Testing the Parlee ESX

Set-up

As I rode my Ritchey Breakaway, with its mix of SRAM Rival, Force, and Red components, to Bespoke Cycles in San Francisco I felt a bit of a fraud. It was Parlee demo week at the shop, part of Parlee's program to publicize its new aero bike, the ESX (web page). More info is in this RoadCycling review.

Designed by Parlee, the frame is built in Taiwan, sold with frame, fork, seatpost, and direct-mount brakes for $5400. That's a lot, making this particular slot in the demo schedule, the weekly Bespoke shop ride, a potentially valuable one for both Parlee and Bespoke Cycles. But it didn't seem probable I was going to buy it. True, nobody had ever asked me if I was planning to buy the bike, but there's surely some moral obligation to have some minimum intent to do so. As I pedaled my Ritchey through the nearly empty San Francisco streets for the 8 am meeting at the shop, however, I wondered what fraction of high-end bike purchases went to those who had been "unlikely to buy". It's a bit like when I feel guilty about taking free samples in Whole Foods, then walk out of the market with $75 more on my credit card bill.

Ari at the shop set me up with the bike. There were four of us with demos, one to one of the shop employees, two more to other customers. Ari used the Ritchey as a template to match my position. He apologized for not being able to match the handlebar drop on the Ritchey due to the big-stack setup on the size-small Parlee. That could have been fixed: the Parlee has a head tube extender (a spacer molded to appear to be part of the frame) and an additional shaped spacer on top of that. But time was limited, with the ride scheduled to leave @ 8:30 am.

So I felt like I was riding a comfort bike. Some amount of increased bar height can be compensated with bent elbows, but this was a bit excessive. But it was a good test, because the result was surely going to be more, rather than less, wind drag relative to the Ritchey.

Digression: Parlee wind tunnel data

Here's Parlee's wind drag data:

If I assume an air density of 1.2 kg/m³, and a test speed of 30 mph (standard for the MIT tunnel, and consistent with power data from other plots on the Parlee page), then the drag of 1.75 pounds of force corresponds to a CdA value of 0.072 meters squared, while the Parlee Z5 with a drag of approximately 2.25 pounds of force corresponds to a CdA of 0.093 meters squared. The difference is 0.021 meters squared.

These results can be compared to the tests done by Scott Bicycles, also with waterbottles and also without a frame. I described those results here. In that test, the foil came out with a CdA of around 0.074 m², while the "round-tube bike" was around 0.087 m². In extracting these values, I eyeball the variation from a yaw of 0 to 10 degrees, since due to ground sheer on terrain with ground features like trees and buildings, typically low yaw values dominate at the height of a bicycle frame. This difference is 0.013 m².

I've also looked at results from VeloNews. There a Cervelo S5 compared to a Masi, both bikes with Fulcrum wheels, the near-zero-yaw CdA values were 0.088 m² for the Cervelo versus 0.104 for the Masi, a difference of 0.016 m₂. Note the Cervelo S5 was also in the Parlee tests and tested similar to the ESX.

So in summary, the comparison of the ESX and the Cervelo S5 compared to the Parlee Z5 shows a greater advantage to the aerodynamic frames than does the VeloNews test comparing the S5 to the Masi, or the Scott comparison of the Foil to what is presumably an Addict or similar road frame.

Conclusion: the claimed aerodynamic advantage of this bike is considerable. A typical pro racer has a CdA of around 0.32 m² according to Tour magazine tests (Veloclinic has reported 0.35 m² is a good match to climbing data). A difference of 0.021 is 6.6% of this. If wind drag is 90% of total power, as it is on the flats, that's a 6% net power advantage, which is the same advantage as has been reported in experiments using EPO on runners, according to the Science of Sport blog. Nevertheless, I wasn't going to see any such advantage on this ride due to my upright position.

The bike

The bike was set up with the following:

1. Ultegra Di2 "Ui2" electronic shifting.
2. Ultegra direct-mount brakes, a conventional position on the fork for the front, a chainstay mount for the rear
3. Ultegra 165 mm crank. I was surprised and pleased when I realized it was 165, which I didn't check until most of the way through my ride.
4. Shallow drop & reach 38.5 cm/40 cm handlebars (narrower at the hoods, wider on the drops).
5. Ultegra Al-rim clincher wheels with 25 mm Michelin tires. I would have liked to use my Powertap wheel I had mounted on the Ritchey but that was 10-speed and wasn't compatible with the 11-shifting on the Parlee.
6. Barfly forward mount for my Garmin Edge 500.

The ride

We rolled out from the shop on Clay Street and crossed the Golden Gate Bridge. There's almost always some cross-wind on the bridge, but the real cross-wind test came immediately after: the descent of Alexander to Bridgeway. With gusting winds sweeping across different relative headings as the road curves, this section can get a bit sketchy. A common concern about aero frames is how they'll handle cross-winds, but the Parlee passed this test no problem. The key is to avoid "stalling", where the wind force increases substantially when yaw angle increases a relatively small bit, and poorly designed aero bikes can experience this which yields instability. The Parlee was great, though. If anything it was better than my Ritchey with its 1-inch steel tubes. Besides the aerodynamics of the frame, geometry (trail) and wheels also play a large role in cross-wind handling.

The next test was on the same section of roads: rough pavement. Aero frames have a reputation of being harsh. No problem here: while perhaps not as plush as some carbon bikes, including perhaps the Parlee Z5, it was fine. Perhaps part of this was due to the seat-post design, which has a double cantilever design near the clamp to provide for vertical compliance. The 25 mm tires were if anything slightly narrower than my usual 26 mm Grand Bois tires on the Ritchey, so it wasn't the tires. Part of this is acoustic: the Ritchey tends to rattle with the cable couplers hitting the frame on rough roads. But the feeling was the Parlee was taking out the high frequency component of the vibrations.

The next test was climbing. We hit Camino Alto from Mill Valley, my favorite 5-minute fitness test, and I left the group to try for a good time. The wind felt fairly neutral (best times come from a strong tailwind, but a strong headwind is fatal). The bike felt really good going uphill. I was in the drops due to the high bars and shallow drop. The 165 mm cranks spin easier than 170 mm. But there were two additional factors which were nice. One was the electronic shifting. On the rear, no real difference, but on the front it was easier to drop from the big ring into the little ring on the one steep section, then shift back when it was done, relative to my usual SRAM 10-speed mechanical shifting. This wasn't that big a deal: on the SRAM instead of going big-small-big in the front I might have gone big-small then upshifted in the rear instead after the steep bit was passed. But getting back up into the big ring right away was preferable.

Another advantage on the climb was how it handled the rough pavement. Bouncing over a rough surface disruptive to power generation and I felt as if I could keep the power higher on the Camino Alto pavement.

Then there was the bike mass. I unfortunately didn't weigh it, but it was almost certainly lighter than the steel Ritchey, since the Parlee frame is spec'ed at 950 grams for size medium, and I was on a small. Everything other the frame and fork and rear wheel (Powertap on the Ritchey) was likely heavier on the Parlee, but the Ritchey is around 1600 grams for the frame, 8.1 kg with components, so I'm going to guess the Parlee was maybe 7.5 kg. One downside on the climb was enormous bike traffic: I had to work my way around or through several large groups of riders. Despite this I posted an above-average time (13/34 of my efforts on Strava) for the climb, despite less than impressive fitness right now and despite having run the day before. So it definitely passed the climbing test.

After the climb came a technical descent of Camino Alto to Corte Madera.

I waited for the others to reach the top, then followed them on the descent. It seemed to descend fine. This was a combination of the smooth wind flow and the handling of the rough pavement, I'd suggest. The trail, which I calculated from the geometry data to be 62 mm, was unexceptional: between my Ritchey and my Fuji. No-hands it rode fine.

The rest of the ride included pacelining out near China Camp over rolling terrain, then riding back via Larkspur using bike paths and a relatively new bike/ped tunnel. There was some delay due to multiple flats, including Ari blowing a tire off his carbon rims twice. But that was all resolved and after a peak-tourist-season crossing of the Golden Gate Bridge, we were soon back at the shop, a too-soon end to what was a wonderful ride.

General impressions

1. The Bar-Fly mount is great. I really liked having the Garmin in a more forward position. It reduced the perturbation to my normal position to look at the Edge display, keeping the road ahead in my peripheral vision, which I view as a major safety advantage as well as a convenience. I'd get one for each of my bikes except my road bikes have 26 mm bars. The Garmin forward mount has an adaptor for 26 mm, so I may look into that, but it's considered inferior to the K-Edge or Bar-Fly.
2. The electronic shifting was nice, but I don't think it was substantially better than my mechanical system in the rear. My pre-"yaw" SRAM front shifting, however, is inferior. Most important is the worry about throwing the front chain. That happens a lot less since I got a Deda chain keeper, but the chain keeper isn't perfect, and in particular it shifted on me when I recently traveled with the bike, something I didn't realize until I threw the chain on the Wednesday noon ride last week. For a time trial bike, I think the electronic shifting would be a greater, but still small advantage.
3. The 165 mm crank was nice for spinning, but I like having that little bit of extra range of motion on the 170's. At one time I used 167.5 mm, which work well, although 170 may offer a small advantage on steep roads. I'll stick with 170.
4. I like 38-39 cm bars: they fit my shoulder width well. The flared bars with 38.5 cm on the hoods and 40 cm on the drops, were okay, but I felt slightly spread out when on the drops. I like my arms to extend out straight from my shoulders. I feel this is the most stable position.
5. I didn't like the shallow drop. Why bother moving to the drops if my position barely changes? And as I noted I'd have preferred the bars being several cm lower.
6. The brakes were fine. In theory the direct mount brakes are stiffer and lighter. All I know is they worked as well as I could want, including the chainstay-mounted rear brakes.

Post-ride

I was curious what it would be like going from this top-end $5k+ frame with electronic shifting back to my modest Ritchey Breakaway, the frame/fork purchased in 2008 for around $1.1k including a travel case, topped off with a mix of SRAM parts of various vintage. Would I get on it, take a few pedal strokes, and think "oh, my, what a dog! I can't believe I'm riding this thing?"

Quick answer: No. The Parlee was a wonderful experience, for sure, but the Ritchey was fine as well. The better position helps, of course, but I don't feel there's much sacrifice in riding the Ritchey except for the additional mass and theoretical higher wind resistance, both of which show up only primarily in Strava segment times but not so much in the ride experience.

Super-thanks to Ari and Bespoke for hosting the Parlee demo and for setting me up with the bike! I'm definitely a Parlee fan now. They know their stuff.