Why straight fork
ROMERO
What is the advantage adn the disadvantage of straight fork?
Is straight fork better or worse than curve fork?
Thanks
Cesar
--
César Augusto
Jason Morrill
My understanding is that a curved fork flexes more and allows for a
cushier ride. A straight fork would tend to send all of the shock from
bumps right up your arms, unless your straight forks had shocks on them.
-Jason
Joshua_Putnam
>What is the advantage adn the disadvantage of straight fork?
>Is straight fork better or worse than curve fork?
Assuming you're manufacturing enough forks to have a
made-to-order fork crown anyway, and all the forks have exactly
the same steering geometry, a straight fork is a little easier
and cheaper to build. Steering is essentially identical to a
curved fork with the same geometry. When only one or two
companies were using straight forks, their distinctiveness was a
marketing advantage.
--
Jo...@WolfeNet.com is Joshua Putnam / P.O. Box 13220 / Burton, WA 98013
"My other bike is a car."
http://www.wolfenet.com/~josh
Graeme Woodward
The straightness, and angle, of the fork can also effect stability
and steering responsiveness. I remember seeing an interesting demonstration
on a TV program about bicycles (part of a British series about mankind's
inventions, with "mad-scientist" host with side-kick) with two different fork
configurations. A bike with a large rake could be pushed, riderless, across
a car park and go a long way. The same bike with the handlebars reversed
(i.e. backward rake, the bike frame was set up so this was possible) fell
over pronto.
Roughly, I think straighter => less stable but more responsive. (The 'give'
in suspension forks may change things a bit. ??) I'm sure there are other's
out there who can give a more scientific and detailed explanation.
Graeme.
In article <CESARR-1604...@hawkweed.agcomed.uiuc.edu>,
CES...@uiuc.edu (ROMERO) writes:
>What is the advantage adn the disadvantage of straight fork?
>Is straight fork better or worse than curve fork?
>Thanks
>Cesar
>
>--
>César Augusto
-----------------------------------------------------------------------
Graeme Woodward. (gra...@ee.usyd.edu.au) Ph: +61 2 9351 4767
Communications Science & Engineering Laboratory. Fax: +61 2 9351 3847
Electrical Engineering, J13, Sydney University, NSW, 2006, AUSTRALIA
Bob Lambert
Graeme Woodward wrote:
>
> The straightness, and angle, of the fork can also effect stability
> and steering responsiveness.
The straightness of the fork has negligible effect on stability or
steering responsiveness. The offset from the axis of the head tube,
provided by curved forks, is replaced by an offset provided by angling
the straight forks away from the head tube axis at the crown. Look at a
picture of group of pros taken from the side. Some have straight forks,
some have curved, but the position of the front hub relative to the head
tube is the same. Of course it is not exactly the same on all the bikes
but that is because of different sizes and geometry not the fork shape.
To put it another way, the frame builder knows where he want's to put
the front hub. He can connect it to the steerer by straight or curved
forks but it's in the same place either way. It's the head angle and
trail that affect handling not the fork shape, that's about style, dude.
Bob Lambert
Camiel Rouweler
ROMERO wrote:
>
> What is the advantage adn the disadvantage of straight fork?
> Is straight fork better or worse than curve fork?
>
The curvature of a fork is a very crude form of suspensions. It allows
the fork to bend slightly and absorb small hits. A curved fork should
therefore be more comfortable then a straight one.
--
Camiel Rouweler (cami...@surf.phys.tue.nl)
"I don't want to be called 'boy' anymore. I find that term
sexist and demeaning!". "How do you want to be called then?"
"I want to be called 'chromosomally advantaged youth." (Calvin&Hobbes)
Jeffrey L. Bell
In article <CESARR-1604...@hawkweed.agcomed.uiuc.edu>,
ROMERO <CES...@uiuc.edu> wrote:
>What is the advantage adn the disadvantage of straight fork?
>Is straight fork better or worse than curve fork?
No difference.
Some people will try to claim that the curved fork gives a
smoother ride. This is not true, since most of the flex is at the top.
The tubes of a curved fork do not bend much.
-Jeff
Jobst Brandt
Camiel Rouweler writes:
>> What is the advantage and the disadvantage of straight fork?
>> Is straight fork better or worse than curve fork?
> The curvature of a fork is a very crude form of suspensions. It
> allows the fork to bend slightly and absorb small hits. A curved
> fork should therefore be more comfortable then a straight one.
That's a new one. You repeat this urban legend as though you know
something about it. There being no bending moment at the fork end,
the fork can be reduced in cross section. That is why it is thin
down there. The curl is there to offset the axle forward, thereby
reducing trail that would otherwise be too great with the typical
15 degree rake of most bicycles.
This can be achieved equally well with straight fork blades, but these
require an angled fork crown that has the specific angle to match.
Jobst Brandt <jbr...@hpl.hp.com>
Jeremy Dowdall
Jobst Brandt (jbr...@hpl.hp.com) wrote:
: Camiel Rouweler writes:
:
: > The curvature of a fork is a very crude form of suspensions. It
: > allows the fork to bend slightly and absorb small hits. A curved
: > fork should therefore be more comfortable then a straight one.
:
: That's a new one. You repeat this urban legend as though you know
: something about it. There being no bending moment at the fork end,
: the fork can be reduced in cross section. That is why it is thin
: down there. The curl is there to offset the axle forward, thereby
: reducing trail that would otherwise be too great with the typical
: 15 degree rake of most bicycles.
:
: This can be achieved equally well with straight fork blades, but these
: require an angled fork crown that has the specific angle to match.
:
: Jobst Brandt <jbr...@hpl.hp.com>
true and good to clear it up (thank you also for the call for accurate
wording - "Aluminum Alloy" versus just "Alloy", even if C M Lozier
wouldn't agree :))
with that said, I would like to state that my 1987 Specialized Rockhopper
Comp actually listed the "shock absorbing" feature of its curved fork
blades as a selling point!
just a note to be careful of what marketing tells you...
Camiel Rouweler
Jobst Brandt wrote:
>
> Camiel Rouweler writes:
> > The curvature of a fork is a very crude form of suspensions. It
> > allows the fork to bend slightly and absorb small hits. A curved
> > fork should therefore be more comfortable then a straight one.
>
> That's a new one. You repeat this urban legend as though you know
> something about it. There being no bending moment at the fork end,
> the fork can be reduced in cross section. That is why it is thin
> down there. The curl is there to offset the axle forward, thereby
> reducing trail that would otherwise be too great with the typical
> 15 degree rake of most bicycles.
This may be an urban legend which I am ignorantly (sp?) spreading here,
but I saw this on a TV show where a chief engineer of Gazelle bicycles
tells this. And they have been making bicycles for more than 90 years,
so I guess they know something! But I guess it is not in your book so it
is not true ;-)
Otherwise, I could easily revert your statement and say that the fork
legs are thin in the curl to make them bend more easily.
Bobby
jbr...@hpl.hp.com (Jobst Brandt) wrote:
>Camiel Rouweler writes:
>
>That's a new one. You repeat this urban legend as though you know
>something about it.
Well now, it's either "a new one" or it's a "legend". It can't be both.
If nothing else, a curved fork is longer than a straight one - which
should impart a little more "flex", all else being the same. Also, there
is the concept of "angle of impact". An impact applied parallel to a
rod, at its end, will not induce as much flex in the rod as the same
impact applied perpendicular to the rod. Although the fork end itself
may not bend much, the curved geometry could make the fork bend more in
its mid-portion, depending on the angle of impact. The increased flex in
the fork ought to dissipate more road vibration etc. because that little
stuff imparts impact with some vertical component . The leaf spring makes
use of this concept. On the other hand, if you ride into a tree, a
straight fork might actually give a little more because that impact is
likely to be directed more horizontally.
Dissipation of energy is not always desirable. The guys who ride in
tight pants want all that thunderstomp transmitted to the drivin' wheel.
So it's possible that a straight fork would dissipate less of that
vertically applied stompforce.
I repeat this prairie legend as though I know something about it.
Bobby
John Foster
Jobst Brandt wrote:
> Camiel Rouweler writes:
> That's a new one. You repeat this urban legend as though you know
> the fork can be reduced in cross section. That is why it is thin
Sorry, I don't get it. Please explain why there is no bending moment
at the fork end. I had always assumed that a bump would increase the
upward force on the fork tips, and since they are not directly below
the point of support, that they would bend in the curved part, which I
assumed was reduced in cross-section, and less elongated fore-aft to
allow this.
I recently bent (gradually) the forks on my old raleigh (not an old, old
raleigh, just a 10spd pseudo-racing bike). The steering tube bent
just above the lower race, decreasing the trail and making the steering
funny. I thought this was from curb hopping (as opposed to curb
bashing).
John Foster
Joshua_Putnam
>This may be an urban legend which I am ignorantly (sp?) spreading here,
>but I saw this on a TV show where a chief engineer of Gazelle bicycles
>tells this. And they have been making bicycles for more than 90 years,
>so I guess they know something! But I guess it is not in your book so it
>is not true ;-)
>Otherwise, I could easily revert your statement and say that the fork
>legs are thin in the curl to make them bend more easily.
It they wanted them to bend easily in the curl, they wouldn't
make the walls so much thicker down there than they are further
up the blade.
As an empirical test, check out what happens with a front
touring rack mounted at the dropouts and mid-fork brazeons --
even when there's enough fork flex to make the top of the rack
move more than half an inch with respect to the head tube of the
bike, there's no significant flex in the part of the fork that
holds the rack. You can see this by loosening the mid-fork
screws and seeing if the rack stays mounted there move around
over the braze-ons. At least on my bikes, they don't. (I
looked at this when designing my own front rack, to see if I
needed to allow extra strength or flexibility for the fork
trying to bend the rack stays.)
As another test, look at a bike that has suffered a really severe
head-on impact, like riding into a wall. The curl remains about
the same, but the fork gets bent back at and above the crown, at
least on the trashed bikes I've cut apart for brazing practice.
Jobst Brandt
Camiel Rouweler writes:
>>> The curvature of a fork is a very crude form of suspensions. It
>>> allows the fork to bend slightly and absorb small hits. A curved
>>> fork should therefore be more comfortable then a straight one.
>> That's a new one. You repeat this urban legend as though you know
>> something about it. There being no bending moment at the fork end,
>> the fork can be reduced in cross section. That is why it is thin
>> down there. The curl is there to offset the axle forward, thereby
>> reducing trail that would otherwise be too great with the typical
>> 15 degree rake of most bicycles.
> This may be an urban legend which I am ignorantly spreading here,
> but I saw this on a TV show where a chief engineer of Gazelle
> bicycles tells this. And they have been making bicycles for more
> than 90 years, so I guess they know something! But I guess it is not
> in your book so it is not true ;-)
I suppose you believe all other advertising as well. I gave you a
reason why this is not true, but you can ask someone to measure it.
Why not try a simple test. Hold the front brake on your bike and rock
the bike fore and aft, and notice what's bending. If you doubt what
you see, you might tape a straight edge to the front of the fork to
assure that it is occurring at the top of the fork, not at the curl.
> Otherwise, I could easily revert your statement and say that the
> fork legs are thin in the curl to make them bend more easily.
Yes, you could say that, but there is no bending moment at the tip of
the fork because there is no leverage, that's why it can be thin. The
"rear triangle" (a tetrahedron) of a bicycle also has no bending in it
and is likewise made of slender lightweight tubes.
Jobst Brandt <jbr...@hpl.hp.com>
Jason Morrill
I have found that the fork on my GT flex a bit, and in my mind that
indicates some form of shock absorbtion. However I also think that
straight forks will also flex depending on the angle of the head and the
material used for the fork.
-Jason
Jobst Brandt wrote:
>
> Camiel Rouweler writes:
>
> >> What is the advantage and the disadvantage of straight fork?
> >> Is straight fork better or worse than curve fork?
Jobst Brandt
Bobby snipes anonymously:
>> That's a new one. You repeat this urban legend as though you know
>> something about it.
> Well now, it's either "a new one" or it's a "legend". It can't be both.
No no. it was a new twist to the urban legend, not a new urban legend,
just as your contribution is just another dodge.
> If nothing else, a curved fork is longer than a straight one - which
> should impart a little more "flex", all else being the same. Also,
> there is the concept of "angle of impact". An impact applied
> parallel to a rod, at its end, will not induce as much flex in the
> rod as the same impact applied perpendicular to the rod.
"should impart", "a little more", "all else being the same", what a
string of waffling! You're grasping at straws in the wind. All this
can be reduced to simple beam bending analysis and there is no flex
developed at the fork end. In fact, if you were to flex that end of
the fork more than a few 0.001 of an inch it would cause a permanent
set and subsequent fatigue failure.
> Although the fork end itself may not bend much, the curved geometry
> could make the fork bend more in its mid-portion, depending on the
> angle of impact.
You must be applying New Physics here. A angled straight fork with
the same offset has no more or less flex as a curved on for the same
load.
> The increased flex in the fork ought to dissipate more road
> vibration etc. because that little stuff imparts impact with some
> vertical component.
Ought to schmought to. Give it up.
> The leaf spring makes use of this concept. On the other hand, if
> you ride into a tree, a straight fork might actually give a little
> more because that impact is likely to be directed more horizontally.
> Dissipation of energy is not always desirable. The guys who ride in
> tight pants want all that thunderstomp transmitted to the drivin' wheel.
> So it's possible that a straight fork would dissipate less of that
> vertically applied stompforce.
This is drifting into its own ambience, that of a fairy tale. It has
a definitely non scientific ring to it.
> I repeat this prairie legend as though I know something about it.
But you don't... in spite of the gratuitous and misplaced buzz words.
Jobst Brandt <jbr...@hpl.hp.com>
Jeremy Dowdall
any chance we could switch the flame wars over to email for awhile?
once you guys decide on something, maybe then present it to the group. as
it is, I really don't see how any of us are learning anything here, and
this thread certainly isn't helping to create a comfy atmosphere for new
people to post new ideas...
thanx,
jeremy
TedHas
In article <5jk05c$o...@ratty.wolfe.net>, jo...@WOLFENET.COM (Joshua_Putnam)
writes:
>As an empirical test, check out what happens with a front
>touring rack mounted at the dropouts and mid-fork brazeons --
>even when there's enough fork flex to make the top of the rack
>move more than half an inch with respect to the head tube of the
>bike, there's no significant flex in the part of the fork that
>holds the rack. You can see this by loosening the mid-fork
>screws and seeing if the rack stays mounted there move around
>over the braze-ons. At least on my bikes, they don't. (I
>looked at this when designing my own front rack, to see if I
>needed to allow extra strength or flexibility for the fork
>trying to bend the rack stays.)
Sorry if I missed this part, but how did the taper/curve fork get
its start? Spoke clearance? Aesthetics? Easier to get proper trail?
Anyone know?
--Ted Haskell
Bobby
jbr...@hpl.hp.com (Jobst Brandt) wrote:
>Bobby snipes anonymously:
>
>>> That's a new one. You repeat this urban legend as though you know
>>> something about it.
>
with so little forethought that you can't even remember doing it.
Furthermore, you tell the whole world [they're on the edges of their
seats, just wondering how this matter of supreme importance will be
resolved] that, "Bobby snipes anonymously". It is not possible for a
named person to do anything anonymously. Did you mean, "Bobby offers
logical well-founded critique without first providing us with a
curriculum vitae?"
>>Well now, it's either "a new one" or it's a "legend". It can't be both.
>
>No no. it was a new twist to the urban legend, not a new urban legend,
>just as your contribution is just another dodge.
>
Your original harangue accused the well-meaning gentleman of REPEATING
the legend. There was nothing new or twisted about his statement.
>> If nothing else, a curved fork is longer than a straight one - which
>> should impart a little more "flex", all else being the same. Also,
>> there is the concept of "angle of impact". An impact applied
>> parallel to a rod, at its end, will not induce as much flex in the
>> rod as the same impact applied perpendicular to the rod.
>
>"should impart", "a little more", "all else being the same", what a
>string of waffling!
Yes, this is the New Physics. One of its more radical principles,
apparently not yet embraced by Mr. Brandt, states that the shortest path
between two points lies along a straight line. Thus straight fork blades
connecting a steerer tube to front dropouts will ALWAYS be shorter than
curved ones connecting the same points. The equivocation on my part was
to acknowledge that materials may be different, as well as the amount of
offset.
>All this can be reduced to simple beam bending analysis and there is no >flex developed at the fork end.
No known substance in the material universe is impervious to flex, i.e.
perfectly rigid, except perhaps your brain.
>In fact, if you were to flex that end of
>the fork more than a few 0.001 of an inch it would cause a permanent
>set and subsequent fatigue failure.
>
Which in fact happens. In the meantime, that few 0.001 of an inch makes
the ride sooooooooooo smooth.
>> Although the fork end itself may not bend much, the curved geometry
>> could make the fork bend more in its mid-portion, depending on the
>> angle of impact.
>
>You must be applying New Physics here. A angled straight fork with
>the same offset has no more or less flex as a curved on for the same
>load.
>
That sentence is a work of art. Either fork, under constant load, won't
flex until that load is altered. The alteration in load is "impact".
Impact is a vector - i.e. it has a magnitude and a direction. My point
was that curved fork blades will orient the fork ends so that a vertical
impact [e.g. rolling off a curb] will be directed more perpendicular to
the fork blade at the point of impact. The straight fork and the curved
one experience the same magnitude of impact, but the angle of the impact
with respect to the fork blade is different. As stated above, an impact
directed perpendicular to a rod will cause more flex than one directed
parallel.
>>The increased flex in the fork ought to dissipate more road
>> vibration etc. because that little stuff imparts impact with some
>> vertical component.
>
>Ought to schmought to. Give it up.
>
Very articulate.
>> The leaf spring makes use of this concept. On the other hand, if
>> you ride into a tree, a straight fork might actually give a little
>> more because that impact is likely to be directed more horizontally.
>
>> Dissipation of energy is not always desirable. The guys who ride in
>> tight pants want all that thunderstomp transmitted to the drivin' wheel.
>> So it's possible that a straight fork would dissipate less of that
>> vertically applied stompforce.
>
>This is drifting into its own ambience, that of a fairy tale. It has
>a definitely non scientific ring to it.
>
Yeah, "Schmought to." Now that has a real scientific ring to it.
I have offered an plausible explanation addressing the original question:
what are the potential advantages/disadvantages of a fork with straight
blades? In so doing, I have provided you with an outlet for your main
talent, which consists of avoiding any intelligent discussion of a
technical issue and launching cheap-shot misguided diatribe.
Glad to oblige,
Bobby
Paul Smeulders
Sorry Jeremy, but I vote NO on this. Keep it here, keep it going,
please.
A 'comfy atmosphere' for new people to post new ideas is fine, but when
'new ideas' are
presented on the misinformation superhighway without the kind of
analysis and criticism that
Jobst and others provide, we end up with a bunch of myths being
perpetuated as truth. Jobst is
a great watchdog for this. I enjoy watching (reading, actually) the
thrusts, parries (sp?), and dodges. I also like reading reasoned
arguments based on quantitative results and measurements. The sniping
and flames are now part of the culture, and it adds some entertainment
value.
"ought to, schmought to"... a gem.
So it isn't comfy. Post your new idea anyway. Let it be considered by
experts and (regrettably) non-experts alike. What's so bad about being
wrong (in the unlikely event your idea is found to be poor), anyway?
You might learn MORE by making a mistake.
Paul Smeulders
PS. to add one completely useless comment which relates to the thread:
"I think dem straight forks ROCK"
Jobst Brandt
John Foster writes:
>>> The curvature of a fork is a very crude form of suspensions. It ...
>> That's a new one. You repeat this urban legend as though you know
>> something about it. There being no bending moment at the fork end,
>> the fork can be reduced in cross section. That is why it is thin
>> down there. The curl is there to offset the axle forward, thereby...
> Sorry, I don't get it. Please explain why there is no bending moment
> at the fork end. << stop right there. Are you asking or are you
> telling us how bending moments work? >> I had always assumed that a
> bump would increase the upward force on the fork tips, and since
> they are not directly below the point of support, that they would
> bend in the curved part, which I assumed was reduced in
> cross-section, and less elongated fore-aft to allow this.
Bending force is a lever mechanism, where the leverage increases the
longer the lever on which a force acts. There is zero bending at the
point where the force is applied and the bending moment increases
> along the length of the lever. That is why torque is given in foot
pounds. The "pounds" are the load and the "foot" is the lever length.
There is no bending at the fork tip and because the forces from road
shock enter the fork almost head-on the lever is at most an inch or so
at the curl of the fork. 50 inch pounds on a 6" long piece of 1/2 inch
steel tubing cannot cause more than a thousandth or so deflection.
> I recently bent (gradually) the forks on my old Raleigh (not an old,
> old Raleigh, just a 10spd pseudo-racing bike). The steering tube
> bent just above the lower race, decreasing the trail and making the
> steering funny. I thought this was from curb hopping (as opposed to
> curb bashing).
As you see, a small amount of bending at the fork crown makes a large
change in fork end position. This is also true during riding without
plastic deformation of the fork. The action is at the fork crown, not
at the curl. It is no different for straight forks where the offset
is in the crown angle (the one your clunker experienced).
Jobst Brandt <jbr...@hpl.hp.com>
Mike Garrison
Jobst Brandt wrote:
>
>
> can be reduced to simple beam bending analysis and there is no flex
> the fork more than a few 0.001 of an inch it would cause a permanent
> set and subsequent fatigue failure.
What kind of brittle material is this Jobst-fork made out of, anyway? Steel
and aluminum beams of 18-24 inches (like a fork) will certainly bend more
than 0.001 inch without taking "a permanent set."
Note, I am not disputing that the bend of the fork is to reduce trail. You
are obviously right about that. But it's silly to say that a metal beam like
a fork cannot bend "more than a few 0.001 of an inch" without damage.
-Mike
John Roden
So, correct me if I'm wrong. It seems that all my bike race cronies who
tell me the straight fork is "stiffer" and somehow help them "dive
in-ta" the corner are mistaken. What, besides weight, is the difference
in ride handling, etc. between straight/curved and carbon fiber/metal
forks? Does it really matter a hill of beans?
Vincent Cheng
John Foster (jfo...@zercom.net) wrote:
: Sorry, I don't get it. Please explain why there is no bending moment
: at the fork end. I had always assumed that a bump would increase the
: upward force on the fork tips, and since they are not directly below
: the point of support, that they would bend in the curved part, which I
: assumed was reduced in cross-section, and less elongated fore-aft to
: allow this.
OK, I'm just an engineering student, but I will give this one a shot. You
can not have a moment at the fork end because there is nothing there that
can support a moment. Ok, imagine yourself holding a pen on one end, now
you can easily apply a vertical/horizontal/angled force at the other end,
but you cannot apply a torque to that end without support it, ie holding
on to the other side. Now, the fork is simply supported at the "crown",
where it will have a bending momement, an unaxial torque and a shear
force. On the other hand, the other tip of the fork can only give you
shear force and maybe torque if the thing is attached to the hub.
: I recently bent (gradually) the forks on my old raleigh (not an old, old
: raleigh, just a 10spd pseudo-racing bike). The steering tube bent
: just above the lower race, decreasing the trail and making the steering
: funny. I thought this was from curb hopping (as opposed to curb
: bashing).
Like I said, the fork can only support a moment at the "crown" or the
steerer tube, that's why it bent there.
Vince
--
***************************************************************************
Vincent Cheng**2nd Year Mechanical Engineering Co-op**University of Alberta
vcc...@gpu.srv.ualberta.ca http://gpu.srv.ualberta.ca/~vccheng/
Creative Web Catchers HTML Designer*http://www.cwc.cban.com
Edmonton Bicycle Commuters Society*http://freenet.edmonton.ab.ca/ebc/
Columnist-Gearhead MTB e-zine*http://www.gearhead.com/
Columnist-Edmonton Oilers Hockey*http://www.ualberta.ca/~mkozak/oilers.htm
***************************************************************************
ROMERO
Hello everybody, I was the one that asked the question. I am glad the
interested in the discussion. All I need to know which has the advantage
or disadvantage.
I thank you again.
--
César Augusto
Jeremy Dowdall
Paul Smeulders (sme...@THISPARTBOGUS.nortel.ca) wrote:
: Jeremy Dowdall wrote:
: >
: > any chance we could switch the flame wars over to email for awhile?
: >
: > once you guys decide on something, maybe then present it to the group. as
: > it is, I really don't see how any of us are learning anything here, and
: > this thread certainly isn't helping to create a comfy atmosphere for new
: > people to post new ideas...
: >
: > thanx,
: > jeremy
:
: Sorry Jeremy, but I vote NO on this. Keep it here, keep it going,
: please.
:
: A 'comfy atmosphere' for new people to post new ideas is fine, but when
: 'new ideas' are
: presented on the misinformation superhighway without the kind of
: analysis and criticism that
: Jobst and others provide, we end up with a bunch of myths being
: perpetuated as truth. Jobst is
: a great watchdog for this. I enjoy watching (reading, actually) the
: thrusts, parries (sp?), and dodges. I also like reading reasoned
: arguments based on quantitative results and measurements. The sniping
: and flames are now part of the culture, and it adds some entertainment
: value.
fair enough I spose...I think its very much needed to have "watchdogs" to
keep track of truth and correct analysis, I just don't enjoy the personal
slanders and insult that don't really seem to answer the questions posted.
:
: "ought to, schmought to"... a gem.
:
: So it isn't comfy. Post your new idea anyway. Let it be considered by
: experts and (regrettably) non-experts alike. What's so bad about being
: wrong (in the unlikely event your idea is found to be poor), anyway?
: You might learn MORE by making a mistake.
:
I agree whole-heartedly in this! mistakes are the best way to learn, and
if someone catches you, you'll almost certainly never make it again...when
Jobst posted the correction _way_ back at the beginning of this thread,
that was good and necessary. Once you start insulting one each other and
getting personal, you loose the point, which was to help someone answer a
question.
When I asked to take it to email, the thread just seamed to be going in
circles with personal comments being the only new additions...oh well
: Paul Smeulders
:
: PS. to add one completely useless comment which relates to the thread:
: "I think dem straight forks ROCK"
my manitou is a straight fork, and it flexes more than any curved blade :)
BC Holicky
>
> As another test, look at a bike that has suffered a really severe
> head-on impact, like riding into a wall. The curl remains about
> the same, but the fork gets bent back at and above the crown, at
> least on the trashed bikes I've cut apart for brazing practice.
>
Having enter a few bikes into severe head-on impacts while powering them
(cars, not walls. well....one wall....), I can tell you that that is
exactly what happens - the bend occurs near the crown.
While it screws up the head tube (understatement) it does create a
REALLY quick handling bike.
by the way - straight forks can be made lighter. Less material. Almost
insignifigant difference, however. I do think that curved forks are
stronger since the legs enter the crown parallel to the direct of the
force rather than on an angle, however slight.
BC Holicky,
Boulder, CO
Ted and Jennifer Nichols-Payne
Okay here it is, the reason to buy a straight fork..
Because it just looks so damn cool.
BTW, I've read this entire thread, and the award for the most amusing
quote goes to... "No known substance in the material universe is
impervious to flex, i.e. perfectly rigid, except perhaps your brain."
Hee hee! Tagged!
Jobst Brandt
John Roden writes:
If the geometry is otherwise the same, they are imagining it. It
makes no difference and that is why it can be done either way, other
than being ugly to the conventional image.
Jobst Brandt <jbr...@hpl.hp.com>
Jeffrey L. Bell
In a previous article, Mike Garrison <mike.g...@boeing.com> wrote:
>Jobst Brandt wrote:
>> can be reduced to simple beam bending analysis and there is no flex
>> developed at the fork end. In fact, if you were to flex that end of
>> the fork more than a few 0.001 of an inch it would cause a permanent
>> set and subsequent fatigue failure.
>
>What kind of brittle material is this Jobst-fork made out of, anyway? Steel
>and aluminum beams of 18-24 inches (like a fork) will certainly bend more
>than 0.001 inch without taking "a permanent set."
I think he meant that there is not much flex in just the curved part
near the end of the fork, rather than along the length of the whole fork.
In that case you've got a 4-7 inch beam.
It will flex a little, but very little.
The vast majority of the flex is at the top,
and in the steerer.
-Jeff Bell
Camiel Rouweler
Vincent Cheng wrote:
> OK, I'm just an engineering student, but I will give this one a shot. You
> can not have a moment at the fork end because there is nothing there that
> can support a moment. Ok, imagine yourself holding a pen on one end, now
> you can easily apply a vertical/horizontal/angled force at the other end,
> but you cannot apply a torque to that end without support it, ie holding
> on to the other side. Now, the fork is simply supported at the "crown",
> where it will have a bending momement, an unaxial torque and a shear
> force. On the other hand, the other tip of the fork can only give you
> shear force and maybe torque if the thing is attached to the hub.
I am not completely convinced yet, as you are only considering
horizontal (or head-on) forces here. I agree that the bend in the fork
does nothing to absorb these. But what about the vertical forces? With
these I mean forces that are acted by the wheel axle to the fork in the
vertical direction. They make an angle significantly greater than 0
degrees with the fork legs and IMHO these forces can make the fork bend.
As I have already displayed no to be an expert in this field, it may be
that the fork is overengineered in a way that these forces have a
negligible bending effect, but as I said, no one in this discussion has
been able to give me the arguments that pulls me to the others side.
BTW is this discussion really that boring? I feel it is a whole lot more
interesting than these roadies vs mtb'ers or what is lighter, X or Y
discussions.
Vincent Cheng
Camiel Rouweler (cami...@surf.phys.tue.nl) wrote:
: I am not completely convinced yet, as you are only considering
: horizontal (or head-on) forces here. I agree that the bend in the fork
: does nothing to absorb these. But what about the vertical forces? With
: these I mean forces that are acted by the wheel axle to the fork in the
: vertical direction. They make an angle significantly greater than 0
: degrees with the fork legs and IMHO these forces can make the fork bend.
: As I have already displayed no to be an expert in this field, it may be
: that the fork is overengineered in a way that these forces have a
: negligible bending effect, but as I said, no one in this discussion has
: been able to give me the arguments that pulls me to the others side.
What do you mean I didn't consider the vertical forces!?! I considered
the perpendicular forces, and that was it. Forks do deflect(not buckle),
but only at the supported end, right below the steerer tube. Why is it so
hard to understand that. Bending in a straight beam can only occur at the
supported end if it's only supported on one side. Draw your basic shear
stress and moment diagrams and you will see what I mean. Even better,
draw your free body diagram for the equilibrium condition, assume you have
a force of (2)^.5 N acting 45 degrees to the horizontal...
/crown
/
/
1N>/
^1N
Ok, as you can see, the force is the same as 1N up and 1N down. To have
an equilibrium, there must be the forces acting opposite, but the same in
magnitude on the other side, ie 1N down and 1N to the left. However, we
must not forget the bending moment at that end, which is 1N*vertical
distance-1N*horizontal distance.
As you can see, bending does not occur at the fork end, but rather, and
the crown.
Vince
--
Eric P. Salathe, Jr.
Vincent Cheng wrote:
> Bending in a straight beam can only occur at the
> supported end if it's only supported on one side.
This is overly simplified. A cantilevered beam, which is what the fork
is, can be tapered so that the bending stress is uniform along its
length (the crossection is reduced to match the decrease in bending
moment). However, for a typical fork, the taper is insufficient to make
the stresses at the tip anywhere as great as those nearer the crown.
Thus, whether the tip is straight or curved does not affect its
performance. If a fork were tapered sufficiently that the bending
stresses were large at its tip, then its curvature could affect its
performance.
--
,
Eric P. Salathe, Jr. sal...@atmos.washington.edu
Seattle WA
Tom Lawrence
Often when considering whether or not a phenomenon exists it is useful
to consider an extreme case. We are comparing a straight fork to a
fork with very minimal deflection, say 1 inch over its length. The
difference between the two is too small to make the answer intuitively
obvious.
Consider a larger deflection. Rather than 1 inch, make it, say, 3 ft.
That is, the fork blade starts at the crown in the usual position,
goes rearward 3 feet, down, and forward 3 feet to the ordinary dropout
positioning. Clearly such a fork would afford greater vertical
compliance than a straight fork. No formulas need be inserted for this
to be obvious.
Why then assert that a fork with 1 inch deflection adds no vertical
compliance? Clearly somewhere between a straight fork and a fork with
3 ft. deflection we reach a minimal deflection where vertical
compliance begins to increase. I see no reason to believe that this
minimal deflection is greater than 1 inch rather than less. Rather, I
would expect this point to be zero, i.e. as soon as you start to bend
the fork you start to add vertical compliance.
As with many threads recently, including the recent thread on whether
or not gyroscopic precession helps to balance a bicycle, the issue
here is not whether or not the effect exists but whether or not it is
significant enough for the rider to actually notice it. It is obvious
to me based on the above mind experiment that a bent fork is more
vertically compliant than a straight fork. That being said I don't
know how to quantify this and based on my purely empirical evidence
with the one straight fork and 3 curved forkes in my garage, I can say
that from my experience, the effect is indeed too small to notice.
Derick Siddoway
In article <5jm5jt$jq4$1...@bigboote.WPI.EDU>, Jeremy Dowdall wrote:
> any chance we could switch the flame wars over to email for awhile?
>
>once you guys decide on something, maybe then present it to the group. as
>it is, I really don't see how any of us are learning anything here, and
>this thread certainly isn't helping to create a comfy atmosphere for new
>people to post new ideas...
September already?
--
Derick Siddoway
Network Archangel
University Hospital
University of Utah
James Deering
I taped a string along the length of a curved fork and then
put all of my weight on the handlebars. The string loosened
(but only slightly).
Theories must be proved by experiments.
Bobby
Genius.
John Foster
Vincent Cheng wrote:
> You can not have a moment at the fork end because there is nothing there that
Thanks for straightening that out. I apologize for the moment that I
bandied the 'bending moment' term out of shape.
But, hey, really, you are assuming a fork is a solid straight bar-
and proving the obvious from there. If only the upper fork were a solid
steel bar, and the forward curved tip (4") were spring steel, THEN could
I talk of a bending moment caused by the upward force on the dropout?
I (like probably everyone who thought that tips flex) always visualized
it as a leading link with a sort of virtual pivot in the center of the
curve.
You and Jobst had me convinced, but then I measured my forks, started
thinking, and did an experiment;
2) 1)
d d
===========d=====d========= 1x2 taped to top tube
d d
crown d d
f d d
f d d
f=fork f d d two vertical dowels measure
f d d vertical displacement of
f d d 1) the dropout, and
. f d d 2) the point where the
. f d d fork curves
. f d d
. +.....d.................
t=tip : :t d 90cm
(curved) : : t +....dropout......
: : :
: :<...>:
: 80mm:
:<........>:
187mm (unloaded dimensions)
I strapped the top tube down to a 2x4 which ran between the two wheels,
then jammed another board under the front wheel and levered it
up.
.... and the results were(repeated twice), that the dropout raised
2.5mm, from unloaded, while the point at the curve raised 1mm. I think
the accuracy was at least +-.5mm. I had no way to measure the force, but
I estimate in the order of 200lbs.
So, if my experiment is accurate, my tips DO account for about twice as
much flex as my crown. This fork is an old 27" of unknown origin. It has
a very pronounced short curve, followed by a straight section to the
dropout - so maybe it's an extreme example. I didn't measure the tire
squash(~70psi), but it looked like it gave more than the forks.
John Foster
************************************************************************
John Foster**6mo No Job Electrical Engineer**The Streets of Montreal
Official keeper of the anarchist bike repair space,
Santropol Roulant meals on wheels
Deliverer of meals on wheels the enviro friendly way
Putterer-about with cargo trailers in hs own basement
Recycler of other people's junk from the garbage
General Shmo about the Plateau
************************************************************************
Ed Chait
I hesitate to enter into a discussion of physics with participants who
are much more knowledgeable than I, but I have a question:
Does the fact that a fork is curved make it a longer lever? If it does,
then would that not make a curved fork more susceptible to flex than a
straight fork, all else being equal?
Ed Chait
Jobst Brandt
Tom Lawrence writes:
> Consider a larger deflection. Rather than 1 inch, make it, say, 3
> ft. That is, the fork blade starts at the crown in the usual
> position, goes rearward 3 feet, down, and forward 3 feet to the
> ordinary dropout positioning. Clearly such a fork would afford
> greater vertical compliance than a straight fork. No formulas need
> be inserted for this to be obvious.
Hold the phone! Straight forks have the same wheel position with
respect to the steer tube that curled forks do. The geometry is
identical. Your example is way off base. Get the facts first, then
tell us how it works.
Jobst Brandt <jbr...@hpl.hp.com>
Jobst Brandt
Eric P. Salathe, Jr. writes:
> This is overly simplified. A cantilevered beam, which is what the
> fork is, can be tapered so that the bending stress is uniform along
> its length (the cross section is reduced to match the decrease in
> bending moment). However, for a typical fork, the taper is
> insufficient to make the stresses at the tip anywhere as great as
> those nearer the crown. Thus, whether the tip is straight or curved
> does not affect its performance. If a fork were tapered sufficiently
> that the bending stresses were large at its tip, then its curvature
> could affect its performance.
That's a big IF and even bigger "sufficiently".
As was pointed out, the relative stress (and therefore, strain) in the
fork is insignificant at the forkend. This is borne out by the lack
of any failures there, whether fatigue or crash. I have observed a
few fork failures of both kinds and have measured deflections. The
compliance in a fork comes from the upper part near the fork crown.
The forkend is made small both to save weight and to gain clearance.
Road bikes have, in spite of tapered forks, little clearance between
spokes and forkend.
Jobst Brandt <jbr...@hpl.hp.com>
Jobst Brandt
James Deering writes:
> I taped a string along the length of a curved fork and then put all
> of my weight on the handlebars. The string loosened (but only
> slightly).
That does not prove where the deflection took place, only that there
was a change in shape.
> Theories must be proved by experiments.
Yes, properly controlled experiments.
Jobst Brandt <jbr...@hpl.hp.com>
David Tolpin
Jobst Brandt (jbr...@hpl.hp.com) wrote:
: Hold the phone!
Good advice. Someone cought a roach, put it on the desk and hit the desk.
The roach ran off. Then the experimentator cut the roach's legs, put
the poor creature on the desk and hit again. The roach didn't run.
'Roaches cannot hear without legs', concluded the wise man.
The fact that a fork doesn't break at the end doesn't mean that the
shape of the end doesn't affect it's stiffness. Actually, the two
extreme cases: a perfectly straight fork and a fork with small appendices
perpendicular to the blades (\ and L) with all other measurements matching
differ in stiffness as 2/3. This is because and L-shaped fork is affected
by a bending torque that is constant through it's length, while \-shaped
is bent by a linearly increasing from zero to the same maximum value
torque. In fact, the difference between the two extreme cases must be
even greature for forks those are thinner at the end. Real forks are
between L and \, but my road bike has a fork, that is (this result
is of sufficiently high precision, I can supply an extensive explanation
on request) 34% more responsive than it's straight analog would be.
Regarding the amount of reaction: it is about one millimeter per 100 kg.
This resistance is good for absorbing vibrations higher than the tyre does.
David
David
Vincent Cheng
Ed Chait (ech...@fix.net) wrote:
: Does the fact that a fork is curved make it a longer lever? If it does,
: then would that not make a curved fork more susceptible to flex than a
: straight fork, all else being equal?
No, it doesn't. The distance from hub to frame is still the same. There
is more metal there, but the linear distance is still the same.
Jobst Brandt
David Tolpin writes:
> The fact that a fork doesn't break at the end doesn't mean that the
> shape of the end doesn't affect it's stiffness. Actually, the two
> extreme cases: a perfectly straight fork and a fork with small appendices
> perpendicular to the blades (\ and L) with all other measurements matching
> differ in stiffness as 2/3. This is because and L-shaped fork is affected
> by a bending torque that is constant through it's length, while \-shaped
> is bent by a linearly increasing from zero to the same maximum value
> torque. In fact, the difference between the two extreme cases must be
> even greature for forks those are thinner at the end. Real forks are
> between L and \, but my road bike has a fork, that is (this result
> is of sufficiently high precision, I can supply an extensive explanation
> on request) 34% more responsive than it's straight analog would be.
You failed to note that straight road forks are identical with curved
ones, are made from the same tubing and are essentially the same
length within a couple of millimeters. It's not as though these
curves protrude at right angles to the fork for several inches. This
hypothesizing with exaggerated models is bizarre. Just measure some
forks and get in touch with reality.
> Regarding the amount of reaction: it is about one millimeter per 100
> kg. This resistance is good for absorbing vibrations higher than
> the tyre does.
Not so. This is a series shock absorption problem and the damping and
compliance of the tire is more than 100x of the steel fork that
resonates for a long time with a nice Hummmmm when rung in its weak
(tuning fork) axis, but rings a high ding in the load axis. Whatever
got though the tire, definitely gets transmitted by the fork. It
might be important to note that the fork is axially loaded by road
shock, not bending as in braking. The bias is forward from road
shock, but that is a small effect.
Jobst Brandt <jbr...@hpl.hp.com>
Stacey Jenkins
Camiel Rouweler wrote:
> The curvature of a fork is a very crude form of suspensions. It allows
> the fork to bend slightly and absorb small hits. A curved fork should
> therefore be more comfortable then a straight one.
>
If this is the case, Why would the Mapei/GB team of last year have
chosen to ride Colnagos with straight forks in Paris Roubaix, and done
so to a 1-2-3 finish? Read Colnago's literature, and you will find that
the shock absorbtion is nearly identical with straight or bent fork legs
Stacey Jenkins
spa...@vegasnet.net
Technical Support Supervisor
Las Vegas Internet Inc.
Tom Lawrence
Tom Lawrence writes:
Jobst Brandt <jbr...@hpl.hp.com>
You'll note in my text above, I mention that the crown and dropouts
are in the ordinary position.
Someone else stated that the path followed by the fork blades is
unimportant, the only thing that matters is the relative position of
the crown and dropouts. This is utter nonsense.
Come on people show a little common sense. If the spring in a car's
suspension went straight from the car's frame to the wheel without
making all of those circular loops, would it still work?
Conversely, if you shaped your fork blades like a coiled spring, using
the same gauge tubing as the straight fork, can you really tell me
with a straight face that the ride would not be more compliant?
And if a large deflection in the forks will make a measurable increase
in vertical compliance, then why won't a small deflection?
Robert Sean Murphy
John Foster wrote:
> You and Jobst had me convinced, but then I measured my forks, started
> thinking, and did an experiment;
...
> I strapped the top tube down to a 2x4 which ran between the two wheels,
> then jammed another board under the front wheel and levered it
> up.
> So, if my experiment is accurate, my tips DO account for about twice as
> much flex as my crown. This fork is an old 27" of unknown origin. It has
> a very pronounced short curve, followed by a straight section to the
> dropout - so maybe it's an extreme example. I didn't measure the tire
> squash(~70psi), but it looked like it gave more than the forks.
As they say, 1 experiment is worth 1000 expert opinions...
R. Sean Murphy
Graduate Research Assistant
Mechanical Engineering
Georgia Institute of Technology
gt3...@prism.gatech.edu
Vincent Cheng
ok, this is the respond I sent when he sent me the post by mail.
From vcc...@gpu.srv.ualberta.ca Mon Apr 28 21:37:01 1997
Date: Sat, 26 Apr 1997 19:48:41 -0600 (MDT)
From: Vince Cheng <vcc...@gpu.srv.ualberta.ca>
To: John Foster <jfo...@zercom.net>
Subject: Re: Why straight fork - experiment
On Sat, 26 Apr 1997, John Foster wrote:
>
> Thanks for straightening that out. I apologize for the moment that I
> bandied the 'bending moment' term out of shape.
>
Nice pun!
> But, hey, really, you are assuming a fork is a solid straight bar-
> and proving the obvious from there. If only the upper fork were a solid
> steel bar, and the forward curved tip (4") were spring steel, THEN could
> I talk of a bending moment caused by the upward force on the dropout?
well, I have been simplifying the situation a lot. If the tip does have
some moment to it, but is minimal compared to the top and since engineers
work on assumptions...:-)
> I (like probably everyone who thought that tips flex) always visualized
> it as a leading link with a sort of virtual pivot in the center of the
> curve.
I really don't think it works that way. Even if we account for the
bending moment along the beam, the moment doesn't act that way.
>
> I strapped the top tube down to a 2x4 which ran between the two wheels,
> then jammed another board under the front wheel and levered it
> up.
> .... and the results were(repeated twice), that the dropout raised
> 2.5mm, from unloaded, while the point at the curve raised 1mm. I think
> the accuracy was at least +-.5mm. I had no way to measure the force, but
> I estimate in the order of 200lbs.
>
> So, if my experiment is accurate, my tips DO account for about twice as
> much flex as my crown. This fork is an old 27" of unknown origin. It has
> a very pronounced short curve, followed by a straight section to the
> dropout - so maybe it's an extreme example. I didn't measure the tire
> squash(~70psi), but it looked like it gave more than the forks.
No, it doesn't account for twice the deflection. If "the point at the
curve" is somewhere in the middle of the fork, shouldn't you account for
the fact that the lever arm is longer at the end? ie, if you pin a pen
down, moving the center 1mm will translate into 2mm of movement at the
end? Now, if the crown is tha part that is flexing, and the middle of the
fork is raised 1mm, then the end will flex more than that.
Vince
Robert Perkins
In article <5k2tjj$5...@hplms2.hpl.hp.com>, jbr...@hpl.hp.com (Jobst Brandt) writes:
|> Whatever
|> got though the tire, definitely gets transmitted by the fork. It
|> might be important to note that the fork is axially loaded by road
|> shock, not bending as in braking. The bias is forward from road
|> shock, but that is a small effect.
|>
|> Jobst Brandt <jbr...@hpl.hp.com>
Since practically all shock is transmitted, how does
bike geometry- head angle and fork offset (= trail?)
affect comfort? Does the trail affect the amount
of shock transmitted to the rider? Or does trail simply
affect handling and stability?
Rob rper...@nortel.ca
--
Rob Perkins rper...@nortel.ca
Jobst Brandt
Tom Lawrence writes:
>>> Consider a larger deflection. Rather than 1 inch, make it, say,
>>> 3 ft. That is, the fork blade starts at the crown in the usual
>>> position, goes rearward 3 feet, down, and forward 3 feet to the
>>> ordinary dropout positioning. Clearly such a fork would afford
>>> greater vertical compliance than a straight fork. No formulas need
>>> be inserted for this to be obvious.
>> Hold the phone! Straight forks have the same wheel position with
>> respect to the steer tube that curled forks do. The geometry is
>> identical. Your example is way off base. Get the facts first,
>> then tell us how it works.
> You'll note in my text above, I mention that the crown and dropouts
> are in the ordinary position.
> Someone else stated that the path followed by the fork blades is
> unimportant, the only thing that matters is the relative position of
> the crown and dropouts. This is utter nonsense.
It isn't when the path length is the same within a few millimeters.
As I pointed out, the straight and curved fork blades are the same
tubing and differ in length by less than 10mm, that is, before the are
bent. Road shocks of any significance act axially on the fork in the
first place, so there isn't any cushioning to speak of.
> Come on people show a little common sense. If the spring in a car's
> suspension went straight from the car's frame to the wheel without
> making all of those circular loops, would it still work?
Your "common sense" is exactly what misplaced technology usually does,
it distorts a physical response to the absurd, taking a secondary
effect to be the principal one. It is similar to the concept that the
hub of a bicycle wheel hangs from the top spokes in a wheel, and that
striking a hard bump rips spokes. This was standard belief until I
exposed it for what it is. Believers of old, are still in that frame
of mind. The same is true for the curl in the forks folks. It is not
a spring, believe me. The straight forks people know this as well.
Jobst Brandt <jbr...@hpl.hp.com>
Jobst Brandt
Robert Perkins writes:
>> Whatever got though the tire, definitely gets transmitted by the
>> fork. It might be important to note that the fork is axially
>> loaded by road shock, not bending as in braking. The bias is
>> forward from road shock, but that is a small effect.
> Since practically all shock is transmitted, how does bike geometry-
> head angle and fork offset (= trail?) affect comfort? Does the
> trail affect the amount of shock transmitted to the rider? Or does
> trail simply affect handling and stability?
The cushioning in current bicycles is in the tire, the rest goes
directly to the handlebars. Frame geometry is for handling, not
comfort. The length of a frame is more important than any other frame
dimension in comfort, because you sit over the rear wheel on the
popular road bikes. Long chainstays reduce that direct coupling by
placing the rider's seat farther from the direct transmission of rear
wheel road motion.
Unless you buy a custom frame or some trekking bike, you'll have no
choice in this dimension, short being IN these days. How else are you
going to have the lightest frame on your block.
Jobst Brandt <jbr...@hpl.hp.com>
CogSet
In article <5k5a54$m...@hplms2.hpl.hp.com>,
jbr...@hpl.hp.com (Jobst Brandt) writes:
>... Long chainstays reduce that direct coupling by
>placing the rider's seat farther from the direct transmission of rear
>wheel road motion.
>
>Unless you buy a custom frame or some trekking bike, you'll have no
>choice in this dimension, short being IN these days. How else are you
>going to have the lightest frame on your block.
>
I'm not sure about curved vs straight fork controversy but I can attest
to the concept that longer chain stays are more comfortable.... and in
fact there do exist a few non-custom frames with longer chainstays.
(Rivendell makes a couple models, perhaps Waterford or some of
the Trek/LeMond frames?)
While few bikes these days have more than 40 or 41cm stays
(many are 38.5 to 39). My 57.5cm Rivendell road has 42.5cm
chainstays with horizontal dropouts. With the wheel pulled all
the way back I'm really getting a 43cm chainstay and I believe it is
noticeably more comfortable than my other bike with 41cm stays.
Perhaps it is my imagination (or due to other geometry
differences - even though the bikes are very similar) but I believe
the longer stays do help in comfort.
Regards,
Peter "how much weight is saved with short stays?" Guyton
Vincent Cheng
CogSet (cog...@aol.com) wrote:
: I'm not sure about curved vs straight fork controversy but I can attest
: to the concept that longer chain stays are more comfortable.... and in
: fact there do exist a few non-custom frames with longer chainstays.
: (Rivendell makes a couple models, perhaps Waterford or some of
: the Trek/LeMond frames?)
OK people, get this, if the LINEAR distance between 2 points is longer,
you will get more flex. If the linear distance is the same, than flex
will be the same since the moment applied to the crown will be EXACTLY the
same....what is so hard about this?
Vince
--
LAM
Can anyone suggest a US steel(lugged) framebuilder who makes straight
forks?
Dave Blake
In article <5k6fql$f...@pulp.ucs.ualberta.ca>, vcc...@gpu5.srv.ualberta.ca
says...
>
>CogSet (cog...@aol.com) wrote:
>: I'm not sure about curved vs straight fork controversy but I can attest
>: to the concept that longer chain stays are more comfortable.... and in
>: fact there do exist a few non-custom frames with longer chainstays.
>: (Rivendell makes a couple models, perhaps Waterford or some of
>: the Trek/LeMond frames?)
>
>OK people, get this, if the LINEAR distance between 2 points is longer,
>you will get more flex. If the linear distance is the same, than flex
>will be the same since the moment applied to the crown will be EXACTLY the
>same....what is so hard about this?
>
I do not believe this to be the effect, but I could be
convinced otherwise. If you apply a vertical displacement
to the rear wheel, the vertical displacement at the seat
is inversely monotonic with the horizontal distance
between the seat and the rear axle. The shorter the
chainstay, the greater the vertical seat movement. This
effect is not related to increases in flex, which I feel
are probably an order of magnitude smaller.
A few years back Merckx was making Hampsten bikes with
short stems and longer top tubes, which Andy claimed
in interviews were more comfy and did not handle
much worse, if at all. Longer wheelbases are more
comfy rides due to leverage between the axles of
the wheels and the position of the seat. The further
the wheel is from your butt, the more comfortable the
ride is.
--
Dave Blake
dbl...@phy.ucsf.edu
http://www.keck.ucsf.edu/~dblake/
Mark Atanovich
In article <3366EA...@Gramercy.ios.com>
LAM <lam...@Gramercy.ios.com> writes:
> Can anyone suggest a US steel(lugged) framebuilder who makes straight
> forks?
Seems like a pretty frivolous reason to choose a frame builder, but to
each their own.
Mark_Atano...@REMOVEemail.sps.mot.com
Remove "REMOVE"
Stacy Byrd
>In article <3366EA...@Gramercy.ios.com>
>LAM <lam...@Gramercy.ios.com> writes:
>
>> Can anyone suggest a US steel(lugged) framebuilder who makes
straight
>> forks?
BREW framesets have straight forks, not sure if frames are lugged.
Located in North Carolina. Custom road frames, limited production and
custom mtn frames. Check out www.cyclery.com/brew
- Taz
Mark McMaster
Vincent Cheng wrote:
>
> Dave Blake (dbl...@phy.ucsf.eduDELETETHISPART) wrote:
> : I do not believe this to be the effect, but I could be
> : convinced otherwise. If you apply a vertical displacement
> : to the rear wheel, the vertical displacement at the seat
> : is inversely monotonic with the horizontal distance
> : between the seat and the rear axle. The shorter the
> : chainstay, the greater the vertical seat movement. This
> : effect is not related to increases in flex, which I feel
> : are probably an order of magnitude smaller.
>
> -Pb(L^2-b^2)^1.5/(9root 3*EIL), where
> L=length of beam,
> P=load
> b=distance from one end
> e=young's modulus
> I=moment of inertia
>
> the longer the beam, the more deflection you will have.(this is for a beam
> that is supported on both ends and P-b>b.
>
> ie, the longer the chain stay, the more it flexes?
The longer chainstay might flex more, but only if it were a
cantilevered beam. It is not; it is supported against vertical
deflection at the dropout by the seatstay. The stiffness of the
seatstay in compression is a couple of magnitudes greater than the
stiffness of the chainstay in bending, so the vertical flex of the rear
triangle is governed by the compressive flex of the seatstay.
Compared to the vertical flex of the other components between the rider
and the road (tires, saddle, handlebar, etc.), the seatstay is so stiff
it might as well be perfectly rigid. Given this, Dave B. is correct
about the length of the chainstay effecting the horizontal offset
between the rear axle and saddle, which in turn effects the proportion
by which a rear wheel bump deflection effects the rider.
Mark McMaster
MMc...@ix.netcom.com
Joshua_Putnam
>Peter "how much weight is saved with short stays?" Guyton
Let's assume that all the shortening is done on the BB end of the
stay, and that we're dealing with traditional 7/8" round stays.
To maximize the benefit, let's say we're using a heavy tube set
with 0.9mm walls. Aircraft Spruce & Specialty lists 7/8"x.035"
4130 tubing as weighing 0.3140 pounds per foot, so if we make
each stay a whole inch shorter, we've saved two inches of tubing,
weighing 0.0523 pounds, or less than 24 grams.
Put two dozen paper clips in your jersey pocket and tell me if
it makes a difference on hills :-)
--
Jo...@WolfeNet.com is Joshua Putnam / P.O. Box 13220 / Burton, WA 98013
"My other bike is a car."
http://www.wolfenet.com/~josh
Vincent Cheng
Dave Blake (dbl...@phy.ucsf.eduDELETETHISPART) wrote:
: I do not believe this to be the effect, but I could be
: convinced otherwise. If you apply a vertical displacement
: to the rear wheel, the vertical displacement at the seat
: is inversely monotonic with the horizontal distance
: between the seat and the rear axle. The shorter the
: chainstay, the greater the vertical seat movement. This
: effect is not related to increases in flex, which I feel
: are probably an order of magnitude smaller.
Isn't the formula for deflection:
-Pb(L^2-b^2)^1.5/(9root 3*EIL), where
L=length of beam,
P=load
b=distance from one end
e=young's modulus
I=moment of inertia
the longer the beam, the more deflection you will have.(this is for a beam
that is supported on both ends and P-b>b.
ie, the longer the chain stay, the more it flexes?
Vince
Jobst Brandt
Vincent Cheng writes:
>> I do not believe this to be the effect, but I could be convinced
>> otherwise. If you apply a vertical displacement to the rear wheel,
>> the vertical displacement at the seat is inversely monotonic with
>> the horizontal distance between the seat and the rear axle. The
>> shorter the chainstay, the greater the vertical seat movement. This
>> effect is not related to increases in flex, which I feel are
>> probably an order of magnitude smaller.
The effect is that a bump at the rear wheel pivots the entire bicycle
around the front wheel axle. If you are sitting in an arc about the
front axle that goes through the rear axle, you get a 1:1 transfer of
displacement or bump. If you ae sitting half way between the two, you
get only half of that. If both wheels go over an identical bump at
the same time, the rider gets the whole bump regardless of geometry
(assuming no suspension elements).
> Isn't the formula for deflection:
> -Pb(L^2-b^2)^1.5/(9root 3*EIL), where
> L=length of beam,
> P=load
> b=distance from one end
> e=young's modulus
> I=moment of inertia
> the longer the beam, the more deflection you will have.(this is for
> a beam that is supported on both ends and P-b>b.
> ie, the longer the chain stay, the more it flexes?
The "rear triangle" is actually a tetrahedron, the most rigid
geometric structure you can devise. That is why these tubes need be
only about 1/2 inch in diameter. There are torques introduces at the
bottom bracket but these are small compared to the bending and torsion
of the large diameter tubes in the forward part of the frame.
The chain stay does not flex.
Jobst Brandt <jbr...@hpl.hp.com>
Jobst Brandt
Vincent Cheng writes:
>> I do not believe this to be the effect, but I could be convinced
>> otherwise. If you apply a vertical displacement to the rear wheel,
>> the vertical displacement at the seat is inversely monotonic with
>> the horizontal distance between the seat and the rear axle. The
>> shorter the chainstay, the greater the vertical seat movement. This
>> effect is not related to increases in flex, which I feel are
>> probably an order of magnitude smaller.
The effect is that a bump at the rear wheel pivots the entire bicycle
around the front wheel axle. If you sit in an arc about the front
axle that passes through the rear axle, you get a 1:1 transfer of
displacement or bump. On many short frames, the rider sits farther
back than that so there is more than a 1:1 jolt. However, sitting
half way between the two, gets only half the bump as on a tandem. If
both wheels go over an identical bump at the same time, the whole bump
is felt regardless of geometry (assuming no suspension elements).
David Tolpin
Jobst Brandt <jbr...@hpl.hp.com> wrote:
> David Tolpin writes:
>
> > differ in stiffness as 2/3. This is because and L-shaped fork is affected
> > by a bending torque that is constant through it's length, while \-shaped
> > is bent by a linearly increasing from zero to the same maximum value
> > torque.
> You failed to note that straight road forks are identical with curved
> ones, are made from the same tubing and are essentially the same
> length within a couple of millimeters. It's not as though these
No, I didn't. Well, I'll repeat, the fork is loaded by a torque from a force
that is directed (usually) toward the headset inside the angle formed by
vertical direction and the direction of the axis of the fork's rotation.
This bending torque is linear through the length of a straight fork,
and grows fast to an almost constant value for a curved fork. The torques at
the crowns are equal, and therefore a rough estimate is 2/3. I don't know
how well strengh of materials is studied in the US, but this should be a simple
calculation for any mechanical engineer.
> curves protrude at right angles to the fork for several inches. This
> hypothesizing with exaggerated models is bizarre. Just measure some
> forks and get in touch with reality.
I supplied (in my previous post) results of measurements of my curved fork,
that was by ~30% more responsive than it's straight replacement would be (i.e.
a fork in which the same positions of the crown and of the axis are connected
by a straight line).
David
Jobst Brandt
David Tolpin writes:
>>> differ in stiffness as 2/3. This is because and L-shaped fork is
>>> affected by a bending torque that is constant through it's length,
>>> while \-shaped is bent by a linearly increasing from zero to the
>>> same maximum value torque.
>> You failed to note that straight road forks are identical with curved
>> ones, are made from the same tubing and are essentially the same
>> length within a couple of millimeters. It's not as though these
> No, I didn't. Well, I'll repeat, the fork is loaded by a torque
> from a force that is directed (usually) toward the headset inside
> the angle formed by vertical direction and the direction of the axis
> of the fork's rotation. This bending torque is linear through the
> length of a straight fork, and grows fast to an almost constant
> value for a curved fork. The torques at the crowns are equal, and
> therefore a rough estimate is 2/3. I don't know how well strength of
> materials is studied in the US, but this should be a simple
> calculation for any mechanical engineer.
Hold it! The bending moment on the fork is the same in both cases.
The geometric location of the axle with respect to the fork crown is
identical. What is different is how the fork blade spans that
distance. As I stated above, the two are structurally identical. It
is merely a visual and aesthetic difference. No calculation can me
made to discern the two from each other. I don't know where you get
your 2/3.
>> curves protrude at right angles to the fork for several inches.
>> This hypothesizing with exaggerated models is bizarre. Just
>> measure some forks and get in touch with reality.
> I supplied (in my previous post) results of measurements of my
> curved fork, that was by ~30% more responsive than it's straight
> replacement would be (i.e. a fork in which the same positions of
> the crown and of the axis are connected by a straight line).
Where did you measure the straight fork and where did you measure the
curved fork, both forks made of similar design? The offset in my fork
is about 25mm over a length of 200mm... not an abrupt curve in any
estimation.
Jobst Brandt <jbr...@hpl.hp.com>
Mark McMaster
Well, I'd stop just short of saying that a curved and a straight fork
are EXACTLY structurally identical. With the curved fork, there is just
a little bit more bending moment in the curved section than in the same
region on the straight fork. Of course we're talking about the
difference between a small moment and a very small moment, with any
additional flex in the curved section being insignificantly small
compared to the flex of the entire fork. But then, when you consider
the compliance of the other components in the load path (like the tire
and the handlebar), it probably doesn't matter what shape the fork is.
Mark McMaster
MMc...@ix.netcom.com
P.S. Yes, I can dig out the formulas for elasticity of curved beams, to
produce a calculation to discern the two from each other (slightly). In
my work on load transducers, I have occasionally taken advantage of
using curved beams to increase load stress/strain/deflection - but with
beams curved to a more significant degree than you would find in any
bicycle fork.
P.P.S. I hesitate a little bit about mentioning the slight difference in
bending moments between curved and straight forks, lest I give the
"curved forks are more comfortable" crew something to latch onto
(however insignificant it may be).
Ron Porat
Hi Forkheads,
I'm convinced there's no more spring in a curved fork compared to a straight
one.
Given that, is it just an aestehtic thing? (Jobst?)
Is the tooling cheaper for a curved fork?
Many traditional, older bicycle parts were manufactured with cheaper
tooling, compared with more modern equivalents. For example,
cottered crank attachments can be manufactured with a grinding wheel,
a drill press, and maybe some blacksmithing equipment, whereas
precision machining is required to fabricate the 4-degree tapers on modern
cotterless cranks, Similarly cups and cones vs. sealed cartridge
bearings, solid axles vs. hollow axles, lugged frames vs. precision
mitered welded frames.. and so on.
I wonder if there's a manufacturing process rationale to the curved fork.
-Ron
Mark Bulgier
Mark McMaster wrote:
> Well, I'd stop just short of saying that a curved and a straight fork
> are EXACTLY structurally identical. With the curved fork, there is just
> a little bit more bending moment in the curved section than in the same
> region on the straight fork. Of course we're talking about the
> difference between a small moment and a very small moment, with any
> additional flex in the curved section being insignificantly small
> compared to the flex of the entire fork. But then, when you consider
> the compliance of the other components in the load path (like the tire
> and the handlebar), it probably doesn't matter what shape the fork is.
I agree that the difference in compliance between straight and curved
forks is insignificant, and to even mention that there *is* a tiny
difference gives the voodoo magic crowd a nail to hang their mythology
on. But what the heck, they'll ignore what the scientists say anyway.
To those who say they are exactly equal: Note that the rider's weight
bends the fork forward, and a brick wall bends the fork back. Between
these two forces is a pothole or object in the road just large enough
to contact the wheel directly in line with the straight fork blade,
loading it in compression only with _no_ bending moment at the crown.
The curved blade in the same scenario will see a bending moment at
every point, zero at the tip and varying non-linearly as you go up the
blade.
Given how much more flexible forks are in bending than in compression,
shouldn't we expect a deflection at least an order of magnitude
greater for the curved blade in this limited case? Though only of
course because the deflection in either case is so small. (A pothole
or object in the road large enough to contact that far up the wheel
will dent the rim and cause a pinch flat unless the rider unweights
properly, obviously using all of the available tire compression and
more. So a few thousandths of an inch of give in the fork won't help
enough to be worth fretting about.)
Mark Bulgier, Seattle
mailto:ma...@ticycles.com
http://ticycles.com
David Tolpin
Jobst Brandt <jbr...@hpl.hp.com> wrote:
> Hold it! The bending moment on the fork is the same in both cases.
> The geometric location of the axle with respect to the fork crown is
> identical. What is different is how the fork blade spans that
> distance. As I stated above, the two are structurally identical. It
> is merely a visual and aesthetic difference. No calculation can me
> made to discern the two from each other. I don't know where you get
> your 2/3.
A fork is not a solid entity, it bends in all points through it's length.
If we draw an imaginary line in the direction of application of the force
(that is roughly parallel to the steering axis and passes through the hub),
the bending moment at a point is computed as product of distance between the
point and the line dran and of the force applied to the free end of the fork.
For a straight fork, this moment grows linearly from zero to the maximum value,
for a curved fork, it grows faster than linearly, and the extreme case is when
the moment momentarilybecomes equal to it's maximum value and then remains constant. Displacement of the end of the fork (in direction perpendicular to the
axis of the fork) is M*l/(3*E*I) for a straight fork, where M is the moment at the crown, l is the
distiance between the crown and the hub; this value approaches M*l/(2*E*I) when
the curve becomes steeper. Since the fork is very stiff axially, the vertical
displacement is estimated as d*cos(70 degree) ~ 0.3d. Thus, vertical displacement
is 0.1Ml/EI for a straight fork,
<= 0.15Ml/EI for a curved fork and approaches the maximum value when the curved
part becomes shorter.
This is indeed very simple.
David
liv2padl
puleeeeze -- this is getting rediculous. structural, inertia, moment,
linearity, steering axis, perpendicular, cosign, entity --blah, blah,
blah, and blah !!!!!! this has been going on for days and you guys haven't
decided anything. this isn't brain surgury here ... the only thing
different and important about a straight blade fork is the COOLNESS
factor. Straight forks look cool, look trick, look different, and that's
the name of that tune Ollie. :-) :-O :-( DC
Joshua_Putnam
>I wonder if there's a manufacturing process rationale to the curved fork.
Let's say you make a fairly wide range of frames, with head tube
angles ranging from 72 to 74 degrees and steering ranging from
very stable touring bikes to criterium bikes. With a curved
fork you can produce the correct steering geometry for each of
those bikes simply by varying the amount of bend in the blade.
With a straight fork, all your fork offset is controlled by the
angle and offset of the crown, so if you want one fork with
1-3/8" of offset for a quick-handling road bike, another fork
with 2" of offset for a loaded touring bike, and another fork
somewhere in between for a general purpose road bike, you need
three different models of crown to produce the three different
offsets.
Eric Topp
Ron Porat, po...@cs.Stanford.EDU writes:
>I wonder if there's a manufacturing process rationale to the curved fork.
It's easier to align a fork with curved blades; if one blade is shorter
than the other you can reduce its curvature to make it longer. It's
faster and easier than filing the dropout. I doubt that many large
manufacturers rely on this technique for their forks, but custom builders
and good mechanics know the technique.
-=E
Jon Fiedler
On 27 Apr 1997 02:24:49 GMT, jbr...@hpl.hp.com (Jobst Brandt) wrote:
>Tom Lawrence writes:
>
>> Consider a larger deflection. Rather than 1 inch, make it, say, 3
>> ft. That is, the fork blade starts at the crown in the usual
>> position, goes rearward 3 feet, down, and forward 3 feet to the
>> ordinary dropout positioning. Clearly such a fork would afford
>> greater vertical compliance than a straight fork. No formulas need
>> be inserted for this to be obvious.
>
>Hold the phone! Straight forks have the same wheel position with
>respect to the steer tube that curled forks do. The geometry is
>identical. Your example is way off base. Get the facts first, then
>tell us how it works.
>
>Jobst Brandt <jbr...@hpl.hp.com>
Jobst,
Tom's example also has the same wheel position as all other forks
being considered.
Jon