together and the second is drafting off the lead rider- does the lead rider have to work harder? There seems to be conflicting information out there on this and I just wanted to find out if anyone on here knows for sure.

Thanks MTT :confused:
 

· Registered
Joined
·
3,083 Posts
harder than what? Harder than the drafting rider to maintain the given speed? obviously, yes. Harder than he would without someone on his wheel to maintain the giving speed (as in, "get off my wheel you wheelsucking vampire, you are slowing me down!")? uh, no. Well, that second answer is purely anecdotal, but if there's an aerodynamic effect to having another rider on your wheel, I'm nearly certain that it's positive, not negative.
 

· Registered
Joined
·
3,230 Posts
I have always heard that there is a slight benefit to having a rider in your draft. I did not do research myself, of course, nor can I say that I ever have noticed. But there is a theoretical basis -- something about smoothing out the turbulence behind you. I think it's the same theory, more or less, why TT helmets have that teardrop shape that's sharper on the trailing edge.
 

· It's all ball bearings
Joined
·
5,285 Posts
There is acutally a significant benefit for the lead rider to being drafted, something to the effect of 5-8% energy savings compared to riding alone. I don't have the refs handy, but I recently read a couple of studies that approximate this by using cylinder-shaped bodies (a grossly-oversimplified analog to human figures on bicycles) set one in front of the other in a wind tunnel. Essentially, the slipstream of the lead cylinder continues somewhat uninterrupted past the following cylinder so that the following cylinder inherits a signifcant amount of the lead cylinder's wake eddies that have an effect of "sucking" the cylinder back against the direction of travel (this is aerodynamically an oversimplified explanation), as depicted below in my preschool Windows paint figure:
 

· Registered
Joined
·
3,375 Posts
MTT said:
If two people are riding together and the second is drafting off the lead rider- does the lead rider have to work harder? There seems to be conflicting information out there on this and I just wanted to find out if anyone on here knows for sure.
I've seen several references saying there is no effect on the lead rider (e.g., http://www.midweekclub.com/articles/broker98.pdf ), but have never seen any data (theoretical or empirical) showing there is. I would welcome it if someone could provide it, but until then I wouldn't say there is any conflict; all the data shows no effect.
 

· Registered
Joined
·
3,375 Posts
BenWA said:
I'm still looking for the paper that I made reference to above, but in the meantime this site has some pretty good info that touches on what I was talking about with wind tunnel cylinder studies: http://www.princeton.edu/~asmits/Bicycle_web/bicycle_aero.html

I don't see any reference to bodies in tandem there, and if as you claim there is a 5-8% reduction in drag for the lead rider, how do you explain Broker et al. being able to model performance to within ~2% without adjusting the drag coefficient on the lead rider (i.e., no effect from the trailing 3 riders). More significantly, why has no one published to refute Broker's results and add the 5-8% correction?
 

· It's all ball bearings
Joined
·
5,285 Posts
couldn't tell ya

asgelle said:
I don't see any reference to bodies in tandem there, and if as you claim there is a 5-8% reduction in drag for the lead rider, how do you explain Broker et al. being able to model performance to within ~2% without adjusting the drag coefficient on the lead rider (i.e., no effect from the trailing 3 riders). More significantly, why has no one published to refute Broker's results and add the 5-8% correction?
Well first of all, I don't mean to suggest that what I read was necessarily more correct than Broker et al.'s findings. I have no idea whether Brokers paper predates or postdates the paper that I read, so I don't know if any studies have directly refuted Broker's results.

On the one hand, the cylinder studies were far less empirical than Broker et al. in the application of cycling, but on the other hand I suspect that Broker's study wasn't nearly as controllable and cleanly quantifyable as something as reduced as the cylinder studies. Granted, I only read Brokers abstract and conclusion, so I don't really know how carefully designed his study was. I'll read thru it when I have a spare few minutes.

The link that I posted has a pretty good explanation of flow about bluff bodies, particularly at higher reynolds numbers (cyclists behave aerodynamically in part as streamlined bodies and in part as bluff bodies, but the net behavior is as bluff body). The higher the reynolds number, the more the total drag of a bluff body is dominated by energetic wake eddies (I hate to use the term "vorticies" here, as that implies a high degree of organization which bluff wakes generally are not)...so it would stand to reason that if these eddies can be transferred to a body following in tandem, the forward body will be relieved of some work.

Of course besides a paper or two, I haven't looked at that many acutal data, so it's very possible that I could be wrong. :)

Still digging for that paper that I read...
 

· Registered
Joined
·
3,375 Posts
BenWA said:
The link that I posted has a pretty good explanation of flow about bluff bodies, particularly at higher reynolds numbers (cyclists behave aerodynamically in part as streamlined bodies and in part as bluff bodies, but the net behavior is as bluff body). The higher the reynolds number, the more the total drag of a bluff body is dominated by energetic wake eddies ...
Right, and these eddies are caused my viscous interactions (generation of vorticity) at the surface of the leading body. Once momentum is transferred to the eddy, and the eddy moves downstream, it doesn't matter if a downstream body is present or not. All the momentum from the leading body has already been transferred. So the only way for the trailing body to affect the leader is to somehow affect the viscous layer at the surface of the leading body. However, if you calculate the boundary layer thickness at the leading edge of bluff body at the Reynolds number typical of cycling speeds, you'll find the boundary layer thickness is on the order of a millimeter, much less than the separation of wheels, much less riders. So it seems the trailing rider can't affect the leader through this mechanism.

BenWA said:
Well first of all, I don't mean to suggest that what I read was necessarily more correct than Broker et al.'s findings. I have no idea whether Brokers paper predates or postdates the paper that I read, so I don't know if any studies have directly refuted Broker's results.
Broker's paper was published in 1999, that's a long time in the scientific literature. Knowing Broker, Kyle, and Burke's body of work, it would be very surprising if there were data extant at the time showing a reduction in drag for the lead rider and they did not address it. Further, it's hard to believe the paper would get through the review and editorial process without citing such data if it existed at the time. As to subsequent papers refuting Broker, I did a citation search within the last month and as of that time, there was none.
 

· Registered
Joined
·
21,834 Posts
Mixed bases

The problem here is that simple cylindrical model studies (actual data or calculations only?) are being compared with measurements on actual riders. A very poor comparison, indeed. If there is any benefical effect from having someone draft you, it is because the eddys form differently due to the combined shape of two riders, and therefore less energy is transferred to them. This is why race cars are faster when another car is drafting. It's not just about boundary layer thickness, it's about the entire flow pattern. That said, 5-8% would definitely be noticeable, and there's no way you can notice a speedup due to someone latching onto your wheel. If there is an effect, it is very small, perhaps in the range of the noise (1-2%) of the tests run, or perhaps less. And to the OP's question, it certainly is not slowing the lead rider.
 

· Registered
Joined
·
3,375 Posts
Kerry Irons said:
If there is any benefical effect from having someone draft you, it is because the eddys form differently due to the combined shape of two riders, and therefore less energy is transferred to them.
But that begs the question what is the mechanism for the eddy's forming differently, In other words, how does the presence of the trailing rider feed forward to affect the momentum transfer from the leading rider?

Kerry Irons said:
It's not just about boundary layer thickness, it's about the entire flow pattern. [sorry to rearrange your post]
Not about boundary layer thickness, but boundary layers. As long as we're talking about eddies, we're dealing only with viscous effects. These are determined entirely in the boundary layer. Therefore for the trailing rider to affect the leader, the trailing boundary layer would have to overlap the leader's (which is where boundary layer thickness enters) and this is not possible unless the separation is less than 1 mm. You're right that this isn't the whole story. There is also the pressure drag which is a purely inertial effect (the idea of filling in the vacuum behind the leader). Scaling arguments for this also show there should be no measurable effect on the leader from the drafting rider.

Kerry Irons said:
This is why race cars are faster when another car is drafting.
That's true but I think it's misleading to draw an analogy between a race car with frontal area of 2 m^2 moving at 200 mph and a cyclist with area 0.5 m^2 moving at 20 mph.

Kerry Irons said:
If there is an effect, it is very small, perhaps in the range of the noise (1-2%) of the tests run, or perhaps less.
Which is still possible and if someone could conduct sensitive enough measurements wouldn't contradict the original paper.
 

· Premium Member
Joined
·
15,996 Posts
Without all the big words about boundary layers and viscous effects, I'll say this much:

In auto racing (when speeds are much higher, of course, and the cars are much bigger), drafting the car in front of you very closely speeds up both cars. It's a noticable advantage at this level.

So, unless something is vastly different in physics for cyclist, there's going to be SOME advantage to being drafted, and it definitely doesn't HURT the lead rider. Whether that advantage is 1 watt, I don't know; I certainly can't feel it.
 

· Registered
Joined
·
735 Posts
Discussion Starter · #13 ·
Thanks for all the references. It does seem you would have to do a controlled study with cyclists, and vary one thing at a time (for example- how would side wind effect the lead rider). The good news is I will worry allot less when people latch onto my wheel, now I know that the extra effort is minimal (if it exists at all). I think the effort is purely in my head, maybe because I automatically try to drop them (not that many can catch me :D ).

Thanks.............MTT
 

· Registered
Joined
·
12,083 Posts
asgelle said:
But that begs the question what is the mechanism for the eddy's forming differently, In other words, how does the presence of the trailing rider feed forward to affect the momentum transfer from the leading rider?
Which is still possible and if someone could conduct sensitive enough measurements wouldn't contradict the original paper.
The eddies are not made very differently. However, instead of being released from the behind of the lead guy, they are released from the one drafting. The front rider will benefit and the rear rider will benifit more from the drafting than the small effect from the eddies is slowing him.
 

· It's all ball bearings
Joined
·
5,285 Posts
could be

Kerry Irons said:
That said, 5-8% would definitely be noticeable, and there's no way you can notice a speedup due to someone latching onto your wheel. If there is an effect, it is very small, perhaps in the range of the noise (1-2%) of the tests run, or perhaps less.
I could probably buy that. I can't say that I have personally made any anecdotal attempts of my own while riding to see if there is a "noticible" difference in performance with someone drafting me versus not. I think this sort of thing would be difficult to qualify, even if there IS a measurable effect on the lead rider....too many other variables to control in that kind of experiment IMO.

Cylinders are indeed distant analogs to cyclists, however that doesn't mean that the aerodynamic behaviors that are manifest in cylinder studies can't be present in a cycling application.
 

· It's all ball bearings
Joined
·
5,285 Posts
splitting hairs

asgelle said:
But that begs the question what is the mechanism for the eddy's forming differently, In other words, how does the presence of the trailing rider feed forward to affect the momentum transfer from the leading rider?...
...Not about boundary layer thickness, but boundary layers. As long as we're talking about eddies, we're dealing only with viscous effects. These are determined entirely in the boundary layer. Therefore for the trailing rider to affect the leader, the trailing boundary layer would have to overlap the leader's (which is where boundary layer thickness enters) and this is not possible unless the separation is less than 1 mm.
Well we could speculate all we want, but my guess is that none of us here are aerodynamicists, computational fluid dynamicists, etc. by profession (or am I wrong?) so we're probably getting into the realm of splitting hairs that are out of our league. My hunch, however, is that the effect, as KI mentioned, has more to do with the entire flow pattern than individual boundary layer separations and whatnot. I don't think eddies form differently per se, I think that there is "less room" for more eddies to form/evolve/persist behind the lead body when a second body downstream is filling the void. Fewer or less-persistant eddies would seem to me mean less drag on the leader. As Argentius put it, if there is a known effect with race cars, why should there not also be a similar effect in with cyclists? The question in my mind is how significant is the effect. 10 watts? 1 watt? 0.1 watts? 0.01 watts? I don't think any of us can answer that with any confidence, even with Brokers paper.
 

· Registered
Joined
·
735 Posts
Discussion Starter · #17 ·
Well if you have a wind tunnel type apparatus, why not scale down objects that look more like a person on a bike? Seems like that would be the first thing to try. Then from there, maybe on a day with out much wind, two people could go out and try and measure if there is a big enough difference on output (maybe measured by the lead rider's KPH on his computer both with and with out another rider drafting). Not as exact as a publishable paper, but this might give us some practical information- no?

MTT
 

· Registered
Joined
·
3,375 Posts
BenWA said:
Well we could speculate all we want, but my guess is that none of us here are aerodynamicists, computational fluid dynamicists, etc. by profession (or am I wrong?)
You're wrong.

BenWA said:
As Argentius put it, if there is a known effect with race cars, why should there not also be a similar effect in with cyclists? The question in my mind is how significant is the effect. 10 watts? 1 watt? 0.1 watts? 0.01 watts? I don't think any of us can answer that with any confidence, even with Brokers paper.
I think we can. The Broker paper says we can bound the effect as less than 2% if we believe their results, and given the reputation of the authors, supporting follow-up research, and lack of any contradictory data this seems reasonable. Then, looking at the race car analogy we can refine this. I'm going to assume that drafting cars increase the speed of the leader by 1% (~2mph) since that represented the difference between first and fifth for Indy 500 qualifying, I suspect that's a lot. Power to overcome drag goes as speed cubed, so that repesents about 3% reduction in power. But a car has about 10 times the frontal area of a bike and rider so lets say that drops the power savings to 0.3%. Also cars draft at about the same separation as bikes do, but we should scale the separation to the square root of frontal area, so that gives a further reduction of a factor of 3 giving a final reduction in power of 0.1%. So for a typical rider puting out 250 W, the trailing rider might save 0.25W for the leading rider. I'd say this is well below any measure of significance. Now maybe I've been overly conservative or just plain wrong in some of my assumptions. Still I doubt I'm off by a factor of 10, but let's say I am, then the savings jumps to a watt or two. Still hardly a noticable effect.
 

· It's all ball bearings
Joined
·
5,285 Posts
asgelle said:
You're wrong.
You are a professional in one of the aforementioned fields then?

Your rationalization/guestimates seem reasonable enough. As I said, I haven't done enough research to really argue too deep one way or the other.

One thing of note though, a quick few googles on the subject of drafting provides dozens of hits on webpages that have quotes expressing that there is some significant advantage to a person being drafted. It appears that this may be a myth (if it indeed is a myth) that is quite broadly believed and accepted.

E.g.:

Suprisingly drafting not only helps the bicyclist following the leader, but the lead cyclist gains an advantage as well. Paul [Doherty] explained, "The interesting thing is by filling in her eddy you improve the front person's performance as well. So two people who are drafting can put out less energy than two individuals (who are not drafting) would covering the same distance in the same time."
from http://www.exploratorium.edu/cycling/aerodynamics2.html

As the riders continually progress toward the front of the paceline, take a short turn at the front, and then soft pedal as they drift to the rear of the line, the average drag coefficient of a TTT rider (assuming a constant rate of rider rotation and ignoring the effect of dropping back) is around 27% lower than experienced by an individual rider. Perhaps the most surprising conclusion is that, despite the full force of the oncoming air, the lead rider experiences lower drag than if he were riding an ITT at the same speed. The drag coefficient of the leading TTT rider is 0.277, while that of an individual rider is 0.285. This occurs because the second place rider reduces the influence of the lead rider’s wake, increasing his base pressure and consequently reducing the drag force
from http://www.deskeng.com/Articles/Cov...ag-Out-of-the-Tour-de-France-20050715558.html
 

· For president!
Joined
·
7,851 Posts
BenWA said:
You are a professional in one of the aforementioned fields then?
In my time in this forum I've found it futile to argue with asgelle about aerodynamics, Dwayne Barry about biology/physiology, and Kerry Irons about supplements...

Not at all surprised he does it for a living,

Silas
 
1 - 20 of 26 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top