Disc-brakes and QR's - An Open Letter from Pace Cycles

Discussion in 'Cycling Equipment' started by James Annan, Apr 16, 2003.

Thread Status:
Not open for further replies.
  1. James Annan

    James Annan Guest

    Some of you may have seen this on singletrackworld.com. I'll post my reply shortly. (For those who
    don't know, Pace are a UK MTB fork manufacturer.)

    James
    ---------------

    Disc-brakes and QR's An Open Letter from Pace Cycles

    The possibility of a quick release wheel disengaging from a dropout is an important issue and
    certainly requires careful scrutiny. This seems and obvious thing to say (and we say this as a fork
    manufacturer, not as a representative of other manufacturers). In this respect Pace feels that the
    responsibility rests equally between the parts manufacturers and the user. Manufacturers have to
    make sure that the design has integrity, fitness for purpose, and meets all acceptable standards,
    whilst the user must fit and use the product correctly, checking the product before each ride.

    I guess Mr Campagnolo laboured over these issues long and hard when he designed the QR, after all we
    all want a quick release and the convenience it offers, but without compromising safety. No
    cross-country rider that I'm aware of has come to Pace asking for a Slow Release. And when a new
    product is introduced into a proven assembly (disc-brakes) then manufacturers must re-examine the
    overall assembly. I guess in this respect the onus here lies with the disc brake manufacturer as any
    manufacturer designing a new product must make sure it is compatible (in every respect) with
    interfacing parts before confidently releasing onto the open market (the wide usage of dropouts
    coming before the wide acceptance of disc-brakes that is).

    Only a cynic would say that manufacturers would prefer to ignore issues such as these and not offer
    an opinion. The truth is I suspect that we would all prefer to stay low profile, as in these
    litigious days offering a view might make them liable ( and I would have to say this statement is
    made without prejudice) however Pace does not base its designs on unproven principles. I think that
    much of the debate and investigative work which has been done and referred to so far ( Bike Magic
    and other sites) has approached the matter correctly i.e. let mathematics and the laws of physics
    support design and application hypothesis. Subjective views can be dangerous.

    >From our past R&D one important influencing factor is brake calliper
    position and this should not be confused with ISO disc-brake mounting standards. Pace forks are
    designed to meet ISO standard but this does not prevent a brake manufacturer from designing a
    calliper so as to pick up on the ISO mounts but actually position the pads anywhere on the disc.
    Most certainly it is the case that the closer the pad is aligned to the dropout angle the lower the
    disengaging force whilst the more the pad is located perpendicular to the dropout angle the higher
    the disengaging force. Generally this means callipers tucked up against the slider leg are a
    positive in this respect, those hung out in the wind are not.

    Specifically then, from our analysis there is a reaction at the wheel spindle which has a
    disengaging force. If we take it that the dropout slot itself is conventional we can expect it to be
    approximately parallel to the
    c/l of the slider with varying degrees of offset designed into the fork crown. This accepted, unless
    the dropout has an unorthodox design where it is angled rearwards following a radius springing
    from the centre of the brake pad, this disengaging factor must be less than unity.

    I think it can be accepted that there is no forward moving reaction.

    To disengage the wheel spindle from the dropout the distance from wheel spindle c/l and brake-pad
    reaction c/l must increase whilst the force resisting this movement across the disc is the brake
    pad reaction.

    Since the disengaging component of this force is less than unity then in our view the spindle
    will not disengage through brake action alone. It follows then that although pad position
    relative to fork c/l does effect the disengaging force the force can never be greater than that
    exerted by the pad.

    This tells us that in itself a disc brake will not disengage a wheel OR which is correctly fitted
    and tightened (not withstanding points above).

    One very important addendum on the last point. Pace have found a few brands of OR that actually
    loosen slightly as lever finally closes. Make sure the OR used works correctly and can be tightened
    to the manufacturers spec' and this meets ISO standards.

    Another important footnote is that because of the dynamics created through 'fluttering' brakes
    strange effects can be created (as can poorly adjusted fork damper settings) such as a form of
    harmonic. Pace has not undertaken any analysis of this phenomenon.

    Personally speaking I have had the experience of a front wheel disengaging and luckily I came away
    with dented pride rather than a dented face. However over many years of mountain biking I have also
    experienced a disengaged rear wheel (OK laughs all round), however the calliper was positioned such
    that there would have been a positive force pushing the spindle back into the dropout. A positive
    engaging force so to speak. Obviously my fault not tightening the QR correctly- not weird science.
    If we as riders make mistakes such as this, with respect, we should hold our hand up- not try and
    place the blame elsewhere.

    In summary I guess Pace are saying there can be a variable depending upon dropout design, calliper
    design, OR design and correct fitting by the rider. But our view is that as long as all products are
    designed correctly and the rider fits the OR and torques it up correctly the wheel will not
    disengage.

    Adrian Carter Pace Cycles Limited
     
    Tags:


  2. James Annan

    James Annan Guest

    My reply, emailed to Pace and also posted on uk.rec.cycling, rec.bicycles.tech, singletrackworld.com

    >
    > Disc-brakes and QR's An Open Letter from Pace Cycles
    >

    [...]

    > Only a cynic would say that manufacturers would prefer to ignore issues such as these and not
    > offer an opinion. The truth is I suspect that we would all prefer to stay low profile, as in these
    > litigious days offering a view might make them liable ( and I would have to say this statement is
    > made without prejudice) however Pace does not base its designs on unproven principles.

    I do think you deserve some credit for being prepared to publically defend your products, even if
    your defence is faulty as you will see below. The Director of R+D at SRAM/Rockshox also emailed me
    after reading my web-page, but seems to take the view that since the lawyers have so far decided
    that all failures are due to operator error, he doesn't need to concern himself any further (he had
    no comment on the content of the page itself). I doubt that the courts will reach the same judgement
    in the future now that the theory and evidence for a design failure has been described.

    > I think that much of the debate and investigative work which has been done and referred to so far
    > ( Bike Magic and other sites) has approached the matter correctly i.e. let mathematics and the
    > laws of physics support design and application hypothesis. Subjective views can be dangerous.

    Indeed. So I recommend you have a look at the mathematics and the laws of physics described on my
    web page: http://www.ne.jp/asahi/julesandjames/home/disk_and_quick_release/

    In a nutshell, the axle slips slightly under the applied load (which exceeds the ISO standard for
    pull resistance), and as it slips, the QR tends to unscrew. Once it has unscrewed enough, the wheel
    can be ejected. On that web page, there are also many descriptions of failures observed in use
    quoted and/or linked, all of which match the theory accurately.

    If you think there are any significant errors in the analysis, I'd be interested in hearing about
    them. No-one else has found any yet(*).

    >
    > Specifically then, from our analysis there is a reaction at the wheel spindle which has a
    > disengaging force. If we take it that the dropout slot itself is conventional we can expect it to
    > be approximately parallel to the
    > c/l of the slider with varying degrees of offset designed into the fork crown. This accepted,
    > unless the dropout has an unorthodox design where it is angled rearwards following a radius
    > springing from the centre of the brake pad, this disengaging factor must be less than unity.
    >
    > I think it can be accepted that there is no forward moving reaction.
    >
    > To disengage the wheel spindle from the dropout the distance from wheel spindle c/l and brake-pad
    > reaction c/l must increase whilst the force resisting this movement across the disc is the brake
    > pad reaction.
    >
    > Since the disengaging component of this force is less than unity then in our view the spindle
    > will not disengage through brake action alone. It follows then that although pad position
    > relative to fork c/l does effect the disengaging force the force can never be greater than that
    > exerted by the pad.

    This is irrelevant. Why do you think this 'ratio' being less than unity has any bearing on the
    matter? You can't be seriously trying to deny that the wheel will slip if not securely restrained,
    this is too much part of everyday experience of vast numbers of riders who have learnt to do up the
    QR as tight as possible, even to the extent of having to stand on them.

    >
    > One very important addendum on the last point. Pace have found a few brands of OR that actually
    > loosen slightly as lever finally closes.

    These will be the QRs with the common over-centre cam action. Far from being a fault, most people
    agree that this is how they should be designed, as it means that they will can never open under
    vibration (NB by 'open', I mean the lever actually opening as in normal use, and it's not the same
    thing as unscrewing).

    > Make sure the OR used works correctly and can be tightened to the manufacturers spec' and this
    > meets ISO standards.

    If you are aware of any skewers that do not meet the ISO standard for pull resistance, it would be
    interesting to hear of them. The only tests I've seen used an apparently arbitrary set of 40-odd,
    and all passed. However even if feeble skewers are part of the problem, the standard is still
    demonstrably inadequate for this application, as the skewer is likely to experience a force
    significantly in excess of it (rather than having a significant safety margin, as we might wish for
    a component performing such a vital function).

    > Personally speaking I have had the experience of a front wheel disengaging and luckily I came away
    > with dented pride rather than a dented face.

    Those who do not learn from history are doomed to repeat it. You might not be so lucky next time.

    > our view is that as long as all products are designed correctly and the rider fits the OR and
    > torques it up correctly the wheel will not disengage.

    Change 'as long as' to 'if' and I'd have no quarrel with you. However, products (in particular the
    quick release wheel/fork attachment method, when a disk brake is mounted in the standard position)
    are currently NOT designed correctly, and there is nothing much the rider can do to prevent failure
    other than trust to luck and hope to spot a loose QR before the wheel detaches.

    I'll close with a couple of quotes from Chris Juden and Jobst Brandt. If you don't know who either
    of them are, then it might be worth your while finding out before you dismiss us all as cranks who
    do not understand what we are talking about.

    CJ:"I've been corresponding with James about this recently and although his tandem fork is an
    oddity, I'm convinced he's exposed a real problem here. That's backed up by all the reports
    coming into this site of ejected disc-braked wheels and loosened fasteners - some obviously
    rotated. MTB Review might be shy of the subject, but I'll be putting something in the next issue
    of CYCLE (CTC mag.), but first I'd like to run my analysis past y'all."

    JB:
    >> The more I see on this the more I find the defense of the status quo stranger than fiction. Why
    >> are writers trying to say that it can't happen? What motivates writers to claim that disc brakes
    >> as currently offered are not a hazard?

    >> The mechanism has been clearly stated, the forces have been identified in magnitude and
    >> direction, and credible descriptions of failures have been presented. What's going on here!
    >> There is no easter bunny. Believe it!

    James

    (*) Jobst Brandt doesn't like the way I've described the unscrewing as 'vibration loosening' when it
    is really the sliding up and down that probably dominates the effect. I think that his view has
    merit and will change the wording soon. Note that he does not dispute (indeed, clearly agrees with)
    the fact that the unscrewing does occur in normal use with a disk brake on standard forks.
     
  3. James Annan <[email protected]> wrote:
    >Some of you may have seen this on singletrackworld.com. I'll post my reply shortly. (For those
    >who don't know, Pace are a UK MTB fork manufacturer.) Disc-brakes and QR's An Open Letter from
    >Pace Cycles

    I don't think their analysis is very clear, to put it mildly. I think it's bogus, but I can't deduce
    exactly what they are saying.

    >One very important addendum on the last point. Pace have found a few brands of OR that actually
    >loosen slightly as lever finally closes.

    Surely all correctly designed QRs do so, as this prevents the lever from coming undone in
    normal use.
    --
    David Damerell <[email protected]> Distortion Field!
     
  4. Jose Rizal

    Jose Rizal Guest

    James Annan:

    > My reply, emailed to Pace and also posted on uk.rec.cycling, rec.bicycles.tech,
    > singletrackworld.com
    >
    > > I think that much of the debate and investigative work which has been done and referred to so
    > > far ( Bike Magic and other sites) has approached the matter correctly i.e. let mathematics and
    > > the laws of physics support design and application hypothesis. Subjective views can be
    > > dangerous.
    >
    > Indeed. So I recommend you have a look at the mathematics and the laws of physics described on my
    > web page: http://www.ne.jp/asahi/julesandjames/home/disk_and_quick_release/
    >
    > In a nutshell, the axle slips slightly under the applied load (which exceeds the ISO standard for
    > pull resistance), and as it slips, the QR tends to unscrew. Once it has unscrewed enough, the
    > wheel can be ejected. On that web page, there are also many descriptions of failures observed in
    > use quoted and/or linked, all of which match the theory accurately.
    >
    > If you think there are any significant errors in the analysis, I'd be interested in hearing about
    > them. No-one else has found any yet(*).
    >

    A contentious issue here is your statement that the QR tending to unscrew as it slips being the
    major contributor for QRs loosening in use. While I have no trouble believing that with enough slips
    up and down the dropout, the QR can loosen, it's unproven that this cyclic sliding in fact occurs in
    most cases where QRs have come loose, especially those with ribbed surfaces. Note that it should
    take several slips at least for a QR to loosen significantly; once is unlikely to do
    it. With dropout lips, QRs cannot loosen enough for the wheel to be ejected by that event alone, as
    the QR needs to disengage from the dropout far enough that it misses the lips and does not make
    contact with the dropout (and hence movement up and down the dropout no longer has any effect).

    Application of a disc brake when the QR is loose enough, however, may eject the wheel after bending
    or breaking the skewer, which can be accomplished easily enough. It's how the QR loosens that is
    still to be proved. I think it's important to get the mechanism for damage correct and accurate, in
    order to be able to be an informed buyer about any solutions manufacturers may put in place.
     
  5. Gary Young

    Gary Young Guest

    James Annan <[email protected]> wrote in message news:<[email protected]>...
    > My reply, emailed to Pace and also posted on uk.rec.cycling, rec.bicycles.tech,
    > singletrackworld.com
    > >
    > > Disc-brakes and QR's An Open Letter from Pace Cycles
    > >
    >
    > [...]
    >
    > > Only a cynic would say that manufacturers would prefer to ignore issues such as these and not
    > > offer an opinion. The truth is I suspect that we would all prefer to stay low profile, as in
    > > these litigious days offering a view might make them liable ( and I would have to say this
    > > statement is made without prejudice) however Pace does not base its designs on unproven
    > > principles.
    >
    > I do think you deserve some credit for being prepared to publically defend your products, even if
    > your defence is faulty as you will see below. The Director of R+D at SRAM/Rockshox also emailed me
    > after reading my web-page, but seems to take the view that since the lawyers have so far decided
    > that all failures are due to operator error, he doesn't need to concern himself any further (he
    > had no comment on the content of the page itself). I doubt that the courts will reach the same
    > judgement in the future now that the theory and evidence for a design failure has been described.
    <snip>

    I doubt you'll get a straight answer out of the fork manufacturers. This Pace guy is tying himself
    in knots worrying about the liability problems. I wonder if this problem could lead to a recall?
    That might be a bigger financial hit than any lawsuit. Have you sent anything to the Consumer
    Product Safety Commission or the UK equivalent? If a recall were ordered, is there some stop-gap
    solution that could make existing forks relatively safe at a reasonable cost?

    I'm not an expert in products liability law (I'm a legal reporter), but I wouldn't take the Rockshox
    reply seriously. For the most part, judges and juries make rulings, not lawyers. Rockshox's lawyers
    may have reviewed the caselaw and come to the conclusion that courts have attributed similar
    accidents to human error. But those court decisions, if they exist, were made before you had done
    your research. Besides, I can't imagine any competent lawyer issuing an opinion on a product's
    safety. Just picture the courtroom scene:

    Plaintiff's lawyer: "Despite James Annan's warnings, you continued to market this fork, the very
    fork that put my client in a wheelchair?

    Rockshox: "We'll, we ran it past the lawyers and they said it was safe...."

    Plaintiff's lawyer: "The lawyers!? The lawyers!? Did it occur to you to hire an engineer?!"

    Judge: "Members of the jury, need I remind you that laughter will not be tolerated during these
    proceedings?"

    My point is that you may have rendered us a service even if you get no acknowledgment from the fork
    manufacturers. If you really want to be sure they address this problem, though, I would suggest
    taking it a step further -- write letters to all fork manufacturers (and perhaps hub and brake
    manufacturers as well?) and to the CPSC. The more publicity this gets, the less likely it is that
    they can squirm their way out of doing the right thing.
     
  6. James Annan

    James Annan Guest

    Gary Young wrote:

    > I doubt you'll get a straight answer out of the fork manufacturers. This Pace guy is tying himself
    > in knots worrying about the liability problems. I wonder if this problem could lead to a recall?
    > That might be a bigger financial hit than any lawsuit. Have you sent anything to the Consumer
    > Product Safety Commission or the UK equivalent?

    I haven't yet, I don't know who the UK equivalent is and it would probably sound better coming from
    someone with more credibility rather than a mere user with a chip on his shoulder and a website that
    'proves' he is right. After all, what would they do but talk to 'experts' who will assure them that
    the QR is safe. However rest assured that the problem is not simply going to be ignored, some people
    with better contacts than me are taking the problem seriously.

    > If a recall were ordered, is there some stop-gap solution that could make existing forks
    > relatively safe at a reasonable cost?

    Not sure. I was thinking along the lines of either some clamp-on device that fixes onto the
    dropouts and physically restrains the axle, or else a QR with a proper threadlock. I can't see
    either being popular with users, but of course anyone who still prefers to not believe me can
    always not use such a device.

    > I'm not an expert in products liability law (I'm a legal reporter), but I wouldn't take the
    > Rockshox reply seriously.

    I don't. I was naive enough to expect that the Director of R+D would be capable of understanding the
    issues but it seems like he is a manager with no technical understanding. He offered a few
    platitudes on the importance of attaching the wheel correctly, and absolutely no comment whatsoever
    on the content of the website. I only hope he showed it to someone who has a clue, although judging
    from Pace's letter these may be in short supply in the bicycle industry.

    James
     
  7. Jobst Brandt

    Jobst Brandt Guest

    Jose Rizal writes:

    > A contentious issue here is your statement that the QR tending to unscrew as it slips being the
    > major contributor for QRs loosening in use. While I have no trouble believing that with enough
    > slips up and down the dropout, the QR can loosen, it's unproven that this cyclic sliding in fact
    > occurs in most cases where QRs have come loose, especially those with ribbed surfaces. Note that
    > it should take several slips at least for a QR to loosen significantly; once is unlikely to do it.
    > With dropout lips, QRs cannot loosen enough for the wheel to be ejected by that event alone, as
    > the QR needs to disengage from the dropout far enough that it misses the lips and does not make
    > contact with the dropout (and hence movement up and down the dropout no longer has any effect).

    > Application of a disc brake when the QR is loose enough, however, may eject the wheel after
    > bending or breaking the skewer, which can be accomplished easily enough. It's how the QR loosens
    > that is still to be proved. I think it's important to get the mechanism for damage correct and
    > accurate, in order to be able to be an informed buyer about any solutions manufacturers may put
    > in place.

    That's mighty wordy. Can you cut that down to the main point. The loosening effect has been clearly
    explained and verified on bicycles with disc brakes. What is your point and whom are you defending?

    Jobst Brandt [email protected] Palo Alto CA
     
  8. x

    x Guest

    RE/
    >Not sure. I was thinking along the lines of either some clamp-on device that fixes onto the
    >dropouts and physically restrains the axle, or else a QR with a proper threadlock. I can't see
    >either being popular with users, b

    My old StumpJumper has, in addition to bolts instead of QR, metal plates that are attached to the
    fork via retaining screws that screw into the luggage rack holes. To remove the wheel, one has to
    take an allen key to those retaining screws.

    Actually, it's not as bad as it sounds and, if the wheel had a skewer I think I could learn to love
    it because I don't get that many front-tire flats.

    But, overall, I'm having trouble visualizing why the current (standard?) skewer setup is such a
    problem. The lawyer lips on both of my forks are such that the skewers have to take 3-5 complete
    additional turns before I can yank the wheel out.
    -----------------------
    PeteCresswell
     
  9. Dion Dock

    Dion Dock Guest

    Is there a problem with the rear wheel falling out? It typically has vertical dropouts and no
    lawyer tabs.

    -Dion
     
  10. Peter

    Peter Guest

    Dion Dock wrote:
    > Is there a problem with the rear wheel falling out? It typically has vertical dropouts and no
    > lawyer tabs.

    But the disc brake calipers are positioned forward of the rear axle. Therefore when the brake is
    applied the calipers apply an upward force on the rotor and this would tend to drive the axle upward
    in the vertical dropouts.
     
  11. David Kunz

    David Kunz Guest

    James Annan wrote:
    > Gary Young wrote:
    ...

    > Not sure. I was thinking along the lines of either some clamp-on device that fixes onto the
    > dropouts and physically restrains the axle, or else a QR with a proper threadlock. I can't see
    > either being popular with users, but of course anyone who still prefers to not believe me can
    > always not use such a device.

    The Salsa QR has a nylon sleeve that acts as a lock-nut (like the nylock nuts that you can buy in a
    hardware store). Only slightly annoying -- not a real problem.

    David
     
  12. James Annan

    James Annan Guest

    David Kunz wrote:

    >
    > The Salsa QR has a nylon sleeve that acts as a lock-nut (like the nylock nuts that you can buy in
    > a hardware store). Only slightly annoying -- not a real problem.

    Unfortunately it also has a completely smooth face at the lever end.

    James
     
  13. x

    x Guest

    RE/
    >But the disc brake calipers are positioned forward of the rear axle. Therefore when the brake is
    >applied the calipers apply an upward force on the rotor and this would tend to drive the axle
    >upward in the vertical dropouts.

    Which begs the question which has been bugging me since day one of this thread's
    precursor....howcome the front calipers aren't positioned forward of the forks?
    -----------------------
    PeteCresswell
     
  14. Nc

    Nc Guest

    "James Annan" <[email protected]> wrote in message news:[email protected]...
    > Gary Young wrote:
    > > If a recall were ordered, is there some stop-gap solution that could make existing forks
    > > relatively safe at a reasonable cost?
    >
    > Not sure. I was thinking along the lines of either some clamp-on device that fixes onto the
    > dropouts and physically restrains the axle, or else a QR with a proper threadlock. I can't see
    > either being popular with users, but of course anyone who still prefers to not believe me can
    > always not use such a device.

    I've followed James' research with interest. Though I don't have discs on my bikes, I'm interested
    in engineering and potential failure modes.

    Thinking about the comment above - James states the requirement is for a QR with a threadlock to
    prevent the unscrewing failure mode - the answer might be some form of split pin and castle nut on
    the nut end of the QR. This would not require a massive change in design of skewers, nor does it
    involve unusual parts manufacture.

    More details: A castle nut is a normal nut with castelations around one edge. When a split put is
    put through a cross-drilled hole in the thread, it engages in a castelation, thus preventing
    rotation. ( One use of such devices is to prevent wheel hubs falling off cars ). To be practical in
    the field on a bike, where a bike wheel may need to be removed for punctures, or for transport, the
    split pin needs to be a sprung re-usable item (fine, such types exist), and ideally held with a
    short tether wire to the QR skewer or nut, or some other convinient fixture (so its not dropped in
    the grass when fixing a puncture).

    Lockwire would also work (as used on many aircraft bolts), though one would have to carry a bit of
    spare wire in a pocket for use each time a wheel was changed.

    Anyway, there it is, my 2p contribution.

    NC
     
  15. Ted Bennett

    Ted Bennett Guest

    It improves the steering a little. Just as with low-rider racks, which position panniers closer to
    the steering axis, it's better to place weight as close as possible to the axis of motion.

    Whether or not that is really significant, I don't know. But it is done for the same reason in
    motorcycles.

    > >But the disc brake calipers are positioned forward of the rear axle. Therefore when the brake is
    > >applied the calipers apply an upward force on the rotor and this would tend to drive the axle
    > >upward in the vertical dropouts.
    >
    > Which begs the question which has been bugging me since day one of this thread's
    > precursor....howcome the front calipers aren't positioned forward of the forks?
    > -----------------------
    > PeteCresswell

    --
    Ted Bennett Portland OR
     
  16. Spider

    Spider Guest

    James Annan <[email protected]m> wrote in message news:<[email protected]>...
    > David Kunz wrote:
    >
    > >
    > > The Salsa QR has a nylon sleeve that acts as a lock-nut (like the nylock nuts that you can buy
    > > in a hardware store). Only slightly annoying -- not a real problem.
    >
    > Unfortunately it also has a completely smooth face at the lever end.

    Does that make a difference? The lever end is usually opposite the caliper, and the caliper/disk
    side is where the offset twisting forces are located. Isn't the non-caliper side just "seeing" axial
    pull forces?

    (Feel free to guffaw if I've totally mucked it up.)

    Spider
     
  17. James Annan

    James Annan Guest

    Spider wrote:
    > James Annan <[email protected]> wrote in message
    > news:<[email protected]>...
    >

    >>
    >>Unfortunately it also has a completely smooth face at the lever end.
    >
    >
    > Does that make a difference?

    Not to the unscrewing, but it does mean that it will slip until it wedges against the retention lip.
    Which is not exactly a perfect state of affairs.

    FWIW, I was using a salsa QR when my front wheel pulled out, but that was without the benefit of a
    retention lip. I'd still prefer the system to be designed such that the QR does not have to handle
    such a massive load.

    I like Jobst Brandt's idea with the conical interlocking. But it seems like it is half-way to the
    bolt-through idea that already exists, and I wonder if the manufacturers will (eventually, when the
    message gets into their thick heads) prefer to move over wholesale to that system, with its much
    touted benefits of greater stiffness. They aren't really much hassle to remove even now, and could
    probably be improved further. I saw one with some sort of hinged clamp over the end of the dropout,
    which looked like it just flipped open with no need for tools. But the cone idea does have the
    obvious advantage of allowing the manufacturers to salvage both their existing stock and, perhaps
    more importantly, to cheaply upgrade the huge number of dangerous forks already in circulation.

    James
     
  18. Nc

    Nc Guest

    "James Annan" <[email protected]> wrote in message news:[email protected]...
    > Gary Young wrote:
    > > If a recall were ordered, is there some stop-gap solution that could make existing forks
    > > relatively safe at a reasonable cost?
    >
    > Not sure. I was thinking along the lines of either some clamp-on device that fixes onto the
    > dropouts and physically restrains the axle, or else a QR with a proper threadlock. I can't see
    > either being popular with users, but of course anyone who still prefers to not believe me can
    > always not use such a device.

    I've followed James' research with interest. Though I don't have discs on my bikes, I'm interested
    in engineering and potential failure modes.

    Thinking about the comment above - James states the requirement is for a QR with a threadlock to
    prevent the unscrewing failure mode - the answer might be some form of split pin and castle nut on
    the nut end of the QR. This would not require a massive change in design of skewers, nor does it
    involve unusual parts manufacture.

    More details: A castle nut is a normal nut with castelations around one edge. When a split put is
    put through a cross-drilled hole in the thread, it engages in a castelation, thus preventing
    rotation. ( One use of such devices is to prevent wheel hubs falling off cars ). To be practical in
    the field on a bike, where a bike wheel may need to be removed for punctures, or for transport, the
    split pin needs to be a sprung re-usable item (fine, such types exist), and ideally held with a
    short tether wire to the QR skewer or nut, or some other convinient fixture (so its not dropped in
    the grass when fixing a puncture).

    Lockwire would also work (as used on many aircraft bolts), though one would have to carry a bit of
    spare wire in a pocket for use each time a wheel was changed.

    Anyway, there it is, my 2p contribution.

    NC
     
Loading...
Thread Status:
Not open for further replies.
Loading...