Missy Giove's QR pops open

[Marc Davison \]

Tim McNamara wrote:
> In article <[email protected]>, "Marc Davison \(london-mtb.com\)"
> <[email protected]> wrote:
>
>> Andy Armstrong wrote:
>>> Chris Phillipo wrote:
>>>>
>>>> Are you kidding me. Any breaking force on the wheel is trying to force the wheel out of
>>>> verticle dropouts.
>>>
>>> No, a braking force applied at the top of the rim tends to force the axle back in the dropouts
>>> rather than down. Unless I've missed your point altogether.
>>
>> indeed.
>>
>> I think the fact that disk brakes can indeed push the wheels out of the drop outs (and rim brakes
>> can't) is something that the most sceptical amongst us will agree can happen.
>
> Well, no, the most skeptical do not agree this can happen. They reject the idea from
> beginning to end.
>



true.

in which case I'll rate my self as middle ground skeptical

>> this however is for a wheel that is unsecured.
>>
>> what there is disagreement over is how this then translated into the real world with a
>> retained wheel.
>
> You are thinking that the act of closing the QR skewer changes the magnitude or directions of the
> forces involved?
>

No. I'm sure all the forces acting on the wheel are the same, with the difference that the wheel is
held in place, until such time as the skewer has some how (and there seems to be the real grey
area) been loosened enough for it to be ejected. You mention in some other posts about the QR
stretching in order for it to get over the drop out, not being much of an engineer, correct me if I
am wrong, but the act of the skewer being stretched is going to require some sort of force/energy
to happen, has anyone added this into the wheel ejection calculation? as the fact that the ejection
action has to possibly stretch the skewer as well as eject the wheel must lessen the ejection force
to an extent?

of course this is just guess work (from my point of view) but is no less guess work than talk of how
the skewer gets past the lawers lips with out the wheel coming loose enough for someone to notice.

>> The big issue here is now to do with whether a skewer can unscrew, without being detectable to
>> the rider and yet flex/or open up enough for the above brake ejecting the wheel action t ocause
>> the wheel to come out. This seems to be the biggest grey area in the whole theory and certainly
>> seems to be one that James has no real evidence of any worthwhile sort to backup. in probably all
>> of his ancedotal evidence there is not enough to say whether it is as james has said, or it's
>> user error in the form of badly positioned/loose skewers or whether the skewer has been flipped
>> open whilst in use.
>
> Except that the unscrewing effects of cyclic stresses on nuts and bolts are well-known mechanical
> events, so that's not anecdotal. The analysis of forces provided by Annan simply follows the laws
> of physics, so that too is not anecdotal. The forces are easily estimated because the geometry is
> simple and the analysis is consistent with known mechanical principles. This is not anecdotal
> either. Frankly this is excellent evidence- scientific rather than anecdotal- which has been
> considered and endorsed by several mechanical engineers so far, and which is officially being
> examined by one shock fork company (and I bet unofficially examined by every other shock fork
> company).
>
again I'm not denying that the cyclic stress thing is not a well known mechanical event. However
there seems to be no hard evidence to suggest that that is what is occuring here. James has found
something that seems to fit the bill and therefore associated it with his ejection theory to get
around the issue of how the qr can undo/open/get past the lawers lps etc.

the next stage is for someone to show that this is indeed what happens in the real world.

> So, we then have the various reports sent to Annan which are of course anecdotes- and thus
> anecdotal- but they merely serve to illustrate the analysis. The anecdotal evidence is not the
> basis of the analysis offered by Annan. At this point, however, virtually all of the arguments
> against Anna's model *are* anecdotal and basically boil down to "I can't believe it."
>

thats an over simplification and I thinks it's fair to say that most of the ancedotal evidence just
serves to cloud the issue whilst proper evidence/proof is not forthcoming.

I'm not saying I can't believe it, I am just saying that as yet there has been nothing shown that
confirms the idea that the skewers do indeed undo.

I would be very worried if everybody sat down and just agreed without question what James has
proposed. Hopefully a certain amount os skepticism will aid James and others into progressing this
issue to a point where it can be seen without a doubt what is happening as far as how the wheel is
being allowed to be ejected, is it plain bad luck/user error etc...

what is beyond doubt is that the QR system, is not fail safe, but then again it never has been...

> Now, one can ride a bike with rim brakes and an unsecured skewer in vertical dropouts- front or
> rear- and not have the wheel fall out when you hit the brakes. There are no forces of sufficient
> magnitude in that direction. I am sure I am not alone with having once been riding along and
> noticing a flopping skewer, stopping to tighten it, and riding on. Even with operator error, there
> is not necessarily a crash- unless I try to lift that wheel off the pavement for any reason
> (hopping a curb, clearing a pothole, etc.) at which point the results will not be pleasant.
>
> But if I do the same thing with a front skewer on a bike with disk brakes, then braking will force
> the axle out of the dropout. Since this is easily correctable- by placing the brake caliper in
> front of the fork leg- the design can be considered defective. This is much in the same light as
> most SUVs having a defective design with narrow stance, high center of gravity and short
> wheelbases leading to lateral instability and a high propensity for rolling over, sometimes in
> normal traffic maneuvers. Note that the rear wheel will *not* be forced out of the dropout because
> the brake caliper is in front of the seatstay.

I agree wholeheartedly with all of that.

all that remains to be proven is how the skewer comes undone in the first place...

--
Marc
-------------------------------------------
http://aurora.homedns.org

 

[Chris Snell]

<[email protected]> wrote in message news:[email protected]...
> Chris Snell writes:
>
> >>>> Until some real hard evidence is produced to show what happens
then
> >>>> I think this final argument will just go round in circles...
>
> >>>> I think that probably sums it up...
>
> >>> Yes, Marc, I think that is a dandy summary.
>
> >> ...except that is is hugely superficial, since you could say that at any point in any
> >> discussion. Save that phrase for future discussions. It'll come in handy. Of course you'll
> >> have to
define
> >> "real hard evidence first."
>
> > Let me make it clearer: I think Marc's ENTIRE POST serves as a
dandy
> > summary, the WHOLE BIT, not just the LAST THREE LINES.
>
> > You say the QR will loosen. I say a ridged one won't. The page James Annan links to says ridged
> > fasteners are effective at preventing loosening. You say they're not. I say ridged QR's deform
> > the fork. You say they don't. You say that I must not
have
> > observed what I observed.
>
> I don't know what you call rigid fasteners. All screws are elastic
to
> some degree and that elasticity enters into this problems when a QR nut loosens. The more rigid
> the bolt/skewer, the smaller the
rotation
> required to loosen the nut, there being less preload length for a given tightness. That is why
> strain bolts are used on vibrating machinery or Belleville washers.
>

Jobst, please read more carefully. RIDGED. As in "having ridges". The horizontal line formed by the
juncture of two sloping planes. Teeth. You call them crenelations. Not RIGID. Is anyone else
laughing at this?

> The difference is that I explain the mechanism that can cause loosening and that there is evidence
> that such loosening occurs,
both
> from people reporting it and the fact that lateral movement of the axle can be shown with normal
> closure tightness. In contrast, you claim it isn't so and your proof is that YOU haven't
> experienced it.
>

The boltscience page explains a mechanism that prevents such loosening. I am saying QR's have such a
mechanism. Crenelations. Ridges. Teeth.

> > What sort of discussion will resolve this? None I can think of.
>
> I think that if you don't take scrutiny of your testimony so
seriously
> you can see that the level of observation may be different in yours, mine and others assessment of
> what occurs and what does not.
>

<sigh> See what I mean? "You say I must not have observed what I have observed." Round and
round we go.

> I see a difference in approach here. Some participants explain a phenomenon factually with
> scientific reasons while those defending
the
> status quo use rude rejoinders and question their opponents loyalty
to
> the American flag, so to speak, in Sen. Joe McCarthy style. It's a classic conservative ploy and
> runs amazingly close to US Politics of the day.
>
> Jobst Brandt [email protected] Palo Alto CA

The difference in approach seems to be that some resort to ad hominem arguments. That would be you,
Jobst Brandt. Twice now.

- Chris.

 

[Jobst Brandt]

Mark Hickey writes:

> You missed what is (to me at least) the most puzzling aspect of this whole thing. How can the
> wheel get loose enough to be ejected without being obvious to the rider? On a bike with disc
> brakes, you'll be experiencing some world-class brake drag long before the skewer is open enough
> to be forced over even the wimpiest lawyer lips.

That is not the case. I visited my local bicycle shop, opened the QR on a disc brake bicycle and
demonstrated the downward movement of the axle when the brake was applied. The wheel did not
disengage because it stopped on the retention lips. The bicycle could be ridden around the shop with
no indication that the QR was open until braking. The motion this caused at the axle in the dropout
would not be noticeable on uneven terrain.

Other than that, I think skewer stretching has been misinterpreted. A change in length of skewer
(and axle) are the elastic compliance felt when closing a QR lever. This amounts to 0.010-0.015 inch
for normal closure. It is not an elongation of the skewer to the point of clearing retention lips.
The length change subject comes up regularly in wheel bearing adjustment because this axle
compression overloads bearings that were adjusted with no clearance before QR closure.

Jobst Brandt [email protected] Palo Alto CA

 

[Jobst Brandt]

Chris Snell writes:

>>> You say the QR will loosen. I say a ridged one won't. The page James Annan links to says ridged
>>> fasteners are effective at preventing loosening. You say they're not. I say ridged QR's deform
>>> the fork. You say they don't. You say that I must not have observed what I observed.

>> I don't know what you call rigid fasteners. All screws are elastic to some degree and that
>> elasticity enters into this problems when a QR nut loosens. The more rigid the bolt/skewer, the
>> smaller the rotation required to loosen the nut, there being less preload length for a given
>> tightness. That is why strain bolts are used on vibrating machinery or Belleville washers.

> Jobst, please read more carefully. RIDGED. As in "having ridges". The horizontal line formed by
> the juncture of two sloping planes. Teeth. You call them crenelations. Not RIGID. Is anyone else
> laughing at this?

I see you are easily amused. I took it as a misspelling. A "ridged" fastener is not a common
mechanical term. Apparently you mean a knurled contact face. From Webster's dictionary: "'knurl' one
of a series of small ridges or beads on a metal surface to aid in gripping."

>> The difference is that I explain the mechanism that can cause loosening and that there is
>> evidence that such loosening occurs, both from people reporting it and the fact that lateral
>> movement of the axle can be shown with normal closure tightness. In contrast, you claim it isn't
>> so and your proof is that YOU haven't experienced it.

> The boltscience page explains a mechanism that prevents such loosening. I am saying QR's have such
> a mechanism. Crenelations. Ridges. Teeth.

They have no effect if they don't make an indentation in the mating part. In this case it's the
coefficient of friction and it remains unchanged by reducing contact area in this mechanism. Only
if features of one element mesh with surface of the other, does the resistance to sliding
increase. These features can be pre-formed, as on cassette retainer nuts, or be embossed by the
closure device. This is not the case on most dropouts or they would develop deep groves with
repeated closure.

>>> What sort of discussion will resolve this? None I can think of.

>> I think that if you don't take scrutiny of your testimony so seriously you can see that the
>> level of observation may be different in yours, mine and others assessment of what occurs and
>> what does not.

> <sigh> See what I mean? "You say I must not have observed what I have observed." Round and
> round we go.

What have you observed? Are there dents in the dropout that you can click over with a screwdriver or
fingernail?

>> I see a difference in approach here. Some participants explain a phenomenon factually with
>> scientific reasons while those defending the status quo use rude rejoinders and question their
>> opponents loyalty to the American flag, so to speak, in Sen. Joe McCarthy style. It's a classic
>> conservative ploy and runs amazingly close to US Politics of the day.

> The difference in approach seems to be that some resort to ad hominem arguments. That would be
> you, Jobst Brandt. Twice now.

My comment refers to the ridicule of other's findings in this matter, findings that are supported by
logic and science of materials. The mischaraterization of their comments is what I refer to as
McCarthy tactics. Contention arises from those who deny that there is a hazard and belittle those
who have taken the time to investigate the likelihood and cause of wheel separation on disc brake
equipped bicycles.

Jobst Brandt [email protected] Palo Alto CA

 

[Chris Snell]

<[email protected]> wrote in message news:[email protected]...
> Chris Snell writes:
>
> > Jobst, please read more carefully. RIDGED. As in "having ridges".
The
> > horizontal line formed by the juncture of two sloping planes.
Teeth.
> > You call them crenelations. Not RIGID. Is anyone else laughing at this?
>
> I see you are easily amused.

Evidently. I find it very funny that we cannot seem to make any progress in this because we can't
understand the words, let alone the ideas the words are trying to convey.

> I took it as a misspelling. A "ridged" fastener is not a common mechanical term. Apparently you
> mean a knurled contact face. From Webster's dictionary: "'knurl' one of a series of small ridges
> or beads on a metal surface to aid in gripping."

It wasn't a misspelling. Knurled is fine. To prevent further misunderstanding, I will, from now
on, not use the word ridges to describe these ridges, although Webster's seems to be quite
comfortable with it.

I was trying to find a better description of what my QR's have than crenelation, which perhaps we
can agree is "knurling only at the outer edge"? My QR's have knurling from the outer to inner edge
on the faces that contact the dropouts.

> >> The difference is that I explain the mechanism that can cause loosening and that there is
> >> evidence that such loosening occurs, both from people reporting it and the fact that lateral
> >> movement
of
> >> the axle can be shown with normal closure tightness. In
contrast,
> >> you claim it isn't so and your proof is that YOU haven't experienced it.
>
> > The boltscience page explains a mechanism that prevents such loosening. I am saying QR's have
> > such a mechanism. Crenelations. Ridges. Teeth.
>
> They have no effect if they don't make an indentation in the mating part. In this case it's
> the coefficient of friction and it remains unchanged by reducing contact area in this
> mechanism. Only if
features
> of one element mesh with surface of the other, does the resistance
to
> sliding increase. These features can be pre-formed, as on cassette retainer nuts, or be embossed
> by the closure device. This is not
the
> case on most dropouts or they would develop deep groves with
repeated
> closure.

As I have said, these indentations are exactly what I have observed. The QR knurling embosses the
dropout on the fork, they do develop groves with closure. Not so much that the lifespan of the fork
is reduced to something unacceptable, and I suspect this is at least partly because as the groves
tend to align the knurling with subsequent closure. (This wear is the 200 year lifespan limit I was
talking about in a previous post).

I have observed this on steel forks, magnesium forks, and aluminum dropouts on carbon forks.

I predict you will now again say that I have not observed what I have observed.

>
> >>> What sort of discussion will resolve this? None I can think of.
>
> >> I think that if you don't take scrutiny of your testimony so seriously you can see that the
> >> level of observation may be different in yours, mine and others assessment of what occurs and
> >> what does not.
>
> > <sigh> See what I mean? "You say I must not have observed what I
have
> > observed." Round and round we go.
>
> What have you observed? Are there dents in the dropout that you can click over with a screwdriver
> or fingernail?
>

There are indentations made by the knurled QR that I believe will prevent the QR from rotating
except under substantial force. Do the indentations seem weak? No. Does it seem to be wearing down
the forks? Not to a degree that I have any concerns the fork will not outlive me. This is especially
true for the steel forks. I suspect the knurling on the QR's reseats into and increases the
indentations already previously embossed by the QR into the steel.

> >> I see a difference in approach here. Some participants explain a phenomenon factually with
> >> scientific reasons while those
defending the
> >> status quo use rude rejoinders and question their opponents
loyalty to
> >> the American flag, so to speak, in Sen. Joe McCarthy style. It's
a
> >> classic conservative ploy and runs amazingly close to US Politics
of
> >> the day.
>
> > The difference in approach seems to be that some resort to ad hominem arguments. That would be
> > you, Jobst Brandt. Twice now.
>
> My comment refers to the ridicule of other's findings in this
matter,
> findings that are supported by logic and science of materials. The mischaraterization of their
> comments is what I refer to as McCarthy tactics. Contention arises from those who deny that
> there is a
hazard
> and belittle those who have taken the time to investigate the likelihood and cause of wheel
> separation on disc brake equipped bicycles.
>
> Jobst Brandt [email protected] Palo Alto CA

I don't think I have ridiculed other's findings. I have very little problem with James' analysis of
the forces on the hub / QR. However, I do not believe that the forces tending to unscrew a knurled
QR have been shown to be sufficient to cause said QR to actually unscrew. Where is the analysis of
this part of James theory? I believe the knurling is there for a reason. I believe the reason is to
prevent unscrewing. I believe this indicates someone has already thought about this.

Do we agree that the web page that James Annan links to says that a circumferential row of teeth
under the screw head is sufficient to prevent loosening?

Do we agree that knurled QR's have a circumferential row of teeth under the screw head?

So can we proceed from here?

- Chris.

 

[Mark Hickey]

[email protected] wrote:

>Mark Hickey writes:
>
>> You missed what is (to me at least) the most puzzling aspect of this whole thing. How can the
>> wheel get loose enough to be ejected without being obvious to the rider? On a bike with disc
>> brakes, you'll be experiencing some world-class brake drag long before the skewer is open enough
>> to be forced over even the wimpiest lawyer lips.
>
>That is not the case. I visited my local bicycle shop, opened the QR on a disc brake bicycle and
>demonstrated the downward movement of the axle when the brake was applied. The wheel did not
>disengage because it stopped on the retention lips. The bicycle could be ridden around the shop
>with no indication that the QR was open until braking. The motion this caused at the axle in the
>dropout would not be noticeable on uneven terrain.

Unless the QR was nearly loose enough to allow the wheel to be ejected under strong braking, I don't
think this proves much. From your description, I'm assuming you loosened the QR enough to get the
initial up/down movement of the axle under braking to happen. Between that point and the point where
the wheel can be forcefully ejected over the lawyer lips is a wide gulf in which the wheel will
easily be loose enough for serious disc brake drag to occur under normal riding conditions. Even if
the terrain is rough enough to make it hard to feel (or the rider is clueless), it's certainly going
to be plainly audible long before the front wheel exits the bike stage left.

>Other than that, I think skewer stretching has been misinterpreted. A change in length of skewer
>(and axle) are the elastic compliance felt when closing a QR lever. This amounts to 0.010-0.015
>inch for normal closure. It is not an elongation of the skewer to the point of clearing retention
>lips. The length change subject comes up regularly in wheel bearing adjustment because this axle
>compression overloads bearings that were adjusted with no clearance before QR closure.

I agree - no skewer is going to stretch enough to clear the lawyer lips under any circumstances not
involving a hydraulic jack!

Mark Hickey Habanero Cycles http://www.habcycles.com Home of the $695 ti frame

 

[Jobst Brandt]

Chris Snell writes:

> However, I do not believe that the forces tending to unscrew a knurled QR have been shown to be
> sufficient to cause said QR to actually unscrew. Where is the analysis of this part of James
> theory? I believe the knurling is there for a reason. I believe the reason is to prevent
> unscrewing. I believe this indicates someone has already thought about this.

I don't believe that all QR faces have knurling on them. My Campagnolo and Shimano QR's have smooth
faces and that leads me to believe others may also. That would be one point to investigate on the
wheel that separated and those of users who noted more looseness after use than before.

> Do we agree that the web page that James Annan links to says that a circumferential row of teeth
> under the screw head is sufficient to prevent loosening?

I'm sorry, I didn't follow that link so I don't know what it said and don't know where that might
be. Could you point me that way?

> Do we agree that knurled QR's have a circumferential row of teeth under the screw head?

They may, I don't have any at hand as I said. Mine have knurling around the lever head and nut but
none on the pressure face. The hubs I have were made for road bicycles where it would not be in the
interest of the rider to have knurling there. The knurling is on the jam nuts but even they do not
make a dent in my steel Campagnolo dropouts. I don't tighten them that tight and am not even sure I
could. I know that I once burst a Campagnolo QR head by too much preload during closure.

> So can we proceed from here?

Of course, but where are we headed? I'm just pursuing the concept that there is a problem and that
picking at nits surrounding the problem is not going to make it go away, especially when failure
mechanisms are dismissed without reasonable cause.

Jobst Brandt [email protected] Palo Alto CA

 

[Jobst Brandt]

Mark Hickey writes:

>>> You missed what is (to me at least) the most puzzling aspect of this whole thing. How can the
>>> wheel get loose enough to be ejected without being obvious to the rider? On a bike with disc
>>> brakes, you'll be experiencing some world-class brake drag long before the skewer is open enough
>>> to be forced over even the wimpiest lawyer lips.

>> That is not the case. I visited my local bicycle shop, opened the QR on a disc brake bicycle and
>> demonstrated the downward movement of the axle when the brake was applied. The wheel did not
>> disengage because it stopped on the retention lips. The bicycle could be ridden around the shop
>> with no indication that the QR was open until braking. The motion this caused at the axle in the
>> dropout would not be noticeable on uneven terrain.

> Unless the QR was nearly loose enough to allow the wheel to be ejected under strong braking, I
> don't think this proves much. From your description, I'm assuming you loosened the QR enough to
> get the initial up/down movement of the axle under braking to happen.

Not at all. I opened the QR fully from the customary closure tightness to demonstrate that there is
a wheel separation force. As I said, the wheel moved to the limit of the retention lips when the
brake was applied and did not separate. However. it is not far fetched to assume that were the
bicycle ridden in this condition for a while, the QR could become looser and the wheel could
separate, especially if thread friction in the adjusting nut were weak... which the plastic stop nut
variety often are.

> Between that point and the point where the wheel can be forcefully ejected over the lawyer lips is
> a wide gulf in which the wheel will easily be loose enough for serious disc brake drag to occur
> under normal riding conditions. Even if the terrain is rough enough to make it hard to feel (or
> the rider is clueless), it's certainly going to be plainly audible long before the front wheel
> exits the bike stage left.

I don't see where that drag should arise. The jam nuts are touching the dropouts on wither side of
the hub while the QR pressure faces are free and have clearance.

>> Other than that, I think skewer stretching has been misinterpreted. A change in length of skewer
>> (and axle) are the elastic compliance felt when closing a QR lever. This amounts to 0.010-0.015
>> inch for normal closure. It is not an elongation of the skewer to the point of clearing retention
>> lips. The length change subject comes up regularly in wheel bearing adjustment because this axle
>> compression overloads bearings that were adjusted with no clearance before QR closure.

> I agree - no skewer is going to stretch enough to clear the lawyer lips under any circumstances
> not involving a hydraulic jack!

Or failure in tension.

Jobst Brandt [email protected] Palo Alto CA

 

[Chris Phillipo]

In article <[email protected]>, [email protected] says...
> Chris Phillipo wrote:
> >
> > Are you kidding me. Any breaking force on the wheel is trying to force the wheel out of verticle
> > dropouts.
>
> No, a braking force applied at the top of the rim tends to force the axle back in the dropouts
> rather than down. Unless I've missed your point altogether.
>
>

It tends to force the wheel backwards with the road surface in the equation. Try it with no QR
skewer and see if your wheel stays on.
--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[Chris Phillipo]

> > what there is disagreement over is how this then translated into the real world with a retained
> > wheel.
>
> You are thinking that the act of closing the QR skewer changes the magnitude or directions of the
> forces involved?

You think that it doesn't? Tell me Tim, how exactly does the wheel pull down on the drop outs whent
here is no QR installed? Or are you now saying it does not do with with the QR installed? That would
completely invalidate your argument.
--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[Chris Phillipo]

In article <[email protected]>, [email protected] says...
> Tim McNamara thoughtfully penned:
> > In article <[email protected]>, "Marc Davison \(london-mtb.com\)"
> > <[email protected]> wrote:
> >
> >> Andy Armstrong wrote:
> >>> Chris Phillipo wrote:
>
>
> Amazing. You guys are still going on about this
>
> 1¢

A thread that doens't last a week is a failure.
--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[Chris Phillipo]

> Not at all. I opened the QR fully from the customary closure tightness to demonstrate that there
> is a wheel separation force. As I said, the wheel moved to the limit of the retention lips when
> the brake was applied and did not separate. However. it is not far fetched to assume that were the
> bicycle ridden in this condition for a while, the QR could become looser and the wheel could
> separate, especially if thread friction in the adjusting nut were weak... which the plastic stop
> nut variety often are.
>

As usual you are stating common sense, but it should be pointed out that this in no way validates
the argument that the lever comes loose in the first place as the result of disc brake design. I
assume you used your hand to open the lever and did not do so simply by braking.
--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[Chris Phillipo]

> > > If that were the case, then no nut or bolt would ever work loose. And yet they do, hence the
> > > market for things like Loctite and Nylock nuts,
> >
> > QR levers have Nylock nuts and serrated edges, are you starting to get the picture yet?
>
> Some do. Some don't.

Alright I'll throw you a bone on this one Tim, if you have a QR that has no nyloc insert and has
smooth faced fasteners, then yes I believe it not only can slip, it will slip and should be illegal
for sale. I certainly wouldn't sell one.
>
> > Tell me oh great one, why in your opinion does only the nut on the right move and not the lever
> > on the left side which is identical in every way that matters? Is your voodoo magic a liberal or
> > a conservative?
>
> The forces involved are conservative, of course. ;-) In fact, they're downright Newt-onian. The
> disk is only on one side of the wheel, being designed by obviously a bunch of Commie leftists.
> Don't even get me started on those pinkos at Cannondale with their Lefty shock fork, nor about
> Mike Burrows and his one-sided front *and* rear wheel mounts (at least those are balanced by
> having one on the right and one on the left). :-D
>
> The reason is simple: the axle us pushed towards the end of the dropout on the left side upon
> braking, which in turn puts a bending force on the right end of the skewer and axle. When the
> brake is released and possibly when you hit a bump, the axle is reseated at the top of the slot.
> Repeat every time you brake, hundreds of times perhaps in an hour. I imagine that a light
> feathering of the brakes has little effect, BTW. Over time, this movement (which I think is likely
> to be exacerbated by uneven fork leg compression) allows the retaining nut on the end of the
> skewer to back off, lowering tension on the skewer.

Well then if this is true, to solve this "problem" one only needs to install the QR so that the
lever is on the right had side and wedged against the fork so that it can not spin. If it did move
it would be blatantly obvious be the direction the lever is pointing.

--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[Chris Phillipo]

> >
> > All brakes produce the same sort of movement, the only difference is that in your flawed
> > experement, there is no road surface to force the wheel past the meager holding force of the
> > brake pads on the rim.
>
> Umm, no, actually all brakes do not produce the same *direction* of movement. Disk brakes, because
> of the location of the caliper relative to the dropout slot, produce a force that pushes the axle
> out of the dropout. Also, because of the relative sizes of disk brakes and the wheel overall, the
> ejection force produced by disk brakes is high.
>
> This of course has already been explained in detail in previous threads as well as on Annan's
> Website, and if you are still unclear on the subject you are welcome to re-read the material.
>

Well, if you are indeed worried about the difference that 10" of distance is making in the in the
overall leverage calculation, I suggest you have a 26" disc rotor made, peraps dual rotors just to
be safe. That should give you the effect you are looking for from your bike upside down with the QR
removed. I'm going to go ride mine, thanks.
--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[Sorni]

"Penny S." <[email protected]> wrote in message news:[email protected]...
>
> Amazing. You guys are still going on about this

I finally gave in to the "right-click/Mark Conversation As Read" option this morning. Knocked out
about 23 windy diatribes.

Bill "Fridays are meant for GETTING OUT AND RIDING" S.

 

[Jobst Brandt]

Chris Phillipo writes:

>> Not at all. I opened the QR fully from the customary closure tightness to demonstrate that there
>> is a wheel separation force. As I said, the wheel moved to the limit of the retention lips when
>> the brake was applied and did not separate. However. it is not far fetched to assume that were
>> the bicycle ridden in this condition for a while, the QR could become looser and the wheel could
>> separate, especially if thread friction in the adjusting nut were weak... which the plastic stop
>> nut variety often are.

> As usual you are stating common sense, but it should be pointed out that this in no way validates
> the argument that the lever comes loose in the first place as the result of disc brake design. I
> assume you used your hand to open the lever and did not do so simply by braking.

I didn't say that. Even when the lever remains closed, the nut end, or the lever end, can
incrementally rotate as the axle moves downward on braking and upward as the wheel encounters the
next bump and so forth. The mechanism for this rotation has been aired in this thread et length.
Retention lips on forks are not a press fit around either the axle or QR ends, they are there to
keep the wheel from dropping out if the QR was improperly closed. If ridden this way for any length
of time, the QR can gradually loosen far enough to allow it to disengage. That it won't occur is
more probable than that it will, but it can occur and therein lies the problem.

Jobst Brandt [email protected] Palo Alto CA

 

[Chris Phillipo]

In article <[email protected]>, [email protected] says...
> In article <[email protected]>,
> Chris Phillipo <[email protected]> wrote:
>
> > > > All I am saying is QRs have been coming undone and wheels falling out for years not because
> > > > of these postulated new forces but simply because people did them up wrongly.
> > >
> > > Erm, that's rather a non sequitor. Rim brakes do not place these forces on the QR. Neither do
> > > drum brakes, roller brakes nor coaster brakes. So of course the problems seen prior to disk
> > > brakes were not due to the forces Annan identifies.
> >
> > Are you kidding me. Any breaking force on the wheel is trying to force the wheel out of verticle
> > dropouts.
>
> Ummm, no, actually that is incorrect.
>
Ok to be more accurate, and force applied by the brakes is trying to shear off the drop outs, that
is true for both types. Because the fulcrum point is closer to the hub with the disc brake it does
make the movement without a skewer installed closer to perpendicular with the direction of (bike)
travel, but I still content that this makes no difference when you have the axle clamped securly to
the drop out. There are only two issues here that I can see, dropout failure and QR lever falure due
to mechanical flaw or improper use. Either one can happen with or without brakes.

If you want to see a wheel slip, try replacing your QR with solid axle and a nut/flat washer
arrangment.

--
_________________________
Chris Phillipo - Cape Breton, Nova Scotia http://www.ramsays-online.com

 

[A Muzi]

<[email protected]> wrote in message news:[email protected]...
> Chris Snell writes:
>
> >>>> Until some real hard evidence is produced to show what happens then I think this final
> >>>> argument will just go round in circles...
>
> >>>> I think that probably sums it up...
>
> >>> Yes, Marc, I think that is a dandy summary.
>
> >> ...except that is is hugely superficial, since you could say that at any point in any
> >> discussion. Save that phrase for future discussions. It'll come in handy. Of course you'll have
> >> to define "real hard evidence first."
>
> > Let me make it clearer: I think Marc's ENTIRE POST serves as a dandy summary, the WHOLE BIT, not
> > just the LAST THREE LINES.
>
> > You say the QR will loosen. I say a ridged one won't. The page James Annan links to says ridged
> > fasteners are effective at preventing loosening. You say they're not. I say ridged QR's deform
> > the fork. You say they don't. You say that I must not have observed what I observed.
>
> I don't know what you call rigid fasteners. All screws are elastic to some degree and that
> elasticity enters into this problems when a QR nut loosens. The more rigid the bolt/skewer, the
> smaller the rotation required to loosen the nut, there being less preload length for a given
> tightness. That is why strain bolts are used on vibrating machinery or Belleville washers.
>
> The difference is that I explain the mechanism that can cause loosening and that there is evidence
> that such loosening occurs, both from people reporting it and the fact that lateral movement of
> the axle can be shown with normal closure tightness. In contrast, you claim it isn't so and your
> proof is that YOU haven't experienced it.
>
> > What sort of discussion will resolve this? None I can think of.
>
> I think that if you don't take scrutiny of your testimony so seriously you can see that the
> level of observation may be different in yours, mine and others assessment of what occurs and
> what does not.
>
> I see a difference in approach here. Some participants explain a phenomenon factually with
> scientific reasons while those defending the status quo use rude rejoinders and question their
> opponents loyalty to the American flag, so to speak, in Sen. Joe McCarthy style. It's a classic
> conservative ploy and runs amazingly close to US Politics of the day.

I've tried to follow this thread and I think I understand the theoretical possibility of a skewer
nut unwinding from precession. I understand and agree that it is theoretically possible for a
fastener to "walk", especially on something like a suspension fork tip, given the elasticity of the
fastener and the flex in the fork and many kinds of forces ( Braking torque, impact, vibration,
flex). And it is a short step from a walking motion to a directional unscrewing motion as the nut
would effectively "ratchet", there being potentially enough force to unscrew but not nearly enough
to tighten That seems sensible. James Annan and others have clearly described this potential and
putative effect.

But is that effect significant? Can a rider allow that vibration/hysteresis/creep to continue for a
good long while without noticing a loud and annoying brake disc rubbing? Or hearing the
ting-ting-ting of a skewer lever flopping around? Has any rider continued like that until a five or
six millimeter gap has opened, enough to pass the raised lips at the bottom of a fork?

I can't imagine a mechanism which would further unscrew the skewer nut after it isn't pressed
against the fork tips any longer. Remember they are nylock. Once it swings free of any pressure
against the fork tip, you need another several millimeters to get past those lips. And besides not
understanding the other 90% of unscrewing, how can that progress unnoticed?? Unnoticed as the wheel
rocks half an inch side to side while listening to the brake disc rub?? I am less sure of all that.
Not saying it can't happen, just unsure up to here.

Will anyone ever find the first case of a gouged fork tip, intimating such a walking motion? Maybe .
We just haven't seen conclusive evidence that it has ever actually happened. I think most of us
pretty much allow that it is certainly possible.

But how the heck did McCarthy and the Flag get involved here? Current politics?? Whatever. I detest
Daschle and Hillary as much as anyone, but I didn't understand the analogy to skewers and discs. I
bet it's not germaine.

--
Andrew Muzi http://www.yellowjersey.org Open every day since 1 April 1971

 

[Tim McNamara]

In article <[email protected]>, Mark Hickey <[email protected]> wrote:

> Tim McNamara <[email protected]> wrote:
>
> > "Marc Davison \(london-mtb.com\)" <[email protected]> wrote:
>
> >> The big issue here is now to do with whether a skewer can unscrew, without being detectable to
> >> the rider and yet flex/or open up enough for the above brake ejecting the wheel action t ocause
> >> the wheel to come out. This seems to be the biggest grey area in the whole theory and certainly
> >> seems to be one that James has no real evidence of any worthwhile sort to backup. in probably
> >> all of his ancedotal evidence there is not enough to say whether it is as james has said, or
> >> it's user error in the form of badly positioned/loose skewers or whether the skewer has been
> >> flipped open whilst in use.
> >
> >Except that the unscrewing effects of cyclic stresses on nuts and bolts are well-known mechanical
> >events, so that's not anecdotal. The analysis of forces provided by Annan simply follows the laws
> >of physics, so that too is not anecdotal. The forces are easily estimated because the geometry is
> >simple and the analysis is consistent with known mechanical principles. This is not anecdotal
> >either. Frankly this is excellent evidence- scientific rather than anecdotal- which has been
> >considered and endorsed by several mechanical engineers so far, and which is officially being
> >examined by one shock fork company (and I bet unofficially examined by every other shock fork
> >company).
>
> You missed what is (to me at least) the most puzzling aspect of this whole thing. How can the
> wheel get loose enough to be ejected without being obvious to the rider? On a bike with disc
> brakes, you'll be experiencing some world-class brake drag long before the skewer is open enough
> to be forced over even the wimpiest lawyer lips.

A good question and one which I am able to answer only with a guess. My guess is that the QR doesn't
have to be loose, as in allowing free play of the axle in the dropouts which ought to be noticeable.
The QR just has to be detensioned enough to allow the ejection forces to be greater than the
clamping forces.

If we're talking about a 2 mm tall laywer lip on the outside of the dropout (similar to most that I
have seen) that probably doesn't take much more force than you can apply by hand to close the lever.
Only one side of the QR has to be forced out, which probably happens almost instantaneously when it
does happen.

 

[Tim McNamara]

In article <[email protected]>, [email protected] wrote:

> What have you observed? Are there dents in the dropout that you can click over with a screwdriver
> or fingernail?

I would add a further question to this for my own edification: are the impressions in the metal or
in the paint? The reason I as is that my road bikes all have impressions in the paint from the
knurling on the faces of the QR nuts and the locknuts on the hubs; they do not have impressions in
the metal, however.

And to Jobst: you mentioned the auguring of the locknuts into the horizontal dropouts due to flex.
Would unequal motion of the two legs of a suspension fork cause similar results?