1. traditional road bikes (eg. (the 1000, 1500, etc.) marked by a
level/horizontal top tube, among other things, and
2. "comfort" road bikes (eg. Pilot 1.0, 1.2, etc.) with a supposedly
"more natural riding position," marked by a top tube sloped up from
seat post to headset, thus raising the handlebars relative to the seat
position.
Since my budget will limit me to either a Trek 1000 (traditional) or
Trek Pilot 1.0 (comfort), I wonder which way to go. In the last 3
months, I've put about 500 miles on a borrowed vintage (1983) Trek 620
touring bike, which obviously has a very traditional geometry. In all
that time, I've never experienced any discomfort or body pain while or
riding (despite not owning any padded bike shorts and the 620 being
equiped with its original seat).
Q. -- If I'm comfortable on the vintage 620, should I stick to a
traditional road bike (e.g. Trek 100)?
Q. -- Am I being short-sighted passing up a "comfort" road bike?
Especially considering that I don't see myself ever racing or trying to
ride for time.
BONUS Q. -- What brand other than Trek should I seek out to buy a new
or late model road bike? It's not that I am a Trek snob. I just have
found them more plentiful and easier to research.
THANKS IN ADVANCE.
Gray Strickland
Tulsa, OK
Comfort is one issue, but any bike can be comfortable if you get your
position correct. Another key difference may be the handling. The more
race inspired bikes might feel very twitchy and be difficult to keep in
a straight line. Perhaps the comfort series has angles more like the
touring bike you have been using which I assume has very stable
handling. I do race and ride for time, but not ever in races that
require quick handling (ie criteriums) so my preferences lean toward
straight line stability. I'd go for the Pilot.
I have no direct experience with them but Bianchi has a new series
called "C-2-C" (coast to coast) which is supposed to be a comfort type
of road bike:
http://www.bianchi.it/en/products2007/introRoad_C2C.aspx
have fun!
Joseph
Hi Gary,
I ride a 1985-89 peugeot roadbike, described, fairly so perhaps, by the road
expert at my LBS as a farm gate with wheels. But personally, i find the
traditional style roadbikes more comfortable, despite the slightly more bent
over riding position. I find the extra pressure on my spine from sitting
more upright is more uncomfortable on both my back and tackle so i guess
whether or not the "comfort" range of bikes are actually more comfortable is
personal, not very helpful i know, but sitting on them is the only decent
way to tell.
In addition, for speed and endurance i seem to perform better on
traditional styles. I also never race but prefer the race style, they're
much nippier when needed.
Also, despite the higher gear ratios i find traditional style kinder to my
knees, this may be unique to me tho.
Perhaps consider a race bike with a compact geometry? They're meant to be a
little more comfortable than standard roadbike style, lighter by virtue of
less frame material and the geometry stiffer. I guess this is a speed
machine. Despite the similarities in frame apperance between this and the
'comfort' range the ride will be very different.
It really depends on needs; if you like to look around, wave at people and
generally 'amble' for A to B then get a tourer/hybrid. Just be careful that
you don't all of a sudden you don't get excited by out and out speed (if
only for leisure or fitness as opposed to competition). Also take typical
road surface in to account, anything but tarmac for my bike a the wheels
would no doubt 'taco'.
Go to LSB and sit on as many as you can, then put the ebay bids in.
cheers
The Pilots are pretty much 'fast tourer' geometry. If you like the tourer,
then the pilot is closest. a I've got a 1.2 as a winter training bike.
Well made, takes mudguards (handy in the UK), and sporty enough to let me
keep up with a racing club run. I think you can fit guards to the 1000 and
1200 now, just. If you don't race and you like the colour, go with a Pilot.
If Treks are readily available where you are, that is good. Find a dealer
you get on with and it could be the start of a wonderful relationship. I'm
not a fan of buying a whole bike of eBay or similar. The up-front price
isn't the whole of the story. For instance, at a dealer they're not going
to charge you to swap a stem to fit you better.
If you can stretch to the 1.2 / 1200 the better components are (to me
anyway) worth it.
Not sure about altenatives. If you don't need guards, then there are plenty
now. Specialized Roubaix, Giant SCR being two.
Skippy E&OE
> I'm in the market for a road bike. Given my limited funds, I'll
> probably buy something off ebay. To keep from being overwhelmed with
> too many options, I've limited my bargain-hunting so far to Treks,
> which has led to the surprising discovery that Trek makes:
>
> 1. traditional road bikes (eg. (the 1000, 1500, etc.) marked by a
> level/horizontal top tube, among other things, and
>
> 2. "comfort" road bikes (eg. Pilot 1.0, 1.2, etc.) with a supposedly
> "more natural riding position," marked by a top tube sloped up from
> seat post to headset, thus raising the handlebars relative to the seat
> position.
I really don't understand the practical difference in terms of riding
position b/n the two. Doesn't matter if the headset is higher than the
seat post and what angle the top tube is at, with the way how people
typically fit it, the seat is invariably significantly higher than the
handle bar. The top tube angle is almost irrelevant as it's the handle
bar level and seat height that determines that so called "comfort"
position.
There may be handling, ride quality differences b/n the two, but I just
can't see what comfort has to do with it.
Of course, this is assuming identical seat and fork angles and other
basic geometries.
--
[ Trek 1000 vs Pilot 1.0 ]
> I really don't understand the practical difference in terms of riding
> position b/n the two. Doesn't matter if the headset is higher than the
> seat post and what angle the top tube is at, with the way how people
> typically fit it, the seat is invariably significantly higher than the
> handle bar. The top tube angle is almost irrelevant as it's the handle
> bar level and seat height that determines that so called "comfort"
> position.
You're right that the seat and handlebar position is what matters.
You're missing the effect that frame design has on the handlebar
position.
The Pilot and similar bikes have a sloping top tube because the head
tube is taller than usual relative to the top tube length and standover.
That, plus the long steerer that's normal on such bikes, means that the
handlebars can be set quite a bit higher.
For instance, before I got my Pilot I used to ride an old Sekai. The two
bikes are similar in basic frame geometry: seat tube is within 1/4",
effective top tube is within 1/2", and standover is close enough that I
don't notice a difference.
For all that similarity, with the Sekai's quill stem at full extension,
the handlebars on the Pilot are about 3 inches higher. It might *just*
be possible to make up that difference with a really tall quill stem,
but as far as I know threadless stems with that kind of rise just don't
exist.
You're right that the top tube angle doesn't have any direct effect on
comfort. However, it is influenced by things that do.
Some bikes will never be comfortable. Early Cinellis (pre mid 1970's)
had a reputation for being "nice" riding bikes. Around 1976 someone
traded in a 56cm Cinelli frame. I'd lusted for a Cinelli for years so I
grabbed it. The frame had relaxed 72° or 73° seat and head tube angles
with about 42cm chainstays. I built it up with a Campy NR gruppo,
Cinelli bars and stem, a Unicanitor seat and sewups.
At almost 23 lbs. it was rather heavy for a 56cm bike. I soon realized
that the frame was made of Columbus SP heavy gage "pipe" tubing which
made for a super stiff frame. After a few days of bone jarring, teeth
rattling riding I pulled off my components and hung it back up for sale.
Most steel frame Treks of that era were made with fairly light gage
tubing. Smaller frames 54cm frames or less are going to be stiffer
riding especially for riders weighing under 150 lbs. Heavy gage tubing
is also going to make for a stiffer riding bike.
I remember a fellow who won our UCSF state championship road race on
year was riding a new Gios Torino. He claimed he won because the bike
was so stiff that he kept saying to himself "I got to get off this bike,
I got to get off this bike.....".
Chas.
[...]
> Since my budget will limit me to either a Trek 1000 (traditional) or
> Trek Pilot 1.0 (comfort), I wonder which way to go. In the last 3
> months, I've put about 500 miles on a borrowed vintage (1983) Trek 620
> touring bike, which obviously has a very traditional geometry. In all
> that time, I've never experienced any discomfort or body pain while or
> riding (despite not owning any padded bike shorts and the 620 being
> equiped with its original seat).
>
> Q. -- If I'm comfortable on the vintage 620, should I stick to a
> traditional road bike (e.g. Trek 100)?
>
> Q. -- Am I being short-sighted passing up a "comfort" road bike?
> Especially considering that I don't see myself ever racing or trying to
> ride for time.
If there's any way that you can ride examples of both, I'd do so. That
will answer your questions better than anything. Failing that, try to
compare the geometry of your current bike to the ones you're
considering. If you're comfortable on what you ride now, try to match
the relationship between the handlebars, saddle, and bottom bracket as
closely as practical.
Don't assume that the 1000 will be the most similar to your current
bike. Touring bikes are usually set up for a more upright riding
position than sport bikes.
> BONUS Q. -- What brand other than Trek should I seek out to buy a new
> or late model road bike? It's not that I am a Trek snob. I just have
> found them more plentiful and easier to research.
Just about any brand that's sold in bike shops, really. Most
manufacturers' entry-level sport bikes are more or less similar in spec
to the 1000. Models similar to the Pilot 1.0 aren't too hard to find
either. The Specialized Sequoia and the Raleigh Cadent 1.0 come to mind.
>Most steel frame Treks of that era were made with fairly light gage
>tubing. Smaller frames 54cm frames or less are going to be stiffer
>riding especially for riders weighing under 150 lbs. Heavy gage tubing
>is also going to make for a stiffer riding bike.
Common wisdom, perhaps, but I have to point out that when Bicycle
Guide magazine did a blind test 15 years ago (give or take), and had
otherwise identical bikes built from every set of tubing in the line
(from Columbus, IIRC), suddenly all the "common wisdom" disappeared...
the reviewers who'd previously waxed eloquent on the relative and
obvious changes that resulted from changing one tube in a frame were
suddenly unable to distinguish any of the frames from each other. In
fact, when they summarized the overall impressions, the lowest-grade
(heaviest) tubing got more votes for the best riding frame.
The position (that frame material, especially subtle variations in the
same kind of material - make a big difference in ride quality) is one
that many on r.b.t. have challenged. Many have disagreed with these
challenges, but none has been able to come up with a plausible
mechanism that supports the contention that somehow a vertically
inflexible structure like the rear end of a bicycle can deflect enough
to make a significant difference when masked by the much, much larger
vertical compliance of the tires, rims, saddle, bars, stem and bar
tape.
Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $795 ti frame
Thanks for confirming it. At one stage I was guessing if the angled top
had some special magical mechanical characteristics.
In any case, I note that Bianchi seemed to be favour this so called
compact design. I quite like it. Just need a longer seat post. :)
--
In spite of you maybe trying a bike at a bike shop where you don't
intend to buy it(bad), why not go to a Trek dealer that will let you
ride each for a fair amount of time...to see.
If the tradional bike feels comfy, it probably is and the high front
end bike won't do anything the standard one won't. 'Bent' over riding
position isn't automatically uncomfortable anymore than a sport
motorcycle position compared to a 'cruiser'.
1000: http://www2.trekbikes.com/bikes/bike.php?bikeid=1413000&f=3
Pilot: http://www2.trekbikes.com/bikes/bike.php?bikeid=1402000&f=4
and the 1983 Trek 620 taken from
http://www.vintage-trek.com/images/trek/83brochurePart1.pdf
Frame Size
1000: 58cm
Pilot: 58cm
620: 22.5in (57.15cm)
Head Angle
1000: 73.0°
Pilot: 72.5°
620: 73.0°
Seat Angle
1000: 73.5°
Pilot: 73.0°
620: 73.0°
Effective Top Tube
1000: 57.3cm / 22.6in
Pilot: 57.0cm / 22.4in
620: 56cm
Actual Top Tube
1000: 56.9 / 22.4
Pilot: 54.8 / 21.6
620: 56cm
Chain Stay
1000: 41.7 / 16.4
Pilot: 41.7 / 16.4
620: 44.0cm
Bottom Bracket (ground clearance? center of BB to ground?)
1000: 26.8 / 10.6
Pilot: 27.2 / 10.7
620: 7.2 (called "drop" in 620 brochure; same?)
Offset (fork offset)
1000: 4.5 / 1.8
Pilot: 5.0 / 2.0
620: 5.5cm
Wheelbase
1000: 100.4 / 39.5 (1.3% < than Pilot; 2.3% < than old 620)
Pilot: 101.7 / 40.0
620: 102.8
Trail (defined at
http://www.slowtwitch.com/mainheadings/techctr/geometry.html)
1000: 5.7 / 2.2
Pilot: 5.5 / 2.2
Stand Over
1000: 80.8 / 31.8
Pilot: 78.6 / 30.9
Seat Tube
1000: 58.0 / 22.8
Pilot: 53.0 / 20.9 (9.4% shorter than 1000; 5.6% < than old 620)
620: 56cm
Head Tube
1000: 14.0 / 5.5 (36% longer than Pilot)
Pilot: 19.0 / 7.5
I agree. I consider the owner, salesmen and mechanics at my LBS (which
is a Trek dealer) to be friends and I wouldn't care to waste their time
when I can't afford the $767 (with tax) for a 1000 or $831.59 (with
tax) for a Pilot 1.0. It wouldn't be right. Then again, if I can buy a
2 or 3 year old model on ebay for $400, I won't be embarassed to bring
it to them to work on. I have four children and an employee for whom I
bought a Trek Fuel 90 (MTB) as a bonus. Over the years, I've spent a
ton -- at least to me -- in that shop, but I wouldn't go test ride
bikes that I don't intend to buy. And, unfortunately for me, the Trek
dealers seem to have price discipline which would have made any 20th
century dictator proud. Maybe... if a 58cm 2006 gets dusty on the
floor, I might get a small discount, but not enough to wait for.
My buddy who has graciously allowed me the loan of his spare 1983 Trek
620 will require it back when my hybrid is finally fixed this week.
Over nearly three months of riding a road bike makes me want my own.
That's where all this investigation comes in. My buddy won't sell me
his bike (sentimental), which I love, so I'd like to find something
which rides like it. Friends have suggested looking at the 1000 and the
Pilot 1.0, so that's where my focus has been so far.
It may be true that some bikes can never be truly comfortable. When I
said any bike could be comfortable, I was more thinking of a
comfortable position. The frame, tires, saddle, etc may conspire to
make some bikes very harsh, but that is a different matter.
I used to ride an SPX frame, and I always wanted a Gios Torino...
Joseph
>Joseph
I always liked the color... Colnago blue. We sold a number of them
during the mid 70's. They had some of the best workmanship of any
Italian bike of the period, especially with the gold medallions in the
crown. They came packed in a great shipping container. We never got one
in my size that wasn't presold so I wasn't too tempted.
On the other hand, I always lusted for an early 70's Colnago Super....
Chas.
Mark,
I agree with you on the effects of all the components on bicycle
flexibility but a modern bike frame is a essentially a space frame type
construction. ALL bike frames can flex in every plane if enough force is
applied. Any vertical flexing is a function of frame size, geometry,
material, tubing diameter, wall thickness and rider weight. A 100 lb.
rider may never flex a "stiff" frame but a 200 lb. rider can experience
vertical frame flexing on a light weight frame made from thin wall
tubing.
Your comment all along has been "prove that frames can flex vertically".
I throw it back to you, prove that they can't! Many people would never
get on an airplane if they knew how much airframes flex in flight or
even while taxiing. Almost all lightweight structures are designed to do
that.
Have you ever ridden a curly stay Hetchins or Bates frame with "pencil"
stays in the rear triangle. You should follow one and watch it as it
goes over bumps. You can see the rear triangle flexing up and down. I
had a 1954 Hetchins with a 44" wheelbase and 46cm chainstays. I could
ride over a speed bump and hardly feel it. Surprisingly the bottom
bracket had very little lateral flex.
Carbon fiber can be up to 12 times stronger than steel and 4 times
stronger than aluminum for the same weight. The rule of thumb in
engineering products is that aluminum has 1/3 the weight, strength and
rigidity of steel. Some aluminum alloys such as heat treated 7075 T6 can
be stronger than annealed low carbon steel: 7075 T6 = 73200 PSI tensile
strength vs. 39900 PI for tempered 1018 steel. Titanium has 2/3 the
weight and rigidity of steel; strength differences depend on the alloys.
A properly designed carbon fiber frame is probably going to be the least
flexible per pound followed by alloy steel then titanium and aluminum.
Rigidity can be controlled by frame geometry and tubing diameter.
Chas.
That's not quite fair. Bicycles are not cars, there are no 'facory'
stores and Trek dealers don't get 'kickbacks' or inventory returns on
not sold items(like your local Toyota dealer). Bicycles have the lowest
margin of anything in a bike store and when you discount, most lose
money. Remember it takes around 35-37% margin to break even in the bike
biz. Bicycles often are around 40% margin when not discounted so taking
$100 off a $800 bicycle means the dealer loses money. Price discipline
is not a bad thing if you want that bike dealer that you are friends
with and respect, to be around next year.
I couldn't agree more.
>I agree with you on the effects of all the components on bicycle
>flexibility but a modern bike frame is a essentially a space frame type
>construction. ALL bike frames can flex in every plane if enough force is
>applied. Any vertical flexing is a function of frame size, geometry,
>material, tubing diameter, wall thickness and rider weight. A 100 lb.
>rider may never flex a "stiff" frame but a 200 lb. rider can experience
>vertical frame flexing on a light weight frame made from thin wall
>tubing.
Thin walls and tubing diameter do make a difference, but it's all
relative to other non-flexible structures. It would be akin to
changing the wheels on your car from steel to alloy - yes, there's a
difference in their flexibility, but there are so many other "squishy
bits" above and below them that the difference would be entirely
indistinguishable (note - this assumes the wheels weigh the same - I'm
well aware of the other differences in changing the unsprung weight in
a wheel).
>Your comment all along has been "prove that frames can flex vertically".
>I throw it back to you, prove that they can't!
It's been measured many times - you can do it easily yourself. Start
loading weight on the saddle of a bike and compare the vertical
movement of the various points. The tires and the saddle will compress
many, many more times than any "normal" rear triangle.
Try to compress the rear structure of a conventional frame by a mere
1mm. You'll be astonished at how difficult that is.
> Many people would never
>get on an airplane if they knew how much airframes flex in flight or
>even while taxiing. Almost all lightweight structures are designed to do
>that.
Long, cantilevered structures flex or the break. But a bike frame is
essentially a bridge truss - a tetrahedron is essentially the least
flexible structure you could find.
>Have you ever ridden a curly stay Hetchins or Bates frame with "pencil"
>stays in the rear triangle. You should follow one and watch it as it
>goes over bumps. You can see the rear triangle flexing up and down. I
>had a 1954 Hetchins with a 44" wheelbase and 46cm chainstays. I could
>ride over a speed bump and hardly feel it. Surprisingly the bottom
>bracket had very little lateral flex.
But that is entirely tangential to our discussion, as are softtail
frames and suspension bikes.
>Carbon fiber can be up to 12 times stronger than steel and 4 times
>stronger than aluminum for the same weight. The rule of thumb in
>engineering products is that aluminum has 1/3 the weight, strength and
>rigidity of steel. Some aluminum alloys such as heat treated 7075 T6 can
>be stronger than annealed low carbon steel: 7075 T6 = 73200 PSI tensile
>strength vs. 39900 PI for tempered 1018 steel. Titanium has 2/3 the
>weight and rigidity of steel; strength differences depend on the alloys.
"Strength" and flex aren't related.
>A properly designed carbon fiber frame is probably going to be the least
>flexible per pound followed by alloy steel then titanium and aluminum.
>Rigidity can be controlled by frame geometry and tubing diameter.
You're chasing a myth. Lateral flex IS important, though the relative
stiffness has more to do with tubing diameter limitations than
material per se.
But this has been discussed here ad nauseum. You can take my word for
it, or you can continue to believe that there's really a lot of
compression going on in those seat stays, contrary to every known
materials principle. It's trivial to test this to see that I'm right.
Essentially, you end up believing the engineers or the ad writers.
It's your call.
Every test I've found on the web or seen quoted here are static tests.
The flexing that I'm talking about comes from road shock and impact
caused by bumps, pot holes, rough road surfaces, and so on.
I agree, the horse is dead. BTW, I like the geometries that you use on
your frames.
This is not because the rear triangle flexed, but because the
chainstays were a super-long 46cm and the overall wheelbase a
super-long 44". The extra-long chainstays move the contact patch back
away from under the arse, resulting in a much smoother ride, regardless
of flexibility. This well-known effect is geometric, not elastic.
Overall wheelbase, bottom bracket height, and wheel size are similarly
involved:
<http://www.classicrendezvous.com/British/cycling_old_articles.htm>
¿
The characteristics of the material in a static test are a very good
indicator of the characteristics at the frequencies we're talking
about (low audio range).
>I agree, the horse is dead. BTW, I like the geometries that you use on
>your frames.
Thanks - I haven't found anything that works better. ;-)
There are some significant differences from your ideal (620) here, most
significant being chainstay length, bottom bracket height/drop,
effective/real top tube length, and it would seem, handlebar height and
trail. The first three affect the response over bumps (see the article
by Davision I linked in my previous post), while the fourth is of
course rather a direct question of fit (as is effective top tube
length) and the last affects handling.
You can calculate bottom bracket height in inches from bottom bracket
drop in centimetres (7.2) if you know the wheel diameter, which
requires knowing exactly what size tire they used when referencing that
bottom bracket height.
Obviously the best solution would be to find a 620 used on eBay. It's
hard to say how big the differences above are, but I do believe they
are significant. I can't guaranteed the geomtries are similar, but you
might try looking at old Centurions, Raleighs, some others. Of course,
the best solution really is to find a 620. If after a quick check, I
were to find one available now on eBay, I wouldn't broadcast it here,
but really you should check for yourself.
ú
You found something you like? Good deal. If you haven't already,
measure the heck out of that bike, including "repeating" the factory
spec sheet for the frame. Record, incl. handlebar brand, model, width,
stem length, crank length, saddle brand, model, height and "setback",
whatever.
I know someone whose son (sent on a recon mission?) found an old steel
Trek at a yard sale for less than $30. Rideable with a little work.
(Understanding "children, budget"): I found a local-sale Ti Litespeed
on ebay recently. Was looking for a Trek or similar, for a sacrificial
(only if unavoidable) crit bike, but the L-d Catalyst wasn't much of a
budget stretch once a repaint was figured in. I got to look at it,
money in hand, and it indeed was very nice, low miles. "Decals already
removed", no worries about rust or disfigurement in normal use. Parts
scrounged successfully and economically, rides great.
Craigslist might be an excellent resource for local items, too. Happy
hunting! --D-y
<http://www.classicrendezvous.com/British/cycling_old_articles.htm>
ż
Interesting article.
In the mid 1970's I built a number of very large frames, 66cm to 83cm. I
was concerned about stiffness vs. flexibility and did a number of static
tests myself. We later had several European bike manufacturers build
frames and complete bikes for us which we sold both retail and
wholesale.
I obtained some special long tube sets and could build a 70cm all
Reynolds 531 (metric only) and a 67cm all Columbus frame.
I've never built a bike with curly stays but I've ridden a number of
them. The idea behind the design was that they were supposed to act as
rear "springs" and flex up and down to absorb road shock and impact.
A bicycle frame is not as rigid as many people think!
Chas.
> I've never built a bike wi th curly stays but I've ridden a number of
> them. The idea behind the design was that they were supposed to act as
> rear "springs" and flex up and down to absorb road shock and impact.
Alf Hetchin and his builders knew bicycles, so I don't think they
actually believed that marketing ploy, although I don't know if anyone
knows for sure. The real reason though was that at the time, no logos
of any kind were allowed on the bicycles used for six day races and so
on, so they came up with this as a brilliant way of making their
bicycles identifiable. Also, the aesthetic matches the Hetchin's lugs
very well.
> A bicycle frame is not as rigid as many people think!
The rear triangle (tetrahedron), made out of somewhat flexible but
effectively incompressible tubes that are triangulated, is effectively
completely rigid. The front triangle is not a triangle and therefore
has some compliance, especially at the head tube/steerer, and depending
too on the size of the frame. But it still pales in comparison with
that of the tires. You might notice that no automotive springs are ever
made out of comparable tubes of steel.
Bicycles have much more compliance laterally. Some work (at Trek) seems
to show that lateral stiffness does indeed have something to do with
perceived ride comfort, although how exactly this works is not clear.
This was reported in the Velo News article on frame stiffness testing
that I mentioned some time ago. I believe it was the May issue.
‡
See <http://www.ransbikes.com/images06/F5XP/7F5XP.jpg> for an example
of a better frame geometry. :)
--
Tom Sherman - Here, not there.
>Mark Hickey wrote:
>> "* * Chas" <dna...@aol.spam.com> wrote:
>>
>> >...BTW, I like the geometries that you use on
>> >your frames.
>>
>> Thanks - I haven't found anything that works better. ;-)
>
>See <http://www.ransbikes.com/images06/F5XP/7F5XP.jpg> for an example
>of a better frame geometry. :)
Wait, I was talking about bikes... ;-)
Some cars us hollow torsion bars which are springs of sorts.
There a special tape used to measure stress and movement. I've never
seen this material applied to testing bicycle frames. I'm going to try
and get some of this tape and prove a point.
Chas.
If you can manage it, it might be instructive. Before doing so, you
might want to dig up the, May issue I believe it was, where they had an
extensive article on frame stiffness testing as done by various
manufacturers. You may note that according to the article, only one of
them (Specialized) even goes to the trouble to test vertical
compliance, and even they are yet to make use of the data in any
production bicycle, zerts or not.
In article <1158207896.3...@i3g2000cwc.googlegroups.com>,
41 <KingGe...@yahoo.fr> wrote:
>
>Bicycles have much more compliance laterally. Some work (at Trek) seems
>to show that lateral stiffness does indeed have something to do with
>perceived ride comfort, although how exactly this works is not clear.
>This was reported in the Velo News article on frame stiffness testing
>that I mentioned some time ago. I believe it was the May issue.
>‡
>
_ One thing I've always pondered is how often a bicycle is actually
completely vertical. If you tilt a bike sideways lateral
compliance in the frame has a vertical component. I'd also
guess that even if the bike is completely vertical, there is
generally some lateral component to the driving forces from
"road noise". In particular there is likely an imbalance of
force between the front and back wheel.
_ I doubt any of this matters, but the mind wanders...
_ Booker C. Bense
-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
iQCVAwUBRQmSO2TWTAjn5N/lAQH7lQQAreaYq4pEyMhq+UGs8RVOUR2fZlW6ixn9
Yrs5g0Vkz4FfaeNV9ueJzpP4nd3PSSg9lJLpZcs/vUIOwxYUDofq690PSvKvkDXG
SHoM7Py5mLWwPZ+D/bwBUJuNS5hepexeGKfJ29V1cITmdMLKIKAmq9mlyO0eU8Mg
xy0GnT5y2Hk=
=qTwC
-----END PGP SIGNATURE-----
Bicycle - noun
1. A vehicle with two wheels in tandem, usually propelled by pedals
connected to a wheel by a chain, and having handlebars for steering and
a seat.
Do we need to explain this further to Mark?