So, one goal for today's rides was to put in a little down payment on that commitment. It was time. I either needed to get back on the bike, and prove to myself I was going to be able to get through this, psychologically, or....well...the alternative isn't pretty. I'd be pretty embarrassed to sell the Carbent just because I'm too lame to develop the skills for it. It was time to 'get back on the horse', so...today, I made myself do it. I was pretty wobbly at first; which was a big surprise to me. I have tens of thousands of recumbent miles in my legs, and really...plenty of hiracer miles (although I'm new to riding a hiracer at a 23 degree angle), and this was just an indicator of how badly this crash shook me up. I literally couldn't start the bike on the very mild uphill slope of the street in front of my house. I had to walk it somewhere flatter and get started. Of course, after a bit of riding, this all went away; it was all in my head, but still....very real, and daunting.
The second goal for today was to try to acquire some evidence to help ascertain how effective aero wheels (specifically, my Zipp 404's) are aerodynamically. I was waxing lyrical in a recent Bentrideronline thread about how I like these wheels, and how much faster I think they are than light wheels. A skeptic (Joel Dickman) piped up online, quoting a some-years-old Pete Penseyres article. The article isn't available on line, but the gist was that Pete found only a .5 mph benefit from the aero wheels of that day at 25 mph speeds. Joel's point was that, if this is so, for most of us, these very expensive aero wheels are a poor investment as a performance aid.
Of course, I knew this was wrong, right? I knew my fancy, expensive Zipp 404 rims were fast. This is actually a 28 spoke 'Clydesdale' set of Zipp 404 clinchers; the rear has been rebuilt onto a PowerTap hub. So I was set with aero wheels front and rear.
Of course, now I needed a set of non-aero 700c wheels, with PowerTap rear, to compare with, right? I got a good deal on Ebay for a PowerTap 'Elite' hub (the cheapest one) at 32 holes, and I got the seller to build it into a Velocity Aerohead rim. So now I had both non-aero and aero wheels with wattage measurement. I borrowed the front wheel from my roadbike; it's a Bontrager paired-spoke affair, very similar to the old Rolf Vector Pro wheels; of course, if I had my druthers, it would have been a 32 spoke 3 cross Aerohead rim, as 'conventional' as I could make it. However, this was what I had handy, so this is what got used.
All four wheels were mounted with Schwalbe Durano Plus tires on them (I don't care if the tire is a little slower than an Ultremo or other sub-250 gram race tire; I don't want flats). All four tires were pumped to exactly 120 psi (the Zipp rims have a max inflation spec of 125 psi).
Unlike my normal practice of riding from my door, I plopped the Carbent on the back of my beautiful 1999 Honda Civic CX with cheapo bike rack strapped to the hatchback, and drove to Leschi (Seattle neighborhood that's a convenient starting point for these tests). I started the runs with the conventional wheels. The test was my same old loop from Dearborne to Seward Park and back. It's about 8 miles or so of indifferent quality road, with a small amount of climbing.
A note about test rigor and repeatability: If I seriously wanted to establish rigorous certainty about this issue, multiple runs for each configuration are a requirement. In fact, possibly a better approach would be to use Robert Chung's Golden Cheetah software in a windless venue. This is one of the many things I haven't gotten around to, mainly because I can't figure out where/how to do windless testing. The thing about this 8 mile run is that it's an out and back, and the wind almost always is either a nearly direct headwind or tailwind. It's along the western edge of Lake Washington, 4 miles south, then turn around and return to the north. So my hope is that wind effects tend to cancel themselves out; not very rigorous or certain, I know. But that's the best I'm offering at this juncture :). This testing is mostly for me, and anyone else is welcome to derive whatever benefit they want to from it, but no warranties are expressed or implied; I'm just trying to make it obvious from the getgo that I'm aware my testing protocol is far from perfect. It is good enough to offer some value to me.
Anyway, as I said.....umm....multiple runs, yeah. I did one each of these 8 mile runs, maintaining 225w average according to the Garmin Edge 705. One run was done with the non-aero wheels, and one with the Zipp's. The Edge was calibrated to the power meter for the rear wheel prior to each run. Multiple runs would've been much better, but I was really happy to have performed these runs on the hiracer without wigging out, and I felt much better about things (as regards my relationship with the hiracer) when I was done, and I was also very ready to get off and get back on a bike I feel safer on :). I'll do a few more miles on the Carbent tomorrow; I think I'm doing well, and I'm keeping the bike (to all the folks in the peanut gallery hoping this was yet another bike I was going to sell at a loss to some lucky beneficiary) :).
So here are the results:
| Time | Speed | Watts | Distance | Ascent |
|---|---|---|---|---|
| 21:35 | 21.7 | 225 | 7.81 | 78 |
| 21:32 | 21.7 | 225 | 7.80 | 84 |
So...I'm not going to try to say this 'proves' anything, but I'm certainly not going to dispute it. I'm sure Joel (Dickman) will be amused, since it tends to support exactly what he was claiming (that aero wheels are a waste of money for most recumbent folks).
I am not convinced that this was enough runs to really establish certainty on this point, but I have to say that it certainly casts serious doubt on the proposition that a pair of Zipp 404 wheels are a hugely dramatic unambiguous performance win; I mean, the runs were a dead heat :). Oh well, this is just one more in a long list of things I've been wrong about in my life, most likely, although I am sure further repeats of this are necessary to understand just exactly what's going on here. And of course, if we really care, a windless venue for testing would be far preferable. But....at some point, when the testing becomes so much trouble to eliminate variables so you can even detect a difference, maybe this means things are so close that it's not worth worrying about, for us mere mortals who are just doing this recreationally (as opposed to folks who have money riding on the results of their race).
Next, I took a deep breath, thankful that I got through this test without mishap, put the Carbent back on my Honda, and drove it back home. I felt good about the threshold riding I'd done, and just didn't feel like pushing it any further.
I got on my EasyRacer's C-Rush (even newer than the Carbent), equipped with carbon/kevlar Double Bubble fairing, and bodysock, and rode from my door down to the same location. I'm an old hand on this style of bike, and am more comfortable riding this style of recumbent in traffic than any other style. So, just for grins, I thought I'd perform the same speed test one more time.
Here are the results:
| Time | Speed | Watts | Distance | Ascent |
|---|---|---|---|---|
| 21:20 | 21.9 | 225 | 7.78 | 87 |
Some people will no doubt say that the C-Rush vs. the Carbent is not apples and apples, indeed, in HPV racing, the C-Rush would go into the 'SuperStreet' class (I believe), and the Carbent is 'stock'. There are no aerodynamic devices on the Carbent at all (unless the little Bentup Cycles seat bag counts as a rear fairing; perhaps in HPV racing I'd have to leave it off). The C-Rush would fare very poorly against other 'SuperStreet' vehicles that are faired to the limit of the HPV rules with no regard for how 'streetable' the vehicle actually is. For example, a hiracer with a Mueller XT fairing, Terracycle scaffold, and bodysock (these are now somewhat hard to come by), is a faster setup than the EasyRacer's socked/faired setup. But it's much less user-friendly, both in traffic, and in high crosswinds. On the EasyRacer's bike, the fairing and sock actually improve its behavior in a lot of ways, to my perception. Without a doubt, at low speeds, these LWB's tend to exhibit that 'uphill wobble' syndrome, modulo the skill and familiarity of the rider. The front fairing alone possibly exaggerates this, since it makes the front end 'heavier', and tend to 'flop' more. But put on the sock, and the steering is dampened such that it's much easier to go straight slowly up a hill. Just as an experiment, subsequent to this speed run, I put the C-Rush in its lowest gear, and toodled up a pedestrian/bike path near my house that's a double-digit grade. It crawled right up the hill with minimal wobble. Of course, I'm an experienced pilot for this platform, and it agrees with me; I know that not all will find it this way.
Still, for me, the EasyRacer's geometry represents an almost ideal tradeoff of all the various factors involved in producing a 'streetable' bike. Whether the bike is a little faster or a little slower than the Carbent, or my M5 TiCa's, the fact remains that, with the bodysock, it's really quite close in speed, close enough that it's going to be tough to figure out which platform is really fastest. And it is, without a doubt, a more 'streetable' platform than any other bike I've owned (which is saying something, I think).
All this is just my perceptions and opinions, and of course yours (dear reader) may be utterly different. Part of the beauty of bents is their diversity. But I was really quite surprised the C-Rush turned in my fastest performance today, for that 225w input.
Does this mean the C-Rush is 'fastest'. No, of course not. Much more testing would be necessary to figure out for certain which of my bikes is the fastest one. But one thing is clear; the Carbent, TiCa, and faired/socked C-Rush are all within spitting distance of each other.
