Tested wide range gears Saturday

[Doug Goncz]

We rode from Bluemont Park in Northern Virginia down parallel to I66 on the Custis Trail, for me a
challenging ride of many short hills and down slopes requiring lots of shifting. All the way to
Rosslyn, then a Quizno's, and the bus back.

The wide range (24 / 35 / 51) x (34 / 24 / 20 / 18 / 16 / 13 / 11) gearing worked pretty well. On
the varying terrain I was able to keep "hooked up" as pitch and speed changed continuously. This is
I think 657% range. The front is a mite touchy with that big jump. Harris recommended an R443 FD and
it seems just right. A few tuning issues...

The cassette version is 34 / 28 / 23 / 19 / 16 / 13 / 11 and may soon be installed on a Lightning
Thunderbolt if I can resolve a few issues.

This gearing may not be for you. I think it is better for riders with limited power, as have I. If
you can push and spin, you might not like those wide (though regular) jumps.

My physics project at NVCC: Google Groups, then "dgoncz" and some of: ultracapacitor bicycle
fluorescent flywheel inverter

 

[Jacques Moser]

On Fri, 05 Mar 2004 12:33:09 +0000, Doug Goncz wrote:

>
> The wide range (24 / 35 / 51) x (34 / 24 / 20 / 18 / 16 / 13 / 11) gearing worked pretty well. On
> the varying terrain I was able to keep "hooked up" as pitch and speed changed continuously. This
> is I think 657% range. The front is a mite touchy with that big jump. Harris recommended an R443
> FD and it seems just right. A few tuning issues...
>

Interesting. I'm no specialist but this is the most extreme range I've heard of. The cassette steps
look a bit unbalanced though: I would have expected something more like 11/13/15/17/20/24/...
because your 13-to-16 step is relatively much larger than the 16-to-18 which is next. Maybe there is
a reason ?

 

[Robert Strickla]

Doug Goncz wrote:
> We rode from Bluemont Park in Northern Virginia down parallel to I66 on the Custis Trail, for me a
> challenging ride of many short hills and down slopes requiring lots of shifting. All the way to
> Rosslyn, then a Quizno's, and the bus back.
>
> The wide range (24 / 35 / 51) x (34 / 24 / 20 / 18 / 16 / 13 / 11) gearing worked pretty well. On
> the varying terrain I was able to keep "hooked up" as pitch and speed changed continuously. This
> is I think 657% range. The front is a mite touchy with that big jump. Harris recommended an R443
> FD and it seems just right. A few tuning issues...
>
> The cassette version is 34 / 28 / 23 / 19 / 16 / 13 / 11 and may soon be installed on a Lightning
> Thunderbolt if I can resolve a few issues.
>
> This gearing may not be for you. I think it is better for riders with limited power, as have I. If
> you can push and spin, you might not like those wide (though regular) jumps.

I set my sport-tourer up with 26/39/50 x 13/14/15/17/19/21/24/27/30. It seems to me to be a better
solution with more practical cassette spacing. Even with 7spd it seems like you could improve the
gearing. If you're not a powerhouse, as you say, then I doubt you really need a 51 x 11. On the
other end, I can't imagine needing a 24 x 34. Even touring I rarely have to resort to my 26 x 30 and
I'm not a strong rider. Check out Sheldon's custom cassettes. I'm sure there's something more
suitable there.

Rob Strickland

 

[Russell Seaton]

> > The wide range (24 / 35 / 51) x (34 / 24 / 20 / 18 / 16 / 13 / 11) gearing worked pretty well.

> I set my sport-tourer up with 26/39/50 x 13/14/15/17/19/21/24/27/30. It seems to me to be a better
> solution with more practical cassette spacing. Even with 7spd it seems like you could improve the
> gearing. If you're not a powerhouse, as you say, then I doubt you really need a 51 x 11. On the
> other end, I can't imagine needing a 24 x 34. Even touring I rarely have to resort to my 26 x 30
> and I'm not a strong rider.

I take it you have not imagined loaded touring in the Alps and Dolomites. A 24x34 is very handy to
have. And a 20x34 is even handier.

Not sure of the point of the original question asker's cassette. I put it into a gear chart and it
seems like its designed to give a complete range of gearing from high to low in each chainring. It
seems designed to minimize front derailleur shifting. At the very high expense of very large and
unnatural jumps between gears.

A better 7 speed cassette would be a 14-32, sold by Nashbar for $20. 14-16-18-21-24-28-32. Or a 13-
32 from Nashbar for $13. 13-15-17-20-24-28-32. Paired up with normal 48 to 52 outside ring, and
normal 38 to 42 middle ring. A nice progression of well spaced gears (with plenty of gears in the 80
to 50 range) on each chainring with the granny 24 as a bailout.

 

[Robert Strickla]

Russell Seaton wrote:
>>> The wide range (24 / 35 / 51) x (34 / 24 / 20 / 18 / 16 / 13 / 11) gearing
>>> worked pretty well.
>
>
>> I set my sport-tourer up with 26/39/50 x 13/14/15/17/19/21/24/27/30. It seems
>> to me to be a better solution with more practical cassette spacing. Even with
>> 7spd it seems like you could improve the gearing. If you're not a powerhouse,
>> as you say, then I doubt you really need a 51 x 11. On the other end, I can't
>> imagine needing a 24 x 34. Even touring I rarely have to resort to my 26 x 30
>> and I'm not a strong rider.
>
>
> I take it you have not imagined loaded touring in the Alps and Dolomites. A
> 24x34 is very handy to have. And a 20x34 is even handier.
>
> Not sure of the point of the original question asker's cassette. I put it into
> a gear chart and it seems like its designed to give a complete range of gearing
> from high to low in each chainring. It seems designed to minimize front
> derailleur shifting. At the very high expense of very large and unnatural jumps
> between gears.
>
> A better 7 speed cassette would be a 14-32, sold by Nashbar for $20. 14-16-18-21-24-28-
> 32. Or a 13-32 from Nashbar for $13. 13-15-17-20-24-28-32. Paired up with
> normal 48 to 52 outside ring, and normal 38 to 42 middle ring. A nice
> progression of well spaced gears (with plenty of gears in the 80 to 50 range)
> on each chainring with the granny 24 as a bailout.

Well, even though I have imagined (with my loaded touring bike - 24x32) touring
in the Alps, it seemed to me that the original poster probably wasn't
encountering those kinds of grades in Virginia. Given his comment about a,
"challenging ride of many short hills and down slopes requiring lots of
shifting. All the way to Rosslyn, then a Quizno's, and the bus back," I'd say a
24x34 is overkill.

Rob Strickland

 

[Dale Benjamin]

"Robert Strickland" <[email protected]> wrote in message
news:[email protected]...
> Russell Seaton wrote:
> >>> The wide range (24 / 35 / 51) x (34 / 24 / 20 / 18 / 16 / 13 / 11) gearing
> >>> worked pretty well.
> >
> >
> >> I set my sport-tourer up with 26/39/50 x 13/14/15/17/19/21/24/27/30. It
> >> seems to me to be a better solution with more practical cassette spacing.
> >> Even with 7spd it seems like you could improve the gearing. If you're not a
> >> powerhouse, as you say, then I doubt you really need a 51 x 11. On the other
> >> end, I can't imagine needing a 24 x 34. Even touring I rarely have to resort
> >> to my 26 x 30 and I'm not a strong rider.
> >
> >
> > I take it you have not imagined loaded touring in the Alps and Dolomites. A
> > 24x34 is very handy to have. And a 20x34 is even handier.
> >
> > Not sure of the point of the original question asker's cassette. I put it
> > into a gear chart and it seems like its designed to give a complete range of
> > gearing from high to low in each chainring. It seems designed to minimize
> > front derailleur shifting. At the very high expense of very large and
> > unnatural jumps between gears.
> >
> > A better 7 speed cassette would be a 14-32, sold by Nashbar for $20. 14-16-18-21-24-28-
> > 32. Or a 13-32 from Nashbar for $13. 13-15-17-20-24-28-32. Paired up with
> > normal 48 to 52 outside ring, and normal 38 to 42 middle ring. A nice
> > progression of well spaced gears (with plenty of gears in the 80 to 50 range)
> > on each chainring with the granny 24 as a bailout.
>
> Well, even though I have imagined (with my loaded touring bike - 24x32) touring
> in the Alps, it seemed to me that the original poster probably wasn't
> encountering those kinds of grades in Virginia. Given his comment about a,
> "challenging ride of many short hills and down slopes requiring lots of
> shifting. All the way to Rosslyn, then a Quizno's, and the bus back," I'd say a
> 24x34 is overkill.

Me too, a 1:1 ratio is low enough. I don't really like any of the proposed
cassettes, a 14 high isn't high enough, an 11 needs a 12 and a 12 needs a 13 or
else there are really large gaps in the ratios. 11, 12, 13, 14, 16, 19, 24 would
be a good 7 gear cassette for touring with a long cage rear deraileur. And a 52,
40, 24 on the front.

My gut always told me that an optimized top end gave more gain than an optimized
low end, loaf going up hills because one doesn't gain or lose much time or
distance anyway, but really work going down, one gets better cooling at speed,
and gains more.

 

[Mark Hickey]

"Dale Benjamin" <[email protected]> wrote:

>My gut always told me that an optimized top end gave more
>gain than an optimized low end, loaf going up hills because
>one doesn't gain or lose much time or distance anyway, but
>really work going down, one gets better cooling at speed,
>and gains more.

Your gut lied. ;-)

You lose a lot more time by taking it easy going up than you
do going down.

When you're going up, putting out 25% more power will result
in you going nearly 25% faster - or going 25% further in the
same amount of time.

If you put that same extra effort into a fast gravity-aided
descent, you'll go only slightly faster (since aerodynamics
will be the chief force to overcome). For example, according
to the excellent calculator at
http://www.analyticcycling.com ...

If a typical cyclist was descending a 10% hill, and
putting out 100 watts, they'd hit a speed of just under
80km/h (or, 50mph).

Increase their output to 125 watts, and the spead "leaps" by
a whopping 0.16km/h (or 1/10th of 1mph).

Mark Hickey Habanero Cycles http://www.habcycles.com Home of
the $695 ti frame

 

[Rick Onanian]

>"Dale Benjamin" <[email protected]> wrote:
>>My gut always told me that an optimized top end gave more
>>gain than an optimized low end, loaf going up hills
>>because one doesn't gain or lose much time or distance
>>anyway, but really work going down, one gets better
>>cooling at speed, and gains more.
On Sat, 06 Mar 2004 08:21:28 -0700, Mark Hickey
<[email protected]> wrote:
>Your gut lied. ;-)
>
>You lose a lot more time by taking it easy going up than
>you do going down.

Mark is right. But, OTOH, going faster downhill is more fun
than going faster uphill -- so, for those of us just trying
to have fun, the optimized top end could provide more.

Personally, I just want to push both ends to further
extremes...
--
Rick Onanian

 

[Doug Goncz]

Eventually the 34-24-20-18-16-13-11 Megarange freewheel will
be a custom 34-28-23-19-16-13-11 cassette. That will give 11
distinct gears.

The purpose is two: wide range. (There's nothing I love
better than sprinting downhill), and an automatic shifter
to come, in which gear spacing must be uniform. The user
would be warned of an impending shift by a high or low tone
and would need to feel the same change from gear to gear
each time.

Gear charts listed in this group under:

Double step gearing

and other posts.

First to come: a lockout to eliminate 6 of the 21 gears,
each of which has a redundant partner preferred in
chainline etc.

Thanks so much for the wide variety of contributions!

My physics project at NVCC: Google Groups, then
"dgoncz" and some of: ultracapacitor bicycle
fluorescent flywheel inverter

 

[David Reuteler]

Rick Onanian <[email protected]> wrote:
> Mark is right. But, OTOH, going faster downhill is more
> fun than going faster uphill -- so, for those of us just
> trying to have fun, the optimized top end could provide
> more. Personally, I just want to push both ends to further
> extremes...

hmmmm .. the fastest i ever remember descending was in white
bird, idaho at around 62mph. my gearing had little to do
with it, tho since i hadn't pedaled for all but the very
beginning of the descent.

actually .. i normally recover on the descent and put most
of my energy into the ascent. that's the new me, tho. i
shaved a lot of time off the cycle by skipping the 15 minute
recovery period/victory dance/brake check/ food & water
break at the summit.

actually that's a lie i still do that.

i love that.

i optimize the low end. for normal use that's a 38/28, for
touring that's a 24/28. my top end is a 48/13. i guess i
wouldn't mind a 12, tho. thanks campag.
--
david reuteler [email protected]

 

[Carl Fogel]

Mark Hickey <[email protected]> wrote in message news:<[email protected]>...
> "Dale Benjamin" <[email protected]> wrote:
>
> >My gut always told me that an optimized top end gave more
> >gain than an optimized low end, loaf going up hills
> >because one doesn't gain or lose much time or distance
> >anyway, but really work going down, one gets better
> >cooling at speed, and gains more.
>
> Your gut lied. ;-)
>
> You lose a lot more time by taking it easy going up than
> you do going down.
>
> When you're going up, putting out 25% more power will
> result in you going nearly 25% faster - or going 25%
> further in the same amount of time.
>
> If you put that same extra effort into a fast gravity-
> aided descent, you'll go only slightly faster (since
> aerodynamics will be the chief force to overcome). For
> example, according to the excellent calculator at
> http://www.analyticcycling.com ...
>
> If a typical cyclist was descending a 10% hill, and
> putting out 100 watts, they'd hit a speed of just under
> 80km/h (or, 50mph).
>
> Increase their output to 125 watts, and the spead "leaps"
> by a whopping 0.16km/h (or 1/10th of 1mph).
>
> Mark Hickey Habanero Cycles http://www.habcycles.com Home
> of the $695 ti frame

Dear Mark,

For such steep descents (10%!) it may be even worse than
you think.

It's awfully hard for the typical cyclist to put any useful
effort into a 50 mph descent because most of the effort goes
into leg-thrashing at a wild cadence:

With a 2124 mm 700c tire, 50 MPH takes:

131 RPM with 53 x 11 gearing 134 RPM with 52 x 11 gearing
143 RPM with 53 x 12 gearing 146 RPM with 52 x 12 gearing

I hear that this kind of cadence is hard to sustain for more
than a few moments, anyway.

In any case, 10% descents that last ten minutes at a steady
50 mph are scarce. They would be 8.3 miles long and drop
4400 feet. Usually, roads like that involve numerous
hairpins and lots of braking.

Carl Fogel

 

[Dale Benjamin]

"Mark Hickey" <[email protected]> wrote in message
news:[email protected]...
> "Dale Benjamin" <[email protected]> wrote:
>
> >My gut always told me that an optimized top end gave more
> >gain than an optimized low end, loaf going up hills
> >because one doesn't gain or lose
much
> >time or distance anyway, but really work going down, one
> >gets better
cooling
> >at speed, and gains more.
>
> Your gut lied. ;-)
>
> You lose a lot more time by taking it easy going up than
> you do going down.
>
> When you're going up, putting out 25% more power will
> result in you going nearly 25% faster - or going 25%
> further in the same amount of time.

125% 0f 6 mph is 7.5 mph. Big deal.

> If you put that same extra effort into a fast gravity-
> aided descent, you'll go only slightly faster (since
> aerodynamics will be the chief force to overcome). For
> example, according to the excellent calculator at
> http://www.analyticcycling.com ...
>
> If a typical cyclist was descending a 10% hill, and
> putting out 100 watts, they'd hit a speed of just under
> 80km/h (or, 50mph).
>
> Increase their output to 125 watts, and the spead "leaps"
> by a whopping 0.16km/h (or 1/10th of 1mph).

I haven't found any hills where 50 mph was realizable, once
I got over 40 mph. Something like 35 mph was generally a
pretty good top speed on any hill around here.

 

[David Reuteler]

Dale Benjamin <[email protected]> wrote:
> 125% 0f 6 mph is 7.5 mph. Big deal.

hmmm, indeed. but you're doing that 7.5mph for a much
longer time. up a 6 mile ascent that would have taken one
hour, you can now do at 7.5mph in 48 minutes. that's 12
minutes. try gaining that much time down the same 6 mile
descent doing 40mph.

> I haven't found any hills where 50 mph was realizable,
> once I got over 40 mph. Something like 35 mph was
> generally a pretty good top speed on any hill around here.

they exist. and they're damn fun! i got one out my
back door.
--
david reuteler [email protected]

 

[Carl Fogel]

"Dale Benjamin" <[email protected]> wrote in message news:<[email protected]>...
> "Mark Hickey" <[email protected]> wrote in message
> news:[email protected]...
> > "Dale Benjamin" <[email protected]> wrote:
> >
> > >My gut always told me that an optimized top end gave
> > >more gain than an optimized low end, loaf going up
> > >hills because one doesn't gain or lose
> much
> > >time or distance anyway, but really work going down,
> > >one gets better
> cooling
> > >at speed, and gains more.
> >
> > Your gut lied. ;-)
> >
> > You lose a lot more time by taking it easy going up than
> > you do going down.
> >
> > When you're going up, putting out 25% more power will
> > result in you going nearly 25% faster - or going 25%
> > further in the same amount of time.
>
> 125% 0f 6 mph is 7.5 mph. Big deal.
>
> > If you put that same extra effort into a fast gravity-
> > aided descent, you'll go only slightly faster (since
> > aerodynamics will be the chief force to overcome). For
> > example, according to the excellent calculator at
> > http://www.analyticcycling.com ...
> >
> > If a typical cyclist was descending a 10% hill, and
> > putting out 100 watts, they'd hit a speed of just under
> > 80km/h (or, 50mph).
> >
> > Increase their output to 125 watts, and the spead
> > "leaps" by a whopping 0.16km/h (or 1/10th of 1mph).
>
> I haven't found any hills where 50 mph was realizable,
> once I got over 40 mph. Something like 35 mph was
> generally a pretty good top speed on any hill around here.

Dear Dale,

Actually, that pitiful 1.5 mph difference works out to a
huge deal in practical situations.

At 6 mph, a six-mile climb takes 60 minutes.

Now strain yourself to 7.5 mph uphill on the same 6 mile
stretch. True, this is only 1.5 mph faster, but it's also a
25% speed increase.

You reach the top in 48 minutes, 12 minutes sooner.

At 30 mph back down the hill (I slowed your descent to make
the arithmetic simple), you cover a mile every two minutes,
so your descent takes 12 minutes.

So the 7.5 mph rider finishes the whole 12 mile ride in 60
minutes, just as the 6 mph rider reaches the top. A six-mile
lead on a twelve-mile ride could be called a big deal.

It's also the explanation for most of Armstrong's advantage
in the Tour de France. He keeps up fine on normal riding,
does well on the individual time trial, and goes maybe a
mile an hour faster up those ugly mountains than whoever's
in second place.

It's a matter of how far, how long, and what the relative
speeds are. The 7.5 mph rider is 12 minutes faster per hour
than the 6.0 mph rider.

To gain the same 12 minutes per hour downhill, you have to
go 37.5 mph against someone going 30 mph--and find a place
where you go downhill that fast for an hour, which is
much, much harder than finding a place to trudge uphill
for an hour.

Carl Fogel

 

[Mark Hickey]

[email protected] (Carl Fogel) wrote:

>Mark Hickey <[email protected]> wrote...

>> Increase their output to 125 watts, and the spead "leaps"
>> by a whopping 0.16km/h (or 1/10th of 1mph).
>
>Dear Mark,
>
>For such steep descents (10%!) it may be even worse than
>you think.
>
>It's awfully hard for the typical cyclist to put any useful
>effort into a 50 mph descent because most of the effort
>goes into leg-thrashing at a wild cadence:
>
>With a 2124 mm 700c tire, 50 MPH takes:
>
> 131 RPM with 53 x 11 gearing 134 RPM with 52 x 11 gearing
> 143 RPM with 53 x 12 gearing 146 RPM with 52 x 12 gearing
>
>I hear that this kind of cadence is hard to sustain for
>more than a few moments, anyway.

Not for Real Men[tm]. ;-) I almost brought up the same point
- that in fact, on such a steep descent you'll likely go
slower as you flail around on the bike trying to pedal at
those cadences, as compared to simply tucking everything in
nice and tight, eliminating aerodynamic drag.

>In any case, 10% descents that last ten minutes at a steady
>50 mph are scarce. They would be 8.3 miles long and drop
>4400 feet. Usually, roads like that involve numerous
>hairpins and lots of braking.

But my, wouldn't that be a fun descent? Think of the time
trial you could have.. suddenly the whippet-thin, toned uber-
riders would be upstaged by those more "girth enhanced".

Mark Hickey Habanero Cycles http://www.habcycles.com Home of
the $695 ti frame

 

[Ritch]

"Dale Benjamin" <[email protected]> wrote in message news:<[email protected]>...
> "Mark Hickey" <[email protected]> wrote in message
> news:[email protected]...
> > "Dale Benjamin" <[email protected]> wrote:
> >
> > >My gut always told me that an optimized top end gave
> > >more gain than an optimized low end, loaf going up
> > >hills because one doesn't gain or lose
> much
> > >time or distance anyway, but really work going down,
> > >one gets better
> cooling
> > >at speed, and gains more.
> >
> > Your gut lied. ;-)
> >
> > You lose a lot more time by taking it easy going up than
> > you do going down.
> >
> > When you're going up, putting out 25% more power will
> > result in you going nearly 25% faster - or going 25%
> > further in the same amount of time.
>
> 125% 0f 6 mph is 7.5 mph. Big deal.
>
It is a big deal: the local hill that is a mile long will
take much longer at 6mph than 7.5mph. Going back down at
50mph won't take long at all. If we're having a race, going
up at 7.5mph and down at 50mph will be much faster than up
at 6mph and down at 50.1mph...

My two cents (discounted to zero today only)

 

[David Damerell]

Dale Benjamin <[email protected]> wrote:
>Me too, a 1:1 ratio is low enough. I don't really like any
>of the proposed cassettes, a 14 high isn't high enough, an
>11 needs a 12 and a 12 needs a 13 or else there are really
>large gaps in the ratios.

I disagree strongly; in the seven-speed world single-tooth
differences are a luxury one must simply do without.

A 1:1 is barely low enough - my lowest is a 1:1 (34:34, as
it happens) but because I'm avoiding the extra
complication of a triple. If I fitted a third 24t
chainring I'd certainly want a lower gear from it than the
1:1 your 11-12-13-14-16-19-24 gives.
--
David Damerell <[email protected]> flcl?

 

[David Damerell]

Dale Benjamin <[email protected]> wrote:
>Me too, a 1:1 ratio is low enough. I don't really like any
>of the proposed cassettes, a 14 high isn't high enough, an
>11 needs a 12 and a 12 needs a 13 or else there are really
>large gaps in the ratios.

I disagree strongly; in the seven-speed world single-tooth
differences are a luxury one must simply do without.

A 1:1 is barely low enough - my lowest is a 1:1 (34:34, as
it happens) but because I'm avoiding the extra
complication of a triple. If I fitted a third 24t
chainring I'd certainly want a lower gear from it than the
1:1 your 11-12-13-14-16-19-24 gives.
--
David Damerell <[email protected]> flcl?

 

[Rick Onanian]

On Sun, 07 Mar 2004 16:46:45 -0700, Mark Hickey <[email protected]>
wrote:
>>In any case, 10% descents that last ten minutes at a
>>steady 50 mph are scarce. They would be 8.3 miles long and
>>drop 4400 feet. Usually, roads like that involve numerous
>>hairpins and lots of braking.
>
>But my, wouldn't that be a fun descent? Think of the time
>trial you could have.. suddenly the whippet-thin, toned uber-
>riders would be upstaged by those more "girth enhanced".

The aerobelly as a tactical advantage...those who are "beer-
enhanced" would be even faster than the "girth-enhanced".
--
Rick Onanian

 

[Dale Benjamin]

"Carl Fogel" <[email protected]> wrote in message
news:[email protected]...
> "Dale Benjamin" <[email protected]> wrote in message
news:<[email protected]>...
> > "Mark Hickey" <[email protected]> wrote in message
> > news:[email protected]...
> > > "Dale Benjamin" <[email protected]> wrote:
> > >
> > > >My gut always told me that an optimized top end gave
> > > >more gain than
an
> > > >optimized low end, loaf going up hills because one
> > > >doesn't gain or
lose
> > much
> > > >time or distance anyway, but really work going down,
> > > >one gets better
> > cooling
> > > >at speed, and gains more.
> > >
> > > Your gut lied. ;-)
> > >
> > > You lose a lot more time by taking it easy going up
> > > than you do going down.
> > >
> > > When you're going up, putting out 25% more power will
> > > result in you going nearly 25% faster - or going 25%
> > > further in the same amount of time.
> >
> > 125% 0f 6 mph is 7.5 mph. Big deal.
> >
> > > If you put that same extra effort into a fast gravity-
> > > aided descent, you'll go only slightly faster (since
> > > aerodynamics will be the chief force to overcome). For
> > > example, according to the excellent calculator at
> > > http://www.analyticcycling.com ...
> > >
> > > If a typical cyclist was descending a 10% hill, and
> > > putting out 100 watts, they'd hit a speed of just
> > > under 80km/h (or, 50mph).
> > >
> > > Increase their output to 125 watts, and the spead
> > > "leaps" by a whopping 0.16km/h (or 1/10th of 1mph).
> >
> > I haven't found any hills where 50 mph was realizable,
> > once I got over
40
> > mph. Something like 35 mph was generally a pretty good
> > top speed on any hill around here.
>
> Dear Dale,
>
> Actually, that pitiful 1.5 mph difference works out to a
> huge deal in practical situations.
>
> At 6 mph, a six-mile climb takes 60 minutes.
>
> Now strain yourself to 7.5 mph uphill on the same 6 mile
> stretch. True, this is only 1.5 mph faster, but it's also
> a 25% speed increase.
>
> You reach the top in 48 minutes, 12 minutes sooner.
>
> At 30 mph back down the hill (I slowed your descent to
> make the arithmetic simple), you cover a mile every two
> minutes, so your descent takes 12 minutes.
>
> So the 7.5 mph rider finishes the whole 12 mile ride in
> 60 minutes, just as the 6 mph rider reaches the top. A
> six-mile lead on a twelve-mile ride could be called a
> big deal.
>
> It's also the explanation for most of Armstrong's
> advantage in the Tour de France. He keeps up fine on
> normal riding, does well on the individual time trial, and
> goes maybe a mile an hour faster up those ugly mountains
> than whoever's in second place.
>
> It's a matter of how far, how long, and what the relative
> speeds are. The 7.5 mph rider is 12 minutes faster per
> hour than the 6.0 mph rider.
>
> To gain the same 12 minutes per hour downhill, you have to
> go 37.5 mph against someone going 30 mph--and find a place
> where you go downhill that fast for an hour, which is
> much, much harder than finding a place to trudge uphill
> for an hour.

Your numbers seem realistic and I can't argue with the
arithmetic, you're entirely correct. Like someone wrote
before, there will be less gain at higher speed for the same
increment of power, so the first guy will probably have an
even larger lead. I don't suppose any realistic
quantification of heat effects on riders in various physical
conditions is feasible, but I think this may sometimes be
significant. Supposing both riders are in the same physical
condition, the one who works harder going uphill will fail
before the one who works harder going down, because their
body will become overheated.