Disc brakes and QRs making headlines

[James Annan]

GigaNews wrote:

> I think a simple force balance will also tell you that the force pushing the wheel out of the
> dropout is dramatically smaller than the force required to move the wheel if the skewer is
> tightened properly or even anywhere near properly.

I think the simple force balance calculation on my web pag contradicts your belief. I think if you
were capable of understanding it, you wouldn't have posted your comment.

James

 

[Andreas Oehler]

Thu, 15 May 2003 07:42:15 -0700, Terry Morse:

>
>> However, bear in mind that the larger the rotor diameter, the greater the load on the fork brake
>> mounts. This is the reason that suspension fork manufacturers (eg Manitou, Fox) void the warranty
>> on their "standard" (ie non-downhill) forks if larger than 6" rotors are used.
>
>Please explain. For a given braking force, a larger diameter rotor will lessen--not increase--the
>force on the brake mounts. There must be some other reason why fork manufacturers don't want large
>diameter rotors.

The only reasons I know against the use of bigger-than-usual discs is that areas in the fork leg are
heated which do not like the heat or that the bigger disc will touch the fork leg.

Andreas (Magura Julie at the tandem on a steal fork)

 

[Jose Rizal]

James Annan:

> GigaNews wrote:
>
> > I think a simple force balance will also tell you that the force pushing the wheel out of the
> > dropout is dramatically smaller than the force required to move the wheel if the skewer is
> > tightened properly or even anywhere near properly.
>
> I think the simple force balance calculation on my web pag contradicts your belief. I think if you
> were capable of understanding it, you wouldn't have posted your comment.
>
> James

I could not see anything on your website which quantitatively compares the force trying to move the
axle off the dropout to the force required to move a "tight" quick release mechanism. If you can
show on a graph the force required to move a QR for a given braking rate, vs the "tightness" with
which the QR is done up, you will be able to have a proper comparison of the forces involved. The
easiest way to determine this is by empirical means.

 

[W K]

"Mike S." <mikeshaw2@coxDOTnet> wrote in message news:GARwa.32722$eJ2.28@fed1read07...

> I've been wondering about this thread. Seems that there are two camps:
one
> is that discs and QRs shouldn't mix, and the other says: properly used the QR should be fine.

Naah. Third camp: this works but looks bad.

So, we need some sensible discussion about whether it is a problem. I'm sure J.A. is not far off
this point, as he is a scientist who would be convinced by decent evidence countering his point.

Please note that one of our usenet colleages has fractured his spine irrepairably recently when the
front wheel came out of his downhill bike.

 

[Jose Rizal]

Andreas Oehler:

> Thu, 15 May 2003 07:42:15 -0700, Terry Morse:
>
> >
> >> However, bear in mind that the larger the rotor diameter, the greater the load on the fork
> >> brake mounts. This is the reason that suspension fork manufacturers (eg Manitou, Fox) void the
> >> warranty on their "standard" (ie non-downhill) forks if larger than 6" rotors are used.
> >
> >Please explain. For a given braking force, a larger diameter rotor will lessen--not increase--the
> >force on the brake mounts. There must be some other reason why fork manufacturers don't want
> >large diameter rotors.
>
> The only reasons I know against the use of bigger-than-usual discs is that areas in the fork leg
> are heated which do not like the heat or that the bigger disc will touch the fork leg.
>
> Andreas (Magura Julie at the tandem on a steal fork)

No, it's everything to do with the increased bending moment on the base of the fork brake mounts
when using larger rotors. This is what Manitou says; see the calculations on my previous post.

 

[W K]

"Doug Taylor" <[email protected]> wrote in message
news[email protected]...
> James Annan <[email protected]> wrote:
>
>
> >> Time will tell whether I am a lame brain or you are Chicken Little.
> >
> >I make no strong claims about the magnitude of the problem, I'm only explaining what goes wrong
> >when it does. I know is it quite rare in absolute terms, but it can have extremely serious
> >consequences. I'm not sure how you get from there to Chicken Little.
>
> By the same token, I don't see how you get to I'm "stupid" because I'm not getting a big hard on
> worrying about my wheel (with snugly fastened QR and lawyer lips on fork dropout) falling off.

From the "math" it seems clear that the QR is only just holding the front wheel in in these
circumstances.

At the very least you are going to have to stop calling those lumps of metal "lawyer lips". Its your
secondary parachute, and how many people would sky-dive if they knew that under fairly standard
conditions they were actually relying on the seldom used secondary safety device.

 

[Jobst Brandt]

Mike Shaw writes:

>> I understand the force diagram, but fail to see what would cause the QR to loosen, unless it was
>> already extremely loose to begin with and vibrated out. My QRs all have serrated faces so this is
>> essentially a non-problem.

This was discussed at length the first time this subject came under discussion. Recognizing that the
disengagement force in the axle is great enough to separate it from the dropout in the absence of
retention lips, even with properly closed QR, it should be apparent that the QR nut will move to its
downward limit in the retention lips. Subsequently, on the next road shock, it will move back to its
upper "normal" position. Repeating this motion, although small, will unscrew the QR, especially if
the closure lever is at the opposite side from the disc brake.

>> I think a simple force balance will also tell you that the force pushing the wheel out of the
>> dropout is dramatically smaller than the force required to move the wheel if the skewer is
>> tightened properly or even anywhere near properly. you = Chicken little.

> I've been wondering about this thread. Seems that there are two camps: one is that discs and QRs
> shouldn't mix, and the other says: properly used the QR should be fine.

> If discs and QRs are a bad combo, why is it that there are only a very, very small minority of
> riders ever having problems with the combo? I know in my experience with QRs and discs, that I
> haven't noticed anything different than if I was running rim brakes.

If the wheel is removed after each ride, to put the bicycle in a car for instance, QR loosening will
not be noticed and will be corrected when the wheel is again installed. If however, if the bicycle
is not disassembled and ridden continuously, incremental loosening will grow. Since MTB's are nearly
all transported to the scene of the crime, they have no problem and believe there is no problem.

> Which led me to thinking about forces on the rear of the bike. Shouldn't the force of pedaling the
> bike do the same thing to the rear wheel as discs do to the front? Since most MTB dropouts are
> semi- to vertical, shouldn't the force of pedaling try to pull the rear wheel out of the dropout
> too? What about discs in the rear?

The rear dropout is always loaded in the same direction, both by chain tension and brake disc. Maybe
if you dig in the RBR-Tech archive you can review all this.

> So, if the rear dropouts are OK, why aren't the fronts?

Apparently you don't understand the force diagram. It depends on the location of the brake caliper.

> I'm still not convinced its not a case of "user error." I detect hints of the Audi "unintended
> acceleration" syndrome here. As y'all know, I'm NOT an engineer, just curious.

There is no parallel to someone stepping on the gas instead of the brake in the bicycle disc
brake problem.

Jobst Brandt [email protected] Palo Alto CA

 

[M-Gineering Imp]

Jose Rizal wrote:

>
> No, it's everything to do with the increased bending moment on the base of the fork brake mounts
> when using larger rotors. This is what Manitou says; see the calculations on my previous post.

V brakes run a pretty big Rotor and didn't fall off much!
--
Marten

 

[Jose Rizal]

M-Gineering import & framebouw:

> Jose Rizal wrote:
>
> >
> > No, it's everything to do with the increased bending moment on the base of the fork brake mounts
> > when using larger rotors. This is what Manitou says; see the calculations on my previous post.
>
> V brakes run a pretty big Rotor and didn't fall off much!
>

The glaringly obvious difference is that V-brake mounts are very close to the pads. Hence, moments
about the V-brake mounts are much smaller.

Imagine that the V-brake mounts are actually located half an inch away from the axle, rather
than half an inch away from the pads, and you might get an insight into the difference between
the two cases.

 

[Amerigo]

"Dave Kahn" <[email protected]> ha scritto nel messaggio
news:[email protected]...
> James Annan <[email protected]> wrote in message
news:<[email protected]>...
>
> (Quoting from http://www.bikebiz.co.uk/web/article.php?id=2899 )
>
> > If the QR problem is ignored, could the bike trade be crippled by similar lawsuits?
>
> Not in the sense that Russ Pinder is, no.
>
> --
> Dave...

I think that the danger is much bigger for the front brake, because the position of the rear brake
is different, so the force generated by the brake is not directed vertically,but it is - almost -
horizontal, how much horizontal it is, it depends from the position of the brake: if it is
vertically above the hub, the force will be horizontal, if it is not, the resultant force will try
to move down the hub

http://www.rothar.com/160303.htm

From: James Subject: Disc brakes and QRs Yes, and I'm in the same position as you. I don't think
it's a big safety issue, but in my case it is a slightly annoying one. It depends on how the
dropouts are angled relative to the disk calliper, and my bike (Ventana El Conquistador tandem) is
particularly bad, with rearward facing dropouts which are at exactly the worst possible angle for
the disc brake. The maker soon realised the error and changed to more conventional vertical ones. If
only major fork manufacturers were so responsive.... That is another reason why I am particularly
aware of the disk brake issue - our rear wheel occasionally slips a bit although now I do the QR up
bloody tight and it's been ok for some time. It's similar to QRs with horizontal dropouts for
singlespeeding - you can just about get away with it, but slippage is a headache (rear wheel slip
while honking up a big hill is not a matter of life and death though). I wouldn't trust my life to
the rear wheel not slipping, but its not a big safety issue in my view since even if the rear wheel
slips sideways a bit or pulls out completely, this is only going to cause a skid and not a headfirst
plummet to the ground (especially on a tandem with the long stable wheelbase). I don't think
different disk calliper types will make a significant difference, it's really just a matter of
geometry. The open fork ends should point directly away from the calliper itself, rather than being
nearly parallel to the tangent.

 

[Amerigo]

"Stergios Papadakis" <[email protected]> ha scritto nel messaggio
news:[email protected]...
> Jon Isaacs wrote:
> >
> > What makes you think they are not thinking about it?
> >
> > One thing to realize is the website in question was the result of a
crash
> > involving a bike with no retaining lips. It took a good deal of effort
to get
> > the site owner to be upfront about this.
> >
> > Jon Isaacs
>
> I think this is a bad argument. The system should NOT rely on the retaining lips to keep the axle
> in position.

you are saying the right thing. The qr will still open and come loose also if there are the
retaining lips!!!! You will probably not loose the wheel, but if you don't recognize soon that the
qr has opened itself, you will loose the wheel anyway

 

[S. Anderson]

"Doug Taylor" <[email protected]> wrote in message >
> I apologize for my stupidity, but until your theories are backed up by statistics which prove
> there is a cost benefit to re-engineering the system , I remain unconcerned and unconvinced. I
> certainly have no plans to change the components on my mountain bike as it is currently set up, or
> stop using it.
>
> Time will tell whether I am a lame brain or you are Chicken Little. --dt

And that's really what it will boil down to. How much to change the system in place versus how much
to pay off the liability claims. I've looked at
Mr. Annan's analysis and although there is some supposition with respect to the magnitudes of
certain forces, it's impossible to ignore the basic design flaw. There will be discussion about
potential corrections, their impact on basic concept of a QR wheel, costs, costs of liablity
etc. and something MAY be changed at some point in the future. There are certainly existing
systems that will adequately keep the wheel in place (just look at any of the through-axle
forks on DH bikes). A knife is sharp and will cut your hand if you slip while cutting
something, but it IS a knife. You trade off convenience for safety. Maybe the incidence of
failure is so insignificant that people and companies will accept the risks involved to be able
to drop their wheel out in 2 seconds instead of 10. As you say, time will tell. But to deny the
problem is faulty.

Cheers,

Scott..

 

[Mark Lee]

"Jose Rizal" <_@_._> wrote in message
news:%[email protected]...
> Andreas Oehler:
>
> > Thu, 15 May 2003 07:42:15 -0700, Terry Morse:
> >
> > >
> > >> However, bear in mind that the larger the rotor diameter, the greater the load on the fork
> > >> brake mounts. This is the reason that
suspension
> > >> fork manufacturers (eg Manitou, Fox) void the warranty on their "standard" (ie non-downhill)
> > >> forks if larger than 6" rotors are used.
> > >
> > >Please explain. For a given braking force, a larger diameter rotor will lessen--not
> > >increase--the force on the brake mounts. There must be some other reason why fork manufacturers
> > >don't want large diameter rotors.
> >
> > The only reasons I know against the use of bigger-than-usual discs is
that
> > areas in the fork leg are heated which do not like the heat or that the bigger disc will touch
> > the fork leg.
> >
> > Andreas (Magura Julie at the tandem on a steal fork)
>
> No, it's everything to do with the increased bending moment on the base of the fork brake mounts
> when using larger rotors. This is what Manitou says; see the calculations on my previous post.

 

[James Annan]

Jose Rizal wrote:
> James Annan:
>
>>
>>I think the simple force balance calculation on my web pag contradicts your belief. I think if you
>>were capable of understanding it, you wouldn't have posted your comment.
>>
>>James
>
>
> I could not see anything on your website which quantitatively compares the force trying to move
> the axle off the dropout to the force required to move a "tight" quick release mechanism. If you
> can show on a graph the force required to move a QR for a given braking rate, vs the "tightness"
> with which the QR is done up, you will be able to have a proper comparison of the forces involved.
> The easiest way to determine this is by empirical means.

Ok, it's true that I did not actually quote directly from the QR test results, but I did provide a
link to the PDF and a few words of summary.

To save you the trouble of downloading it, the QRs did achieve the min ISO pull resistance of 1150N
each side (2300N symmetrical), but the maximum that any of them achieved was about the level of the
force described on my page. Several were actually destroyed in the test due to the torque applied to
close and open the lever.

James

 

[Mike S.]

<[email protected]> wrote in message news:[email protected]...
> Mike Shaw writes:
>
> >> I understand the force diagram, but fail to see what would cause the QR to loosen, unless it
> >> was already extremely loose to begin with and vibrated out. My QRs all have serrated faces so
> >> this is essentially a non-problem.
>
> This was discussed at length the first time this subject came under discussion. Recognizing that
> the disengagement force in the axle is great enough to separate it from the dropout in the absence
> of retention lips, even with properly closed QR, it should be apparent that the QR nut will move
> to its downward limit in the retention lips. Subsequently, on the next road shock, it will move
> back to its upper "normal" position. Repeating this motion, although small, will unscrew the QR,
> especially if the closure lever is at the opposite side from the disc brake.
>
> >> I think a simple force balance will also tell you that the force pushing the wheel out of the
> >> dropout is dramatically smaller than the force required to move the wheel if the skewer is
> >> tightened properly or even anywhere near properly. you = Chicken little.
>
> > I've been wondering about this thread. Seems that there are two camps: one is that discs and QRs
> > shouldn't mix, and the other says: properly used the QR should be fine.
>
> > If discs and QRs are a bad combo, why is it that there are only a very, very small minority of
> > riders ever having problems with the combo? I know in my experience with QRs and discs, that I
> > haven't noticed anything different than if I was running rim brakes.
>
> If the wheel is removed after each ride, to put the bicycle in a car for instance, QR loosening
> will not be noticed and will be corrected when the wheel is again installed. If however, if the
> bicycle is not disassembled and ridden continuously, incremental loosening will grow. Since MTB's
> are nearly all transported to the scene of the crime, they have no problem and believe there is no
> problem.
>
> > Which led me to thinking about forces on the rear of the bike. Shouldn't the force of pedaling
> > the bike do the same thing to the rear wheel as discs do to the front? Since most MTB dropouts
> > are semi- to vertical, shouldn't the force of pedaling try to pull the rear wheel out of the
> > dropout too? What about discs in the rear?
>
> The rear dropout is always loaded in the same direction, both by chain tension and brake disc.
> Maybe if you dig in the RBR-Tech archive you can review all this.
>
> > So, if the rear dropouts are OK, why aren't the fronts?
>
> Apparently you don't understand the force diagram. It depends on the location of the brake
> caliper.
>
> > I'm still not convinced its not a case of "user error." I detect hints of the Audi "unintended
> > acceleration" syndrome here. As y'all know, I'm NOT an engineer, just curious.
>
> There is no parallel to someone stepping on the gas instead of the brake in the bicycle disc brake
> problem.
>
>
No, but there's lots of hysteria about this too.

Mike

 

[x]

RE/
>I think that the danger is much bigger for the front brake, because the position of the rear brake
>is different,

Not to mention that if you lose the rear wheel, you'll still have some semblance of balance/control
- whereas if you lose the front wheel you're going to bite the big one big-time...
-----------------------
PeteCresswell

 

[Tim McNamara]

In article <awUwa.32791$eJ2.6616@fed1read07>, "Mike S." <mikeshaw2@coxDOTnet> wrote:

> No, but there's lots of hysteria about this too.

There's been no hysteria. There's been a presentation by James Annan, backed up by his engineering
analysis, and examined by other mechanical engineers who agree with him. The bike industry has been
made aware of this issue and is considering it- and considering their response. Annan has done the
industry and bike riders a service by bringing up this issue.

There's been disagreement- initially from myself, as a matter of fact, as well as others; but after
careful consideration and having a few questions answered by James and Jobst, I find myself in
agreement. I had been parroting the myth and lore that front wheels don't fall off bikes if the QR
is used properly- which is true *except* with current disk brake designs. QRs were not designed with
disk brakes in mind. There are basically two factors, as far as I can tell.

First, the placement of the brake caliper behind the fork results in a force vector that can eject
the wheel from the dropout. I would imagine this could be readily demonstrated by laying the bike
upside down, loosening the QR, spinning the wheel in the correct direction and applying the brake.
The axle should pop right out of the dropout. The faster the wheel spins, the greater the ejection
force. Note that I haven't tried, this, not having a bike with disk brakes. Somebody give it a try
and let us know how it turns out.

Second, the cyclic application of this ejection force nudges the axle away from the top of the
dropout slot when riding, each time the brake is applied hard. Normal riding knocks the axle back to
the top of the slot. Repeat time and time again. This bends and flexes the axle and QR, eventually
causing the QR nut to begin to loosen. It can eventually loosen enough to allow the axle to be
forced out of the dropout.

As far as I can tell, simply locating the caliper in front of the fork would eliminate this problem.
Or even moving it around the rotor so the resultant force isn't in line with the dropout might help;
surfing around on motorcyle Web sites suggests that those manufacturers have already sussed this
issue- their placement of calipers tends to be at the top of the rotor (e.g., see the Harley Web
site), analogous to the placement of rim brakes, as a matter of fact.

 

[Doug Taylor]

"S. Anderson" <[email protected]> wrote in message
news:[email protected]...

> But to deny the problem is faulty.

For the record, I reiterate that I do not deny the problem. I merely suggest that the "problem"
appears to be one of such low risk that I am not personally concerned and have no plans to trash
my existing equipment or to stop riding my mt. bike. I will, however, take pains to verify that
my QR is soundly fastened before each ride, given the mathematical possibilty (not probability)
of failure.

 

[Stu]

A few more things that may have something to do with the problem, the length of the dropout, its
thickness and the length of axle in the dropout. On my bike the tyre is rubbing against the fork
long before the axle clears the dropout, but if the dropouts were shorter or the axle wasn't as far
through the dropout then the axle could fall out before the tyre hit the fork(so tyre and fork width
have something to do with it to). l don't have discs so l don't know how hard then disc has to push
against the calliper to let the axle clear the dropout. l think you have to have allot of things
working against you before there is a problem, but its a hell of a problem if it happens. So if l
ever get around to discs l want them in front of the fork(l think it would be much nicer to have a
torque triangle with 3 sides, than a torque triangle with 2 sides) at least l think its a torque
triangle????lol

 

[James Annan]

Tim McNamara wrote:

Thanks for the neat summary.

> There's been disagreement- initially from myself, as a matter of fact, as well as others; but
> after careful consideration and having a few questions answered by James and Jobst, I find myself
> in agreement.

I have absolutely no objection to, and indeed welcom, a bit of intelligent scepticism: it was only
due to the (correct) insistence of people like yourself, Jon Isaacs and others that a tight QR
simply cannot pull directly over a lip, that I eventually got round to the conclusion that the QR
loosened first. And without that final detail, I'd have just been one more crank to be dismissed
along with the new revolutionary pedal drive and automatic gear changing.

It is, however, getting a bit tedious seeing the same trivial points over and over again from
bloke-on-a-bike types who imagine that their off-the-cuff opinion is going to unravel several weeks
worth of careful analysis and inspection by a range of experts. That criticism certainly doesn't
apply to all the comments posted here, but it describes most of my mailbox!

James