Fork Shudder



Status
Not open for further replies.
M

Mark Hickey

Guest
Hey, whadaya know?

I'd observed what appeared to be significant amounts of fork shudder when braking with dirty
rims on my road bike - fork deflections that appeared to be the almost an inch in fore/aft
movement... but the
r.b.t. consensus seemed to be that there was no reasonable (non-destructive) way to deflect a fork
that much.

Today I figured out a really easy (5 minute experiment), cheap ($0) way to figure out if what I was
seeing was an optical illusion or reality.

I'd noticed that by locking the front brake solidly I was able to oscillate the rest of the bike
back and forth easily with just a couple fingers by applying a little fore/aft pressure on the top
tube. It sure LOOKED like the fork was flexing, even when sighting down over the handlebars (using
them as a reference point and watching the hub move back and forth, just like that bumpy, wet
descent I mentioned). But how was I going to measure the apparent deflection?

I was thinking of taking a series of high-speed digital images against a background containing a
reference grid (nerdly idea, huh?). Suddenly it occured to me that If I want to know if the back of
the bike is moving, all I have to do is measure how much the REAR wheel rotates.

I rubber-banded a gold paint marker onto the rear triangle of my fixed gear bike, aligned so that
the marker's tip contacted the rear tire's tread just off-center. Then I locked up the front brake
and applied gentle force to the top tube, back and forth with two fingers.

The result was 14mm of movement (or plus / minus 7mm) of the rear wheel (I verified the front wheel
didn't move at all).

I held the top tube a bit more firmly and applied the same sort of motion to the top tube, but with
a little more force. This time I got 19mm of fore/aft movement at the rear tire.

I repeated both tests several times, and got nearly identical results each time.

My conclusion is that with only a few pounds of force applied intermittently AT THE RIGHT RATE, a
significant amount of fork deflection can be produced. It doesn't seem unlikely at all that similar
deflections can occur under certain braking conditions, lending credibility to those who claim to
have seen their front hub dancing back and forth in inch (2.54cm) or so.

Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame
 
I watched my brother descend in the Austrian Alps with a Kinesis MTB alu. fork in the rain and I
swear his fork was oscillating a centimetre or two in the fore-aft direction. I didn't perform any
on the spot anaylysis, but I swore never to ride a Kinesis fork after witnessing that!! ;-)

Cheers,

Scott..
--
Scott Anderson

"Mark Hickey" <[email protected]> wrote in message
news:eek:[email protected]...
> Hey, whadaya know?
>
> I'd observed what appeared to be significant amounts of fork shudder when braking with dirty
> rims on my road bike - fork deflections that appeared to be the almost an inch in fore/aft
> movement... but the
> r.b.t. consensus seemed to be that there was no reasonable (non-destructive) way to deflect a fork
> that much.

<<snip..>>

> My conclusion is that with only a few pounds of force applied intermittently AT THE RIGHT RATE, a
> significant amount of fork deflection can be produced. It doesn't seem unlikely at all that
> similar deflections can occur under certain braking conditions, lending credibility to those who
> claim to have seen their front hub dancing back and forth in inch (2.54cm) or so.
>
> Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame
 
Maybe the frame is not stiff enough. You know, you can't build a really stiff and stable bike from
anything other than steel ;-)

"Mark Hickey" <[email protected]> wrote in message
news:eek:[email protected]...
> Hey, whadaya know?
>
> I'd observed what appeared to be significant amounts of fork shudder when braking with dirty
> rims on my road bike - fork deflections that appeared to be the almost an inch in fore/aft
> movement... but the
> r.b.t. consensus seemed to be that there was no reasonable (non-destructive) way to deflect a fork
> that much.
>
> Today I figured out a really easy (5 minute experiment), cheap ($0) way to figure out if what I
> was seeing was an optical illusion or reality.
>
> I'd noticed that by locking the front brake solidly I was able to oscillate the rest of the bike
> back and forth easily with just a couple fingers by applying a little fore/aft pressure on the top
> tube. It sure LOOKED like the fork was flexing, even when sighting down over the handlebars (using
> them as a reference point and watching the hub move back and forth, just like that bumpy, wet
> descent I mentioned). But how was I going to measure the apparent deflection?
>
> I was thinking of taking a series of high-speed digital images against a background containing a
> reference grid (nerdly idea, huh?). Suddenly it occured to me that If I want to know if the back
> of the bike is moving, all I have to do is measure how much the REAR wheel rotates.
>
> I rubber-banded a gold paint marker onto the rear triangle of my fixed gear bike, aligned so that
> the marker's tip contacted the rear tire's tread just off-center. Then I locked up the front brake
> and applied gentle force to the top tube, back and forth with two fingers.
>
> The result was 14mm of movement (or plus / minus 7mm) of the rear wheel (I verified the front
> wheel didn't move at all).
>
> I held the top tube a bit more firmly and applied the same sort of motion to the top tube, but
> with a little more force. This time I got 19mm of fore/aft movement at the rear tire.
>
> I repeated both tests several times, and got nearly identical results each time.
>
> My conclusion is that with only a few pounds of force applied intermittently AT THE RIGHT RATE, a
> significant amount of fork deflection can be produced. It doesn't seem unlikely at all that
> similar deflections can occur under certain braking conditions, lending credibility to those who
> claim to have seen their front hub dancing back and forth in inch (2.54cm) or so.
>
> Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame
 
Mark Hickey <[email protected]> wrote:
: It doesn't seem unlikely at all that similar deflections can occur under certain braking
: conditions, lending credibility to those who claim to have seen their front hub dancing back and
: forth in inch (2.54cm) or so.

well, i'm one of those people. take a paul neo-retro brake and put it on one of many cyclocross
forks (independent fabrications and the wound-up i've seen) and pull the straddle down low. ride the
bike at 5mph or so and hit the brakes and you'll see shudder of that magnitude.

i didn't realize there wasn't credibility to it, tho? independent fab. knows about it, a few of the
shops more familiar with cyclocross do as well. it's pretty easy to demonstrate (you don't need to
be going fast).

never seen it on a caliper brake, tho.

i rode my bike into the wall (starting at the wall, of course) to see if the deflection was
possible. yep.
--
david reuteler [email protected]
 
On Sun, 02 Mar 2003 19:30:45 -0500, Mark Hickey wrote:

> Hey, whadaya know?
>
> I'd observed what appeared to be significant amounts of fork shudder when braking with dirty rims
> on my road bike - fork deflections that appeared to be the almost an inch in fore/aft movement...
> but the r.b.t. consensus seemed to be that there was no reasonable (non-destructive) way to
> deflect a fork that much.

I don't think that is accurate. The consensus is that the blades themselves do not flex. The
assumption of some was that the purpose of a curved fork blade was to provide some "shock
absorbsion" or spring, which of course it does not.

Forks do flex. At the crown.

--

David L. Johnson

__o | Some people used to claim that, if enough monkeys sat in front _`\(,_ | of enough
typewriters and typed long enough, eventually one of (_)/ (_) | them would reproduce the
collected works of Shakespeare. The internet has proven this not to be the case.
 
"David L. Johnson" <David L. Johnson <[email protected]>> wrote:
: Forks do flex. At the crown.

yea, that's right (as in what i remember the consensus being) .. i can tell you, tho, that the
(straight) fork blades certainly do *APPEAR* to flex. but that doesn't mean they are. it'd be nice
to see with high-speed photography especially since i'm not willing to subject my fork to enuf force
to test otherwise.
--
david reuteler [email protected]
 
David Reuteler writes:

>> Forks do flex. At the crown.

> yea, that's right (as in what i remember the consensus being) .. i can tell you, tho, that the
> (straight) fork blades certainly do *APPEAR* to flex. but that doesn't mean they are. it'd be nice
> to see with high-speed photography especially since i'm not willing to subject my fork to enuf
> force to test otherwise.

What does high speed photography have to do with fork flex? You can assess that standing still. Just
lock the front brake and rock the bicycle back and forth while sitting in the saddle while sighting
over the bars at the front axle. It's not a mystery.

And by the way, there are evening courses for people for whom English is a second language. They
usually cover writing and spelling as well as sentence structure.

Jobst Brandt [email protected] Palo Alto CA
 
In article <[email protected]>, [email protected] says...
> And by the way, there are evening courses for people for whom English is a second language. They
> usually cover writing and spelling as well as sentence structure.
>
>
They may cover writing and spelling even better.

Rick
 
[email protected] wrote:
: What does high speed photography have to do with fork flex? You can assess that standing still.
: Just lock the front brake and rock the bicycle back and forth while sitting in the saddle while
: sighting over the bars at the front axle. It's not a mystery.

the question was whether it was flexing at the crown or in the blade itself.

: And by the way, there are evening courses for people for whom English is a second language. They
: usually cover writing and spelling as well as sentence structure.

there are also classes in reading comprehension.
--
david reuteler [email protected]
 
David Reuteler wrote:
> [email protected] wrote:
> : What does high speed photography have to do with fork flex? You can assess that standing still.
> : Just lock the front brake and rock the bicycle back and forth while sitting in the saddle while
> : sighting over the bars at the front axle. It's not a mystery.
>
> the question was whether it was flexing at the crown or in the blade itself.

The suggested method allows you to determine this for yourself.

Some forks have more flex in the blades than others, but the majority of the flex is in the crown
which is easily assessed by Jobst's method.

--
R.

<> Richard Brockie "Categorical statements <> The tall blond one. always cause trouble." <>
[email protected]
 
Richard Brockie <[email protected]> wrote:
: The suggested method allows you to determine this for yourself.
:
: Some forks have more flex in the blades than others, but the majority of the flex is in the crown
: which is easily assessed by Jobst's method.

god, i realize this. the first thing i did was check the fork flex in the "jobst method" (read my
other post) and yes, there was flex at the crown. that's static, however. the shudder is around 5hz
or so and is probably more like 5cm total (2.5cm fore/aft) in amplitude. when you watch the fork
while riding it *APPEARS* to be flexing the blades (the last 20cm). that may be an optical illusion.
i would like to see that on film.
--
david reuteler [email protected]
 
In article <[email protected]>, "David L. Johnson" <David L. Johnson
<[email protected]>> wrote:

> On Sun, 02 Mar 2003 19:30:45 -0500, Mark Hickey wrote:
>
> > Hey, whadaya know?
> >
> > I'd observed what appeared to be significant amounts of fork shudder when braking with dirty
> > rims on my road bike - fork deflections that appeared to be the almost an inch in fore/aft
> > movement... but the r.b.t. consensus seemed to be that there was no reasonable (non-destructive)
> > way to deflect a fork that much.
>
> I don't think that is accurate. The consensus is that the blades themselves do not flex. The
> assumption of some was that the purpose of a curved fork blade was to provide some "shock
> absorbsion" or spring, which of course it does not.
>
> Forks do flex. At the crown.

Wouldn't the flex be along much of the length of the blade? Note that fork and aft flex is not the
same as compression and would absorb no road shock; a "rigid" fork can't compress.

Now, I have frequently experienced fork flex on bikes with light-gauge forks and cantilever
brakes. I can't recall it happening under braking on a road bike (but that may just say more about
my memory).
 
"David L. Johnson" <David L. Johnson <[email protected]>> wrote:

>On Sun, 02 Mar 2003 19:30:45 -0500, Mark Hickey wrote:
>
>> Hey, whadaya know?
>>
>> I'd observed what appeared to be significant amounts of fork shudder when braking with dirty rims
>> on my road bike - fork deflections that appeared to be the almost an inch in fore/aft movement...
>> but the r.b.t. consensus seemed to be that there was no reasonable (non-destructive) way to
>> deflect a fork that much.
>
>I don't think that is accurate. The consensus is that the blades themselves do not flex. The
>assumption of some was that the purpose of a curved fork blade was to provide some "shock
>absorbsion" or spring, which of course it does not.
>
>Forks do flex. At the crown.

I think you misread my post. I'm talking about fore/aft movement of hub. My testing method doesn't
identify whether that's happening at the crown, in the legs, or a combination of both. I certainly
wouldn't suggest that there's any compressive capability in a curved fork (in fact, just wrote
another post stating just that 5 minutes ago).

Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame
 
>I rubber-banded a gold paint marker onto the rear triangle of my fixed gear bike, aligned so that
>the marker's tip contacted the rear tire's tread just off-center. Then I locked up the front brake
>and applied gentle force to the top tube, back and forth with two fingers.

>The result was 14mm of movement (or plus / minus 7mm) of the rear wheel (I verified the front wheel
>didn't move at all).

I threw some numbers off the top of my head into the basic cantilever beam, 1 inch steel steerer
tube, 0.060 in wall, assumed the entire fork had the same cross-section equation and was 26 inch
inches long.

The length is critical because the deformation is the third power of the length. Since most of the
deflection is caused by the bending of the section near the crown, I figured the tapering probably
did not change much and that the effective moment on the fork was caused by the contact patch, thus
L=26inches (should have been 28 inches)

Then figuring RBTers are a bunch of cyclists, I limited the load to 50 lbs.

Plug it in, and out pops 12.622 mm as the deflection.

For a rough and dirty calculation, I would say this is close given the crudeness of the calculation
but it does indicate that the deflection measured is reasonable.

Previously it has been suggested to remove the wheel and just push on the fork tips. This reduces
the effective length and cuts the deflection roughly by a factor of 8 to about 1.5 mm.

One source of error is the deflection of the brake under the load, some brakes are pretty sloppy.

jon isaacs
 
On Mon, 03 Mar 2003 09:17:53 -0500, Mark Hickey wrote:

> "David L. Johnson" <David L. Johnson <[email protected]>> wrote:
>
>>On Sun, 02 Mar 2003 19:30:45 -0500, Mark Hickey wrote:
>>
>>> Hey, whadaya know?
>>>
>>> I'd observed what appeared to be significant amounts of fork shudder when braking with dirty
>>> rims on my road bike - fork deflections that appeared to be the almost an inch in fore/aft
>>> movement... but the
>>> r.b.t. consensus seemed to be that there was no reasonable (non-destructive) way to deflect a
>>> fork that much.
>>
>>I don't think that is accurate. The consensus is that the blades themselves do not flex. The
>>assumption of some was that the purpose of a curved fork blade was to provide some "shock
>>absorbsion" or spring, which of course it does not.
>>
>>Forks do flex. At the crown.
>
> I think you misread my post. I'm talking about fore/aft movement of hub.

Right.

> My testing method doesn't identify whether that's happening at the crown, in the legs, or a
> combination of both. I certainly wouldn't suggest that there's any compressive capability in a
> curved fork (in fact, just wrote another post stating just that 5 minutes ago).

Agreed. But your reference to consensus above was that they weren't supposed to do that at all. And
it is the case that most of the motion you see is coming from the brake, which is very flexy, and
thus wheel motion allowed by the brake -- along with the tire. Removing the wheel will cut down this
flex considerably, but some still does occur at the crown.

--

David L. Johnson

__o | Deserves death! I daresay he does. Many that live deserve _`\(,_ | death. And some that die
deserve life. Can you give it to (_)/ (_) | them? Then do not be too eager to deal out death in
judgement. -- J. R. R. Tolkein
 
On Sun, 2 Mar 2003 19:29:51 -0800, [email protected] wrote (in message
<[email protected]>):

> And by the way, there are evening courses for people for whom English is a second language. They
> usually cover writing and spelling as well as sentence structure.
>
> Jobst Brandt [email protected] Palo Alto CA

Do you have a supervisor that slaps you down at work, so that you need to take out your
frustrations on some random newsgroup poster? I guess that's preferable to your snapping ala
Michael Douglas in "Falling Down" and laying waste to the general public..
 
David Reuteler wrote:
> god, i realize this. the first thing i did was check the fork flex in the "jobst method" (read my
> other post) and yes, there was flex at the crown. that's static, however. the shudder is around
> 5hz or so and is probably more like 5cm total (2.5cm fore/aft) in amplitude. when you watch the
> fork while riding it *APPEARS* to be flexing the blades (the last 20cm). that may be an optical
> illusion. i would like to see that on film.

Compared to the resonant frequency, 5Hz is rather low, so the static behaviour is a good
approximation of what is going on. However, I agree that it would be an interesting thing to
view a film.

--
R.

<> Richard Brockie "Categorical statements <> The tall blond one. always cause trouble." <>
[email protected]
 
"David L. Johnson" <David L. Johnson <[email protected]>> wrote:

>On Mon, 03 Mar 2003 09:17:53 -0500, Mark Hickey wrote:

>> My testing method doesn't identify whether that's happening at the crown, in the legs, or a
>> combination of both. I certainly wouldn't suggest that there's any compressive capability in a
>> curved fork (in fact, just wrote another post stating just that 5 minutes ago).
>
>Agreed. But your reference to consensus above was that they weren't supposed to do that at all.

I recall lots of people being told that it was an optical illusion that their hub was moving back
and forth significantly under uneven braking.

> And it is the case that most of the motion you see is coming from the brake, which is very flexy,
> and thus wheel motion allowed by the brake -- along with the tire. Removing the wheel will cut
> down this flex considerably, but some still does occur at the crown.

I considered this, but agree with Jon's analysis. If the brake was flexing, there would be some
rotation of the front tire. If the tire were flexing, you'd see movement of the rim relative to
the ground. Yes, there may be some extremely small amount of movement from both, but it's very,
very small - and the rotation of the rear tire is very, very easy to see. Try it and you'll see
what I mean.

Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame
 
Richard Brockie <[email protected]> wrote:
: Compared to the resonant frequency, 5Hz is rather low, so the static behaviour is a good
: approximation of what is going on. However, I agree that it would be an interesting thing to
: view a film.

5hz is a pretty rough guess, obviously. when watching it shudder roughly a third the blade from the
brake boss to the hub appears to be relatively fixed while the last two-thirds appear to flex the
distance (2.5cm or more). it's this very difference between what i can see statically and what my
eyes seem to be showing me when it's shuddering that makes me want to see it on film. i think you
get that, however.

adding a brake booster increased the frequency (not sure about the amplitude) noticeably maybe as
much as twice. and tire pressure also (at least with wider tires, 32mm) makes a difference. high
pressure and the fork will tend to just flex backwards (no shudder), if i lose pressure (get a slow
leak) it'll start to shudder. it's always dependent on the power of the brake with more being worse.
--
david reuteler [email protected]
 
> >On Mon, 03 Mar 2003 09:17:53 -0500, Mark Hickey wrote:
>
> >> My testing method doesn't identify whether that's happening at the crown, in the legs, or a
> >> combination of both. I certainly wouldn't suggest that there's any compressive capability in a
> >> curved fork (in fact, just wrote another post stating just that 5 minutes ago).
> >
> Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame

My experience indicates, to me at least, that there is a fair amount of flex that occurs in the
blades themselves. I have a couple of forks that are identical in all aspects except for the fork
blades. Both use a Henry James fork crown and the same True Temper steerer tube. One fork uses RCX
blades whereas the other uses thicker OXRCX blades. Of these two forks, the RCX fork is quite a bit
more flexable. Clearly the thicker fork blades make a difference in overall fork stiffness. I think
this is why the tube manufactuers make different models of fork blades with a varity of thickness;
so the builder can taylor the ride. No different than building a frame.

Ed
 
Status
Not open for further replies.