pm124

How fast you can go at any given power output is determined by your wind resistance, your rolling resistance, and the stiffness of the frame.

This speed calculator is the best that is out there, but it is a bit out of date. Bikes, tires, and wheels have all improved since then, but it is still the only way to figure out your speed at any give power output or vice versa.

Can people post some data showing how far off the calculator is? That is, if you are running a power meter and just had a run at 25MPH at 230 watts, but the calculator states that you should have been going 23 MPH (or putting out 250 watts) can you post the discrepancy? Please also post your bike and tires and your riding position.

Recumbent riders are posting speeds of 30MPH at 200 watts on some bikes. This calculator has them going just 25MPH, but the cda and rolling resistance have changed considerably on modern bikes.

Doge

TTT today averaged over 35mph.

RChung

I'm not terribly fond of that calculator. The assumptions it makes about CdA tend to run high for me (and it appears to underestimate CdA for recumbents): for example, for riding on bar tops, it estimates that my CdA would be around 0.49 m^2, while on bar tops my CdA is close to 0.4 (i.e., 20% lower than the estimate). Because CdA is less important when going slow, its power estimates while climbing are better than its estimates on the flat. OTOH, its estimate about Crr tend to be optimistically low for the kinds of roads I ride.

pm124

Yes, indeed. That's why I'm hoping to get people to post their data. With these data, it might be possible to improve the calculator!

pm124

Can you provide data on your power and whether you were on the drops of a road bike? What tires?

RChung

The problem isn't in the calculations per se -- those are well understood if not well known. The problem is that he used very rough physiological models to get from height and weight to CdA. Since I do a fair bit of CdA measurement, I've seen guys whose CdA has dropped by more than 20% from positional and equipment changes. I've also seen guys with the same nominal height and weight with a range of CdA on TT bikes from .20 to .30; more than that for road bike positions. As it happens, I've also done some measurement of CdA (and Crr) for recumbents. There is a lot of range in drag there, too (though, as I said, his assumptions about recumbent drag appear optimistic, and his assumptions about conventional bike drag appear pessimistic, so his calculations of power for speed tend to be lowish for recumbents and tend to be highish for road bikes).

As more riders have been measuring their drag parameters (both CdA and Crr) it has become less necessary either to make broad brush assumptions about their drag or to use out-moded models to ballpark them. To improve the Kreuzotter calculator, you should remove parts of it.

pm124

@RChung--So, just to be clear--are you saying that the recumbent calculations are over-estimates or under-estimates of the actual CdA? And what about modern day Z bikes with carbon frames? What about tailbones?

Also, are you suggesting that the assumptions need to be changed?

Doge

The TTT today was the Team Time Trial in the Tour de France. The speed is a matter of record, but I do not have their power data.

Doge

I do have one with power data tied to position and tires - only as it is in video. We use the hub Power Tap at the hub which measures a bit lower than at the pedals.


Rider is male age 16 (by 2 months) ~130#. The kick here is 1,400W, power tap hub, zipp 404 tubular on Veloflex 23mm tires. Front is most likely a 16 spoke Nimble fly with same tire.

https://www.strava.com/segments/764232 You can click through to the power profile.

RChung

I don't think I've ever measured the CdA of a LWB USS recumbent -- they're pretty rare nowadays. As you may know, there's a lot of variability in how SWB recumbents are set up; accordingly, there's a lot of variability in CdA for SWB recumbents -- the estimates given by Kreuzotter for both "racing" and "commuter" set ups appear to be lower than what I've measured. The Quest velomobile estimate of CdA happens to be pretty much right on what I've measured. I've measured a couple of lowracers, and they're slightly higher than Kreuzotter. I've never measured a lowracer with a tailbox.

As an aside, the CdA of riders on UCI-compliant TT bikes I've measured is lower than what I've measured for lowracer recumbents, but those riders have been highly optimized. (The CdA of riders on non-UCI-compliant TT bikes is lower than that).

I haven't looked at the code so I don't know if these are assumptions or calculations of the recumbent CdA.

[Edited to add:] Not that this has anything to do with Kreuzotter (may he RIP) I estimated the CdA of the Varna Diablo, the last of the record-holding non-camera streamliners. Its CdA was < .03 m^2. The VeloX and Aerovelo eta, being camera bikes, would be lower than that.

HTupolev

Not to a degree that's very significant compared with the huge inaccuracy of guessing CdA based on height, weight, and rough bicycle type.


Differences in power transfer between frames are so small that, to my knowledge, nobody has ever published measurements of them.

Frame flex behaviors appear to have some physiological impact on pedaling, but it's more complicated than stiffer-is-better. Otherwise, steel touring bikes designed for heavy loads ought to pedal nicer than lightweight skinny-tubed steel racing frames, but people tend to have the opposite experience.

I'd guess it depends a lot on the particular setup. One of my friends recently adjusted his Bacchetta CA2 high-racer to a slightly more upright comfort posture... the difference wasn't visually striking, but the little draft pocket behind him was suddenly much easier to use. When he had the CA2 in the more aggressive posture, it at least *seemed* like he got a larger gain from it over his road bikes than people usually get between their road bikes and TT bikes.

RChung

Yup. Exact setup and exact equipment can make a pretty big difference. For UCI-compliant TT bikes and positions, your upper arms and lower legs are pretty much cylinders, and cylinders are notoriously non-aero. It turns out that the right sleeves on your arms and the right socks on your legs can make a big difference. The right fabric is way more aero than bare arms or legs. The right socks alone can make a difference of .01 m^2 in drag area, because the legs aren't just cylinders, they're churning cylinders. That's why the Tour was measuring sock height and enforcing sock height rules for yesterday's TTT.