Tri Talk Triathlon Podcast, Episode 76

You can listen to the audio by clicking above, or download it here

Wind tunnel testing on 10 hydration systems, and an interview with Ironman World Champion Chris McCormack. Today, on Tri Talk.
Welcome to Tri Talk your podcast source for triathlon tips, training, news and more. Welcome to new listeners from Singapore and Germany, as those two countries just moved into the top 10 countries subscribing to Tri Talk. And hats off to California and Texas, who not only have significant electoral college votes, but who represent the primary markets for Tri Talk’s US subscription base. Thank you for listening! My goal at Tri Talk is to help you swim, bike, and run faster, to meet your personal triathlon goals. Whether you are an elite or amateur triathlete, we cover sprint distance to Ironman distance. I’m your host, David Warden, and this is Tri Talk Episode 76.


Today on Tri Talk I’m excited to share the preliminary results on hydration systems from our wind tunnel testing. You’ll recall in episode 75 I gave a small report on helmets from the wind tunnel. Today we dive deeper into hydration systems with multiple configurations. After that report, I’m thrilled to have Chris McCormack on the show. If you are a long-time listener to the show, you know that I am not into interviews, but when Chris McCormack’s agent contacts you and asks if you want to interview the IM world champion, you say “yes!”.

This is part 2 in a 3 part series on our recent wind tunnel test. In episode 75 we gave you a teaser on aero helmets and what we discovered, and we’ll continue that in episode 77 with our apparel tests.

In 2003, I read an article by John Cobb that changed my view of cycling. A wind tunnel test he performed on various hydration systems and their placement revealed a wide range of aerodynamic consequences.  Those consequences ranged from 2.5 minutes faster to 1.5 minutes slower over an Ironman distance ride. The concept that aerodynamic elements outside of simply training hard could provide me with this “free speed” was not new to me, but somehow seeing objective numbers with equipment I actually used blew my mind.


For 7 years I’ve wanted to replicate John Cobb’s test on a wider range of configurations, and with more and modern hydration systems. John’s original test covered 6 different hydration systems in 9 different configurations. A few weeks ago, I had the opportunity to initiate the first phase of this goal, by testing 10 modern hydration systems in 18 different configurations. In addition to current hydration systems, we also tested the consequence of coming out of the aero position to drink. Some systems allow you to stay aero when drinking, while some require you to get out of the aero position, and we quantified this result. I felt that this is a critical consideration when selecting a hydration system (spoiler alert!…I was wrong).


The test was administered by Dr. Roy Denoon of CPP engineering, overseen by Joe Friel, with myself as the project manger and test rider. The baseline position was a 2010 SCOTT Plasma with HED Black Dog aerobars, SRAM R2C Shifters, with a mixed SRAM Force and Shimano Ultegra grouping. The rider was 70 kilograms and 170 centimeters tall with shaved legs. The baseline helmet was a LG Rocket with attached wind screen, the apparel was a 2009 De Soto LiftFoil, and the shoes were the 2010 LG Carbon Tri HRS. No socks.


I cannot emphasize enough that this is a preliminary test. A work in progress. This test was done at 0 degrees of yaw, which is an unrealistic, if not impossible scenario for a cyclist (yaw being the angle at which the wind is striking the rider, in this case 0 degrees of yaw would be no wind at all, just the drag generated by the rider). John Cobb did the test with an elegant distribution of yaw from 0 to 30 degrees, making his test much more accurate. This, of course, is our ultimate goal, and subsequent tests will include these additional angles to complete the project.


The positioning of the rider was a relaxed, Ironman position. In an aero position, but fairly upright and comfortable, with a constant cadence of 90rpm maintained throughout the tests. You can see a picture of the rider in Figure A. When pictures of the test leaked out on, I was taken to task regarding the baseline positioning, with several contributors suggesting it was horrible (I was compared to the Wicked Witch of the West, with apologies to L. Frank Baum). After reviewing the drag data supplied by Dr. Denoon, and comparing that to other baseline results, I am confident that the baseline position used in our test represents the typical age-group Ironman position. I’m not suggesting that triathletes could not nor should not achieve a more aggressive position for a 5-7 hour ride, but rather the upright, relaxed position represents the typical age-group triathlete of today. Ultimately, the purpose of the test was not refining rider position, but hydration system performance for a given, static rider position.


I present this preliminary data similar to John Cobb’s results from 2003. Both in a 180k and 40k distance, with the 180k IM distance having the rider output at 150 watts, and the 40k Olympic distance at 225 watts. Unless otherwise noted, the systems were installed exactly as they would be for full use, including water bottles, straws and tubes inserted, where applicable.





40k in Minutes

Seconds from 40k Baseline

180k (112 mi) in Minutes

Seconds from 180k Baseline

Baseline No Bottles 1:07:46.80 0 5:56:31.80 0
Profile Design Aquacell Aero Bracket 1:07:20.40 -26.4 5:54:15.00 -136.8
Profile Design Aerolite Aero Bracket 1:07:46.80 0 5:56:31.80 0
Xlab Carbon Wing – without bags- with CO2- two 20 oz. standard water bottles 1:08:00.00 +13.20 5:57:39.00 +67.20
Xlab Turbo Wing – without bags- with CO2- two 20 oz. standard water bottles 1:08:00.00 +13.20 5:57:39.00 +67.20
Xlab Torpedo, standard 20 oz. water bottle – between aerobars- one 20 oz. standard water bottle 1:08:30.60 +16.80 5:57:55.80 +84.00
Gorilla Cage – seat tube only- standard 20 oz. water bottle 1:08:60.60 +19.80 5:58:12.60 +100.80
Profile Design Aerodrink Aero Bracket 1:08:10.20 +23.40 5:58:29.40 +117.60
Inviscid Design Speedfill seat tube 1:08:10.20 +23.40 5:58:29.40 +117.60
Profile Design Razor down tube 1:08:13.20 +26.40 5:58:46.20 +134.40
Xlab Carbon Wing – with bags- with CO2- two 20 oz. standard water bottles 1:08:29.40 +42.60 6:00:09.60 +217.80
Hydrotail H.5 – with CO2- two 20 oz. standard water bottles 1:08:42.60 +55.80 6:01:15.60 +283.80
Inviscid Design Speedfill down tube 1:08:52.20 +65.40 6:02:05.40 +333.60
Profile Design Razor seat tube 1:08:58.20 +71.40 6:02:37.80 +366.00
Gorilla Cage – down tube only- standard 20 oz. water bottle 1:09:10.80 +75.00 6:02:54.00 +382.20
Gorilla Cage – down tube + seat tube- two 20 oz. standard water bottles 1:09:14.40 +87.60 6:03:59.40 +447.60
Xlab Carbon Wing – without bags- without CO2- two 20 oz. standard water bottles 1:09:27.00 +100.20 6:05:04.20 +512.40

*Must add 0.47 seconds for “down” drinking position, or 1.7 seconds for “up” drinking position

**Must add 2.5 seconds for “down” drinking position, or 9.8 seconds for “up” drinking position


As discussed earlier, I questioned how aero a hydration system is when the rider must come out of an aero position in order to drink. I felt this was a “flaw” in John Cobb’s test from 2003, in that his posted times of the standard water bottles were only valid if the rider never, ever drank during his 6 hour ride. To address this, we tested the drag when the rider was drinking from a water bottle in 2 positions. One position with the rider “down”, leaving one arm on the aero pads, with the other arm drinking from the bottle. And a second position with rider “up”, one hand on the hoods and the other drinking from a bottle. See Figures B and C. I assume drinking 3 times per hour, with the rider out of aero the position for 5 seconds each drink, or a total of 90 seconds out of the aero position when drinking during a 6 hour IM ride, and 15 seconds during a 1 hour 40k ride. I also assume equal time in refilling all systems, so although all systems require the rider to come out of aero to refill, I consider that a wash.


What can we learn from this initial wind tunnel run? Again, not much until we complete the additional yaw tests, but there are a few things that stand out:


–         The baseline position results are eerily similar to John Cobb’s 2003 test. Only 21 seconds difference between his baseline 40k and ours, and only 5 seconds difference between his 180k baseline and ours (John’s calculation was not on 180k, but 112 miles, which is technically 180.25k, thus an explanation for the smaller delta between our 40k and 180k results). Unfortunately, this can be attributed to nothing more than a coincidence. Although we based our calculations on the same wattage and distance, it is very unlikely that John’s baseline rider and I were in the exact same position, and also unlikely that our equipment resulted in identical drag. Plus, we also must have used different drag calculators to compute total time, which although would be almost identical, there would be slight variations.


Does the fact that we have an almost identical baseline to John’s mean that the additional yaw tests are not necessary? Not at all. Again, the similar baselines must be more coincidence than anything else, and the additional yaw angles are essential to complete the project. Although 0 degrees of yaw will always be the primary attack angle of a rider, the other angles complete the true aerodynamic experience.


–         No, it is not an error. The huge, 50 oz. Aquacell performed better than any other system at 0 degrees of yaw. We have to consider that this was the result on a SCOTT Plasma. It is possible that this same system would have performed terribly on a different bike, perhaps a Cervelo. I’ll comment more on this later in the article, but I was surprised.


–         It is interesting that only 1 system performed better than the clean baseline. In John’s test, 3 of 8 configurations performed better than the clean frame, and in our test it was only 1 of 16, with a second component the same as the clean frame, so perhaps 2 of 16. Regardless, a significant difference from John’s test, where the Aerodrink actual improved aerodynamics, unlike our test. This is likely due to the advancement of frame and component technology in the past 7 years. Modern frames have been engineering so precisely, that it truly has become a risk to add anything onto the frame. This is consistent with the claims of bike manufactures, like Cervelo, who have long said their bikes do better without typical hydration systems.


–         Why is the large Aquacell the best performer, the small Aqualite a neutral performer, and the classic, medium, and original Aerodrink the worst of the 3? It does not seem to make any sense. Should not the results be somewhat sequential based on size? I guess that is why we go to the tunnel. Wind works in funny ways. On the Plasma, it just worked out better with the largest and smallest up-front systems.


–         The difference in the Xlab Carbon Wing with and without CO2 is remarkable. We ran the test specifically with and without CO2, confident that the CO2 would increase drag, and we were wrong. The CO2 was positioned “perpendicular” to the ground, with the CO2 cartridges pointing up and down, not sideways. Does this help the airflow around the evil cylindrical water bottles somehow? Do they act as some sort of strange wing as the air flows off the back?  John Cobb took the time to comment on the article and said, “I believe the Co2 placement and shape just happen to work well for this rider by giving a shadow extension to the body at the hips and helps stabilize the airflow off the rear.” The overall distribution of the Carbon Wing alone (with CO2, without CO2, with bags, without bags) is unsettling at best.


–         I was also wrong at how significant it was to come out of aero position to drink. Yes, the drag numbers are horrible when coming out of aero to drink, but not as horrible as I anticipated. You instantly slow down by at least a kilometer per hour for the same wattage. However, since you are only out of aero for a very short time, the result is minimal. 10 seconds over a 180k ride is hardly a deal-breaker when considering a hydration system, and I’ll have to stop my soap-boxing on coming out of aero to drink. What we did learn is that it is much better to keep one arm in the aeropads when drinking from a bottle.


–         There has been heated debate over water bottles in the seat tube or down tube position. John’s results indicated a standard bottle on the down tube improved aerodynamics, while our test had the opposite result. The truth is it depends on the frame, rider, and bottle. For example, our test had standard water bottles perform better in the seat tube, the Speedfill also performed better in the seat tube, but the Razor performed better in the down tube. We could not find consistency between multiple systems on whether seat or down tube is better. I’m confident this debate will rage on forever.


–         Would the Speedfill have performed better without the tubing? I regret not running that additional test. We did our best to keep the tubing close to the frame, and a minimal amount of tubing sticking up on the front end. However, note that the Speedfill did better on the seat tube, where there was more tubing threading its way along the frame then when positioned in the down tube position.


–         This is the 3rd time I’ve said it, but I’ll say it again. This is a preliminary result. Kind of like a top 25 ranking before the season is finished. However, I’ll be very surprised if the rear-mounted systems do any better at the additional yaw. I can see a potential improvement in the front-end and high-tech frame-mounted systems, but I just don’t see how the rear systems will do better with those huge cylinder water bottles sticking out for the wind to abuse them. I’ve been wrong before. I should also note that the rear systems will be the most influenced by rider position, compared to front-end and frame-mounted systems. How upright the rider is will have minimal effect on hydration systems mounted on the first 2/3 of the bike, but would significantly change the rear-end systems’ performance.


It would be tempting to use this information to justify stripping your bike of all hydration systems. Although not really an option for 70.3 and IM racing, it has some appeal for Sprint and Olympic, as it appears that almost any hydration system will slow down you high-end bike. However, the athlete must carefully weigh the aerodynamic advantage of this move compared to the risk of taking in no fluid for 30-75 minutes of their race. It also presents the challenge of getting in nutrition outside of a liquid source. This is not difficult, but it is tough to do without compromising aerodynamics even further. GU flask on the stem? Gel packs on the top tube? Nutrition in the pockets of your tri suit? Clif bar pieces stuffed in your cheeks? While you might reduce anxiety that your hydration system isn’t messing with your aerodynamics, the alternative method could be just as bad, with the additional risk of no fluids during the bike.


In defense of the systems that came in at the bottom of the list, let me close with a commentary on selecting a hydration system. I hope that the reader will pause before choosing a hydration system on aerodynamics alone. Aerodynamics is just one element of several, including capacity, torque, frame compatibility, rider positioning, cost, and pride.


For example, you may be tempted to go out and buy an Aquacell based on these results. However, the Aquacell may be the wrong choice for many riders, regardless of aerodynamic performance. For example:


–         Riding with a Zipp 808 or 3-spoke up front, with a stem of 100mm or greater, plus a large front-end system is setting the rider up for a bad day. That is a lot of torque, should you experience a cross wind, and hadling becomes a serious concern.

–         The ease of filling the Speedfill (in my opinion) is superior to the front-end systems.

–         Have you ever managed to not get completely soaked on a 100+ mile ride using one of the Profile Design front-end systems? Don’t get me wrong, I love Profiles systems. I ride exclusively with the Profile Aqualite. But even with the new design, I can’t go more than 25 miles without looking like one of those Gatorade commercials. I’m dripping with florescent fluid. My last Ironman at Coeur D’Alene, right after a set of nasty railroad tracks that gave you a good jolt, there was a graveyard of yellow sponges off to the side of the road.

–         The Aquacell is so big, you may have to change your positioning in order to get it to fit between your aerobars. Getting less aerodynamic in order to put on a system that might be more aerodynamic seems odd at best.

–         At the risk of sounding like a snob (I am a snob), I just can’t stand the thought of riding with that thing, and having my picture taken with it on the bike course. The Aquacell is HUGE. It just may not be worth the extra 25 seconds over a 40k to sacrifice my pride. I’ll just run faster.


In the sprit of fairness, I should also pick on another system. The Xlab Carbon Wing without bags and with CO2 performed pretty well, but consider these drawbacks:


–         The flying dismount becomes perilous with the additional required clearance due to a rear-mounted system. This is less of a concern for IM racing, but for 40k racing, you better make sure you can clear the system with your leg.

–         While the risk of losing some fluid is small with the front-end systems, the risk of losing the entire bottle is greater with the rear end systems. It takes practice and the right kind of cages to keep those bottles in there, especially since you can’t see where you are putting them. I guess if Faris Al-Sultan can stick it in his speedo without looking, we can learn to do the same (note to self: add water bottle placement in speedo for second round of hydration testing).

–         While I have already calculated the minimal cost of coming out of aero to drink, it is possible double those numbers for the rear mounted systems.

–         The carbon wing is pricey compared to other hydration systems, and you have to supply your own cages and bottles even after you buy the system. It can easily be more than $200 for a complete Carbon Wing setup, compared to $20 for an up-front system.


I could continue this analysis for each and every hydration system, but I hope I have made the point that aerodynamics is simply one element to consider.


We look forward to completing this project with the additional tests. Look for another installment regarding our tests on apparel coming out next month. I want to thank John Cobb for being the inspiration for this project, and although I have not done justice to his original and accurate study, and I will never come close to matching his expertise and credibility, I’m grateful for his influence.


Moving on. Last month the Ironman World Championships were rebroadcast on NBC, and just before that broadcast I was able to give Chris McCormack a call. In researching his career, I found many things I did not know about Chris. Love him or hate him, he’s arguably the most successful all-round triathlete in the history of the sport. From ITU to Ironman, his dominance at multiple distances makes him a fascinating individual.


That’s all for this episode. I’ll be back next month with a review of our apparel tests in the wind tunnel. Thanks for letting me spend some time with you, and I’ll see you next time.