The zero wind tunnel option for serious cyclists

Yo, Joseph, as I promised, I have published an article showing how a
cyclist can discover his power output and Cd with no tools except his
bike and a road, yet with a very high degree of accuracy. See
http://members.lycos.co.uk/fiultra/B...arameters.html

HTH.

Andre Jute
http://members.lycos.co.uk/fiultra/B...20CYCLING.html

On May 9, 3:47 pm, Andre Jute wrote:

with a very high degree of accuracy.

Hmmm. I suppose that depends on how one interprets "very high."

On May 10, 12:47*am, Andre Jute wrote:
Yo, Joseph, as I promised, I have published an article showing how a
cyclist can discover his power output and Cd with no tools except his
bike and a road, yet with a very high degree of accuracy. Seehttp://members.lycos.co.uk/fiultra/BICYCLE%20TECH%20--%20basic%20cycl...

HTH.

Andre Jute
*http://members.lycos.co.uk/fiultra/B...20CYCLING.html

Cyclist power is highly irregular. Chis Hoy powering to a world record
standing start kilometer puts out way more power than he could out
training on some long hill climb. He probably doesn't care what his
sustainable aerobic power is, likewise Leonardo Piepoli probably
doesn't care what his stanidng start power is. Muscle strength,
gearing, and a whole bunch of other factors make an acceleration test
for cyclists problematic.

Perhaps a more suitable way would be to use a hill of known slope and
coast down from a standing start, and measure elapsed time and if
possible speed at the end of the course.

That way you get to use the nice and consistent gravitational force
instead of the variable pedal power.

I have a constant slope hill with a clear 300m or so that would be
good for such a test. I just don't know what to do with the info I
could gather there.

Joseph

wrote in message
...
[Snip]

Perhaps a more suitable way would be to use a hill of known slope and
coast down from a standing start, and measure elapsed time and if
possible speed at the end of the course.

That way you get to use the nice and consistent gravitational force
instead of the variable pedal power.

I have a constant slope hill with a clear 300m or so that would be
good for such a test. I just don't know what to do with the info I
could gather there.

Hi Joseph,

If you really want to try this and accept the inaccuracies in your knowledge
of the slope and other inputs then I would recommend you try using the
formula below. All it does is balance the forces between those of gravity
and both wind and rolling resistance to estimate your CdA. In order to get
the best approximation you need to reach terminal velocity on the run down
the slope, i.e. you reach your maximum steady speed. The best way to do this
is to do several runs whereby your pedal up to as close to the maximum speed
you reached on your previous run before hitting the slope as this will give
you the best chance at achieving a steady terminal velocity.

CdA = (9.8 x kg x (gradient - Crr)) / (0.5 x air density x speed squared)

whe
kg is the weigth in kilogrammes of you and your bike
gradient is the fractional gradient of the slope
Crr is rolling resistance plus if you like a very small addition for hub
bearing resistance
speed is in m/s (sorry about the SI units)

To take an example from a local hill where I tried this on the drops

kg = 90 gradient = 0.1(10%) speed = 21m/s(47mph) air density = 1.23kg/m
cubed.

CdA = (9.8 x 90 x (0.1-0.006)) / (0.5 x 1.23 x 21 x 21)

CdA = 0.31

Whilst the result looks to be in the right ball park it all hinges on how
accurate the 10% figure is for the gradient of the hill. I took an average
from a digital map. Have fun and let us know your results.

Graham.

On May 10, 10:59*am, "graham" wrote:
wrote in message

...
[Snip]

Perhaps a more suitable way would be to use a hill of known slope and
coast down from a standing start, and measure elapsed time and if
possible speed at the end of the course.

That way you get to use the nice and consistent gravitational force
instead of the variable pedal power.

I have a constant slope hill with a clear 300m or so that would be
good for such a test. I just don't know what to do with the info I
could gather there.

Hi Joseph,

If you really want to try this and accept the inaccuracies in your knowledge
of the slope and other inputs then I would recommend you try using the
formula below. All it does is balance the forces between those of gravity
and both wind and rolling resistance to estimate your CdA. In order to get
the best approximation you need to reach terminal velocity on the run down
the slope, i.e. you reach your maximum steady speed. The best way to do this
is to do several runs whereby your pedal up to as close to the maximum speed
you reached on your previous run before hitting the slope as this will give
you the best chance at achieving a steady terminal velocity.

CdA = (9.8 x kg x (gradient - Crr)) / (0.5 x air density x speed squared)

whe
kg is the weigth in kilogrammes of you and your bike
gradient is the fractional gradient of the slope
Crr is rolling resistance plus if you like a very small addition for hub
bearing resistance
speed is in m/s (sorry about the SI units)

To take an example from a local hill where I tried this on the drops

kg = 90 gradient = 0.1(10%) * speed = 21m/s(47mph) air density = 1.23kg/m
cubed.

CdA = (9.8 x 90 x (0.1-0.006)) / (0.5 x 1.23 x 21 x 21)

CdA = 0.31

Whilst the result looks to be in the right ball park it all hinges on how
accurate the 10% figure is for the gradient of the hill. I took an average
from a digital map. Have fun and let us know your results.

Graham.

That is perfect. I am unfortunately grounded for the weekend, so I
won't be able to test until later next week.

What type of gradient? Is that distance travelled over rise?

Joseph

wrote in message
...
On May 10, 10:59 am, "graham" wrote:
wrote in message

...
[Snip]
If you really want to try this and accept the inaccuracies in your
knowledge
of the slope and other inputs then I would recommend you try using the
formula below. All it does is balance the forces between those of gravity
and both wind and rolling resistance to estimate your CdA. In order to get
the best approximation you need to reach terminal velocity on the run down
the slope, i.e. you reach your maximum steady speed. The best way to do
this
is to do several runs whereby your pedal up to as close to the maximum
speed
you reached on your previous run before hitting the slope as this will
give
you the best chance at achieving a steady terminal velocity.

CdA = (9.8 x kg x (gradient - Crr)) / (0.5 x air density x speed squared)

whe
kg is the weigth in kilogrammes of you and your bike
gradient is the fractional gradient of the slope
Crr is rolling resistance plus if you like a very small addition for hub
bearing resistance
speed is in m/s (sorry about the SI units)

To take an example from a local hill where I tried this on the drops

kg = 90 gradient = 0.1(10%) speed = 21m/s(47mph) air density = 1.23kg/m
cubed.

CdA = (9.8 x 90 x (0.1-0.006)) / (0.5 x 1.23 x 21 x 21)

CdA = 0.31

Whilst the result looks to be in the right ball park it all hinges on how
accurate the 10% figure is for the gradient of the hill. I took an average
from a digital map. Have fun and let us know your results.

Graham.

That is perfect. I am unfortunately grounded for the weekend, so I
won't be able to test until later next week.

What type of gradient? Is that distance travelled over rise?

For the above formula you want rise over horizontal distance travelled. For
modest road gradients the error is very small if you use rise over distance
travelled.

Graham.

On May 9, 4:47*pm, Andre Jute wrote:
Yo, Joseph, as I promised, I have published an article showing how a
cyclist can discover his power output and Cd with no tools except his
bike and a road, yet with a very high degree of accuracy.

Use terminal speed on a downhill to get CdA, and a climb to get power.
More accurate and simpler. The power part will always have a time
attached to it (ie 300 watts for 10 minutes).

wrote:

On May 10, 12:47*am, Andre Jute wrote:
Yo, Joseph, as I promised, I have published an article showing how a
cyclist can discover his power output and Cd with no tools except his
bike and a road, yet with a very high degree of accuracy. See http://members.lycos.co.uk/fiultra/B...20basic%20cycl...

HTH.

Andre Jute
*http://members.lycos.co.uk/fiultra/B...20CYCLING.html

Cyclist power is highly irregular. Chis Hoy powering to a world record
standing start kilometer puts out way more power than he could out
training on some long hill climb. He probably doesn't care what his
sustainable aerobic power is, likewise Leonardo Piepoli probably
doesn't care what his stanidng start power is. Muscle strength,
gearing, and a whole bunch of other factors make an acceleration test
for cyclists problematic.

The method I'm suggesting is exceedingly subtle, so it takes a while
to understand how it overcomes all these difficulties you raise. My
method uses repeated surplus traction measurements over your speed
range (that's those iterative acceleration readings) to measure very
closely your power *on the day*, and then, without any assumptions or
manufactured constants -- fudge factors, guesses, kludges --
approximates very closely all the other factors lumped together that
influences Cd (that's the coastdown tests) to determine your Cd.

This business of adjusting constants -- fudge factors, guestimates,
street myths, wishful thinking -- is important. Notice for instance
that I made -- it took me two days of hard thought to get there -- a
formula that entirely obviates the necessity for working with the
rolling resistance of some notional tire because I saw too wide a
range in the data you referred me to, and didn't trust the idea of a
lab test with a drum substituting for the road. Instead, I made my
formula include the actual tyres you use on the test, with a real
measurement, not a guesstimate, no matter how distinguished the
guesser. In fact, I made my formula work so that it can operate as a
check on the Cr guess!

But if you think my suggestion is too much work, then that's it;
someone else will take it up sooner or later and then we'll find out
who's right. All I can say is that my method has worked for a quarter-
century for special car builders who bought my book (they write to me
to tell me so) and before that, back into the nineteenth century, for
automobile engineers and before them railway engineers, whose methods
I adapted in the light of modern requirements and knowledge. (It isn't
like I invented anything weird: I just rearranged and reapplied widely
known physics to overcome practical difficulties in cyclist
measurements.)

Perhaps a more suitable way would be to use a hill of known slope and
coast down from a standing start, and measure elapsed time and if
possible speed at the end of the course.

Sure, if that's what you want to do. It sounds a lot easier and
quicker than my method, but it only gets you one reading; even
averaging several runs gets you only one data-point (my suggestion
gets you averages on many data points -- you could for instance use
the data gathered for my method to calculate your optimum gearset). To
exclude the other factors from your reading of downhill speed to
arrive at Cd, you must then have instruments not available to you, or
make all kinds of assumptions about conditions and mechanical
reactions. My method, while more tedious, excludes these sources of
error.

That way you get to use the nice and consistent gravitational force
instead of the variable pedal power.

The second, coastdown part of my suggested test also uses
gravitational force. The iterative acceleration tests overcomes the
perceived problem of variable pedal power. You keep trying to solve
the problem with one big bang, by measuring top speed and trying to
deduce power from that; that is obviously a very fallible method. My
method argues that you exhibit maximum power on acceleration, and by
repeated tests over various ranges with results averaged, it will give
a more reliable final reading. What's more, my method separates the
distribution of the power between the resistances without making any
assumptions and without any fudge factors.

I have a constant slope hill with a clear 300m or so that would be
good for such a test. I just don't know what to do with the info I
could gather there.

Graham has already supplied a formula

We'll find out after you do the downhill test whether the Cd you
calculate predicts your maximum speed pedaling flat out along a flat
road, which is the point of having a Cd number. One thing is for su
if you merely want a cafe Cd closer to the 0.3 you dream of than the
c0.5 average (for 80 per cent of racing cyclists, say) that I suspect,
you're more likely to have your wish fulfilled with the downhill
shortcut than my method!

For those who want to look it up, we're referring to my article at:

http://members.lycos.co.uk/fiultra/B...arameters.html

Good luck with the test.

Andre Jute
No such thing as a free lunch -- Hayek
Never ate lunch in my life -- Armstrong

On May 10, 7:20 am, Andre Jute wrote:

The method I'm suggesting is exceedingly subtle, so it takes a while
to understand how it overcomes all these difficulties you raise.

I'm more interested in your claim of "a very high degree of accuracy."
Would you kindly provide an example that shows either the accuracy or
precision of your exceedingly subtle method?

On May 10, 4:20*pm, Andre Jute wrote:
wrote:
On May 10, 12:47*am, Andre Jute wrote:
Yo, Joseph, as I promised, I have published an article showing how a
cyclist can discover his power output and Cd with no tools except his
bike and a road, yet with a very high degree of accuracy. Seehttp://members.lycos.co.uk/fiultra/BICYCLE%20TECH%20--%20basic%20cycl...

HTH.

Andre Jute
*http://members.lycos.co.uk/fiultra/B...20CYCLING.html

Cyclist power is highly irregular. Chis Hoy powering to a world record
standing start kilometer puts out way more power than he could out
training on some long hill climb. He probably doesn't care what his
sustainable aerobic power is, likewise Leonardo Piepoli probably
doesn't care what his stanidng start power is. Muscle strength,
gearing, and a whole bunch of other factors make an acceleration test
for cyclists problematic.

The method I'm suggesting is exceedingly subtle, so it takes a while
to understand how it overcomes all these difficulties you raise. My
method uses repeated surplus traction measurements over your speed
range (that's those iterative acceleration readings) to measure very
closely your power *on the day*, and then, without any assumptions or
manufactured constants -- fudge factors, guesses, kludges --
approximates very closely all the other factors lumped together that
influences Cd (that's the coastdown tests) to determine your Cd.

This business of adjusting constants -- fudge factors, guestimates,
street myths, wishful thinking -- is important. Notice for instance
that I made -- it took me two days of hard thought to get there -- a
formula that entirely obviates the necessity for working with the
rolling resistance of some notional tire because I saw too wide a
range in the data you referred me to, and didn't trust the idea of a
lab test with a drum substituting for the road. Instead, I made my
formula include the actual tyres you use on the test, with a real
measurement, not a guesstimate, no matter how distinguished the
guesser. In fact, I made my formula work so that it can operate as a
check on the Cr guess!

But if you think my suggestion is too much work, then that's it;
someone else will take it up sooner or later and then we'll find out
who's right. All I can say is that my method has worked for a quarter-
century for special car builders who bought my book (they write to me
to tell me so) and before that, back into the nineteenth century, for
automobile engineers and before them railway engineers, whose methods
I adapted in the light of modern requirements and knowledge. (It isn't
like I invented anything weird: I just rearranged and reapplied widely
known physics to overcome practical difficulties in cyclist
measurements.)

Perhaps a more suitable way would be to use a hill of known slope and
coast down from a standing start, and measure elapsed time and if
possible speed at the end of the course.

Sure, if that's what you want to do. It sounds a lot easier and
quicker than my method, but it only gets you one reading; even
averaging several runs gets you only one data-point (my suggestion
gets you averages on many data points -- you could for instance use
the data gathered for my method to calculate your optimum gearset). To
exclude the other factors from your reading of downhill speed to
arrive at Cd, you must then have instruments not available to you, or
make all kinds of assumptions about conditions and mechanical
reactions. My method, while more tedious, excludes these sources of
error.

That way you get to use the nice and consistent gravitational force
instead of the variable pedal power.

The second, coastdown part of my suggested test also uses
gravitational force. The iterative acceleration tests overcomes the
perceived problem of variable pedal power. You keep trying to solve
the problem with one big bang, by measuring top speed and trying to
deduce power from that; that is obviously a very fallible method. My
method argues that you exhibit maximum power on acceleration, and by
repeated tests over various ranges with results averaged, it will give
a more reliable final reading. What's more, my method separates the
distribution of the power between the resistances without making any
assumptions and without any fudge factors.

I have a constant slope hill with a clear 300m or so that would be
good for such a test. I just don't know what to do with the info I
could gather there.

Graham has already supplied a formula

We'll find out after you do the downhill test whether the Cd you
calculate predicts your maximum speed pedaling flat out along a flat
road, which is the point of having a Cd number. One thing is for su
if you merely want a cafe Cd closer to the 0.3 you dream of than the
c0.5 average (for 80 per cent of racing cyclists, say) that I suspect,
you're more likely to have your wish fulfilled with the downhill
shortcut than my method!

For those who want to look it up, we're referring to my article at:

http://members.lycos.co.uk/fiultra/B...20basic%20cycl...

Good luck with the test.

Andre Jute
No such thing as a free lunch -- Hayek
Never ate lunch in my life -- Armstrong

I have no doubt your method works, and I expect I will use some parts
to check the Crr values. I think the hill roll down will work well
because it uses a constant force of gravity working on the constant
mass of the rider/bike instead of the variable force avialable form
the rider.

Joseph