More on disk brakes and quick release failure
James Annan
in December following a crash I had when my disk brake pulled my front
wheel out of the fork. The aftermath of that crash was discussed here,
and there are pictures of it on
http://www.ne.jp/asahi/julesandjames/home/fork/
The new web pages is on:
http://www.ne.jp/asahi/julesandjames/home/disk_and_quick_release/
and I'd appreciate any comment or criticism.
What prompted me to finish it off after a few months of
procrastination is hearing about yet another serious crash which seems
to have been due to the same basic problem: a disk brake puts a huge
force on the front wheel, in the direction of the open fork ends,
which may be far in excess of anything the QR is designed to handle.
James
Tim McNamara
1. reports of this type of failure would be more widespread (as you
note you couldn't find any)
2. manufacturers- who don't want to get sued and who pay huge
liability insurance premiums- would have changed the design.
Since neither 1 or 2 seem to be true, I can only speculate that either
1. you hadn't properly tightened your QR (e.g., the wheel was cocked
in the dropout when you clamped it down, then straightened out while
riding) or
2. you were using a poorly designed QR with inadequate clamping force
(e.g., most of the aftermarket "cool" QRs with Ti skewers and/or
aluminum jam nuts).
I'll leave others who know more to judge your engineering analysis.
In the interests of disclosure, I should note that I think "disk"
brakes on bikes are a silly idea and offer only one significant
benefit (the braking surface is harder to get muddy) and a number of
disadvantages (weight, increased loads on spokes, etc). Bikes already
have disk brakes if they use a rim brake.
Jon Isaacs
>
I looked over both your pages.
Some comments and questions:
1. It appears to me that there were no "Lawyer Lips" on the fork in question
and that they might have been filed off at some time.
Is this the case? This is an issue that should be addressed up front because
it is a critical part of any analysis.
2. I think your diagram of how a QR might slip past a retention lip is faulty.
The forces required to over come the friction of the QR are quite different
that the forces required to either bend the QR which is supported by the axle
within a few mm of the end or to shear the retention tab.
3. It appears to me that this is a standard cantilever MTB fork that someone
fitted disk brake mounts to. Besides the issue of the tab, it is also
important to remember that disk brakes put considerably more bending moments
and stresses on the crown. This means a cantilver/Vbrake fork might not be
adequate for use with a disk brake.
4. I believe that commerically made bikes equipped with disk brakes do have
substantial retention tabs and beefy forks.
On your web page your wrote:
"So was this just a one-off dangerous design? Perhaps it is worse than most,
but I'm not convinced that 'mainstream' forks are so much better."
I think you hit the nail on the head here. When doing any design, whether it
is one-off or something that will be used for a million bikes, one needs to
understand the forces involved and make sure there are no problems.
It seems to me that the problem here is not with the QR but rather with the
lack of adequate retension lips.
Jon Isaacs
Mike Jacoubowsky
suspension fork that I know of; if there was any retention device on the
dropout, it's not evident in the photos. Current mountain bike forks not
only have a "lawyer lip" (as they're affectionately called) but typically a
recessed area in which the quick release resides. Short of complete failure
(breakage) of the quick release, I do not think it likely that any force
created by a disc brake could liberate the wheel from the fork.
As another poster pointed out, this method of failure appears to be rare
(virtually unknown) in the real world, and that's a world in which I have
customers who often exert enough force on a wheel to destroy a fork, but
leaving the wheel in place.
Could we have a close-up photo of the dropout itself? I'd like to see what
type of retention device it actually had, and compare it to others.
--Mike-- Chain Reaction Bicycles
http://www.ChainReactionBicycles.com
"James Annan" <still_th...@hotmail.com> wrote in message
news:c96ea403.03032...@posting.google.com...
James Annan
> Well, my first thought is that if this was a real problem:
>
> 1. reports of this type of failure would be more widespread (as you
> note you couldn't find any)
Did you actually read the web pages? I mentioned 3 other failures that
happened on standard MTB forks, as well as my rather unusual case. One
of the other cases was a QR shearing, the other 2 were almost
certainly also wheel loss of some sort of another, but the precise
cause has not been diagnosed. These are all in the last few months, by
the way. I've only been interestd in the problem since it happened to
me.
James
Jon Isaacs
>the way. I've only been interestd in the problem since it happened to
>me.
>
>James
>
Were there retention tabs ie Lawyer lips?? Was this a caliper fork modified to
take a disk brake??
jon isaacs
James Annan
> 1. It appears to me that there were no "Lawyer Lips" on the fork in question
> and that they might have been filed off at some time.
>
> Is this the case? This is an issue that should be addressed up front because
> it is a critical part of any analysis.
In my case, the builder chose to use fork ends without the lips. This
obviously is directly relevant to my crash, but not to the other 3 I
know of where lips were present on standard MTB suspension forks. In
one of these other cases, the QR sheared, but in the other 2, this
appears to have not happened (being on the other side of the world
makes it hard to examine the remains in person, and I'm relying on
reports from witnesses who don't necessarily know what they should be
looking for).
I don't consider that the sheared QR is a significantly different
problem from the others - it was still demonstrably being put under a
load substantially in excess of that which it is designed to handle,
due to the positioning of disk and fork end slot.
> 2. I think your diagram of how a QR might slip past a retention lip is faulty.
I wouldn't be at all surprised to find out that it is. It was only a
random idea. _Something_ happened to allow the wheel to separate from
the fork, apparently without any breakage. Obviously I'd like to have
a look at the parts myself, but that't not going to be possible.
The particularly horrifying thing about the most recent crash, apart
from the resulting injuries (which are very bad), is the fact that the
victim is not just some random cyclist who has a friend with net
access, but is one of the small handful of people with whom I had
recently discussed this very 'problem'. He was an experienced cyclist,
he understood the theoretical problem of the wheel extraction force,
and he knew how to use a QR. If wheel loss can be ruled out as a cause
in his case, that would be great. Otherwise, it seems to me that a
method that fails on him is likely to be insufficiently safe for
everyone.
> 4. I believe that commerically made bikes equipped with disk brakes do have
> substantial retention tabs.
This is definitely false, as only a few days ago I saw a set of rigid
no-name forks in a Tokyo bike shop with very small retention tabs
(with a rounded paint-filled corner) and somewhat rearward-facing
slots. I will try to go back and take some close-up photos. Of course,
even with the lip, there's still the possible problem of the QR
breaking, and the resulting crash will be just as bad.
What do you think of the Surly forks shown on my 'fork' web page? Also
on
http://www.surlybikes.com/frames/images/karatemonkeyLG.jpg
FWIW, I've heard a report of front wheel slippage when using disk
brakes on this frame, but not yet a full wheel extraction. Clearing
off the paint and using 'a high quality QR' probably helps, but how is
a consumer supposed to know that, and what measure is available of
'high quality' beyond the ISO standard?
I realise that I'm talking about something that is quite rare, and the
hypothesis is somewhat improbable. This seems entirely consistent to
me. These relatively rare crashes must of course be explained by
something that only happens occasionally, perhaps under a combination
of circumstances. Otherwise the crashes wouldn't be rare! Therefore, a
hypothesis cannot be dismissed simply on the grounds that it seems
'unlikely'.
James
James Annan
Jon Isaacs wrote:
>
> Still wondering what was the history of your fork..
>
> Were there retention tabs ie Lawyer lips?? Was this a caliper fork modified to
> take a disk brake??
Um, I think I just answered, but hasn't propagated yet. It was
specifically built for the disk brake, and had no lips from the outset.
Not my deliberate choice, I just trusted the builder.
I realise that my case is a specific and unusual combination, but the 3
other recent cases known to me, that appear to be also due to front
wheel loss, were all using standard MTB forks.
James
Jon Isaacs
I can't see it well enough to say. But I am quite sure it has retention tabs.
>FWIW, I've heard a report of front wheel slippage when using disk
>brakes on this frame, but not yet a full wheel extraction. Clearing
>off the paint and using 'a high quality QR' probably helps, but how is
>a consumer supposed to know that, and what measure is available of
>'high quality' beyond the ISO standard?
>
>I realise that I'm talking about something that is quite rare, and the
>hypothesis is somewhat improbable. This seems entirely consistent to
>me. These relatively rare crashes must of course be explained by
>something that only happens occasionally, perhaps under a combination
>of circumstances. Otherwise the crashes wouldn't be rare! Therefore, a
>hypothesis cannot be dismissed simply on the grounds that it seems
>'unlikely'.
>
>James
I think it is really on possible for me to comment on your crash because yours
is the only one which is well documented. The lack of Retention tabs on your
fork is clearly the important issue. I do hope you make this clear on your
page so that someone does not have to guess this fact.
I personally wonder why they just don't mount the brake on the front of the
fork so that applying the brake forces the wheel into the dropout rather than
out of the dropout. Rear mounting might reduce the forces on the fork a bit.
Jon Isaacs
James Annan
Jon Isaacs wrote:
>>What do you think of the Surly forks shown on my 'fork' web page?
>
>
> I can't see it well enough to say. But I am quite sure it has retention tabs.
Indeed it does, of some sort. However I'm sure there are plenty who
scoff at them, at least before they ride the bike.
> I think it is really on possible for me to comment on your crash because yours
> is the only one which is well documented.
What about the case of the snapped QR? I suppose I could say that the
rider in that case was definitely using a standard cannodale fork with
lips, the QR was the Coda-branded one that came with the bike. I'm not
sure what more documentation would be helpful. Possibly part of the QR
could be examined closely (I think one end was lost), but it broke
anyway - what more is there to say?
> The lack of Retention tabs on your
> fork is clearly the important issue.
If my fork had had a lip and the QR had snapped instead of pulling out,
the crash would hardly have been any better. A changed dropout angle
would make the lips irrelevant though, and also eliminate the massive
loading on the skewer.
> I do hope you make this clear on your
> page so that someone does not have to guess this fact.
Sorry, I'll change that. The page was originally put up following
conversations with several people who already knew about the lack of
lips - I hadn't realised it was not obvious to new readers.
James
David Kunz
>
>
> Jon Isaacs wrote:
>
>>> What do you think of the Surly forks shown on my 'fork' web page?
>>
>>
>>
>> I can't see it well enough to say. But I am quite sure it has
>> retention tabs.
>
>
> Indeed it does, of some sort. However I'm sure there are plenty who
> scoff at them, at least before they ride the bike.
>
>> I think it is really on possible for me to comment on your crash
>> because yours
>> is the only one which is well documented.
>
>
> What about the case of the snapped QR? I suppose I could say that the
> rider in that case was definitely using a standard cannodale fork with
> lips, the QR was the Coda-branded one that came with the bike. I'm not
> sure what more documentation would be helpful. Possibly part of the QR
> could be examined closely (I think one end was lost), but it broke
> anyway - what more is there to say?
Didn't Cannondale recall this QR?
Also, some people put titanium QRs in front -- and this is what happens
(why not to use titanium there).
James Annan
>
> Didn't Cannondale recall this QR?
I haven't heard that, and presumably neither did the victim. Do you have
any more details of which skewers were recalled, and when?
> Also, some people put titanium QRs in front -- and this is what happens
> (why not to use titanium there).
Well, I agree that a Ti skewer seems like a bad idea, but this
Cannondale one certainly wasn't that. I don't believe it is reasonable
to put the onus on the customer to work out which skewers are fit for
purpose, and which are not - AIUI there is a single ISO standard for
pull retention, and it is provably inadequate for disk brake-equipped bikes.
I do understand the scepticism on whether a QR can really get pulled out
of a fork with a solid lawyer lip. It's hard to understand how that can
happen without some sort of mechanical failure. Nevertheless, wheel/fork
separation of some sort continues to occur occasionally, even when
installed by experienced cyclists who are aware of the danger. And even
without a direct pull-out, skewers do snap under the force which is far
in excess of that which they are designed to handle. With the spread of
disk brakes to road bikes, there's a whole new set of frame builders
(and cyclists) who have been brought up to believe that retention lips
are entirely unnecessary, and who may not even realise there is the
potential for a problem.
(NB the person who built my forks was actually an experienced MTB
builder, although it seems like he had limited experience of disk brakes
and tandems - I didn't actually communicate with him at all, he was
subcontracted by my frame builder.)
James
Jon Isaacs
>If my fork had had a lip and the QR had snapped instead of pulling out,
>the crash would hardly have been any better
The point is that the QR slipped. You have no evidence that the QR would have
snapped had it not slipped.
Had you had retention tabs I suggest that you would not have crashed.
>
>What about the case of the snapped QR? I suppose I could say that the
>rider in that case was definitely using a standard cannodale fork with
>lips, the QR was the Coda-branded one that came with the bike. I'm not
>sure what more documentation
>would be helpful. Possibly part of the QR
>could be examined closely (I think one end was lost), but it broke
>anyway - what more is there to say?
There is a lot more to say and a lot more to look at. In order to comment on a
failure it is necessary to have more than just the word of those involved.
Your webpage is a perfect example of this. By looking at the photos it was
clear that you did not have Retention tabs and that this was the major issue.
But from what you wrote, this was not clear, one had to infer from the photo
plus the fact that you tried to claim that the QR could slipout of a dropout
with retention tabs.
It is possible for example that the QR in question had been previously abused
or that in fact someone had filed off the tabs on the fork.
Your website allowed me to understand what actually happened and comment on
that rather than take your word that there was a problem with the QR.
It seems to me that whoever built that fork for you did not understand the
issues involved and since it already had cantilever pads on it, it seems likely
that this was modified fork rather than a new fork specifically built for this
purpose.
I am still curious about the exact history of your fork.
jon isaacs
James Annan
>>If my fork had had a lip and the QR had snapped instead of pulling out,
>>the crash would hardly have been any better
>
>
> The point is that the QR slipped. You have no evidence that the QR would have
> snapped had it not slipped.
No, but I do have evidence that no QR is designed to handle this stress.
At least, a Shimano engineer has emailed me to say that their skewers
are designed to satisfy the ISO standard, and by testing and popular
acclaim, they are as good as any and better than most, probably
including the one I was using. That doesn't mean that my QR would
necessarily have broken in that incident, of course. It might have
waited until I was going faster over rougher ground.
The person with a snapped skewer said that it snapped, while he was
descending a steep slope. One end was lost in the crash. It was all as
supplied with the bike. Not clear what he might have been doing to
'abuse' the QR, it was almost certainly above the recommended torque
because - ta da! - it has to be to keep the wheel in place! Ah, off the
hook again, that's his fault for doing it up too tight, and if it had
slipped, it would have been his fault for doing it too loose. When one
starts off from the presumption that the QR is always innocent, it's
easy to 'diagnose' any failure...however that does not actually _cure_
the problem, does it?
> But from what you wrote, this was not clear, one had to infer from the photo
> plus the fact that you tried to claim that the QR could slipout of a dropout
> with retention tabs.
As I said, the origin of the web-page (written several months) ago meant
that this was known to the original audience. It was clearly visible on
the pictures too. I certainly wasn't trying to deceive anyone, it would
have been a rather feeble effort to show a close-up of the fork end if I
was.
> It seems to me that whoever built that fork for you did not understand the
> issues involved and since it already had cantilever pads on it, it seems likely
> that this was modified fork rather than a new fork specifically built for this
> purpose.
No, it was specifically built for the disk brake, and this bicycle.
> I am still curious about the exact history of your fork.
It had no 'history'. It had been ridden for about an hour previous to
failure. It was supplied with the frame, complete with disk mounts, a
fresh coat of paint and no retention tabs. Oh, much the same goes for
the skewer, except it was just a standard widely availabe one and not in
any way custom. The skewer I now realise is not the best available.
Frankly, I'm glad it wasn't slightly better, as it would in that case
have failed on a faster and steeper hill a couple of miles further along
my route.
James
Jon Isaacs
>> The point is that the QR slipped. You have no evidence that the QR would
>have
>> snapped had it not slipped.
>
>No, but I do have evidence that no QR is designed to handle this stress.
>
>At least, a Shimano engineer has emailed me to say that their skewers
>are designed to satisfy the ISO standard, and by testing and popular
>acclaim, they are as good as any and better than most, probably
>including the one I was using. That doesn't mean that my QR would
>necessarily have broken in that incident, of course. It might have
>waited until I was going faster over rougher ground.
Again, your QR only slipped, it did not fail. It is not uncommon for QRs to
slip. One only has to go to the velodrome where people are doing pursuits or
kilos to see a rash of QRs slipping.
But failure of the QR is a whole different issue. You show a calculation in
your page that the force on the dropout could be as high as 1875 newtons which
is about 420 lbs force.
A rough calculation shows the force required to shear a standard 5 mm diameter
QR would be much higher than that, something on the order of 13000 newtons, a
factor of 7. And of course the shear load should not be taken by the QR but
rather by axle.
So, in my view, it is unlikely that in your case the QR would have sheared or
other wise failed and furthermore it is important for you to clearly understand
and ackknowledge this.
>The person with a snapped skewer said that it snapped, while he was
>descending a steep slope. One end was lost in the crash. It was all as
>supplied with the bike. Not clear what he might have been doing to
>'abuse' the QR, it was almost certainly above the recommended torque
>because - ta da! - it has to be to keep the wheel in place! Ah, off the
>hook again, that's his fault for doing it up too tight, and if it had
>slipped, it would have been his fault for doing it too loose.
Again, without knowing what actually happened, it is impossible to say whether
it was a QR failure or something else.
Consider your own situation. The QR was not a fault, it was being used
improperly. It was going to slip sometime or another, it was a stupid design
and if you were hurt, the fellow who built that fork and who built that bike
ought to be treating you very nicely. They should have known better than to
build a bike with a fork without retention tabs.
In the other case(s) you mention it does seem that the QR failed. But the
cause of this failure has not been established.
Understanding a failure is not a simple task and cannot be done on heresay and
guesses.
It may be that there indeed is some problem here, but from the evidence
supplied, it is not possible to say.
When one
>starts off from the presumption that the QR is always innocent, it's
>easy to 'diagnose' any failure...however that does not actually _cure_
>the problem, does it?
Sorry but I have no presumptions, I want evidence. The evidence you provided
me was sufficient for me to see that the problem was not with the QR.
You are the one pointing the finger.
I think you are the one who presumes that the disk brake-QR issue is the cause
when indeed there may be some other cause.
It may or may not be but without evidence, one cannot understand what actually
happened.
You asked for an evalution of your web page and I gave you one.
If you really want to understand this, you need to be unbiased and open minded.
What you have clearly proved is that a bike equipped with a front disk brake
better have retention tabs.
I think this is a significant point that should not be overlooked because of
the number of people that routinely file these off so that it is easier to take
the wheel on and off.
If you want to establish that QRs can fail because of the loads applied by a
disk brake mounted on the rear side of the fork, then IMHO, you need to attack
this with an open mind and work on it.
Your apparent belief that the QR would have failed had the QR not slipped
indicates to me that you do not have an open mind.
Think about it. These are not just words I am writing or flames.
This is scientific/technical issue that you have raised and it is an
interesting one.
But to resolve this issue you will have to resolve it in a scientific/technical
manner and that means forgetting about your unfortunate experience and actually
understanding the issue objectively.
Jon Isaacs
Eric Murray
This morning I tried a test. I've got a bike with Shimano XT disc
brakes and a Marzzochi "bomber" front fork, lawyer lips filed off.
This fork has a dropout with the opening parallel to the fork tube.
My test consisted of loosening the QR until the wheel was barely
held in. Then I held the front end off the ground, spun the tire
and clamped on the brake. The wheel didn't budge.
I tried running alongside the bike (I wasn't going to ride
a bike with a loose QR) and braking. The wheel didn't budge
then either.
I'm not going to worry about my wheel coming out on braking.
Eric
Dallas
"Eric Murray" <sp...@lne.com> wrote in message
news:b5iu0c$nn9$1...@slack.lne.com...
My Rock Shox Psylo sounds like a similar set-up, but with an Avid brake. The
fork in the photos seems to have dropouts that angle rearward with respect
to the fork blades.
Incidentally I also noticed that the postitioning of the caliper on the
Psylo is much higher on the fork blade as well, torquing the wheel axle
against the side of the dropout instead of the opening. I wonder though if
using a larger than stock disc might increase the risk of a wheel pulling
out (the mounts requiring the caliper to be positioned further out from the
fork rather than higher up the blade)?
Mike Dallas
James Annan
> Again, your QR only slipped, it did not fail.
Its job was to keep the wheel in place, and it failed to do so. Of
course it was not designed to do that job.
> But failure of the QR is a whole different issue. You show a calculation in
> your page that the force on the dropout could be as high as 1875 newtons which
> is about 420 lbs force.
No, I don't think that 'as high as' is fair. It's a ballpark figure
that could surely be momentarily exceeded, for example by a juddering
brake. Wouldn't a stick-slip scenario imply that the peak force could
be roughly double the average force?
(Actually, we were riding a tandem, so the force in our case might
well have been much higher again. However I think the calculations I
have done for a single bike are still worrying enough.)
> A rough calculation shows the force required to shear a standard 5 mm diameter
> QR would be much higher than that, something on the order of 13000 newtons, a
> factor of 7.
The QR is already in tension to the tune of perhaps half that level
(I've seen a value of 5600N quoted). The force due to the disk might
be 3600N (I'm assuming that my 'doubling' above is plausible - please
correct if I'm out of line here). Of course, the disk force isn't a
direct tension, it's applied perpendicular to the skewer and will bend
it putting the stress disproportionately on the edge. Consider a beam,
maybe 3cm long (distance from disc to the QR nut, at which end the
beam is supported) and only 5mm diameter with a perpendicular force
applied to the free end. Or maybe, the left hand end of the hub is
pushed down in the dropout, and then the stress is focussed on the
lower edge of the skewer, at the point where it exits the hub. Now
what is the safety margin for the QR, which is already notched with
threads?
> And of course the shear load should not be taken by the QR but
> rather by axle.
By 'should', do you mean that you think it actually is, or that you
think the fork end should be designed to ensure this? If the latter,
then I agree, but it's not the case with current designs. The axle is
not supported by the fork end, the force is (near enough) aligned with
the slot.
> So, in my view, it is unlikely that in your case the QR would have sheared or
> other wise failed and furthermore it is important for you to clearly understand
> and ackknowledge this.
Oh, I'm happy to agree that it's 'unlikely', if QRs can routinely
withstand the forces described above. I'm not sure what value of
'unlikely' you are using, and I'm not convinced that it will be
adequate for such a critical component where a failure is so
dangerous.
> Again, without knowing what actually happened, it is impossible to say whether
> it was a QR failure or something else.
In that case, can you think of any way that someone who suffers a QR
failure (break) could convince you of that fact? It really seems like
you are the one with a closed mind here. I don't see why it is so hard
to believe, and it is certainly the most plausible explanation I can
think of for this crash. It's not as if skewer breakage is
particularly rare even, but it tends to happen when a wheel is being
installed rather than ridden.
> I think you are the one who presumes that the disk brake-QR issue is the cause
> when indeed there may be some other cause.
Well all the 4 recent wheel separation failures I know of are on disk
brake + QR bikes, all of them occurred on steep descents, and in my
case it certainly seems very clear that it wouldn't have happened
without the disk brake. Even if lips were present, the skewer is still
not designed to withstand the force although perhaps it would
'usually' be ok in practice.
> What you have clearly proved is that a bike equipped with a front disk brake
> better have retention tabs.
>
> I think this is a significant point that should not be overlooked because of
> the number of people that routinely file these off so that it is easier to take
> the wheel on and off.
Yes, I think this is certainly a conclusion that we can all agree on.
Furthermore, I hope you will agree that the retention tabs had better
be designed to handle the large forces applied by a disk brake, rather
than simply be high enough to stop a loose QR from falling out due to
gravity. Of course, the same goes for the QR skewer.
[It seems to me that changing the dropout angle would be a more
sensible alternative, since it removes the problem at source.]
> If you want to establish that QRs can fail because of the loads applied by a
> disk brake mounted on the rear side of the fork, then IMHO, you need to attack
> this with an open mind and work on it.
>
> Your apparent belief that the QR would have failed had the QR not slipped
> indicates to me that you do not have an open mind.
It's not a 'belief that the QR would have failed'. I agree that quite
possibly it would not have done so. I'm just pointing out that lips or
not, it was not designed for this application, and that is as true on
any MTB with standard forks as it is on my badly designed custom fork.
> Think about it. These are not just words I am writing or flames.
>
> This is scientific/technical issue that you have raised and it is an
> interesting one.
>
> But to resolve this issue you will have to resolve it in a scientific/technical
> manner and that means forgetting about your unfortunate experience and actually
> understanding the issue objectively.
I am trying to understand it objectively. I agree that is it difficult
to see how a properly adjusted QR can pull over a substantial
retention lip. It seems to imply the QR stretching by about 1mm = 1%,
even for a 1mm lip. That implies a tension of 40kN (I think - 2kN per
sq mm, for 20sq mm x-sectional area) which is 3 times the force you
quote to break it. Nevertheless, wheels come out, even apparently for
experienced cyclists who know how to use a QR. I do think it's
important to have a good look at the aftermath of these crashes, but I
would rather stop them from happening in the first place. It is
certainly clear that quick release skewers are not designed for this
application, even if they 'usually' cope.
James
Mark Hickey
>I am trying to understand it objectively. I agree that is it difficult
>to see how a properly adjusted QR can pull over a substantial
>retention lip. It seems to imply the QR stretching by about 1mm = 1%,
>even for a 1mm lip. That implies a tension of 40kN (I think - 2kN per
>sq mm, for 20sq mm x-sectional area) which is 3 times the force you
>quote to break it. Nevertheless, wheels come out, even apparently for
>experienced cyclists who know how to use a QR. I do think it's
>important to have a good look at the aftermath of these crashes, but I
>would rather stop them from happening in the first place. It is
>certainly clear that quick release skewers are not designed for this
>application, even if they 'usually' cope.
Why would skewer tension increase under braking? Even if the skewer
slips in the droput, the preload should remain nearly the same unless
the dropout width isn't uniform.
If the axle slips enough for the QR nut to contact the lawyer lips, it
seems to me that an effective impasse has been reached
And personally I don't put a lot of faith in the fact that an
"experienced cyclist who knows how to use a QR" can't simply forget to
tighten it up. The one front wheel departure I helped clean up
(called in the ambulance) was clearly due to the rider not tightening
the skewer properly.
Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame
S. Anderson
news:20030322005423...@mb-fn.aol.com...
<<snip..>>
> I personally wonder why they just don't mount the brake on the front of
the
> fork so that applying the brake forces the wheel into the dropout rather
than
> out of the dropout. Rear mounting might reduce the forces on the fork a
bit.
>
> Jon Isaacs
>
Usually calipers are mounted behind the fork blade to put their mass as
close as possible to the steering axis so that handling isn't affected to
any great degree. It's a small point but that's why it was done on
motorcycles and probably copied on bicycles.
Cheers,
Scott..
Jon Isaacs
>This morning I tried a test. I've got a bike with Shimano XT disc
>brakes and a Marzzochi "bomber" front fork, lawyer lips filed off.
>This fork has a dropout with the opening parallel to the fork tube.
>
>My test consisted of loosening the QR until the wheel was barely
>held in. Then I held the front end off the ground, spun the tire
>and clamped on the brake. The wheel didn't budge.
The moment on the QR contact that causes it to come free is caused by the
position of the brake and the contact patch on ground. The inertial loading
you applied is a different situation and for that reason is not a valid.
>
>I tried running alongside the bike (I wasn't going to ride
>a bike with a loose QR) and braking. The wheel didn't budge
>then either.
>
>I'm not going to worry about my wheel coming out on braking.
>
>Eric
You ought to be concerned because James Annan has raised a valid issue here and
if your disk brake are mounted behind the fork, then there is indeed a force
trying to slip the front wheel out of the drop out.
Those "Lawyer Lips" are there for a reason and it is not just to keep the
companies out of law suits.
I suggest reading his pages because they should raise some concerns.
Jon Isaacs
jobst....@stanfordalumni.org
> What prompted me to finish it off after a few months of
> procrastination is hearing about yet another serious crash which
> seems to have been due to the same basic problem: a disk brake puts
> a huge force on the front wheel, in the direction of the open fork
> ends, which may be far in excess of anything the QR is designed to
> handle.
To make this more direct, lets assume the bicycle is capable of
braking with 1g acceleration. That means, for a low rider position,
the entire weight of rider and bicycle is bearing on the front
wheel... and that the rotational torque of the front wheel is its
radius times bicycle and rider weight. The radius at which the brake
pads act is about a 6 inch while wheel periphery lies at about 13
inches.
That means the upward force on the brake pads is slightly greater than
twice (13/6) the horizontal force (which is equal to the vertical
force at 1g) on the front wheel. Therefore, there is a net
disengagement force of half the rider-and-bicycle weight. This is not
a safe condition when repeated often, especially on rough terrain
where this force is hammered onto the wheel retention.
This example is accurate because the ratio of forces is equally true
for lesser braking forces. The example of 1g is to simplify
visualization by avoiding arithmetic.
Jobst Brandt
jobst....@stanfordalumni.org
Palo Alto CA
Jon Isaacs
>course it was not designed to do that job.
Likewise it failed to put a million dollars in my bank account, but then it was
not designed to do that either. <g>
The failure in your case was the failure on the part of the fork builder and
frame build to understand that disk brakes are different than cantilever style
brakes.
>No, I don't think that 'as high as' is fair. It's a ballpark figure
>that could surely be momentarily exceeded, for example by a juddering
>brake. Wouldn't a stick-slip scenario imply that the peak force could
>be roughly double the average force?
I think the peak force is determined by the tire slip on the ground or the
overturning moment, (0.6 gs is what I remember you using.) Doubling the force
might occur were this a true dynamic situation but the force is really limited
the conditions. In this case if the force increases either the tie slips or
the rear end lifts off.
>Consider a beam,
>maybe 3cm long (distance from disc to the QR nut, at which end the
>beam is supported) and only 5mm diameter with a perpendicular force
>applied to the free end.
The maximum bending stress on the QR would occur at the point it entered the
axle assuming a reasonable fit between the axle and the QR. Assuming the QR
extends a quarter inch past the (and this is a lot in my view) the maximum
"nominal" bending stress on the QR axle is about 6000 psi which of course is
nothing.
Something else to consider: Assuming a 100ksi yield (probably closer to
200ksi) that the nominal tensile failure of the QR is something like 26,000
newtons.
>By 'should', do you mean that you think it actually is, or that you
>think the fork end should be designed to ensure this? If the latter,
>then I agree, but it's not the case with current designs. The axle is
>not supported by the fork end, the force is (near enough) aligned with
>the slot.
It is true the QR does have shear loading against the "Lawyer lip."
>Oh, I'm happy to agree that it's 'unlikely', if QRs can routinely
>withstand the forces described above. I'm not sure what value of
>'unlikely' you are using, and I'm not convinced that it will be
>adequate for such a critical component where a failure is so
>dangerous.
I think that they routinely can. Unlike the case with your fork, manufacturers
actually engineer their forks and someone with specific numbers looks at these
issues. And given the number of disk brakes out there, it is pretty clear to
me that were there a real issue, you would be seeing significant numbers of
crashes.
In many cases, back yard engineering results in a reasonable product. Frame
builders are a good example of this. The engineering in a bike frame is
typically done by the manufacturer of the tubing and the frame builder only has
to choose the right tubing.
>> Again, without knowing what actually happened, it is impossible to say
>whether
>> it was a QR failure or something else.
>
>In that case, can you think of any way that someone who suffers a QR failure
(break) could convince you of that fact? It really seems like >you are the one
with a closed mind here. I don't see why it is so hard to believe, and it is
certainly the most plausible explanation I can
>ink of for this crash.
My point is this: There is a major difference between a plausible explanation
and understanding what actually happened. Just because something seems
plausible does not mean that is actually what happened.
It may be that indeed the QR did fail because of loads applied by the disk
brake. But there may well be other explanations and without proper
investigation and analysis, one cannot conclude what actually happened. There
are a number of possibilities, two that immediately come to mind are an
improperly installed wheel and a QR that failed at some other time. If one
were riding along with a broken QR it is clear that applying the brake may
cause the front wheel to pop free.
>> I think you are the one who presumes that the disk brake-QR issue is the
>cause
>> when indeed there may be some other cause.
>
>Well all the 4 recent wheel separation failures I know of are on disk brake +
QR bikes, all of them occurred on steep descents, and in my case it certainly
seems very clear that it wouldn't have happened without the disk brake.
As I continue to point out, your case has to do with rear mounted disk brakes
without retention tabs, the disk brakes are part of the issue but I have
acknowledged that and we are now looking at disk brakes with retention tabs.
>Even if lips were present, the skewer is still
>not designed to withstand the force although perhaps it would
>'usually' be ok in practice.
I don't think this has been established at all. The only analysis I have seen
is what you and I have done, back of the envelope sort of stuff that can show
if there is a problem or not. So far, I have seen nothing that indicates that
QR is over taxed for this.
>> What you have clearly proved is that a bike equipped with a front disk
>brake
>> better have retention tabs.
>>
>> I think this is a significant point that should not be overlooked because
>of
>> the number of people that routinely file these off so that it is easier to
>take
>> the wheel on and off.
>Yes, I think this is certainly a conclusion that we can all agree on.
>Furthermore, I hope you will agree that the retention tabs had better
>be designed to handle the large forces applied by a disk brake, rather
>than simply be high enough to stop a loose QR from falling out due to
>gravity. Of course, the same goes for the QR skewer.
I agree they should be designed for this, but look at the shear strength of
steel and the forces involved, it doesn't take much.
>[It seems to me that changing the dropout angle would be a more
>sensible alternative, since it removes the problem at source.]
Mounting the brake on the front would achieve the same result, though at some
expense.
>It's not a 'belief that the QR would have failed'. I agree that quite
>possibly it would not have done so.
Good.
>> But to resolve this issue you will have to resolve it in a
>scientific/technical
>> manner and that means forgetting about your unfortunate experience and
>actually
>> understanding the issue objectively.
>
>I am trying to understand it objectively. I agree that is it difficult
>to see how a properly adjusted QR can pull over a substantial
>retention lip. It seems to imply the QR stretching by about 1mm = 1%,
>even for a 1mm lip. That implies a tension of 40kN (I think - 2kN per
>sq mm, for 20sq mm x-sectional area) which is 3 times the force you quote to
break it.
Actually few steels will handle 1% elastic strain but almost all of them will
handle 1% strain easily. But my calculation was simply based on the diameter
of a measured QR, and the assumption fo 100ksi shear strength.
But 40 Knewtons is a big force, about 9000 lbs, not likely in a QR.
Nevertheless, wheels come out, even apparently for experienced cyclists who
know how to use a QR. I do think it's
>important to have a good look at the aftermath of these crashes, but I
>would rather stop them from happening in the first place. It is
>certainly clear that quick release skewers are not designed for this
>application, even if they 'usually' cope.
>
>James
>
>Nevertheless, wheels come out, even apparently for
>experienced cyclists who know how to use a QR.
Yes this is true. I have never had this happen but I know another "experienced
cyclist" who lost a front wheel. QR didn't fail, no disk brake to blame it on,
for some reason, when he "lifted" the front wheel over a curb, it fell out and
he ended up with some broken bones.
>I do think it's
>important to have a good look at the aftermath of these crashes, but I would
rather stop them from happening in the first place. It is
>certainly clear that quick release skewers are not designed for this
application, even if they 'usually' cope.
I would rather stop them from happening as well. I think realizing the
importance of the "Lawyer Lips" is a important step in that direction. It may
also be that the QRs are not sufficient to handle the shear and bending loads
imposed by the disk brakes but my rough calculations show that they ought to be
plenty strong.
I am sure that the bike companies have had engineers look at this problem on a
more sophisticated level than you or I. I would not be concerned about buying
a disk brake equippend bike but I certainly would never file off the Lawyer
Lipps and after reading your pages I would make sure that my front wheel was
properly installed.
With cantilever brakes, the braking force keeps the wheel in place, with the
disk brake, this is not the case.
One thing that ought to be addressed here is the issue of objectivity.
In your case you have a hypothesis. It is important to remain objective and
refrain from jumping to conclusions. You have found three other people who
have lost a front wheel when using a disk brake. But it is important to
realize that you went looking for people who believe that the brake was
responsible for their crash and that just because you found people who believe
that this was the case, it is important for you to establish that indeed this
was the case.
Front wheels can be lost for a variety of reasons, QRs can pop for a variety of
reasons, the connection may seem plausible but it does need to be established.
Consider that you were trying to show the connection for a Masters Thesis in
engineering or in front of a group of fork designers.
These are the expectations I have of you and your webpage and the standards I
am of you.
So realize where you are out on a limb and taking a leap and realize I expect
those points to be on solid ground.
So far I see that the relying on the friction of the QR for rear mounted disk
brakes is a poor idea. This is important and should not be neglected, from my
point of view, your contention that the QR is cannot handle these loads when
used with (properly designed) retention tabs has yet to be established.
I also think that this might be better carried on as private email so if you
agree, just email me at my aol address.
jon isaacs
Jose Rizal
> To make this more direct, lets assume the bicycle is capable of
> braking with 1g acceleration. That means, for a low rider position,
> the entire weight of rider and bicycle is bearing on the front
> wheel... and that the rotational torque of the front wheel is its
> radius times bicycle and rider weight. The radius at which the brake
> pads act is about a 6 inch while wheel periphery lies at about 13
> inches.
>
> That means the upward force on the brake pads is slightly greater than
> twice (13/6) the horizontal force (which is equal to the vertical
> force at 1g) on the front wheel. Therefore, there is a net
> disengagement force of half the rider-and-bicycle weight. This is not
> a safe condition when repeated often, especially on rough terrain
> where this force is hammered onto the wheel retention.
>
If you look at the geometry of the dropouts in front suspension forks
(which arguably most disc brakes are used with), and the position of the
disc brake caliper (regardless of model and rotor size, as the caliper
locations are all behind and above the QR), the QR will tend to be
pushed towards the rear half of the dropout rather than along the slot
where the QR slides in and out at removal. In other words, the
retention of the QR axle in place when using disc brakes is not at all
dependent on the tightness of the QR, but rather on the strength of the
dropout. The safety of this arrangement is not as questionable as you
imply.
A Muzi
As I understand this problem, the fork that began thuis discussion was
erroneously built with the opening in the tip exactly lined up with the
direction the axle wants to go when the brake is applied. No quick release
could hold against those forces, being amplified and directed out the slot!
If your fork tips are pointing along the fork blade's centerline, the axle
wants to press against the side of the tip when the brake is applied, not
out the slot. There's no particular reason for a Q/R to fail in that setup
because it isn't the only thing holding the axle in the fork. I wouldn't
worry about your fork at all. Tens of thousands of forks are running just
fine in that format.
--
Andrew Muzi
http://www.yellowjersey.org
Open every day since 1 April 1971
jobst....@stanfordalumni.org
>> To make this more direct, lets assume the bicycle is capable of
>> braking with 1g acceleration. That means, for a low rider
>> position, the entire weight of rider and bicycle is bearing on the
>> front wheel... and that the rotational torque of the front wheel is
>> its radius times bicycle and rider weight. The radius at which the
>> brake pads act is about a 6 inch while wheel periphery lies at
>> about 13 inches.
>> That means the upward force on the brake pads is slightly greater
>> than twice (13/6) the horizontal force (which is equal to the
>> vertical force at 1g) on the front wheel. Therefore, there is a
>> net disengagement force of half the rider-and-bicycle weight. This
>> is not a safe condition when repeated often, especially on rough
>> terrain where this force is hammered onto the wheel retention.
> If you look at the geometry of the dropouts in front suspension
> forks (which arguably most disc brakes are used with), and the
> position of the disc brake caliper (regardless of model and rotor
> size, as the caliper locations are all behind and above the QR), the
> QR will tend to be pushed towards the rear half of the dropout
> rather than along the slot where the QR slides in and out at
> removal.
Only the vertical component of these forces that is important in wheel
retention. Whether there is a rearward resultant is immaterial, the
axle not being retained by its friction against the vertical dropout
slot. Whether the brake caliper is above or at the dropout does not
change the vertical component of its action which it perpendicular to
the radial line to the axle from the caliper centerline... essentially
horizontal.
> In other words, the retention of the QR axle in place when using
> disc brakes is not at all dependent on the tightness of the QR, but
> rather on the strength of the dropout. The safety of this
> arrangement is not as questionable as you imply.
It is exactly as I and others have stated.
jobst....@stanfordalumni.org
> If your fork tips are pointing along the fork blade's centerline,
> the axle wants to press against the side of the tip when the brake
> is applied, not out the slot. There's no particular reason for a Q/R
> to fail in that setup because it isn't the only thing holding the
> axle in the fork. I wouldn't worry about your fork at all. Tens of
> thousands of forks are running just fine in that format.
I think you could convince yourself of the situation with a simple
test at low speed by riding on pavement with the QR open and apply the
brake. If it lifts out, then there are forces that bear on the hold
of the QR and that become larger with braking force. I don't believe
this will cause damage to the QR but it will disengage the wheel.
James Annan
jobst....@stanfordalumni.org wrote:
>
> I think you could convince yourself of the situation with a simple
> test at low speed by riding on pavement with the QR open and apply the
> brake. If it lifts out, then there are forces that bear on the hold
> of the QR and that become larger with braking force. I don't believe
> this will cause damage to the QR but it will disengage the wheel.
The problem with that test, is that it does not always not reliably
disengage the wheel. In which case, the rider might conclude that the
dangers must have been greatly exaggerated.
James
James Annan
>
> To make this more direct, lets assume the bicycle is capable of
> braking with 1g acceleration. That means, for a low rider position,
> the entire weight of rider and bicycle is bearing on the front
> wheel... and that the rotational torque of the front wheel is its
> radius times bicycle and rider weight. The radius at which the brake
> pads act is about a 6 inch while wheel periphery lies at about 13
> inches.
>
> That means the upward force on the brake pads is slightly greater than
> twice (13/6) the horizontal force (which is equal to the vertical
> force at 1g) on the front wheel. Therefore, there is a net
> disengagement force of half the rider-and-bicycle weight. This is not
> a safe condition when repeated often, especially on rough terrain
> where this force is hammered onto the wheel retention.
>
> This example is accurate because the ratio of forces is equally true
> for lesser braking forces. The example of 1g is to simplify
> visualization by avoiding arithmetic.
I think it's a bit worse than you calculated, since you have used wheel
radius but disk diameter in the calculation. Even my largeish disks are
only 185mm diameter, 3.6" to the outer edge and a bit less than that at
the centre of the pad. Also, the braking force acts primarily on one
dropout whereas the body weight is shared across two. Nevertheless, the
point is clear enough.
James
James Annan
Mark Hickey wrote:
>
> And personally I don't put a lot of faith in the fact that an
> "experienced cyclist who knows how to use a QR" can't simply forget to
> tighten it up.
I'm sure you're right that this must be the most likely explanation for
most front wheel loss. Several apparently experienced and competent
riders have mentioned finding their QR inexplicably loose in the middle
of a ride. I wonder if they have managed to get the coil spring (or
something else) caught on closing the lever? If people repeat even a
simple task frequently enough, then they are bound to make mistakes
occasionally. After all, the QR is supposed to be a fast and convenient
method, not something that needs a great deal of care and attention.
Then with a disk brake, there's a big force to push the skewer off
whatever obstruction was present, and disengage the wheel, especially
when braking hard down a steep slope...
James
James Annan
> As I understand this problem, the fork that began thuis discussion was
> erroneously built with the opening in the tip exactly lined up with the
> direction the axle wants to go when the brake is applied.
Well perhaps, but a quick look at a range of suspension forks in my LBS,
along with the picture of my Dirt Jumper fork on
http://www.ne.jp/asahi/julesandjames/home/disk_and_quick_release/marzocchi.jpg
indicates that many standard forks pretty much share this property (the
picture is the 'QR20' model, but the standard quick release model has
the same geometry and the slot aligned with the stanchion). I don't
think a minor misalignment of the order of 20 degrees is going to be
enough to solve the problem, given that cos(20) = 0.94.
I've recently seen some rigid forks with disk mount and rearward-facing
slot in a not-so-local LBS, I may go back and check the angles more
carefully.
> No quick release
> could hold against those forces, being amplified and directed out
the slot!
Well, that's my point. Although I would say that a QR certainly 'could',
and frequently does, but with insufficient reliability given the
consequences of failure.
James
M-Gineering import & framebouw
>
> Mark Hickey wrote:
> >
> > And personally I don't put a lot of faith in the fact that an
> > "experienced cyclist who knows how to use a QR" can't simply forget to
> > tighten it up.
>
> I'm sure you're right that this must be the most likely explanation for
> most front wheel loss. Several apparently experienced and competent
> riders have mentioned finding their QR inexplicably loose in the middle
> of a ride. I wonder if they have managed to get the coil spring (or
> something else) caught on closing the lever?
Not wiping the dropouts clean of mud and sand works for me ;)
--
Marten
James Annan
Jon Isaacs wrote:
> I think that they routinely can. Unlike the case with your fork, manufacturers
> actually engineer their forks and someone with specific numbers looks at these
> issues.
I think this is unrealistically optimistic. There are some major
manufacturers who probably do lots of testing, plenty of smaller ones
who do a bit (and who maybe do a certain amount of copying, without
necessarily knowing what needs to be copied precisely, and why). There
are also a lot of craftsmen who have learnt their trade by practice and
custom, and some of them are starting to use disk brakes. There's a
continuum of engineering skill.
> And given the number of disk brakes out there, it is pretty clear to
> me that were there a real issue, you would be seeing significant numbers of
> crashes.
Even a modest number is significant to those who crash. It's also pretty
likely to me that these crashes must be under-reported. Considering the
most recent case, maybe there's a limited amount of benefit to be gained
by speculating. However, those closely connected with it had described
it as a catastrophic fork failure (snapped somewhere at the crown). It
was only after I asked about the wheel, that it was mentioned that it
was found separately from the fork, at the top of the slope where the
crash occurred, some distance from rider and bike. Surely this crash was
caused by wheel separation, whatever the fundamental reason for that is.
But would they have realised, had I not asked the question?
>>Even if lips were present, the skewer is still
>>not designed to withstand the force although perhaps it would
>>'usually' be ok in practice.
>
>
> I don't think this has been established at all.
What about the ISO 1996 standard of 2300N symmetric pull? I am not
aware of any other design standards for the performance of a quick
release under load. And are you confident that if you use a more elastic
Ti skewer in the calculations, it definitely could not pull over a lip?
>
> I agree they should be designed for this, but look at the shear strength of
> steel and the forces involved, it doesn't take much.
I was thinking more of preventing the skewer nut, with a somewhat
rounded corner, of slipping over the bump under the application of
extremely high forces. That's not quite the same problem as a loose QR
falling out, although it shouldn't be too hard to design a lip for both
problems.
> In your case you have a hypothesis. It is important to remain objective and
> refrain from jumping to conclusions. You have found three other people who
> have lost a front wheel when using a disk brake. But it is important to
> realize that you went looking for people who believe that the brake was
> responsible for their crash and that just because you found people who believe
> that this was the case, it is important for you to establish that indeed this
> was the case.
I don't think that is fair, these cases were effectively thrust in front
of my face. The last one, was someone with whom I was discussing this
very issue only a few months ago following my fork failure! I should
repeat again that in this case I do not yet know for sure what caused
the failure, however the evidence for a wheel pull-out of some sort
seems extremely strong to me.
In the same space of time (only a few months), I've not heard of any
other crashes due to wheel loss with rim brakes, or in fact any other
serious crash due to any equipment failure at all, despite the massive
preponderance of non-disk bikes around (people have dredged up anecdotes
from their childhood, but nothing recent). To me, that points at the
disk brake as being a specific high risk factor.
> So far I see that the relying on the friction of the QR for rear mounted disk
> brakes is a poor idea. This is important and should not be neglected, from my
> point of view, your contention that the QR is cannot handle these loads when
> used with (properly designed) retention tabs has yet to be established.
I think I can probably agree with that. I do agree that it is hard to
see how a correctly fastened QR can either come loose, or slip over a
proper retention lip. And clearly the failures are quite rare in
absolute terms. Even if this is true, the question still remains in my
mind as to whether they are rare enough, and whether even if every
failure is due to 'operator error', it is reasonable to just blame the
person who did not install the QR correctly, and not do something to
stop it from happening again. Several people have mentioned finding QRs
loose mid-ride, which suggests to me that they probably caught something
under the QR on installing it. Of course that could and does happen
occasionally with rim brakes too, but in that case there is not the
massive force firstly pulling the nut off the obstruction, and then
pulling the loose QR over the lip. Even if the probability of doing it
wrong is small, if an action is repeated often enough, mistakes will be
made.
> I also think that this might be better carried on as private email so if you
> agree, just email me at my aol address.
Well, if you think so, however I also appreciate comments from other
people, who may have some ideas that neither of us would come up with on
our own. So I've posted and mailed it...
James
Jose Rizal
It's not a problem, it's a finding. In that case, your latter statement
will be true.
Mike Jacoubowsky
> see how a correctly fastened QR can either come loose, or slip over a
> proper retention lip. And clearly the failures are quite rare in
> absolute terms. Even if this is true, the question still remains in my
> mind as to whether they are rare enough, and whether even if every
> failure is due to 'operator error', it is reasonable to just blame the
> person who did not install the QR correctly, and not do something to
> stop it from happening again. Several people have mentioned finding QRs
> loose mid-ride, which suggests to me that they probably caught something
> under the QR on installing it. Of course that could and does happen
> occasionally with rim brakes too, but in that case there is not the
> massive force firstly pulling the nut off the obstruction, and then
> pulling the loose QR over the lip. Even if the probability of doing it
> wrong is small, if an action is repeated often enough, mistakes will be
> made.
A much simpler case for failure of the quick release system comes to mind.
We've been spending a lot of time analyzing a mechanism used properly, and
assuming there was nothing about it that might encourage improper use for an
experienced person. This is not always the case.
Some years ago, a customer of ours, who'd been riding for years on quick
release-equipped bicycles, claimed that it was quick release failure that
had caused the loss of the front wheel on their new road bike when they hit
a pot hole.
It didn't take too long to see what had actually happened. Front forks,
when painted, are typically held with their dropouts down, allowing paint to
flow down the fork and potentially pool on the dropout prior to it
completely drying. The fork in question had accumulated enough paint on the
dropout that the quick release was actually clamping on a very thick coating
of paint, so thick that it negated the effectiveness of the retention tab.
The combination of moving the clamping surfaces of the quick release out to
the width of the retention tabs, as well as the fact that it was clamping on
thick layers of paint instead of metal, allowed the quick release to
essentially tear away paint and almost float over the retention device.
Anytime I see paint that is even questionably thick in a front dropout, I
remove it. It's easy to do so; just clamp the quick release in place, but
not so tightly that it cannot be rotated. Then simply rotate the lever
around, grinding away any paint. Obviously this works best when using a
quick release that has some "teeth" on the surface to dig into the paint and
remove it.
--Mike-- Chain Reaction Bicycles
http://www.ChainReactionBicycles.com
Jose Rizal
The location of the brake caliper is important, since it determines the
magnitude of the component of the force acting on the axle parallel to
the dropout slot. The radial line from the axle to the caliper may be
almost horizontal, but the dropout slot is not vertical (it approaches
the head tube angle in most suspension forks, or about 30 degrees from
the vertical).
A practical test similar to the one you outlined on another post showed
that when the skewer was removed and disc brake applied at slow speed,
the caliper tended to push the axle on the rotor side down along the
slot, but the wheel rotates laterally due to the caliper applying force
on only one side of the wheel; the brake rotor and tire hitting the left
fork leg prevents the axle on this side from coming off the dropout.
The other end of the axle did not slide out of the slot at all. With
the skewer and QR screwed in, I expect either the axle or the skewer
will have to bend, or the QR will have to break, before the rotor end of
the axle will move along the dropout slot. The brake rotor and the
wheel will hit the fork leg before the axle will come off the dropout,
however. Of course different tire widths and brake rotor diameters will
influence how much movement the axle will have along the slot on the
rotor side.
M Gagnon
20030321100352...@mb-cr.aol.com...
> >and I'd appreciate any comment or criticism.
> >
>
> I looked over both your pages.
>
> Some comments and questions:
>
> 1. It appears to me that there were no "Lawyer Lips" on the fork in
question
> and that they might have been filed off at some time.
>
> Is this the case? This is an issue that should be addressed up front
because
> it is a critical part of any analysis.
>
....
> It seems to me that the problem here is not with the QR but rather with
the
> lack of adequate retension lips.
>
But just one comment: Why do you consider "lawyer lips" essential for fork
retention with disk brakes? Current bikes use lawyer lips or some other
positive retention system for the front wheel, but don't use any for the
rear wheel, and unless I'm mistaken, some bikes are designed with disk
brakes fitted on rear wheels, and these bikes don't use lawyer lips on the
rear wheel.
So if a rear wheel with disk brake is expected to remain in place without
lawer's lips, why would lawyer's lips be essential to keep the front wheel
in place?
BTW, I'm not questionning the safeguarding factors provided by lawyer's
lips, nor am I questionning the possibility of misusing these lips; I'm just
wondering why they would be essential to the disk-braking of a front wheel
but not of a rearwheel...
Regards,
Michel Gagnon
James Annan
> But just one comment: Why do you consider "lawyer lips" essential for fork
> retention with disk brakes? Current bikes use lawyer lips or some other
> positive retention system for the front wheel, but don't use any for the
> rear wheel, and unless I'm mistaken, some bikes are designed with disk
> brakes fitted on rear wheels, and these bikes don't use lawyer lips on the
> rear wheel.
>
> So if a rear wheel with disk brake is expected to remain in place without
> lawer's lips, why would lawyer's lips be essential to keep the front wheel
> in place?
>
> BTW, I'm not questionning the safeguarding factors provided by lawyer's
> lips, nor am I questionning the possibility of misusing these lips; I'm just
> wondering why they would be essential to the disk-braking of a front wheel
> but not of a rearwheel...
Interesting you should ask, and I was intending to put something on my
web page on this subject. It seems that the angle between a vertical
dropout and standard rear disk calliper is usually enough to prevent
the wheel from being pushed out. I say 'usually', because rear wheel
slip is a known issue on older models of Ventana El Conquistador
tandems which had slightly odd rearward-facing dropouts. As soon as
the maker became aware of the problem (which basically became obvious
when people started using decent disk brakes) he changed the design.
There some more info on this at:
http://home.att.net/~debbie.livingood/Ventana/02VentanaDropouts.html
although there is not a clear picture showing the old dropout directly
aligned with the calliper.
However, it's worth pointing out that this is a much lesser problem
than front wheel ejection, since the maximum forces from rear wheel
braking are necessarily much lower, the dropouts are properly
triangulated and so cannot move relative to each other (eg hop over
the axle end), when the wheel slips a little it will immediately start
to rub on the chainstay, and even if it wedges fully or comes out, a
rear wheel loss is unlikely to cause such a serious accident. The
ventana dropout also has a fairly long slot, the axle would have to
move a long way to have any chance of disengaging.
We have one of these Ventana El Conquistador tandems. I've had the
wheel slip a few times, before I worked out exactly what was causing
it. I now use a Shimano skewer, tightened to the point at which my
wife cannot operate it, and it hasn't slipped in a while. It's not
very a good design, and I certainly wouldn't want to trust my front
wheel to it!
[Singlespeed MTBers also suffer similar problems with horizontal
dropouts, a few cope with skewers but the cost of a failure (in this
case, while honking slowly up a steep hill) is pretty low.]
James
Tim McNamara
James Annan <still_th...@hotmail.com> wrote:
> >>Even if lips were present, the skewer is still not designed to
> >>withstand the force although perhaps it would 'usually' be ok in
> >>practice.
> >
> >
> > I don't think this has been established at all.
>
> What about the ISO 1996 standard of 2300N symmetric pull? I am not
> aware of any other design standards for the performance of a quick
> release under load. And are you confident that if you use a more
> elastic Ti skewer in the calculations, it definitely could not pull
> over a lip?
You're talking, what, 4 mm of elongation (total of height of lawyer
lips on both sides, possibly 8 mm if there are inner and outer lips on
both dropouts) over a 100 mm span? Can a skewer elongate that much
without fracturing?
> > I agree they should be designed for this, but look at the shear
> > strength of steel and the forces involved, it doesn't take much.
>
> I was thinking more of preventing the skewer nut, with a somewhat
> rounded corner, of slipping over the bump under the application of
> extremely high forces. That's not quite the same problem as a loose
> QR falling out, although it shouldn't be too hard to design a lip
> for both problems.
Skewer nuts that are aluminum for weight savings are notorious for
slipping. They simply don't "bite" into the dropout enough. This is
complicated even further by the use if Ti skewers which are rather
more elastic than steel ones. That was why I raised the question in
the first place as to what type of skewer the OP was using.
jobst....@stanfordalumni.org
> But just one comment: Why do you consider "lawyer lips" essential
> for fork retention with disk brakes? Current bikes use lawyer lips
> or some other positive retention system for the front wheel, but
> don't use any for the rear wheel, and unless I'm mistaken, some
> bikes are designed with disk brakes fitted on rear wheels, and these
> bikes don't use lawyer lips on the rear wheel.
Rear wheels have no forces that would make them disengage, a disk
brake caliper lying in front of the axle and adding engagement force
when in use. As most riders have noticed, a rear wheel does not
easily drop out even when the QR is open, the derailleur and chain
being an impediment. Beyond that a rear wheel displacement is a
relatively benign condition, skidding the rear wheel not being as
critical as a front wheel failure.
> So if a rear wheel with disk brake is expected to remain in place
> without lawyer's lips, why would lawyer's lips be essential to keep
> the front wheel in place?
I think this was explained in much detail in earlier postings. The
caliper on front wheels is behind the axle, on rear wheels it is in
front of the axle. Therefore, there is no disengagement force from a
rear disk brake.
> BTW, I'm not questioning the safeguarding factors provided by
> lawyer's lips, nor am I questioning the possibility of misusing
> these lips; I'm just wondering why they would be essential to the
> disk-braking of a front wheel but not of a rear wheel...
Because the prevent front wheel separation from the fork.
James Annan
>
> You're talking, what, 4 mm of elongation (total of height of lawyer
> lips on both sides, possibly 8 mm if there are inner and outer lips on
> both dropouts) over a 100 mm span? Can a skewer elongate that much
> without fracturing?
It only has to pull off one (LH) end, then the RH end falls out
freely. As probably happened in my case, where the knurled nut had
clearly bitten well into the paint but did not strip it off. So make
that 2mm (I haven't measured, but some of the lips look smaller than
that to me), and if there's some dirt on the fork end, or uneven
painting, then it could be less than that.
Perhaps a properly fastened QR skewer cannot be pulled off even in
these situations, but it is clear that for whatever reason, a lot of
cyclists have admitted to finding their front QR a little loose during
a ride, including ones with plenty of experience and mechanical
ability. The disk brake must make it that much more likely that the
moment of carelessness will turn into a serious crash, as it will try
to force the wheel out. The current system seems to be like a very
badly designed 'bolt through' or radially restrained axle, where the
radial restraint is just the 1 or 2mm of the skewer nut, with a
somewhat curved edge, catching on the two edges of lip with a big
force trying to eject the wheel through the gaping hole. Changing the
dropout angle would enable the QR fastening to return to its original
function of merely holding the fork ends over the axle, with no
significant forces applied to it.
> Skewer nuts that are aluminum for weight savings are notorious for
> slipping. They simply don't "bite" into the dropout enough. This is
> complicated even further by the use if Ti skewers which are rather
> more elastic than steel ones.
Well, maybe, but as far as the designers are concerned, they either
pass the ISO standard or they don't. I don't know what was used in all
cases, ours was a steel skewer and probably aluminium nuts, but with
the smooth surface at the important end the aluminium can't really
have been a factor.
James
Jon Isaacs
>But just one comment: Why do you consider "lawyer lips" essential for fork
>retention with disk brakes? Current bikes use lawyer lips or some other
>positive retention system for the front wheel, but don't use any for the
>rear wheel, and unless I'm mistaken, some bikes are designed with disk
>brakes fitted on rear wheels, and these bikes don't use lawyer lips on the
>rear wheel.
>So if a rear wheel with disk brake is expected to remain in place without
>lawer's lips, why would lawyer's lips be essential to keep the front wheel
>in place?
The magnitude of the force on the dropout is the same regardless of the
position of the brake but the direction of the force is a function of the brake
placement. The common location for the brake on the front wheel is behind the
fork, this forces the wheel down as the brake is applied.
Were the brake mounted to the front of the fork, then the force would be
pushing the wheel up against the dropout.
Also, the braking force in the rear is limited to much lower levels than in the
front because of tire slip.
>I'm just
>wondering why they would be essential to the disk-braking of a front wheel
>but not of a rearwheel...
Good question. But it is clear why it is not important in the back and it is
in the front.
Jon Isaacs
Jon Isaacs
>manufacturers who probably do lots of testing, plenty of smaller ones
>who do a bit (and who maybe do a certain amount of copying, without
>necessarily knowing what needs to be copied precisely, and why).
>There
>are also a lot of craftsmen who have learnt their trade by practice and
>custom, and some of them are starting to use disk brakes. There's a
>continuum of engineering skill.
I think that the manufacturers of the disk brakes have certainly looked at this
problem and certainly anyone building a fork for commercial sale in any number
must have. Some small builders such as the guy that built your frame obviously
did not understand that disk brakes are a different bird.
>
>Even a modest number is significant to those who crash. It's also pretty
>likely to me that these crashes must be under-reported. Considering the
>most recent case, maybe there's a limited amount of benefit to be gained
>by speculating. However, those closely connected with it had described
>it as a catastrophic fork failure (snapped somewhere at the crown). It
>was only after I asked about the wheel, that it was mentioned that it
>was found separately from the fork, at the top of the slope where the
>crash occurred, some distance from rider and bike. Surely this crash was
>caused by wheel separation, whatever the fundamental reason for that is.
>But would they have realised, had I not asked the question?
Hard to say. But you still, even in this case, established that the problem was
caused by the disk brake forcing the wheel out of the dropout past some
retention tabs.
>>>Even if lips were present, the skewer is still
>>>not designed to withstand the force although perhaps it would
>>>'usually' be ok in practice.
>> I don't think this has been established at all.
>What about the ISO 1996 standard of 2300N symmetric pull? I am not
>aware of any other design standards for the performance of a quick
>release under load.
The 2300 KN does not include the retention tab. From the calculations I have
made it appears that with tabs there is a large safety factor.
>And are you confident that if you use a more elastic
>Ti skewer in the calculations, it definitely could not pull over a lip?
>
You are the one making the claims that it could happen. So far I have seen no
calculation to indicate that is reasonable. The forces are just not there.
>I don't think that is fair, these cases were effectively thrust in front
>of my face.
Any of these people who contacted you because of your experience or because of
your website?
I imagine UFO websites get quite a few people who have seen UFOs. <g>
"The last one, was someone with whom I was discussing this
>very issue only a few months ago
>following my fork failure! I should
>repeat again that in this case I do not yet know for sure what caused
>the failure, however the evidence for a wheel pull-out of some sort
>seems extremely strong to me.
>In the same space of time (only a few months), I've not heard of any
>other crashes due to wheel loss with rim brakes, or in fact any other
>serious crash due to any equipment failure at all, despite the massive
>preponderance of non-disk bikes around (people have dredged up anecdotes
>from their childhood, but nothing recent). To me, that points at the
>disk brake as being a specific high risk factor.
Please realize that your website is one for people who have disk brakes. Why
would anyone respond otherwise?
Did I respond and tell you that my friend lost a front wheel going over a curb
because the QR was not tight?
If there a significant number of these, guys like Mike J, Andrew Muzi and
Sheldon Brown would be seeing them.
>> So far I see that the relying on the friction of the QR for rear mounted
>disk
>> brakes is a poor idea. This is important and should not be neglected, from
>my
>> point of view, your contention that the QR is cannot handle these loads
>when
>> used with (properly designed) retention tabs has yet to be established.
>
>I think I can probably agree with that. I do agree that it is hard to
>see how a correctly fastened QR can either come loose, or slip over a
>proper retention lip. And clearly the failures are quite rare in
>absolute terms. Even if this is true, the question still remains in my
>mind as to whether they are rare enough, and whether even if every
>failure is due to 'operator error', it is reasonable to just blame the
>person who did not install the QR correctly, and not do something to
>stop it from happening again. Several people have mentioned finding QRs
>loose mid-ride, which suggests to me that they probably caught something
>under the QR on installing it. Of course that could and does happen
>occasionally with rim brakes too, but in that case there is not the
>massive force firstly pulling the nut off the obstruction, and then
>pulling the loose QR over the lip. Even if the probability of doing it
>wrong is small, if an action is repeated often enough, mistakes will be
>made.
>
It is certainly not clear if all such failures are Operator Error or if some
are related to other issues. Certainly one possibility that is important to
address is whether someone had filed off the "Lawyer Lips" in order to make the
wheel easier to install and remove. Some riders make this a routine practice
and certainly might do so to a disk brake equipped bike without considering the
consequences.
THere have been several threads on this group where the preponderance of
posters suggest that there is no reason not to file the lips off, most often I
am in a small small minority suggesting it is wise to leave them in place.
Certainly now, anytime anyone asks, it should be made clear that this is a
no-no with a bike equipped with a front disk offset to the rear.
When investigating such crashes, I suggest that you make sure you establish via
photograph whether the fork in question did have "Lawyer Lips". In my mind
this is the critical issue.
Regarding private Email, I think it is best to keep it open, others seem to
have important things to add.
Anyway, I have certainly learned a fair amount from this thread. I had never
considered the issues you have raised and I think they are important ones.
Jon Isaacs
Tim McNamara
still_th...@hotmail.com (James Annan) wrote:
> Tim McNamara <tim...@bitstream.net> wrote in message
> news:<timmcn-B5FD88....@gemini.visi.com>...
>
> > You're talking, what, 4 mm of elongation (total of height of
> > lawyer lips on both sides, possibly 8 mm if there are inner and
> > outer lips on both dropouts) over a 100 mm span? Can a skewer
> > elongate that much without fracturing?
>
> It only has to pull off one (LH) end, then the RH end falls out
> freely.
Which has to overcome not only the clamping force but has to bend the
right hand segment of the skewer significantly.
> Perhaps a properly fastened QR skewer cannot be pulled off even in
> these situations, but it is clear that for whatever reason, a lot
> of cyclists have admitted to finding their front QR a little loose
> during a ride, including ones with plenty of experience and
> mechanical ability.
But this is not operator error, rather than necessarily a design
deficiency? We're not talking- for example- about SUV rollover
problems which can occur even in normal, proper use of the vehicle
because the design itself is deficient.
> The disk brake must make it that much more likely that the
> moment of carelessness will turn into a serious crash, as it will try
> to force the wheel out. The current system seems to be like a very
> badly designed 'bolt through' or radially restrained axle, where the
> radial restraint is just the 1 or 2mm of the skewer nut, with a
> somewhat curved edge, catching on the two edges of lip with a big
> force trying to eject the wheel through the gaping hole. Changing the
> dropout angle would enable the QR fastening to return to its original
> function of merely holding the fork ends over the axle, with no
> significant forces applied to it.
So the design should be changed because of operator error? I'm not
saying that's not reasonable, since people have been using QRs wrongly
for 50+ years. Nor am I finding fault with your engineering analysis
(I don't have the skills to do that, for one thing). A passing
acquaintance with Newton's laws explains the fundamental principle
even without being able to do the math.
I personally think that disc brakes are a solution to a nonexistent
problem and won't ever buy them for that reason. But is this a
problem if the QR is *properly* used? It's harder for a business to
justify the expense of redesign and retooling if the problem is caused
by the consumer improperly using the equipment- simply printing a
leaflet that accompanies the bike on "how to use your QR properly and
why, and if you don't then don't come crying to us" is a lot cheaper.
They also don't have to look at recalls and retrofits for existing
consumers.
Rather than changing the dropout angle, which may only partially solve
the problem, why not simply mount the brake caliper on the front of
the fork as is done on the rear wheel? End of problem.
If you're serious about this, you need to bring it to the attention of
the industry and/or the CPSC. This newsgroup- and your Web site- is
going to have no impact outside of the small circle of people who read
it.
James Annan
> >I don't think that is fair, these cases were effectively thrust in front
> >of my face.
>
> Any of these people who contacted you because of your experience or because of
> your website?
No, of course not. I'm not sure where you get this impression of my
website trawling for anecdotes. I only put that page up a few days
ago, and I can't imagine it will be found yet by any search engines.
The only sense in which you might claim I've been trawling for cases
is that having heard (on a discussion board that I have read for some
time, I wasn't searching) two anecdotes about MTBers' 'catastrophic
fork failure', I guessed correctly that they both involved disk brakes
and quick release front wheels, and that the wheel was found to be
separated from the fork. Although it is not yet proven (to my
knowledge), I hypothesise that the wheel separation was the root cause
of the crashes, as it seems by far the most plausible explanation of
those facts to me. All 4 crashes I know about in recent months took
place on steep descents where one might expect hard braking. In the
other 2 crashes for which I have more clear details, the wheel
separation was definitely the root cause and the disk brake was
definitely being used.
As part of these and other discussions, a lot of people have mentioned
the difficulty of keeping the front wheel straight and the need to use
a good skewer at an extremely high tension. Enough that I don't really
consider this bit to be speculative at all: wheels do slip,
frequently, under the loads applied by a disk brake.
I think there are two distinct issues here. The first issue is that
craftsmen working in their sheds (and the users of their forks) need
to be aware of the danger, and I think we all agree on that. A
freshly-painted dropout at the wrong angle with no retention lip is a
recipe for disaster.
The second issue is whether the design of current MTB forks is
sensible or adequate. Here I think I do agree that a properly fastened
QR skewer should not be able to pull over a decent retention lip,
although it may (and does) break, especially under the extremely high
tension that MTBers routinely use to stop the wheel slipping. However,
people continue to lose their front wheels. There are lots of
anecdotes about people finding their QR loose mid-ride, and the most
plausible explanation I can think of for this is that somehow it was
not securely fixed, due perhaps to getting a bit of dirt under it.
Although that is equally likely with rim brakes, the massive downward
force of a disk must be more likely to turn a slightly loose skewer
into a crash. In many of the cases, the person certainly 'knew how to
use a QR': surely it was a careless error rather than complete
ignorance, but given a lot of people regularly removing and
reinstalling their wheels, often in dirty conditions, there are bound
to be some careless mistakes made. The whole point of the QR is to be
fast and simple, and if it works fine 99.9% of the time, people will
inevitably get a bit careless. I don't know if it is reasonable to
just tell the unlucky few that it is their fault for doing it wrong,
when a simple change like altering the fork dropout angle would
completely eliminate the loading from the QR. You might still get some
people losing their wheels, but I bet it would be a lot fewer. It
would have to be loose enough to fall right out, rather than loose
enough to be pulled over the lip by a 2kN force. You wouldn't also
have people having to overstress their skewers to keep the wheel in
place.
I do find it a little strange that some skewers (Hope, for example)
seem to have a reputation for coming loose. Surely if it was the
user's fault, then changing the skewer would not solve the problem.
I've not seen a Hope skewer and have no idea how or why this could
occur.
> Did I respond and tell you that my friend lost a front wheel going over a curb
> because the QR was not tight?
Was this a recent crash? All the ones I refer to are in the last few
months (maybe 6). Given the ratio of disk to non-disk bikes, it seems
like a lot. I haven't ruled out pure coincidence as a cause though...
> If there a significant number of these, guys like Mike J, Andrew Muzi and
> Sheldon Brown would be seeing them.
Yes, obviously the number of crashes is quite small in absolute terms.
However, my experience does suggest that people with 'catastrophic
fork failure' might not even work out that the crash was due to wheel
loss. When someone says that their fork broke, will the bike shop
staff think to ask the pertinent questions, or just sell them a new
fork?
James
James Annan
> But is this a
> problem if the QR is *properly* used? It's harder for a business to
> justify the expense of redesign and retooling if the problem is caused
> by the consumer improperly using the equipment- simply printing a
> leaflet that accompanies the bike on "how to use your QR properly and
> why, and if you don't then don't come crying to us" is a lot cheaper.
I don't think it is reasonable to think that the problem is always due
to someone actually not knowing how to use a QR. A lot of people with
a wide range of experience and mechanical ability have reported having
found their front QR inexplicably loose in the middle of the ride, and
if you write them all off as incompetent (rather than just the ones
who crashed), then it would seem that 100% reliable use of a QR may
indeed be beyond the competence of a typical rider in typical riding
conditions. You've done it hundreds of times, you're in a rush to get
going, so you slap on the wheel and don't notice the mud on the
dropout, or something. Just a guess. The difference the disk makes is
(probably) to turn a larger proportion of these careless errors into
serious crashes.
> If you're serious about this, you need to bring it to the attention of
> the industry and/or the CPSC. This newsgroup- and your Web site- is
> going to have no impact outside of the small circle of people who read
> it.
Certainly, but I also like to hear the input of the range of people
here with different experience and knowledge. I'm sure the industry
already hears plenty from people who failed to do their skewers up
properly! I don't expect to turn the world around by myself, and I am
happy to admit I don't really know how big a problem the QR + disk
brake issue is, but the basic design seems obviously flawed to me. It
does appear that the more robust axle clamping methods are filtering
down from the heavier-duty forks to lighter models, and I'm sure that
trend will continue. I don't expect the fork manufacturers to admit
that this is due to wheel loss incidents though.
James
James Annan
James Annan wrote:
> A lot of people with
> a wide range of experience and mechanical ability have reported having
> found their front QR inexplicably loose in the middle of the ride
I've just had some more thoughts about this while cycling home...
Just about every bolt connected with my disk brake has come loose
occasionally (with some of them, it's annoyingly frequent). This
phenomenon is certainly far from rare, other people have just advised me
to use loctite (which I now do, sometimes). So why should the QR skewer
be any different?
Of course we all 'know' that they don't unscrew, it is impossible. Yet
from the first link I found on the subject,
http://www.boltscience.com/pages/vibloose.htm :
---
"Pre-loaded bolts (or nuts) rotate loose, as soon as relative motion
between the male and female threads takes place. This motion cancels the
friction grip and originates an off torque which is proportional to the
thread pitch and to the preload. The off torque rotates the screw loose,
if the friction under the nut or bolt head bearing surface is overcome,
by this torque.
There are three common causes of the relative motion occurring in the
threads:
[1,2 deleted]
3. Applied forces on the joint can lead to shifting of the joint
surfaces leading to bolt loosening."
---
So why can this not apply to QR skewers? It certainly applies to all the
other bolts in the vicinity.
Presumably, for rim braked bikes, the forces involved are never high
enough to lead to any shifting of the joint surfaces. I think we've
shown that this is not necessarily the case where a disk brake is
concerned. Moreover, as well as the large forces applied, a disk brake
can generate a massive amount of vibration, and it is directly applied
in the hub area. It unscrews the other bolts frequently enough.
So, is vibration-induced loosening a plausible explanation for skewers
unscrewing? If not, what have I missed?
I realise that many skewers have sufficient knurling to bite into many
dropouts - I'm looking for possible explanations rather than inevitable
or even probable outcomes. But this idea also would help to explain why
some skewers have a better reputation than others, which otherwise seems
impossible according to the 'infallible' theory.
James (just lobbing up ideas, waiting for someone to smash them out of
the park)
James Annan
James Annan wrote:
> A lot of people with
> a wide range of experience and mechanical ability have reported having
James Annan
James Annan wrote:
> A lot of people with
> a wide range of experience and mechanical ability have reported having
Jon Isaacs
>website trawling for anecdotes.
It is not that it trawling for anecdotes, rather that people with such
experiences will find it.
But apparently that is not the case.
>Enough that I don't really
>consider this bit to be speculative at all: wheels do slip,
>frequently, under the loads applied by a disk brake.
Even with retention tabs? Pretty hard to see how they would slip if they were
properly installed.
>I think there are two distinct issues here. The first issue is that
>craftsmen working in their sheds (and the users of their forks) need
>to be aware of the danger, and I think we all agree on that. A
>freshly-painted dropout at the wrong angle with no retention lip is a
>recipe for disaster.
Yeah, I think that is pretty much a given.
>The second issue is whether the design of current MTB forks is
>sensible or adequate. Here I think I do agree that a properly fastened
>QR skewer should not be able to pull over a decent retention lip,
>although it may (and does) break, especially under the extremely high
>tension that MTBers routinely use to stop the wheel slipping.
Don't know about this one. You have evidence of QRs breaking. This should
happen as the QR is being installed so it should be obvious to the person
installing the wheel.
> There are lots of
>anecdotes about people finding their QR loose mid-ride, and the most
>plausible explanation I can think of for this is that somehow it was
>not securely fixed, due perhaps to getting a bit of dirt under it.
>The whole point of the QR is to be
>fast and simple, and if it works fine 99.9% of the time, people will
>inevitably get a bit careless.
Personally I have long been leary of QRs and understand how tenuously they hold
our safety. I believe it is important to take great care with installing the
front wheel and think that many people are not quite aware of both the proper
procedure and the potential for disaster.
>Was this a recent crash? All the ones I refer to are in the last few
>months (maybe 6). Given the ratio of disk to non-disk bikes, it seems
>like a lot. I haven't ruled out pure coincidence as a cause though...
I still think that part of the reason you have discovered these is because of
your experience with loosing a front wheel. This is not to say that this is
bad or that you shouldn't be working on this, rather simply that you probably
don't see a representative sample.
>> If there a significant number of these, guys like Mike J, Andrew Muzi and
>> Sheldon Brown would be seeing them.
>
>Yes, obviously the number of crashes is quite small in absolute terms.
>However, my experience does suggest that people with suggest that people with
'catastrophic fork failure' might not even work out that the crash was due to
wheel
>loss. >When someone says that their fork broke, will the bike shop
>staff think to ask the pertinent questions, or just sell them a new
>fork?
A good possibility. This would be a good time to ask this:
Has anyone reading this thread seen or even heard of incidents of Catatrophic
Fork Failure or wheel loss for riders riding bikes with front disk brakes?
Any other incidents that might be note worthy?
I must say I am curious about what the outcome of your crash was and how the
fork builder and frame builder reacted. Did they assume responsibility and
immediately admit that they had screwed up big time? I would hope so.
Jon Isaacs
Steve
8" downhill brake designed for downhill forks with 20mm axles, not
ridged QR forks. Also, ridge forks are designed to have some flex in
them to provide a small amount of suspension, while suspension forks
are very stiff.
Scott Godfrey
> Don't know about this one. You have evidence of QRs breaking. This should
> happen as the QR is being installed so it should be obvious to the person
> installing the wheel.
> Any other incidents that might be note worthy?
Relating these two statements together, I've had two steel skewers
break on me at different times. One during installation and the other
while riding an "MTB road bike" years back. Both unrelated to disc
brakes and fork failures, etc., but illustrating the point that
failures of this sort do not necessarily occur at install. The
failure in motion did wedge the wheel into the fork and bent one of
the blades (Cro-mo).
-S
Jose Rizal
> To be fair, the rotor is 185mm (7.3"), which is in the middle of the
> range of the more common sizes (165, 185, 205mm). It's true that
> consumers are advised by most, if not all disc brake manufacturers,
> not to use the larger disc rotors with standard QR dropouts.
I haven't seen any warnings at all from any disc brake manufacturers
about this, nor any cautionary advice. Where have you got your warning
from, and what is the basis for it?
> Many
> people (especially teenagers) do it anyway and there are certainly
> reports of the axle pulling out of the dropout where they didn't with
> the smaller rotors - and this is on modern suspension forks with
> lawyer lips.
Where are the reports being published, and by whom?
Karl Frisch
>
> Chris B.:
>
> > To be fair, the rotor is 185mm (7.3"), which is in the middle of the
> > range of the more common sizes (165, 185, 205mm). It's true that
> > consumers are advised by most, if not all disc brake manufacturers,
> > not to use the larger disc rotors with standard QR dropouts.
>
> I haven't seen any warnings at all from any disc brake manufacturers
> about this, nor any cautionary advice. Where have you got your warning
> from, and what is the basis for it?
<snip>
Just to ad fuel to the fire:
Avid does warn against using their 203's with a standard quick release.
If you look at their fork compatibility chart it says "Be aware that
fork manufactures discourage the use of anything larger than a 185mm
rotor on a standard quick-release." Nice to see that they put the onus
on the fork manufacturer.
James Annan
> Why are you using a Hope M4 on the front wheel of a tandem? That's an
> 8" downhill brake designed for downhill forks with 20mm axles, not
> ridged QR forks.
Um....oh, _rigid_.
The Hope brake is excellent for tandems, we have used them since they
first came out, and many tandem MTBers agree with our assessment. They
are one of only a very few models of disk brake that are up to the
job. We had used it on the front of that tandem for some time with a
suspension fork, and didn't see any reason to buy a new brake when we
switched to the rigid fork, since the fork was built for use with
them. I'm not aware of any recommendation from Hope that they should
not be used with QR forks.
> Also, ridge forks are designed to have some flex in
> them to provide a small amount of suspension, while suspension forks
> are very stiff.
I think perhaps you haven't looked carefully at your 'stiff'
suspension forks while braking hard!
James
jobst....@stanfordalumni.org
>> Why are you using a Hope M4 on the front wheel of a tandem? That's
>> an 8" downhill brake designed for downhill forks with 20mm axles,
>> not ridged QR forks.
> To be fair, the rotor is 185mm (7.3"), which is in the middle of the
> range of the more common sizes (165, 185, 205mm). It's true that
> consumers are advised by most, if not all disc brake manufacturers,
> not to use the larger disc rotors with standard QR dropouts. Many
> people (especially teenagers) do it anyway and there are certainly
> reports of the axle pulling out of the dropout where they didn't
> with the smaller rotors - and this is on modern suspension forks
> with lawyer lips.
The smaller the disk diameter, the greater the disengagement force for
the same braking effect. Therefore, I see no connection between
larger disks and more wheel disengagements. What is less apparent,
is that a disk brake puts the same bending force that appears at the
fork crown on the end of the single fork blade at the caliper.
This severe bending moment is the reason not to put a disk brake on a
conventional bicycle (road) fork with a slender blade end, an end that
is not intended to carry bending loads like those at the large
diameter end at the fork crown.
> Of course, some of these people might not be tightening their QR
> skewers tightly enough but surely some of them are and this would
> suggest that there is not a huge safety margin. Of course these
> anecdotes are not worth very much but perhaps it is worth looking at
> simply as a way of approaching the issue differently.
I find amazing how, by insinuation, this whole matter is n=being
relegated to op0erator error. Nearly all of these replies seem like
so much whistling in the dark, for fear that someone will recognize
that disks are inappropriate for front wheels as used today. Face it,
this is not a reasonable design.
> A 20mm _through-axle_ fork would have been a safe bet with an 8"
> disc for sure. However, a lot of the 20mm suspension forks don't
> use a through-axle but rather a similar (but larger) QR setup - in
> fact, the axle I use with my Marzocchi QR20 uses a skewer of the
> _same diameter and pitch_ as a regular QR skewer nested inside the
> 20mm axle - no advantage is taken to increase the clamping force.
> Of course there is greater retention anyway by virtue of the
> proportionally larger axle, dropout and "lawyer lips" but for
> 2001(?), Marzocchi chose to add separate retention tabs which
> spanned across the bottom of the dropouts and were bolted down
> separately ("QR20+") though I understand that this is no longer the
> case, probably because people found it very inconvenient. I'm not
> aware of people having problems with the original QR20 system with
> 8" discs but there must have been some reason for them to have
> played around with the design.
So what is a larger skewer or axle supposed to do? The implication is
that there are forces here that are not present with rim brakes. So!
What are these forces if they aren't disengagement forces. Obviously
it is not an axle or bearing overload, because these are not failing.
What is failing is the attachment of the wheel to the dropout. I
think we've gone through the simple force analysis to show that there
is a large downward (wheel separating) force on the axle.
>> Also, ridge forks are designed to have some flex in them to provide
>> a small amount of suspension, while suspension forks are very
>> stiff.
> I'm not sure that this is a rigid vs suspension issue so much as an
> issue of whether the fork was designed to deal with the loads of rim
> brakes or disc brakes or both.
I don't know where the flex in "rigid" forks is supposed to come from.
If you mean conventional tapered fork blades as on road bicycles, then
that is an additional hazard that will cause fork failure at the
caliper attachment.
jobst....@stanfordalumni.org
> Avid does warn against using their 203's with a standard quick
> release. If you look at their fork compatibility chart it says "Be
> aware that fork manufactures discourage the use of anything larger
> than a 185mm rotor on a standard quick-release." Nice to see that
> they put the onus on the fork manufacturer.
What reason is given for the warning? It seems like another
misunderstanding of the bicycle industry. The larger the disk, the
lower the forces. In fact when the contact diameter of the disk pads
reach that of the rim (as in rim brakes) braking forces are minimized.
What is going on here?
Karl Frisch
>
> Karl Frisch writes:
>
> > Avid does warn against using their 203's with a standard quick
> > release. If you look at their fork compatibility chart it says "Be
> > aware that fork manufactures discourage the use of anything larger
> > than a 185mm rotor on a standard quick-release." Nice to see that
> > they put the onus on the fork manufacturer.
> What reason is given for the warning?
No reason was given.
> It seems like another
> misunderstanding of the bicycle industry. The larger the disk, the
> lower the forces. In fact when the contact diameter of the disk pads
> reach that of the rim (as in rim brakes) braking forces are minimized.
Notice that I didn't comment on the validity of the statement. I only
pointed out that the "warning" exists.
Of course the warning is based on a misunderstanding in the bicycle
industry. And by noting that the warning comes from "fork manufacturers"
the company supplying the brakes can happily propagate the
misunderstanding, and free itself from doing any research on the matter.
> What is going on here?
Bicycle industry business as usual.
Tim McNamara
jobst....@stanfordalumni.org wrote:
> I find amazing how, by insinuation, this whole matter is being
> relegated to operator error.
As you have pointed out in many posts, it is very frequent for what is
in fact operator error to be misconstrued as faulty design, especially
in lawsuits. You've testified in court about that very topic on
multiple ocasions, from what you've written over the years. I recall
that you've said that in almost every case of blaming the design-
especially as related to quick releases- the cause was in fact
operator error.
So, given that, I think it was a reasonable first question to ask: is
the problem reported by the OP caused by operator error rather than
poor design? If it is from poor design, why are we not flooded with
similar reports by experienced cyclists who know how to use a quick
release? This is quite unlike SUV rollovers occuring in normal
traffic maneuvers, where it is abundantly clear that the design of the
vehicles is faulty simply from the frequency of occurrence under
normal conditions.
> Nearly all of these replies seem like so much whistling in the
> dark, for fear that someone will recognize that disks are
> inappropriate for front wheels as used today. Face it, this is not
> a reasonable design.
If we are not seeing many such failures, is the design still
necessarily a bad one? Or does a standard Campagnolo or Shimano type
QR have adequate clamping force to keep the wheel in the dropouts, if
properly used? I can't answer either of those questions, not having
the technical knowledge. I can predict that this would be the
industry stance, however, because it's cheaper to blame the operator
than to recall 100,000 or more forks and brake calipers and replace
them with a corrected design.
It appears from the available information and explanations that the
current crop of front disc brakes for bicycles has a significant flaw,
in that the front wheel braking force attempts to force the axle out
of the dropout slot. The harder you brake, the greater the force
pushing the axle out without much increase in load on the wheel to
counteract the ejection of the axle. As I mentioned earlier, a side
view photograph and a basic comprehension of action and reaction are
enough to explain the problem succinctly without resorting to
mathematics. The obvious cure is to design the system to put the
caliper in front of the fork; current forks and calipers should be
recalled and replaced.
This should be brought to the attention of fork and brake
manufacturers for immediate correction, and should probably also be
brought to the attention of the CPSC to enforce a recall. A Usenet
newsgroup with a few hundred relatively sophisticated readers-
compared to the tens of thousands of people who have purchased such
bicycles- is not going to resolve the problem.
M Gagnon
>
> > > > So if a rear wheel with disk brake is expected to remain in place
without
> > lawer's lips, why would lawyer's lips be essential to keep the front
wheel
> > in place?
> >
"James Annan" <still_th...@hotmail.com> replied
>
> Interesting you should ask, and I was intending to put something on my
> web page on this subject. It seems that the angle between a vertical
> dropout and standard rear disk calliper is usually enough to prevent
> the wheel from being pushed out. ....
Thanks.
Michel Gagnon
James Annan
> So, given that, I think it was a reasonable first question to ask: is
> the problem reported by the OP caused by operator error rather than
> poor design? If it is from poor design, why are we not flooded with
> similar reports by experienced cyclists who know how to use a quick
> release? This is quite unlike SUV rollovers occuring in normal
> traffic maneuvers, where it is abundantly clear that the design of the
> vehicles is faulty simply from the frequency of occurrence under
> normal conditions.
>
It's obvious that the failure is quite low in absolute terms, and the
failure mode only applies to cyclists with disk brakes, which are
still relatively rare. Moreover, most of these cyclists are fairly
experienced, and since they 'know' that the quick release is
infallible, they are unlikely to complain.
>
> If we are not seeing many such failures, is the design still
> necessarily a bad one?
> This should be brought to the attention of fork and brake
> manufacturers for immediate correction, and should probably also be
> brought to the attention of the CPSC to enforce a recall. A Usenet
> newsgroup with a few hundred relatively sophisticated readers-
> compared to the tens of thousands of people who have purchased such
> bicycles- is not going to resolve the problem.
It's not for me to say whether the design problem is so serious that a
recall is required - maybe it is cheaper and simpler to pay off those
who crash (and who can successfuly claim).
"Take the number of vehicles in the field (A), multiply it by the
probable rate of failure (B), then multiply the result by the average
out-of-court settlement (C). A times B times C equals X...
If X is less that the cost of a recall, we don't do one."
Does anyone have anything to say about my comments on
vibration-induced loosening? I consider this to be the crux of the
matter, as it explains how a QR can escape even where there is a
retention lip (and also the anecdotes of mysterious mid-ride skewer
loosening).
James
jobst....@stanfordalumni.org
>> I find amazing how, by insinuation, this whole matter is being
>> relegated to operator error.
> As you have pointed out in many posts, it is very frequent for what
> is in fact operator error to be misconstrued as faulty design,
> especially in lawsuits. You've testified in court about that very
> topic on multiple occasions, from what you've written over the years.
> I recall that you've said that in almost every case of blaming the
> design- especially as related to quick releases- the cause was in
> fact operator error.
This never involved disk brakes, it was always with reference to
bicycles with rim brakes and conventional QR's. These do not loosen
from use as John Howard claimed in his expert testimony that brought
"lawyer lips" to bicycling. I sense that you are pressing the
operator error scenario in spite of irrefutable analysis of large
disengagement forces caused by disk brakes.
> So, given that, I think it was a reasonable first question to ask: is
> the problem reported by the OP caused by operator error rather than
> poor design? If it is from poor design, why are we not flooded with
> similar reports by experienced cyclists who know how to use a quick
> release?
That is because most riders over tighten QR's anyway as we see from
frequent bearing failures, a subject discussed under another thread.
As it is, riders err on the safe side but when a failure occurs, it is
most likely mis-attributed to skewer failure that occurs if there is
brake induced wheel separation.
> This is quite unlike SUV rollovers occurring in normal traffic
> maneuvers, where it is abundantly clear that the design of the
> vehicles is faulty simply from the frequency of occurrence under
> normal conditions.
The SUV rollover occurs because people believe they are driving a
vehicle as safe and fool proof as a sedan that will not turn over on a
dry pavement spin-out. SUV's and similar vehicles were formerly
driven mainly by people aware of their lateral acceleration
limitations. Our newspapers are reporting SUV rollovers more
frequently these days, ones that occurred as a result of a spin-out,
an otherwise benign incident. The latest involved a fatality of a
noted athlete who was a passenger.
>> Nearly all of these replies seem like so much whistling in the
>> dark, for fear that someone will recognize that disks are
>> inappropriate for front wheels as used today. Face it, this is not
>> a reasonable design.
> If we are not seeing many such failures, is the design still
> necessarily a bad one?
Yes, especially because one can demonstrate that it causes wheel
separation if the QR is not unduly tight. Another reason could be
that these incidents usually occur when a rider is performing a
difficult descending maneuver that people generally consider risky.
That would suppress many complaints, especially if the rider was
unaware why the wheel came out.
> Or does a standard Campagnolo or Shimano type QR have adequate
> clamping force to keep the wheel in the dropouts, if properly used?
With knurled jam nuts, sufficient retention is possible but I would
not like to place my safety on that condition. It's like asking
someone to hang a bicycle from the front wheel and then hang his
entire weight on the bicycle by swinging from the rear wheel... over a
substantial drop off. If I had my feet near the ground, OK but not if
I was to risk my well being on it.
> I can't answer either of those questions, not having the technical
> knowledge. I can predict that this would be the industry stance,
> however, because it's cheaper to blame the operator than to recall
> 100,000 or more forks and brake calipers and replace them with a
> corrected design.
I don't see why "we" should be apologists for the industry and offer
other explanations for the reported failures. It is more our place to
analyze the effect, suggest solutions, and to be fully aware of the
hazard.
> It appears from the available information and explanations that the
> current crop of front disc brakes for bicycles has a significant
> flaw, in that the front wheel braking force attempts to force the
> axle out of the dropout slot. The harder you brake, the greater the
> force pushing the axle out without much increase in load on the
> wheel to counteract the ejection of the axle. As I mentioned
> earlier, a side view photograph and a basic comprehension of action
> and reaction are enough to explain the problem succinctly without
> resorting to mathematics. The obvious cure is to design the system
> to put the caliper in front of the fork; current forks and calipers
> should be recalled and replaced.
That sounds reasonable to me but the problem remains that the bicycle
industry should be more technically astute and discover these things
before the get on the market.
> This should be brought to the attention of fork and brake
> manufacturers for immediate correction, and should probably also be
> brought to the attention of the CPSC to enforce a recall. A Usenet
> newsgroup with a few hundred relatively sophisticated readers-
> compared to the tens of thousands of people who have purchased such
> bicycles- is not going to resolve the problem.
I am amazed that bicycle companies do not lurk on wreck.bike just to
see what ails the users and what they discover. This is by and large
typical of the bicycle business. The few people from the business who
participate here are not in much of a position to change manufacturing
trends.
Tim McNamara
jobst....@stanfordalumni.org wrote:
> Tim McNamara writes:
>
> >> I find amazing how, by insinuation, this whole matter is being
> >> relegated to operator error.
>
> > As you have pointed out in many posts, it is very frequent for
> > what is in fact operator error to be misconstrued as faulty
> > design, especially in lawsuits. You've testified in court about
> > that very topic on multiple occasions, from what you've written
> > over the years. I recall that you've said that in almost every
> > case of blaming the design- especially as related to quick
> > releases- the cause was in fact operator error.
>
> This never involved disk brakes, it was always with reference to
> bicycles with rim brakes and conventional QR's. These do not
> loosen from use as John Howard claimed in his expert testimony that
> brought "lawyer lips" to bicycling. I sense that you are pressing
> the operator error scenario in spite of irrefutable analysis of
> large disengagement forces caused by disk brakes.
No, not at all. I have thought disc brakes were a silly thing- after
all, bikes already *had* disc brakes all along. Hydraulic disc brakes
seemed to me to be nothing more than a well-marketed better mousetrap
borrowed from automobiles. It turns out they're more than a silly
thing but in fact are a design that is poor and potentially dangerous.
I was just commenting on how and why the issue of operator error
arose- since I was one of the first posters to ask about it.
I had some initial skepticism about the forceful disengagement
proposition, mainly because it didn't seem to me that the
disengagement forces could be large enough to overcome a properly
tightened QR. That's due to my inability to understand the math in
question, I suspect. The wages of working in a soft science...
> >> Nearly all of these replies seem like so much whistling in the
> >> dark, for fear that someone will recognize that disks are
> >> inappropriate for front wheels as used today. Face it, this is
> >> not a reasonable design.
>
> > If we are not seeing many such failures, is the design still
> > necessarily a bad one?
>
> Yes, especially because one can demonstrate that it causes wheel
> separation if the QR is not unduly tight. Another reason could be
> that these incidents usually occur when a rider is performing a
> difficult descending maneuver that people generally consider risky.
> That would suppress many complaints, especially if the rider was
> unaware why the wheel came out.
And, as you or the OP pointed out, riders experiencing this sort of
failure may also blame themselves inappropriately- after all,
"everyone knows" that a properly used QR doesn't spontaneously release.
> > Or does a standard Campagnolo or Shimano type QR have adequate
> > clamping force to keep the wheel in the dropouts, if properly
> > used?
>
> With knurled jam nuts, sufficient retention is possible but I would
> not like to place my safety on that condition. It's like asking
> someone to hang a bicycle from the front wheel and then hang his
> entire weight on the bicycle by swinging from the rear wheel...
> over a substantial drop off. If I had my feet near the ground, OK
> but not if I was to risk my well being on it.
Thanks.
> > I can't answer either of those questions, not having the technical
> > knowledge. I can predict that this would be the industry stance,
> > however, because it's cheaper to blame the operator than to recall
> > 100,000 or more forks and brake calipers and replace them with a
> > corrected design.
>
> I don't see why "we" should be apologists for the industry and offer
> other explanations for the reported failures. It is more our place to
> analyze the effect, suggest solutions, and to be fully aware of the
> hazard.
Agreed, but it is beneficial to predict and thus be able to refute the
"operator error" defense from the get-go.
> > It appears from the available information and explanations that
> > the current crop of front disc brakes for bicycles has a
> > significant flaw, in that the front wheel braking force attempts
> > to force the axle out of the dropout slot. The harder you brake,
> > the greater the force pushing the axle out without much increase
> > in load on the wheel to counteract the ejection of the axle. As
> > I mentioned earlier, a side view photograph and a basic
> > comprehension of action and reaction are enough to explain the
> > problem succinctly without resorting to mathematics. The obvious
> > cure is to design the system to put the caliper in front of the
> > fork; current forks and calipers should be recalled and replaced.
>
> That sounds reasonable to me but the problem remains that the
> bicycle industry should be more technically astute and discover
> these things before the get on the market.
This is certainly true! How many threads have we had in the past 10
years exposing major design and materials flaws in the bike industry?
Probably thousands. And people are still riding marginal designs that
can't be readily repaired or reasonably jury-rigged on the roadside
far from home.
> > This should be brought to the attention of fork and brake
> > manufacturers for immediate correction, and should probably also
> > be brought to the attention of the CPSC to enforce a recall. A
> > Usenet newsgroup with a few hundred relatively sophisticated
> > readers- compared to the tens of thousands of people who have
> > purchased such bicycles- is not going to resolve the problem.
>
> I am amazed that bicycle companies do not lurk on wreck.bike just
> to see what ails the users and what they discover. This is by and
> large typical of the bicycle business. The few people from the
> business who participate here are not in much of a position to
> change manufacturing trends.
Free market research and free engineering analysis! Seems like a
resource that bike industry manufacturers would love to take advantage
of. Of course, if they're lurking we may never know.
Jose Rizal
> jobst....@stanfordalumni.org wrote:
> >
> > Karl Frisch writes:
> >
> > > Avid does warn against using their 203's with a standard quick
> > > release. If you look at their fork compatibility chart it says "Be
> > > aware that fork manufactures discourage the use of anything larger
> > > than a 185mm rotor on a standard quick-release." Nice to see that
> > > they put the onus on the fork manufacturer.
>
> > What reason is given for the warning?
>
> No reason was given.
Well, Answer (Manitou forks) gives the reason as the cast lower legs of
their non-DH specific forks are not designed to handle the "bigger
leverage of the larger rotors". Not sure what this means, but it may
have something to do with the strength of the mounts for the calipers,
which are integral to the lower legs.
> > It seems like another
> > misunderstanding of the bicycle industry. The larger the disk, the
> > lower the forces. In fact when the contact diameter of the disk pads
> > reach that of the rim (as in rim brakes) braking forces are minimized.
>
> Notice that I didn't comment on the validity of the statement. I only
> pointed out that the "warning" exists.
What of the forces in the mounts, though? My understanding of it is
that the moment about the axle of the caliper force on the rotor (say
Fc) must equal the sum of the moments about the axle of the forces
acting on the two mounts (F1 and F2). That is, Mc = M1 + M2. If a
larger diameter rotor is used, Fc is less and its moment arm is longer,
while keeping Mc the same. Although Mc = M1 + M2 still, the mounts are
longer (ie the caliper is further away from the axle) and this will
result in a greater bending moment for each of the two mounts. What
have I missed in this analysis?
> Of course the warning is based on a misunderstanding in the bicycle
> industry. And by noting that the warning comes from "fork manufacturers"
> the company supplying the brakes can happily propagate the
> misunderstanding, and free itself from doing any research on the matter.
I think this is only fair, since the brake manufacturers can justifiably
put the onus on the fork manufacturers to design forks which are able to
use their products. If the brake calipers need to be shifted to the
front of the forks to increase their safety, or any other geometrical
changes, it'll have to be at the initiative of the fork manufacturers to
re-design their forks. Disc brakes can be used in bicycles, and fork
manufacturers must be the ones to provide suitable fork designs to be
able to use these safely.
> > What is going on here?
Worth finding out, that's for sure.
James Annan
jobst....@stanfordalumni.org wrote:
>
> This never involved disk brakes, it was always with reference to
> bicycles with rim brakes and conventional QR's. These do not loosen
> from use as John Howard claimed in his expert testimony that brought
> "lawyer lips" to bicycling.
Just out of interest, I wonder if the much reviled John Howard had used
disk brakes himself prior to his testimony.
James
jobst....@stanfordalumni.org
>>> What reason is given for the warning?
>> No reason was given.
> Well, Answer (Manitou forks) gives the reason as the cast lower legs
> of their non-DH specific forks are not designed to handle the
> "bigger leverage of the larger rotors". Not sure what this means,
> but it may have something to do with the strength of the mounts for
> the calipers, which are integral to the lower legs.
Whoa! The larger the disc, the smaller the forces, aka rim brakes
(also disk brakes o a sort).
>>> It seems like another misunderstanding of the bicycle industry.
>>> The larger the disk, the lower the forces. In fact when the
>>> contact diameter of the disk pads reach that of the rim (as in rim
>>> brakes) braking forces are minimized.
>> Notice that I didn't comment on the validity of the statement. I
>> only pointed out that the "warning" exists.
> What of the forces in the mounts, though? My understanding of it is
> that the moment about the axle of the caliper force on the rotor (say
> Fc) must equal the sum of the moments about the axle of the forces
> acting on the two mounts (F1 and F2). That is, Mc = M1 + M2. If a
> larger diameter rotor is used, Fc is less and its moment arm is longer,
> while keeping Mc the same. Although Mc = M1 + M2 still, the mounts are
> longer (ie the caliper is further away from the axle) and this will
> result in a greater bending moment for each of the two mounts. What
> have I missed in this analysis?
Whoa! The larger the disc, the smaller the forces, aka rim brakes
(also disk brakes o a sort).
>> Of course the warning is based on a misunderstanding in the bicycle
>> industry. And by noting that the warning comes from "fork
>> manufacturers" the company supplying the brakes can happily
>> propagate the misunderstanding, and free itself from doing any
>> research on the matter.
> I think this is only fair, since the brake manufacturers can
> justifiably put the onus on the fork manufacturers to design forks
> which are able to use their products. If the brake calipers need to
> be shifted to the front of the forks to increase their safety, or
> any other geometrical changes, it'll have to be at the initiative of
> the fork manufacturers to re-design their forks. Disc brakes can be
> used in bicycles, and fork manufacturers must be the ones to provide
> suitable fork designs to be able to use these safely.
That doesn't get them off the hook in the least. When designing a
brake, the brake must be tested and its installation and care
reviewed. I would have no trouble attacking both brake and fork
manufacturers in court on this basis.
>>> What is going on here?
> Worth finding out, that's for sure.
The attraction of the disk brake is the same in bicycling as for motor
vehicles. The disk brake, with its highly concentrated pad contact
pressure is relatively immune to water and contamination. Most
readers here may not recall drum brakes on cars, but driving through a
large puddle often caused water intrusion into the drum rendering the
brake almost useless at the onset of the next brak application.
The same is more or less true on bicycles.
Tim McNamara
James Annan <still_th...@hotmail.com> wrote:
That was quite a while ago- I doubt he'd had the opportunity.
Jose Rizal
> Jose Rizal writes:
> > Well, Answer (Manitou forks) gives the reason as the cast lower legs
> > of their non-DH specific forks are not designed to handle the
> > "bigger leverage of the larger rotors". Not sure what this means,
> > but it may have something to do with the strength of the mounts for
> > the calipers, which are integral to the lower legs.
>
> Whoa! The larger the disc, the smaller the forces, aka rim brakes
> (also disk brakes o a sort).
>
> > What of the forces in the mounts, though? My understanding of it is
> > that the moment about the axle of the caliper force on the rotor (say
> > Fc) must equal the sum of the moments about the axle of the forces
> > acting on the two mounts (F1 and F2). That is, Mc = M1 + M2. If a
> > larger diameter rotor is used, Fc is less and its moment arm is longer,
> > while keeping Mc the same. Although Mc = M1 + M2 still, the mounts are
> > longer (ie the caliper is further away from the axle) and this will
> > result in a greater bending moment for each of the two mounts. What
> > have I missed in this analysis?
>
> Whoa! The larger the disc, the smaller the forces, aka rim brakes
> (also disk brakes o a sort).
Yes, but the moment about the axle is the same. Consider a free body
diagram, and assume that the caliper mounting posts are parallel to the
radial line from the axle to the center of the caliper contact area.
This does not affect the validity of the analysis since the mounting
posts can be looked upon as moment arms. Also assume that the caliper
center is exactly between the tips of the mounting posts.
Now in the case of a small rotor with diameter Ro, the force exerted by
the caliper Fo is taken up by the two mounting posts. Assume that each
post takes up exactly half the caliper force, or 0.5*Fo. The moment
about the base of each post will then be
1) 0.5 * Fo * Lo,
where Lo is the length of the mounting post.
Now consider the case of a larger rotor, say R1. Assume that the
mounting posts are lengthened by the same amount as the increase in
radius for the larger rotor, that is,
2) R1 - Ro = L1 - Lo
and the posts are still parallel to the axle-caliper radial line.
Since the moment about the axle needs to be the same as with the smaller
rotor, then
3) F1 * R1 = Fo * Ro.
and so
4) F1 = Fo * Ro/R1.
The moment about the base of each post will now be
5) Moment = 0.5 * F1 * L1 = 0.5 * Fo * Ro/R1 * L1
But L1 = (R1 - Ro + Lo) from 2) above, so the moment on each post is
6) Moment = 0.5 * Fo * Ro/R1 * (R1 - Ro + Lo).
Plugging in typical numbers show clearly that the larger the disc brake
rotor, the greater the moment exerted on the fork leg brake mounts.
Example: let Ro = 80mm (160mm rotor), Lo = 15mm, and R1 = 330mm (26" rim
diameter)
Plug into 1) and 6) above and it's clear that the moment on each
mounting post is about 6 times greater for the larger rotor than for the
smaller rotor.
Hence I believe the fork manufacturers' recommendation against using
larger rotors.
Unless someone shoots holes in the above analysis.