"Actually you are the first person to bring up this issue"

tcmedara wrote:

Could it be that the cases that have occurred are
due to failure of substandard QR's?

What is a "substandard QR"? Are you aware of any designs that are
"substandard"? What, if anything, should be done about this?

Why hasn't this come out of the
statistical noise level, despite the attention of the cycling public?

It has. It came out of the statistical noise level sufficiently to draw
my attention to it, and enough to generate an overwhelming weight of
consistent evidence pointing to the simple and uncontroversial
explanation of a bolt loosening under a large transverse load.

Likewise, if the theoretical problem proves to be a statistically
significant cycling hazard then the various countries consumer agencies will
hopefully step in. Why hasn't it happened yet?

Because of all the people who have experienced the problem, some have
been convinced that it was "probably" their own fault, some have been
assured that something is being done and there is no need to make a
fuss, some don't want to face the vitriol and abuse that will
undoubtedly come their way if they do make a complaint in public, and
some have simply upgraded to a bolt through fork and thought nothing
more of it. A few weeks ago, someone decided to actually do something,
hence this thread.

James

Russ wrote:

"James Annan" wrote in message
...

It's now a year since the QR/disk brake problem hit the headlines, and I
thought some of you might be interested in hearing how the manufacturers
are dealing with it.


It's actually 1 Yr and 3 days :-(

But unfortunately a little under a year since the problem was explained...

Will be interesting to hear how you get on if you do take it further. I
should make clear that this thread was certainly not aimed at you in
particular. However, there are dozens of people who have all ducked out
of taking any personal responsibility for the issue, probably largely
out of naive belief in the good faith of the manufacturers and an
admirable dislike of lawyers.

James

in message , Slacker
') wrote:

Did you try the test of pushing the bicycle forward
with an open QR and applying the disc brake? If so, did you not
notice that the fork lifts off the axle. You dismiss "line drawings
and vector calculations" although you are surrounded by machines that
are designed by these methods and find them reliable. The test I
offer does not rely on such derivative methods and gets directly to
the issue. How about trying that and apply your own analysis to it.

I done this test (by accident) before this whole issue ever surfaced.
Indeed, it does want to pull out, which only proves "our" point; An
improperly installed front wheel (disc + QR) is a very, very dangerous
thing.

So what's your point again?

That safety equipment should not have unsafe modes of failure.

--
(Simon Brooke) http://www.jasmine.org.uk/~simon/

[ This .sig subject to change without notice ]

James Annan wrote:

| tcmedara wrote:
|
| Could it be that the cases that have occurred are
| due to failure of substandard QR's?
|
| What is a "substandard QR"? Are you aware of any designs that are
| "substandard"? What, if anything, should be done about this?
|
| Why hasn't this come out of the
| statistical noise level, despite the attention of the cycling public?
|
| It has. It came out of the statistical noise level sufficiently to draw
| my attention to it, and enough to generate an overwhelming weight of
| consistent evidence pointing to the simple and uncontroversial
| explanation of a bolt loosening under a large transverse load.

Just wondering whether the unscrewing effect on QRs could also apply
to "normal" nutted axles. The effect on QRs seems to need an initial
slippage, but that could happen with nuts too, surely? Is it simply
greater friction or is there a design difference? Apos if I've missed
this point being gone over before.

--
Patrick Herring, Sheffield, UK
http://www.anweald.co.uk

Patrick Herring wrote:

Just wondering whether the unscrewing effect on QRs could also apply
to "normal" nutted axles.

Yes, although nutted axles do have the advantage that the RH end clamps
independently of the LH, and also I believe that they can generate
larger clamping forces (hence used with horizontal rear dropouts). They
however have the disadvantage (according to Chris Juden) that it is easy
to not notice when they are loose, as there is no lever to flop around.
Anyway, none of this addresses the basic problem of a large braking
torque being opposed by nothing more than thin air and a bit of friction.

James

Pete Biggs wrote:
tcmedara wrote:

Where are all the injuries?

Are you familiar with Russell Pinder's accident?

~PB

Yup. Are you familiar with basic logic and statistical analysis?
http://www.datanation.com/fallacies/hasty.htm

I have nothing but sympathy for Mr Pinder. If his situation can improve the
safety of our sport then somthing good can come from his tragedy. However,
conclusions drawn from supposition and based on emotion aren't going to do
that. I'd hate to see everyone point to disks as the new scourge in
mountain biking and miss the real culprit that caused his accident. Sure no
one wants to admit human error. But in the absence of firm evidence,
there's no way to draw firm conclusions about what actually happened.

Unfortunately the only way to determine if this is a real problem is to
study it, not gather a bunch of cases where it might have been the cause and
then begin prosteltyzing about it. Data collection is the first step
obviously, but you've got to draw sound conclusions from the data.

Tom

James Annan wrote:
tcmedara wrote:

Could it be that the cases that have occurred are
due to failure of substandard QR's?

What is a "substandard QR"? Are you aware of any designs that are
"substandard"? What, if anything, should be done about this?

I'm suggesting them as an alternative hypothesis. They could be bad design
or poor manufacturing. What should be done is that the potential danger of
wheel ejection be studied to conclude if it poses an actual threat to
cyclists.

Why hasn't this come out of the
statistical noise level, despite the attention of the cycling public?

It has. It came out of the statistical noise level sufficiently to
draw my attention to it, and enough to generate an overwhelming
weight of consistent evidence pointing to the simple and
uncontroversial explanation of a bolt loosening under a large
transverse load.

You misunderstand. Has the incidence of injury, wheel ejection, or QR
failure increased as a result of the growth of disk brake usage? Is the
rate of incidents more, less, or equal to the rate prior to the advent of
the current disk design. Just because you notice a problem and assert
"overwheming evidence" doesn't make it a valid basis to draw a conclusion.
It makes it a hypothetical problem that can't be differentiated from a
variety of possible intervening variables.

Likewise, if the theoretical problem proves to be a statistically
significant cycling hazard then the various countries consumer
agencies will hopefully step in. Why hasn't it happened yet?

Because of all the people who have experienced the problem, some have
been convinced that it was "probably" their own fault, some have been
assured that something is being done and there is no need to make a
fuss, some don't want to face the vitriol and abuse that will
undoubtedly come their way if they do make a complaint in public, and
some have simply upgraded to a bolt through fork and thought nothing
more of it. A few weeks ago, someone decided to actually do something,
hence this thread.

So you've got it figured out, but everyone else has their head in the sand,
even those who might have been affected by the situation are too unaware?
I'm not buying it, it's just outright rationalization. Again, if the CPSC
will issue recalls for the slightest potential hazard even with no injuries,
I find it hard to believe they are going to patently reject this issue after
they have, in fact, looked at it. Hey, maybe with better data they'll be
forced to address it, but it appears that right now all you've got is
anecdotes, theory, and assertion. Show some data that demonstrates a real
problems. A couple of unresolved cases where qr failure is suspected and a
few letters from dissatisfied customers is just not a great basis for
engineering or public policy

Tom

James

Tim McNamara wrote in message ...
(Jonesy) writes:

James Annan wrote in message ...

So there you have it. At this rate, by the time next year's
complaint comes in, they will presumably have forgotten this first
one. How convenient for them. Those who thought that it wouldn't do
to kick up a fuss because the poor manufacturers were doing their
best, may wish to re-examine their approach. Or else studiously
ignore this post in the vain hope that the problem will go away.

Assuming, of course, that there actually *is* a problem.

There's no doubt that there is a problem.

That depends on how you define "problem," and where you draw the line.

I draw the line in real-world application, not at the force-diagram
level. I just plain don't accept anecdotal evidence as proof. As
support of hypothesis, yes. As proof? No. There is a difference,
and it is real.

The only question is how
often it happens.

That's a very pertinent question. If it doesn't happen often, then
it's hard to define it as a problem.

I agree that to the poor folks who experience catastrophic failure, it
is a very serious problem. But to those folks who go year after year
after year without any sort of evidence of difficulty, then the
problem just plain doesn't exist.

It'll be nice to see all that wonderful, properly-controlled data
from the laboratory testing.

Think of it like Microsoft Windows: *you* are the testing lab.

This is a diversionary argument. Controlled testing will answer the
question. Until then, proof doesn't exist. If one were to give
weight to the anecdotal evidence, then I would have to judge "no
problem" to be the norm, and "problem" to be in the vast minority.

When that comes out for public view, then we'll all be able to avoid
drawing conclusions from anecdotal evidence.

It would be nice to have something other than a detailed engineering
analysis quantifying the force and magnitude of the ejection force

It is not all-inclusive. I notice that there is an assumption of
indepence on the part of the fork legs which does not exist in the
real world. The situation is more complicated than Mr. Annan's
drawings indicate.

created by front disk brakes mounted behind the shock fork leg, paired
with a detailed engineering analysis of how the QR can be unscrewed by
the repeated ejection force resulting from normal use.

Actually, not. The mechanism of unscrewing does not address skewer
and axle nut knurling, nor the material from which the skewer is made.

In order for unscrewing to occur, there must be movement. If there is
no movement, no unscrewing can occur.

I have anecdotal evidence from two different forks with disk brakes
that show that movement does not occur, DESPITE Mr. Brandt's absolute
prediction otherwise.

How do I know? Register marks, made by me, on both my DO and my
skewer nut/lever. They have not moved at all, after literally
thousands of braking cycles.

This experiment is on-going.

In addition, the indentations on my Al DOs match the knurling on my
skewers and axle nuts, and they show no signs of tear-out. In any
axis.

Ummm, oh yeah,
that's *not* anecdotal evidence.

But it is NOT a complete analysis. If the skewers used are not all of
uniform type, then generalizing is not possible.

That's objective evidence, which the
anecdotes (including video of a front wheel ejection) serve to
support.

Evidence proof.

It may be that ejection/QR loosening only happen under very specific
condtions. It may not be a general function of front disk braking
systems.

The fact that the brake creates an ejection force at all is evidence
of the design flaw.

That's an opinion, not a fact.

It's really quite simple.

It is easy for folks to imagine this to be true. But I think it is
demonstrated that the situation is not as simple as you pretend. If
it were, then there would be more evidence, and serious carnage. But
there isn't, which points to a flaw in your simplicity hypothesis.

The arguments against
this citing improper use of a QR are just desperate smoke and mirror
attempts to obfuscate.

Nonsense. You can't separate human error from the equation, any more
than you can eliminate gravity from Mr. Annan's calculations. I
consider Mr. Brandt's dismissal strategy to be smoke and mirrors:
eliminate the QR and grab the brake - et viola! Everything is proven!
Except it isn't. The QR loosening part *is* the rub. If it doesn't
loosen, then the ejection force is moot, from a functional standpoint.

Designing an experiment to eliminate the bone of contention is
disingenous at best, and borders on intellectual dishonesty.
--
Robert Jones

On Sat, 20 Mar 2004 15:50:23 -0600, Tim McNamara
wrote in message
:

Try diagramming out the forces. You'll see that they work out to be
about the same whether the brake is in front of or behind the fork
leg.

I was consideringg the interface between the mounting lugs and the
fork itself; the mounting is (on my bike anyway) welded onto the
forks. The geometry being what it is the forces exerted on the
mounting itself by the brake would be similar.

--
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

On Sat, 20 Mar 2004 14:43:26 -0800, Slacker
wrote in message :

I done this test (by accident) before this whole issue ever surfaced.
Indeed, it does want to pull out, which only proves "our" point; An
improperly installed front wheel (disc + QR) is a very, very dangerous
thing.

Try again with a rim-braked wheel. That's the point. The dropout
angles are the same, which indicates that the manufacurers are simply
ignoring the issue.

--
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