Wigans goes there. Cadence!

[Dear Wiggo]

Re: Re:

plus the yaw angle of riding on the road and the CdA is not possible to measure. Especially in the gusty lowlands where you might be subject to winds switching or the road switching around and reversing back on itself.

We are talking about indoor velodrome / hour record.

I test aerodynamics for a living. We readily attain CdA values with a standard error of 0.001m^2 (which is ~ 0.5%). With experienced riders on the track we can nail it a bit tighter.

As for Boardman, I'll look at posting W/m^2 with error range since it seems to be of so much interest. That'll have to wait until later when I have my modelling handy.

My point was a relative one. Boardman's CdA was lower than for other hour record riders. Superman positioning was significantly more aerodynamic, that's a known fact for track, and one that's pretty obvious when you compare Boardman's previous hour record performed in a more Rominger-like position/bike when compared with his Superman position record.

I made no comment about anyone's doping status, other than to say doping affects the supply side of the hour record equation, not the demand side.

Put another way, speed for the hour record is primarily a function of a rider's W/m^2, and air density.

Doping only affects the W, not the m^2 nor air density. Not sure what's so contentious about this pretty straightforward fact.

How did you arrive at a value of 0.18 for Boardman?

Did he give you his power files?

Do you know if his power meter was calibrated properly?

Do you know the temperature, pressure and humidity of all his training and hour attempt?

 

[Fearless Greg Lemond]

Re: Re:

My point was a relative one. Boardman's CdA was lower than for other hour record riders. Superman positioning was significantly more aerodynamic, that's a known fact for track, and one that's pretty obvious when you compare Boardman's previous hour record performed in a more Rominger-like position/bike when compared with his Superman position record

What distance would Boardman have done using Rominger's position? Can you do the math for that one?

 

[Alex Simmons/RST]

Re: Re:

My point was a relative one. Boardman's CdA was lower than for other hour record riders. Superman positioning was significantly more aerodynamic, that's a known fact for track, and one that's pretty obvious when you compare Boardman's previous hour record performed in a more Rominger-like position/bike when compared with his Superman position record

What distance would Boardman have done using Rominger's position? Can you do the math for that one?

I'll give it a go, but make some comments in prep as well.

TLDR version:

Without the aerodynamics improvement made through a lot of position and equipment experimentation and aero testing, Boardman would have fallen short of Rominger's record by a substantial margin (1.6km +/- 0.3km).

The aerodynamics improvement from the Superman position and better bike gained Boardman something like 2.0 - 2.5km of the 4.1km difference between his 1993 and 1996 rides, with the majority of the balance (1.6 - 2.1km) being from an improvement in power, and some smaller variations due to environmental differences.


Long version:

Two riders of course can't ride in exactly the same position, just similar positions, and of course even if positions are exactly the same, riders have different aerodynamics due to morphology / body shape etc. An object's shape affects aerodynamics a lot.

Plus the bikes were different, e.g.:

Rominger's bike had narrow steel tubes (narrower than today's regs permit) and disks front and rear.

Boardman's 1993 "normal" pursuit position was on Corima bike and Corima 4 spoke wheels front and back, while his "superman" position in 1996 was on monocoque (Lotus/Merckx) bike, rear disk, Mavic 5 spoke front. Even in a normal position, the 1996 bike was somewhat aerodynamically better than the 1993 bike in low yaw indoor track conditions.

Very few riders have formally tested superman v "normal" aero (pointless given regulations effectively prevented such positions not long after they appeared), but one I know of that tested in variable yaw conditions attained a reduction in CdA of 7% at zero yaw (and as yaw increased, so did CdA. At 10 degree yaw, the normal position was better). This is a rider with CdA in the 0.20m^2 range, a similar CdA to Rominger, or Boardman's 1993 set up. IOW I'm not talking about a rider who had crappy aero to start with, they were already very aero.

Boardman did 52.27km in 1993 riding a more normal aero position (i.e. similar to Rominger's position) on his Corima. For his 1996 Superman position record on the Lotus he attained 56.375km. Boardman's 1993 power was less than his 1996 power, so a proportion of the gain was from more power, and a proportion due to aerodynamics improvement, and some variability due to environmental differences: air temps were similar, but I don't have data on air pressure differences (there will be a typical range) and a little due to track/tyres but these are small (I think same tyres but someone can correct me - in any case the tyre Crr differences at the pointy end is small).

Based on best info I have right now, Rominger's CdA was ~10-11% (~0.02m^2) higher than Boardman's Superman position CdA, but it was lower than Boardman's 1993 set up by ~5% (0.01m^2).

Using Boardman's 1996 known repeatable SRM measured power from training, and with known environmental conditions on the day of his ride, we have his hour power pegged at ~442W (I'll leave error bars out for now, and add them later into distance estimates).

Assuming same track, environmental conditions and power as for his 1996 ride:

If Boardman the same CdA as Rominger had, Boardman would have ridden ~54.5km +/- 0.3km, or ~1.9 km less, and of course ~0.8km less than Rominger's ride.

Had he used his 1993 set up, Boardman would ride ~53.7km +/- 0.3km, or ~2.7km less, and about 1.6km less than Rominger.

The aerodynamics improvement from the Superman position and better bike gained Boardman something like 2.0 - 2.5km of the 4.1km difference between his 1993 and 1996 ride, with the majority of the balance (1.6 - 2.1km) being from an improvement in power, and some smaller variations due to environmental differences.

Without the aerodynamics improvement made through a lot of position and equipment experimentation and aero testing, he'd have fallen short of Rominger's record by a substantial margin (1.6km +/- 0.3km).

 

[Alex Simmons/RST]

BTW

I'd put Boardman's 1996 superman CdA in the 0.181m^2 - 0.184m^2 range.

 

[Dear Wiggo]

Re:

BTW

I'd put Boardman's 1996 superman CdA in the 0.181m^2 - 0.184m^2 range.

Based on what?

 

[Alex Simmons/RST]

Re: Re:

BTW

I'd put Boardman's 1996 superman CdA in the 0.181m^2 - 0.184m^2 range.

Based on what?

Speed we know, wind we know, air density based on air pressure range as temp was known with a couple C (controlled environment track), track surface we know, Crr range, known hour power from training and drive train allowance of 1.5%.

Hence why I put it in 2420W/m^2 give or take 30W/m^2.

I note that Padilla, Mujika, Angulo and Goiriena, from this paper estimated Boardman at 0.1838m^2 although they list no error range in their methodology which was a combination of morphological assessment combined with wind tunnel, power and environmental data known for another hour record holder - IOW the W/m^2 for a given speed and air density can be nailed down fairly tightly in an indoor track.

 

[Dear Wiggo]

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

 

[Mr.38%]

Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

 

[Alex Simmons/RST]

Re:

How was speed determined?

Er, it was an hour record. Do you really need me to do the maths?

The splits are also available if you are that interested to look into it, although his pacing was excellent. Here are the 5km splits:
http://www.wolfgang-menn.de/hourrec.htm

Are you sure the PM was calibrated properly?

Peter Keen was looking after the track SRMs used by Boardman. Keen was fastidious in this respect, ultimately doing the scientific research and publishing the methodology used for calibrating the meters, and demonstrating the strong validity of the methodology.

I have no reason to doubt that the person that eventually did the research on the SRM calibration method used and was looking after the equipment used by Boardman, managed to calibrate his SRM properly.

Indeed it was Keen's work that helped set the standard in this respect. It's a shame more don't follow this basic protocol.

Where is the info for temperature on the day recorded, or any of this data that has been used?

Indoor the track temperature was recorded at 24.5 C although it may have lifted up to 2C during. This was as reported in Edward Pickering's book The Race Against Time. 24-26C is pretty normal for the climate control system at Manchester, and they have used 28C for some attempts.

Barometric pressure on the day was 1023hPa - 1024hPa, that's in the weather records.

Altitude of track is 38m.

Humidity ranged from ~55-85% over that broader period of the day, although this only has a tiny affect on air density. It was reported at the time as dry air, so likely more at the lower end of that range.

That puts air density at 1.185kg/m^3 give or take a fraction.

Thanks for asking me to check, that's a fraction (i.e. not much) higher to what I had used today.

The power was known from Peter Keen directly.

My mass value might be out by ~1kg, as I have minor differences in total masses reported (body from Peter Keen). In any case, this has very little effect, with a difference of 1kg only changing overall average wattage demand by 0.3 of a watt (less than 0.1%).

The maths of the physics is the same as used and validated in the paper by Martin et al, including wind tunnel comparisons, and subsequent validation of track data.

So depending on what % you want to allow for drivetrain efficiency (a good road bike will be 2-3%, and track bike somewhat better given lack of guide pulleys and cross chaining), then I don't think my numbers are going to be all that far out.

 

[Alex Simmons/RST]

Re: Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

Unfortunately for Dear Wiggo I actually have solid data and scientific methodologies on which to base my estimates of power and aerodynamics. Which blows out of the water his claims of me being a hypocrite. An apology from him for his unfounded statement would be appreciated, but I'm not holding my breath.

 

[Dear Wiggo]

Re: Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

Report my posts or pull your head in.

 

[Dear Wiggo]

Re: Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

Unfortunately for Dear Wiggo I actually have solid data and scientific methodologies on which to base my estimates of power and aerodynamics. Which blows out of the water his claims of me being a hypocrite. An apology from him for his unfounded statement would be appreciated, but I'm not holding my breath.

Unforunately?

I think not.

If I had not asked you'd be expecting all these fine upstanding forumites to just take your word for the claims you are making.

If me asking for the information to back up your assertions is annoying then poor diddums Mr38% needs to go suck his thumb elsewhere.

 

[Dear Wiggo]

Re: Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

Unfortunately for Dear Wiggo I actually have solid data and scientific methodologies on which to base my estimates of power and aerodynamics. Which blows out of the water his claims of me being a hypocrite. An apology from him for his unfounded statement would be appreciated, but I'm not holding my breath.

Bottom line: you listed data (CdA = 0.18) with no error bars, explanation or backup data - everything you have whinged and belittled the so called "pseudo scientists" for. If you cannot see the hypocrisy then I cannot help you.

You didn't see the power files from Keen.

@vayerism does see the power files from riders on the road.

 

[mrhender]

Okay folks...

Keep the discussion civil, and quit the personal remarks..

Thank you in advance...

 

[oldcrank]

In the last few pages of this thread Mr. Simmons has contributed
some of the most informative, interesting and intelligent posts
in the history of the CN forums.

As we approach the London World Championships and the Rio
Olympics I would welcome more of Alex's wit and wisdom with
regard to the science behind the most beautiful of all cycling
disciplines, la persecucion por equipos.

 

[Knutsen]

In the last few pages of this thread Mr. Simmons has contributed
some of the most informative, interesting and intelligent posts
in the history of the CN forums.

Ditto. Simmons, please keep posting, your contributions are very much appreciated.

 

[ScienceIsCool]

Rominger had a ridiculous record. He used to be a childhood hero of mine, but looking back at Bicisport articles on how he overcame his "pollen allergy" thanks to the great Dottore' Ferrari; nothing short of hilarious. On the other hand, Wiggins, how am I supposed to take his attempt serious?

They are both a tad far fetched for me, but to say one is worse than the other?

How about Boardman? He beat the pollen man.

While doping can/does impact the energy supply, it doesn't affect the energy demand side of the equation.

With a CdA of 0.18m^2, Boardman did get very good speed for his power. At those speeds, each 0.01m^2 ~= 23W.

Ah. The science of CdA. Something Boardman did not have measured, agreed? Something you are guessing at, agreed? Based on an unknown power, agreed?

The irony is stunning.

Why are you not mentioning any error bars?

So very hypocritical.

WTF?

I'm giving exactly an indication of such variability in power estimates by pointing out the impact on power of differences of coefficient aero drag area, the biggest factor with such estimates at same venue.

Boardman's power was well known at the time since all his lead up training and testing, including his aero development work, was done at the track with an SRM track power meter. So we already know he was in the 0.18m^2 range.

It's also why I talk about W/m^2 values when it comes to hour records, since that range can be much more narrowly defined, while power can potentially vary over a wide range, e.g. as in this example for Jens Voigt:

Well it would be interesting to know what model you're using. Also, each and every measured parameter (temp, humidity, weight, Crr, etc) will have an uncertainty. These will propagate and yeah, you need some error bars.

John Swanson

 

[Opalius]

I'll third the opinion about Alex Simmons/RST's posts. Very informative.

Remember that there's an "ignore" function in this forum, it can have its advantages

 

[Alex Simmons/RST]

Well it would be interesting to know what model you're using. Also, each and every measured parameter (temp, humidity, weight, Crr, etc) will have an uncertainty. These will propagate and yeah, you need some error bars.

John Swanson

Yes, they will propagate if those values are unknown or have a level of uncertainty that is significant.

Hence it's also a matter of sensitivity analysis.

For example, as I showed earlier, the estimates of indoor W/m^2 are pretty insensitive to uncertainty with the mass value.

In that example chart on Voigt's W/m^2 I listed precisely the assumptions used, so that the range of variation in power/CdA ratio could be seen for a given speed - it was a sensitivity analysis for this specific element to demonstrate that if you know those other measurements, you can nail the W/m^2 value down to a reasonably narrow range. Naturally if plotting a level of uncertainty with any of the assumptions, you can then add error bars accordingly.

So even if you don't know either of power or CdA, you can at least nail W/m^2 down pretty well if you know air density and speed and couple of other factors. It's the same principle in play for climbing speed and estimates of W/kg, except in an indoor velodrome we can be far more certain about the (lack of) wind variable, as both climbing W/kg estimates and flat land W/m^2 estimates are quite sensitive to air movement.

Even indoors there is the tiniest of air movement in a velodrome caused by the rider themselves (think of using a toothpick to swirl a large pool of water). I've actually measured it at one track (Sydney DGV) and have data on its maximal extent. It takes about 10-15 minutes to reach equilibrium and it's not much. Building design will affect this element.

Elsewhere I have shown how uncertainty in the other factors can affect such calculations, e.g. as per this chart when I blogged about impact of air pressure on the distance a rider may attain for a given power output and CdA:

I've provided the numbers in the past on many of the individual components in various places, but it's an idea for a future blog post to do up a combined sensitivity table for each of the various inputs. i.e. how does uncertainty in "X" affect the W/m^2 required to attain a given speed on an indoor velodrome, or given a W/m^2 value, what's the impact on distance attainable?

 

[Alex Simmons/RST]

Re: Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

Unfortunately for Dear Wiggo I actually have solid data and scientific methodologies on which to base my estimates of power and aerodynamics. Which blows out of the water his claims of me being a hypocrite. An apology from him for his unfounded statement would be appreciated, but I'm not holding my breath.

Bottom line: you listed data (CdA = 0.18) with no error bars, explanation or backup data - everything you have whinged and belittled the so called "pseudo scientists" for. If you cannot see the hypocrisy then I cannot help you.

You didn't see the power files from Keen.

@vayerism does see the power files from riders on the road.

I gave a quick response pointing out that a large part of why Boardman went faster was because of aerodynamics. It's a statement of fact, and I gave a quick indication of why, i.e. that his CdA was very low and provided a value to give an indication of just how low, and that it was somewhat lower than Rominger.*

You then immediately assumed this was complete crap, that Boardman didn't test aerodynamics (wrong, he did), and that I couldn't possibly know such things about Boardman's power and aerodynamics, and then conflated this with my comments elsewhere about estimates of climbing power which are specifically claiming a level of precision whereas I was not.

Here's a tip for future reference: aerodynamics is real and it matters. It's not some made up nonsense.

Keen told us directly the power numbers (via my coaching business partner and we are aware of other data via power tests performed). I have no reason and nor do you to think he lied at the time. In any case, even if the power number was wrong, the W/m^2 value range calculated from other known data is pretty narrow and shows the power value and hence the CdA value quoted makes perfect sense.

And just who have I belittled?


* I included it for those that have an understanding of what a CdA number means as they would have recognised the relative contribution it made, but in the end I spelled it out for you. I don't mind really, because as often happens on threads like this, others find the information interesting.

 

[Dear Wiggo]

Re: Re:

How was speed determined?

Are you sure the PM was calibrated properly?

Where is the info for temperature on the day recorded, or any of this data that has been used?

Your annoying participation on the last pages does not add anything useful to the discussion and is just - annoying.

Unfortunately for Dear Wiggo I actually have solid data and scientific methodologies on which to base my estimates of power and aerodynamics. Which blows out of the water his claims of me being a hypocrite. An apology from him for his unfounded statement would be appreciated, but I'm not holding my breath.

Bottom line: you listed data (CdA = 0.18) with no error bars, explanation or backup data - everything you have whinged and belittled the so called "pseudo scientists" for. If you cannot see the hypocrisy then I cannot help you.

You didn't see the power files from Keen.

@vayerism does see the power files from riders on the road.

I gave a quick response pointing out that a large part of why Boardman went faster was because of aerodynamics. It's a statement of fact, and I gave a quick indication of why, i.e. that his CdA was very low and provided a value to give an indication of just how low, and that it was somewhat lower than Rominger.*

You then immediately assumed this was complete crap, that Boardman didn't test aerodynamics (wrong, he did), and that I couldn't possibly know such things about Boardman's power and aerodynamics, and then conflated this with my comments elsewhere about estimates of climbing power which are specifically claiming a level of precision whereas I was not.

No, I didn't assume it was crap. Unlike the ongoing dimissal of the VAM power estimation guys -- and you can deny you dismiss them but you do -- I didn't dismiss what you were saying at all.

I just pointed out that spitting out numbers without any errors, or explanation, or backup data, is hypocritical.

That's it.

As for not claiming a level of precision? Well, saying flatout that Boardman's CdA is 0.18 certainly gives the impression of an absolute level of precision. You can't see how that is implied by writing

With a CdA of 0.18m^2, Boardman did get very good speed for his power.

and that's fine, but to someone not swallowing everything you're saying without question, it certainly appears to be claiming a level of precision that needs at least some caveats, given the ongoing dismissal of the VAM guys.

Out of curiosity -- what was Boardman's measured CdA? And which wind tunnel did he do it in?

 

[Alex Simmons/RST]

Boardman in the wind tunnel doing some preliminary work on the Lotus, which eventually morphed into the bike design and position used for his 1996 hour record ride.

Source:
http://www.roadandtrack.com/car-culture/features/a7708/lotus-sport-108-secret-history-bicycle/

Then of course there was the six months of wind tunnel work Boardman did leading up to the 1992 Olympics:

Here's a video talking about the testing that was done with Boardman and the aerodynamicist Richard Hill and how that progressed to the Superman position:
https://www.youtube.com/watch?v=RMpMP1nu5jc

Sort of belies the claim that Boardman never did aero testing. He probably spent more hours doing aero testing in tunnels and at the track that just about any other rider on the planet (although I can think of perhaps one or two). I can't provide error bars for that statement though, so take it for what it is.

 

[Alex Simmons/RST]

Well it would be interesting to know what model you're using.

I missed this bit before. As mentioned and linked to in an earlier post, I use the Martin et al model (since it's been very well validated).

 

[Dear Wiggo]

So do we know his CdA of him on the bike he rode for the hour?

 

[Alex Simmons/RST]

Re: Re:

Context Dear Wiggo. Your response was immediately questioning the validity of my statement that aero had a big role to play in Boardman going faster than Rominger. Of course that doesn't fit with doping is the only answer to why performance improved rhetoric and hence your comment:

Ah. The science of CdA.

Then going on to say Boardman never measured CdA when he's done more aero testing than most.

Note that I never claimed aero was solely responsible (or that you said I did). Just that it played a large part.


I don't dismiss VAM at all. VAM is a straightforward measurement. Well of course that assumes one has well established timing points, that's actually been a bit problematic at times as well.

Where it gets a bit more variable is in the interpretation of VAM values.

Either not considering important factors that influence VAM on a given day/ride, or in particular when it's converted to W/kg in a systematic way (rather than used as a one off simple illustrative point like I was doing with Boardman's CdA). I was one of those to point this out to Ross Tucker when he began writing about power and cycling six or so years ago.

There is no hypocrisy in pointing out that when a systematic approach is used to converting VAM values from many riders over many climbs over different days into W/kg numbers, there is in fact variable error worth including. This was well established in the only scientific study that tested the validity of this very process (linked in the Froome-Ventoux thread IIRC).

In any case, given this is the clinic, it really doesn't matter what estimates one derives, nor the error bars, confirmation bias will likely dictate what one believes it tells them.