Cannondale's tests of disks and QRs

On Tue, 28 Sep 2004 00:59:05 GMT, SuperSlinky
wrote:

I have seen only one accident demonstrated to be due to disc brakes
ejecting a wheel--that of James Annan, and his was a very unusual bike,
to say the least.

You think that Russ Pinder's opinion of his crash is wrong, then?

I do believe that the CPSC would be fully justified in
issuing a recall of forks like his, especially those attached to
tandems, but the rest of us are riding very different bikes and dropping
an a-bomb on the industry is simply not justified.

Who said anythign about pressing the nuclear button? I think all
KJames wants - and I'm with himon this - is some serious investigation
which actually quantifies the issue.

Although to be fair the engineering change required to fix the problem
at source does appear to be very small, so I'm somewhat surprised it
is not being embraced. Here I think the threat of lawyers is to
blame.

Couldn't a QR work loose because of something besides a disc brake?

Who knows? Nobody's tried the tests. James' proposed mechanism for
QR loosening makes sense from a mechanical perspective, and at the
very least needs to be properly followed up.

Much more important than Cannondale's test is the test that many
thousands of riders do each day.

That's what DeHavilland said about the first few Comet crashes, IIRC.

On Tue, 28 Sep 2004 04:06:07 +0100, Tony Raven
wrote:

That same article goes on to talk about preventing vibration loosening
and names serratation on the nut, nyloc type friction locking and
loctite type chemical locking as the three recognised preventative
measures. So give that two of these are being used on a standard Shimano
QR, why do you still think it is vibration loosening?

There are a lot of techniques to prevent vibration loosening of
fasteners, and in some cases multiple techniques are applied. That
tends to result in a much longer mean time between failures, but
rarely makes vibration loosening impossible. You won't lose a Nyloc
nut, but I have seen them lose torque, and I have seen various other
vibration-resistant techniques fail (including thread locking
compound). The only thing I can't recall seing come loose is a
Nord-Lock washer set, and I bet even they will in the right (or wrong)
circumstances.

But hey, if one of the major manufacturers wants to set up a jig which
simulates repeated heavy braking on some good old-fashioned rough
stuff, and it reveals that there is no problem, I'll probably be quite
happy. I don't suppose it would be difficult for them, I'm sure they
have the jigs already for testing fork designs. My only quibble here
is that the Cannondale investigation seems specifically designed to
prove there is not a problem, rather than to find out if there is a
problem.

On Mon, 27 Sep 2004 21:24:47 -0700, jim beam
wrote:

you miss the point: if the downhill fork is necessary,
then there is a problem!

from what i understand, the big axles on d/h forks are more to do with
preventing fatigue in "standard size" weeny little 9mm axles than
anything else. d/h bikes run 40lbs or more - a substantial increase in
fatigue demand.

By "downhill fork" I understood you to mean a solid axle with a closed
fork, i.e. the axle passes through holes in the ends of the fork
instead of having dropouts. It would be kind of hard for these to
eject the wheel :-)

What did you mean?

So you think that a front wheel ejecting is no more likely to cause
you to land head first than any other component failure? That's an
odd point of view. It sounds like your riding is far from typical.

i suppose most "mountain bikes" spend more time in urban environs than
on hills, so i guess technically, you're right; my riding /is/ far from
typical. not that i understand why you're still going on about
ejection. from what i've seen, it's /way/ less likely than chain
failure, crank failure, pedal failure, stem failure, bar failure, brake
failure, fork failure, frame failure..................

I've told you why. I've had chain failure - you stop. I've stayed on
the bike following two chain failures. I've had front brake failure.
I stayed on the bike. I don't know how you can stay on the bike if
the fron wheel comes out. Frame failure is the only thing which
really compares, and actually I've had two of those and not crashed
either time.

Just zis Guy, you know? wrote:
compound). The only thing I can't recall seing come loose is a
Nord-Lock washer set, and I bet even they will in the right (or wrong)
circumstances.

googles. How do you undo them when you want to undo the bolt, given
that http://www.nord-lock.com/default.asp?id=3&language=4 says "Here you
see what happens when a bolt is untightened with a wrench. The pair of
washers expand more than the corresponding pitch of the thread allows
the bolt/nut to rise"?
With a locking wire or spilt pin and castellated nut, you can remove the
locking element and then undo the thread normally.
If the answer is "you don't undo them", what's wrong with a spot weld? :-)

On 28 Sep 2004 13:34:19 +0100 (BST), (Alan
Braggins) wrote:

How do you undo them when you want to undo the bolt

With a big f**k-off spanner ;-)

Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

88% of helmet statistics are made up, 65% of them at Washington University

Just zis Guy, you know? wrote:
On Tue, 28 Sep 2004 00:59:05 GMT, SuperSlinky
wrote:


I have seen only one accident demonstrated to be due to disc brakes
ejecting a wheel--that of James Annan, and his was a very unusual bike,
to say the least.


You think that Russ Pinder's opinion of his crash is wrong, then?


I do believe that the CPSC would be fully justified in
issuing a recall of forks like his, especially those attached to
tandems, but the rest of us are riding very different bikes and dropping
an a-bomb on the industry is simply not justified.


Who said anythign about pressing the nuclear button? I think all
KJames wants - and I'm with himon this - is some serious investigation
which actually quantifies the issue.

Although to be fair the engineering change required to fix the problem
at source does appear to be very small, so I'm somewhat surprised it
is not being embraced. Here I think the threat of lawyers is to
blame.


Couldn't a QR work loose because of something besides a disc brake?


Who knows? Nobody's tried the tests. James' proposed mechanism for
QR loosening makes sense from a mechanical perspective, and at the
very least needs to be properly followed up.

no, it _doesn't_ make sense from the mechanical perspective. just
because there is a resolved pullout force does _not_ mean it exceeds the
pullout force necessary to cause slippage, let alone ejection. have you
done the math for brake cable pullout? i mentioned this before but it
doesn't seem to have sunk in.

bottom line, if the pullout force necessary to cause slippage [ignoring
lawyer lips] exceeds the resolved force of braking by a factor of 3 [and
that's a very conservative calculation, unlike the maximized braking
force calc] then this is all a wild goose chase.



Much more important than Cannondale's test is the test that many
thousands of riders do each day.


That's what DeHavilland said about the first few Comet crashes, IIRC.


you're well senior to me if you recall all that first hand.

failure analysis is all about omissions. in the comet case, it was
omission of research into all the stress concentrations caused by a
relatively small window corner radius in conjunction with a pressurized
fuselage. omissions in territory for which there is no prior
experience, like the comet, is the way engineering has evolved. on the
other hand, and this is something materials folks see all the time,
there are omissions where the territory /is/ well known, but the
engineer hasn't done their homework. in this case, we have one side of
the equation, the resolved braking force, and the other, the force
necessary to cause slippage. this latter has been ignored - a
rudimentary omission. the /real/ debate is whether this omission was
selective to further some other agenda.

On Tue, 28 Sep 2004 06:44:07 -0700, jim beam
wrote:

James' proposed mechanism for
QR loosening makes sense from a mechanical perspective, and at the
very least needs to be properly followed up.

no, it _doesn't_ make sense from the mechanical perspective. just
because there is a resolved pullout force does _not_ mean it exceeds the
pullout force necessary to cause slippage, let alone ejection.

And that is precisely the question which I am saying should be
answered: does it exceed the pullout force, and if so, under what
conditions? But nobody with the wherewithal to test it seems terribly
interested in doing so.

have you
done the math for brake cable pullout? i mentioned this before but it
doesn't seem to have sunk in.

Have you tried riding a bike without a front brake? And without a
front wheel? I mentioned this before but it doesn't seem to have sunk
in.

bottom line, if the pullout force necessary to cause slippage [ignoring
lawyer lips] exceeds the resolved force of braking by a factor of 3 [and
that's a very conservative calculation, unlike the maximized braking
force calc] then this is all a wild goose chase.

You are begging the question. Who says it does? And even if it did,
to what extent can we rely on that given that there is a credible
mechanism advanced for the loosening of the QR under repeated heavy
braking?

So, unresolved questions. Being old-fashioned, my preferred solution
for unresolved questions is to set up some experiments.

Much more important than Cannondale's test is the test that many
thousands of riders do each day.

That's what DeHavilland said about the first few Comet crashes, IIRC.

you're well senior to me if you recall all that first hand.

Ah, so we are not allowed to know history, then? I used to volunteer
at a DeHavilland museum.

failure analysis is all about omissions. in the comet case, it was
omission of research into all the stress concentrations caused by a
relatively small window corner radius in conjunction with a pressurized
fuselage.

Precisely. And in this case it's omission of tests of repeated heavy
braking on a fork / dropout whose geometry was clearly designed for
rim brakes and has not been reworked for the different forces involved
in disc brake use. Nobody thought to check. It's very obvious from
the initial reactions that nobody thought to work out the resultant
force. Having been prompted to do so they are now relying on the fact
that skewers are "obviously" not going to come undone. But "obvious"
things have a habit of being wrong, and no amount of Usenet discussion
will substitute for a realistic test. I am perfectly prepared to
admit that such a test may well show that only a fork such as James',
where the dropout was almost in line with the effective force of the
disc brake, is ever likely to be affected. I'd be quite happy with
that, since I ride a disc-braked bike every day. Until then I'm gogin
to be very paranoid about testing my QR.

in this case, we have one side of
the equation, the resolved braking force, and the other, the force
necessary to cause slippage. this latter has been ignored - a
rudimentary omission. the /real/ debate is whether this omission was
selective to further some other agenda.

The force required to cause slippage has not been ignored by James,
and not by me either. Who are you saying is ignoring it?

Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

88% of helmet statistics are made up, 65% of them at Washington University

Pieter wrote:
contact with the ground. As far as I can see when the wheel is off the
ground there is space for the wheel to move but the forces are
determined just by the spinning energy of the front wheel, and hence
will be much less than you calculated. When the bike is on the ground
is on the ground the QR axle can't move (it is being pushed up)

The ejection force is enormously larger than the weight of the bike+rider.
--
David Damerell Distortion Field!

On 23 Sep 2004 05:53:24 -0700, (Pieter) wrote in
message :

When the bike is on the ground
is on the ground the QR axle can't move (it is being pushed up)and the
forces you calculate are being absorbed in distortion of the rim and
spokes (I think). So what am I missing ?

The component of the braking force which is acting to eject the wheel
can easily be greater than the force exerted by the rider's weight
tending to hold the wheel in.

Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

88% of helmet statistics are made up, 65% of them at Washington University

Man, you guys can blow more wind than a Floridian hurricane!

When are you gonna blow out of town (or at least stop x-posting)?!?!?!

--
Slacker