Carlton Reid on QR safety

On Mon, 06 Feb 2006 01:23:22 +0000, jobst.brandt wrote:

I think that calculation accurately states the problem and should make
apparent why this is a bad design.

The racing car design world went through this in the 1970s when the
argument was that leading calipers loaded the hub bearings under braking
but trailing calipers unloaded the hub. So the effect is there without
doubt. For racing cars (and modern motorbikes) it turned out that brake
cooling was more important than bearing loads.


Forget about GC and other calculations that muddy the straight forward
relationship between downward and upward forces on the dropout in
question.

I am not questioning the direction of the load, what I _am_ questioning
is its magnitude in relation to the other loads present. To find the
value of the ejection force and the value of the retaining forces we
need to know the geometry of the whole bike and rider.


I think caliper position is an obvious main item for discussion.

If the caliper is forward of the axle it will not try to eject the
spindle, but it will obstruct cooling of the disc. If the caliper is
aft of the axle the cooling will improve, but it will try to eject the
spindle. If the cooling _is_ needed are the ejection forces outweighed
by the existing retention forces?

As I have not yet seen a bicycle brake disk that looked as if any effort
had been made to improve its cooling (funky slots don't do a lot [1]), I
suppose that brake overheating is not a significant factor in bike
brakes, so the short-duration loading on emergency stops should take
precedence and the caliper would preferably be to the front. If I
lived in the Alps instead of the Fens I might have a different view.



[1] Slots and holes take the "fire-band" off the pad material. To get
the pads that hot the disk will be glowing red. To have the slots
doing anything to cool the disk there needs to be an airflow _though_
them, which needs air ducts or some other method of moving air at right
angles to the direction of travel.



Mike

James Annan wrote:
Carlton Reid has a puff piece about a new "Secure QR system" on bikebiz:
http://www.bikebiz.co.uk/daily-news/article.php?id=6427

While promoting this new mechanism as "safer" than the existing system,
he also insists that "industry experts say QRs are safe, when used
correctly".

This assertion is backed up with a quote from "industry expert" Bob
Burns (actually Trek's *lawyer*), which is nothing more than a
boilerplate denial dating to a few years ago when the QR/disk issue
first surfaced.

Strangely, alongside this there is no space in his article for these
quotes from people who actually have some relevant engineering and
technical experience:

Gee, James, some pretty slick editing on your part, too, being that the
quote from Bob Burns is three years old, and the last time around, he
was praised for being the only representative from a large manufacturer
who was willing to look into your claims. See
http://www.singletrackworld.com/article.php?sid=1005 . Also, there is
nothing sinister about Mr. Burns being Trek's *laywer*. General
counsel usually handles liability claims and knows what claims have
(and have not) been made. Imagine that -- no *conspiracy*.

Caveat: I am an evil defense lawyer. Oddly enough, me and my cabal of
evil defense lawyers here in Oregon -- land of some of the best
mountain biking in the world -- have yet to see a QR/disc brake claim.
Not that the design cannot be improved (a question I am not competent
to answer), but there is no plague of claims as far as I can tell. --
Jay Beattie

Jay Beattie wrote:
James Annan wrote:
Carlton Reid has a puff piece about a new "Secure QR system" on bikebiz:
http://www.bikebiz.co.uk/daily-news/article.php?id=6427

While promoting this new mechanism as "safer" than the existing system,
he also insists that "industry experts say QRs are safe, when used
correctly".

This assertion is backed up with a quote from "industry expert" Bob
Burns (actually Trek's *lawyer*), which is nothing more than a
boilerplate denial dating to a few years ago when the QR/disk issue
first surfaced.

Strangely, alongside this there is no space in his article for these
quotes from people who actually have some relevant engineering and
technical experience:

Gee, James, some pretty slick editing on your part,

_MY_ part? Did you not realise that the bikebiz article was written by
Carlton Reid, not me?


This is a _direct_ cut and paste, completely unedited, from the middle
of Carlton's article:

---
[...] industry experts say QRs are safe, when used correctly.

In 2003, Bob Burns, Trek's US-based General Counsel, told BikeBiz.com:
---

James

Helen Deborah Vecht writes:

I am neither a physicist nor engineer; many here are.

I will leave comments to those more qualified.

The diagrams are so simple that you can understand the principles
involved and understand the issue, even if you can't do the math.
Take responsibility for knowing for yourself.

Mike Causer writes:

On Mon, 06 Feb 2006 01:23:22 +0000, jobst.brandt wrote:

I think that calculation accurately states the problem and should
make apparent why this is a bad design.

The racing car design world went through this in the 1970s when the
argument was that leading calipers loaded the hub bearings under
braking but trailing calipers unloaded the hub. So the effect is
there without doubt. For racing cars (and modern motorbikes) it
turned out that brake cooling was more important than bearing loads.

FYI that guy you're talking to designed suspensions for Porsche. He's
probably aware of that.

Forget about GC and other calculations that muddy the straight
forward relationship between downward and upward forces on the
dropout in question.

I am not questioning the direction of the load, what I _am_
questioning is its magnitude in relation to the other loads present.
To find the value of the ejection force and the value of the
retaining forces we need to know the geometry of the whole bike and
rider.

You're muddying the waters unnecessarily. You need to know the
orientation of the force compared to the dropout slot, and the
difference between the distance from the axle to the braking surface
and from the axle to the tire contact patch. The first will tell you
whether the reaction force from braking will push the axle out of the
dropout; the second will tell you how much the force driving the wheel
will be multiplied. The smaller the brake disk, the higher the
mutiplication of force- so a wheel with a 6" disk will result in a
stronger ejection force that an 8" disk (a 26" MTB wheel being
actually about 24" in diameter, so the difference is 4:1 for a 6" disk
and 3:1 for an 8" disk).

The only number you need to add is how much force is resulting from
forward motion of the bike, and for that you only need velocity and
mass, not the rest of the stuff you're claiming is necessary.

I think caliper position is an obvious main item for discussion.

If the caliper is forward of the axle it will not try to eject the
spindle, but it will obstruct cooling of the disc. If the caliper
is aft of the axle the cooling will improve, but it will try to
eject the spindle. If the cooling _is_ needed are the ejection
forces outweighed by the existing retention forces?

The caliper is behind for fork leg to protect it from hitting rocks,
bits of tree and undergrowth. The location has nothing to do with
cooling.

Mike Causer writes:

I think that calculation accurately states the problem and should
make apparent why this is a bad design.

The racing car design world went through this in the 1970s when the
argument was that leading calipers loaded the hub bearings under
braking but trailing calipers unloaded the hub. So the effect is
there without doubt. For racing cars (and modern motorbikes) it
turned out that brake cooling was more important than bearing loads.

Forget about GC and other calculations that muddy the straight
forward relationship between downward and upward forces on the
dropout in question.

I am not questioning the direction of the load, what I _am_
questioning is its magnitude in relation to the other loads present.
To find the value of the ejection force and the value of the
retaining forces we need to know the geometry of the whole bike and
rider.

I don't see why. All that is required is what I stated, the ratio of
disk diameter to tire OD and the position of the caliper. The fore
that the caliper puts on the fork relative to the wheel is as I
stated, only caliper location is the matter at hand. You'll note this
requires no further information about rider position or other bicycle
dimensions.

I think caliper position is an obvious main item for discussion.

If the caliper is forward of the axle it will not try to eject the
spindle, but it will obstruct cooling of the disc. If the caliper
is aft of the axle the cooling will improve, but it will try to
eject the spindle. If the cooling _is_ needed are the ejection
forces outweighed by the existing retention forces?

Well its about time we got rid of the silly holes in the disk and
realize that in most uses, there isn't much wind for cooling. It's
not as if we are talking about road bicycle descending from one
hairpin turn tot he next, coasting at high speed between.

As I have not yet seen a bicycle brake disk that looked as if any
effort had been made to improve its cooling (funky slots don't do a
lot [1]), I suppose that brake overheating is not a significant
factor in bike brakes, so the short-duration loading on emergency
stops should take precedence and the caliper would preferably be to
the front. If I lived in the Alps instead of the Fens I might have
a different view.

[1] Slots and holes take the "fire-band" off the pad material. To
get the pads that hot the disk will be glowing red. To have the
slots doing anything to cool the disk there needs to be an airflow
_though_ them, which needs air ducts or some other method of moving
air at right angles to the direction of travel.

Holes in the current disks reduce surface area and do nothing positive
for cooling. As I mentioned earlier, the holes seem to be a holdover
from auto drum brake mystique.

Jobst Brandt

In article ,
Helen Deborah Vecht wrote:

Michael Press typed



The problem is not statistical.

The problem is anecdotal.

Many here in the UK will know of a single catastrophic event that might
have been the result of wheel ejection.

A single catastrophe does not a statistic make but it does not mean
there is not a problem.

That IS the problem.

Please comment upon the force diagram for front disc brakes.

I am neither a physicist nor engineer; many here are.

I will leave comments to those more qualified.

You owe it to yourself to see for yourself. Jobst Brandt
has already posted a clear word picture of what is going
on. The braking force of the disk caliper on the disk
generates a force. At the fork tips the braking force
translates into a force on the axle in the direction out
of the fork tips, and this force is opposed only by the
clamping of a quick release on the fork tips.

--
Michael Press

In article ,
Werehatrack wrote:

On Sun, 05 Feb 2006 04:46:49 GMT, Michael Press wrote:

In article ,
"Andy H" wrote:

"David Martin" wrote in message
oups.com...

Werehatrack wrote:
Those of us who have seen your prior postings about the issue of disc
brake ejections are fully aware of your position on the matter. Is it
possible for you to accept the fact that for the majority of the
readers, the evidence thus far published is not persuasive that there
is a serious problem here which is not related to user error?

Why do you claim to speak for the majority of readers, most of whom
have expressed no public opinion on the matter?

..d

Does the fact that the majority of people have expressed no public opinion
(read; interest) speak volumes as to the severity of the problem? Do YOU
know the relevant statistics to say that this is a major problem or design
flaw?

Life is inherently risky and I for one would rather check my qr's before a
ride and have disk brakes than try to do what I do with rim brakes.

I draw inferences from the fact that those who claim there
is no problem refuse to comment upon the force diagram.

You confuse inaction with refusal; they are not the same.

Thus far, the number of incidents documented for which no user error
could be identified is small, and there has been no analysis that I've
seen showing that the rate of failure varies from what is found on
non-disc-brake bikes. Yes, the force diagram indicates that a force
can be present which could lead to this event, but the fact remains
that in the direct experience of the majority of disc-brake-equipped
riders, this force does not have the described effect. Their wheels
are not leaping out, and their brakes are, in the main, stopping them
safely and surely; as a result, they have a hard time seeing that
there's a problem here.

To make matters worse, this is not like the flaming Pinto syndrome, in
which the result was easily duplicated in a test setting. Of course,
as far as I can tell, no one seems to have tried to duplicate the
wheel ejection event in a lab, either, or if they have tried, perhaps
the results have not been conclusive or predictable; I haven't seen
results published in any event.

It is very hard to convince people that a problem is both real *and
serious* when you don't have anything but math and a few isolated
phenomena to offer as evidence. That the problem is real they may
accept if they are math-literate, but since nearly everything has risk
of one sort or another, it's also necessary to convince them that the
problem is serious enough (not just in terms of potential harm should
it occur, but more specifically in terms of the potential for the harm
to come *to them* at all) before they will be persuaded that action is
warranted or necessary. The paucity of demonstrated failures speaks
volumes to the masses.

Most of the reason I'm taking this position is that it is essential,
if the problem is to be addressed, for those advocating change to
understand that the task is not merely to show that something *can*
occur, but that it is *likely*, and most importantly to demonstrate
when, and how, that event is most likely to take place so that a
genuinely repeatable demonstration of the risk is possible. Ford did
nothing about the Pinto fuel tank until the crash test film was shown
on national TV...and the Pinto was a popular car, with a large number
of users who were potentially at risk. Bikes with disc brakes are not
all that common to begin with, and a flaw in them will be of direct
concern to fewer people, so the proponents of change will have to work
even harder to develop a compelling argument in favor of mandatory
action before the existing and potential consumers can be reached and
convinced that they should not buy or use the product until the design
has been made resistant to the described fault.

There are many people who do not crash an are constantly
retightening the quick release nut. It loosens under load
and vibration for which it is not designed.

True or False? You know the results of the free body
analysis and say that it is not a problem.

--
Michael Press

On 5 Feb 2006 18:08:19 -0800, "Jay Beattie"
wrote:


Werehatrack wrote:
In some states, the impetus to settle is being reduced by
defendant-friendly changes to statute, often made under the guise of
"ending lawsuit abuse". Sometimes, what's billed as an abuse-control
measure turns out to be a PLI-defense attorney's nightmare...because
the defense lawyers don't get paid the big bucks for doing the
slam-dunk early dismissal filings, they only make the big bucks when
the case goes on long enough to rack up some worthwhile billable
hours.

You have been reading too many Grisham novels. Every time some tort
reform package is put together, it usually gets smushed like a bug. All
I see are statutes creating new claims for relief and not ending them,
particularly in the employment field.

You're not in Texas. Here, the trend is decidedly in the other
direction, with the effect that it's often difficult to get an
attorney to take a liability case on contingency anymore. The most
significant effect is that even a middle-income plaintiff may not be
able to afford to bring an action for recovery of damages against a
company.

Of course, Texas also has the most selectively gullible electorate in
the nation IMO.
--
Typoes are a feature, not a bug.
Some gardening required to reply via email.
Words processed in a facility that contains nuts.

Tim McNamara writes:

I think that calculation accurately states the problem and should
make apparent why this is a bad design.

The racing car design world went through this in the 1970s when the
argument was that leading calipers loaded the hub bearings under
braking but trailing calipers unloaded the hub. So the effect is
there without doubt. For racing cars (and modern motorbikes) it
turned out that brake cooling was more important than bearing
loads.

FYI that guy you're talking to designed suspensions for Porsche.
He's probably aware of that.

Yes but that was not an issue. Porsche, in an effort to circumvent
patents by Girling and Dunlop, designed a peripherally supported disk
with an inside grasping caliper. This was soon dumped and work
continued with ATE-Dunlop.

Forget about GC and other calculations that muddy the straight
forward relationship between downward and upward forces on the
dropout in question.

I am not questioning the direction of the load, what I _am_
questioning is its magnitude in relation to the other loads
present. To find the value of the ejection force and the value of
the retaining forces we need to know the geometry of the whole bike
and rider.

You're muddying the waters unnecessarily. You need to know the
orientation of the force compared to the dropout slot, and the
difference between the distance from the axle to the braking surface
and from the axle to the tire contact patch. The first will tell
you whether the reaction force from braking will push the axle out
of the dropout; the second will tell you how much the force driving
the wheel will be multiplied. The smaller the brake disk, the
higher the multiplication of force- so a wheel with a 6" disk will
result in a stronger ejection force that an 8" disk (a 26" MTB wheel
being actually about 24" in diameter, so the difference is 4:1 for a
6" disk and 3:1 for an 8" disk).

Dropout slot orientation is a secondary consideration, entirely
missing the first problem which is that the disk brake causes
reversing (up and down loads) on the axle. If you use a motorcycle
type clamped axle, as some BMX bicycles do, then the problem goes away
entirely.

The only number you need to add is how much force is resulting from
forward motion of the bike, and for that you only need velocity and
mass, not the rest of the stuff you're claiming is necessary.

No you don't. You only need to know the parameters I gave. If the
wheel skids, you have the force.

I think caliper position is an obvious main item for discussion.

If the caliper is forward of the axle it will not try to eject the
spindle, but it will obstruct cooling of the disc. If the caliper
is aft of the axle the cooling will improve, but it will try to
eject the spindle. If the cooling _is_ needed are the ejection
forces outweighed by the existing retention forces?

The caliper is behind for fork leg to protect it from hitting rocks,
bits of tree and undergrowth. The location has nothing to do with
cooling.

That's an excuse. You don't need to believe these convenient excuses.

Jobst Brandt