Cycling Equipment · Public discussion

Interesting Hub Flange Failure

Started by Arthur Shapiro · · Last activity · 36 posts · 449 views

Thread navigation

Jump through the discussion

Go to the original post, the replies on this page, or the latest preserved contribution.

Thread details

What we know about this thread

Original section
Cycling Equipment
Published
23 November 2007
Last activity
27 November 2007
Original author
Arthur Shapiro
Posts
36
Discussion status
Public discussion
Total views
449
Views / 30 days
0

The navigation and discussion metadata provide context. Posts remain in their original chronological order.

Showing posts 1–20 of 36
Posts remain in their original chronological order.

Text size
  1. I was riding with a fellow club member on a Thursday club ride, and heard some
    light pinging noises coming from the rear of his bicycle. They got a little
    bit worse as we were on one climb. It sounded as if his rear derailleur might
    be just barely touching one spoke, but visually that didn't appear to be the
    case.

    The noise got louder and more frequent as we started up a fairly healthy
    climb. Then suddenly something broke. It appeared to be one of the spokes on
    his Bontrager rear wheel.

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    and the subsequent "Next Picture" link to picture #02. (Others in the set are
    of no interest beyond the club. The pictures will be up for one week.)

    Is there enough visual evidence to speculate on the cause / nature of this
    flange failure?

    Art

  2. Arthur Shapiro said:

    I was riding with a fellow club member on a Thursday club ride, and heard some
    light pinging noises coming from the rear of his bicycle. They got a little
    bit worse as we were on one climb. It sounded as if his rear derailleur might
    be just barely touching one spoke, but visually that didn't appear to be the
    case.

    The noise got louder and more frequent as we started up a fairly healthy
    climb. Then suddenly something broke. It appeared to be one of the spokes on
    his Bontrager rear wheel.

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    and the subsequent "Next Picture" link to picture #02. (Others in the set are
    of no interest beyond the club. The pictures will be up for one week.)

    Is there enough visual evidence to speculate on the cause / nature of this
    flange failure?

    Art


    bad design: on a normal hub the aluminium flange survives because each
    leading and trailing spokepair fight each other. The aluminium in
    between is under compression and the hub can cope with the residual
    stress in tension. Here two spokes work together in tearing the hub apart

    --
    /Marten

    info(apestaartje)m-gineering(punt)nl

  3. In article <[email hidden]>,

    (Arthur Shapiro) said:

    I was riding with a fellow club member on a Thursday club ride, and
    heard some light pinging noises coming from the rear of his bicycle.
    They got a little bit worse as we were on one climb. It sounded as
    if his rear derailleur might be just barely touching one spoke, but
    visually that didn't appear to be the case.

    The noise got louder and more frequent as we started up a fairly
    healthy climb. Then suddenly something broke. It appeared to be one
    of the spokes on his Bontrager rear wheel.

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    and the subsequent "Next Picture" link to picture #02. (Others in
    the set are of no interest beyond the club. The pictures will be up
    for one week.)

    Is there enough visual evidence to speculate on the cause / nature of
    this flange failure?

    My guess is just too much material removed from the flange. Note that
    the spoke heads sit not in individual spoke holes but in slots.

  4. On Fri, 23 Nov 2007 15:46:18 GMT, [email hidden] (Arthur

    Shapiro) may have said:

    I was riding with a fellow club member on a Thursday club ride, and heard some
    light pinging noises coming from the rear of his bicycle. They got a little
    bit worse as we were on one climb. It sounded as if his rear derailleur might
    be just barely touching one spoke, but visually that didn't appear to be the
    case.

    The noise got louder and more frequent as we started up a fairly healthy
    climb. Then suddenly something broke. It appeared to be one of the spokes on
    his Bontrager rear wheel.

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    and the subsequent "Next Picture" link to picture #02. (Others in the set are
    of no interest beyond the club. The pictures will be up for one week.)

    Is there enough visual evidence to speculate on the cause / nature of this
    flange failure?

    I would add to the observations above that the low-spoke-count wheels
    have loading delta during rotation, and may be operating at a higher
    intial tension. If the Bontrager hubs are machined from billets
    instead of forged, that might be a contributing factor as well. It is
    often the case that the closer you get to the bleeding edge of
    technology, the more fragile the tech gets.

    --
    My email address is antispammed; pull WEEDS if replying via e-mail.
    Typoes are not a bug, they're a feature.
    Words processed in a facility that contains nuts.

  5. Werehatrack said:


    I would add to the observations above that the low-spoke-count wheels
    have loading delta during rotation,

    make that "higher loading delta during rotation relative to that
    typical of a high-spoke-count wheel."

    --
    My email address is antispammed; pull WEEDS if replying via e-mail.
    Typoes are not a bug, they're a feature.
    Words processed in a facility that contains nuts.

  6. In article <[email hidden]>,

    (Arthur Shapiro) said:

    I was riding with a fellow club member on a Thursday club ride, and heard some
    light pinging noises coming from the rear of his bicycle. They got a little
    bit worse as we were on one climb. It sounded as if his rear derailleur might
    be just barely touching one spoke, but visually that didn't appear to be the
    case.

    The noise got louder and more frequent as we started up a fairly healthy
    climb. Then suddenly something broke. It appeared to be one of the spokes on
    his Bontrager rear wheel.

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    It literally exploded!

    Quoted message said:

    and the subsequent "Next Picture" link to picture #02. (Others in the set are
    of no interest beyond the club. The pictures will be up for one week.)

    Is there enough visual evidence to speculate on the cause / nature of this
    flange failure?

    --
    Michael Press

  7. Werehatrack said:
    Werehatrack said:

    I would add to the observations above that the low-spoke-count wheels
    have loading delta during rotation,

    make that "higher loading delta during rotation relative to that
    typical of a high-spoke-count wheel."

    --
    My email address is antispammed; pull WEEDS if replying via e-mail.
    Typoes are not a bug, they're a feature.
    Words processed in a facility that contains nuts.

    I do not understand. Please explain loading delta.

    B.D

    http://cozybeehvie.blogspot.com

  8. (Arthur Shapiro) said:

    I was riding with a fellow club member on a Thursday club ride, and heard some
    light pinging noises coming from the rear of his bicycle. They got a little
    bit worse as we were on one climb. It sounded as if his rear derailleur might
    be just barely touching one spoke, but visually that didn't appear to be the
    case.

    The noise got louder and more frequent as we started up a fairly healthy
    climb. Then suddenly something broke. It appeared to be one of the spokes on
    his Bontrager rear wheel.

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    and the subsequent "Next Picture" link to picture #02. (Others in the set are
    of no interest beyond the club. The pictures will be up for one week.)

    Is there enough visual evidence to speculate on the cause / nature of this
    flange failure?

    Art

    This will surely make it to my blog.

    B.D

    http://cozybeehive.blogspot.com

  9. bicycle_disciple said:
    Werehatrack said:
    Werehatrack said:

    I would add to the observations above that the low-spoke-count wheels
    have loading delta during rotation,

    make that "higher loading delta during rotation relative to that
    typical of a high-spoke-count wheel."

    --
    My email address is antispammed; pull WEEDS if replying via e-mail.
    Typoes are not a bug, they're a feature.
    Words processed in a facility that contains nuts.

    I do not understand. Please explain loading delta.

    B.D

    http://cozybeehvie.blogspot.com

    Dear B.D.,

    The greek delta letter is often used to indicate change in equations.

    On a low spoke count wheel, the change in load may be greater.

    That is, on a 36-spoke wheel, the bottom 5 spokes lose considerable
    tension and then regain it. The total tension loss (or change or
    delta) is spread over 5 spokes.

    On an 18-spoke wheel, the same tension loss is spread over only 3
    spokes, so the total tension loss _may_ be greater.

    Unfortunately, we have only FEA's for 36-spoke wheels and don't know
    what fewer spokes do. Here's an FEA for 36-spokes:

    http://www.astounding.org.uk/ian/wheel/index.html

    Ian's FEA suggests that the 5 spokes lose tension at roughly these
    relative rates (67~69, 244~250, ~350):

    / / | \ \ 36-spoke wheel
    -7 -25 -35 -25 -7 total ~99

    But what happens to an 18-spoke wheel with the same load?

    It could produce a much greater change in the middle spoke:

    / | \ 18 spokes?
    -7 -85 -7 total ~99 much greater change

    Or the same change:

    / | \ 18 spokes?
    -32 -35 -32 total ~99 same change

    Or even slight less change:

    / | \ 18 spokes?
    -33 -33 -33 total ~99 slightly less change

    The assumption is that the correct model tends toward the greater
    change example, meaning that a low spoke-count wheel's spokes see a
    greater delta or change in tension, from o to say 85 rather than just
    from 0 to 35.

    The greater the range of loading delta (spoke tension change), the
    sooner the flange will fatigue and fail.

    Cheers,

    Carl Fogel

  10. Quoted message said:

    On Fri, 23 Nov 2007 15:40:53 -0800 (PST), bicycle_disciple

    Quoted message said:
    Werehatrack said:

    On Fri, 23 Nov 2007 12:46:30 -0600, Werehatrack
    <[email hidden]> said:

    Quoted message said:
    Quoted message said:

    >I would add to the observations above that the low-spoke-count wheels
    >have loading delta during rotation,

    Quoted message said:
    Quoted message said:

    make that "higher loading delta during rotation relative to that
    typical of a high-spoke-count wheel."

    Quoted message said:
    Quoted message said:

    --
    My email address is antispammed; pull WEEDS if replying via e-mail.
    Typoes are not a bug, they're a feature.
    Words processed in a facility that contains nuts.

    Quoted message said:

    I do not understand. Please explain loading delta.

    Quoted message said:

    B.D

    Quoted message said:

    http://cozybeehvie.blogspot.com

    Dear B.D.,

    The greek delta letter is often used to indicate change in equations.

    On a low spoke count wheel, the change in load may be greater.

    That is, on a 36-spoke wheel, the bottom 5 spokes lose considerable
    tension and then regain it. The total tension loss (or change or
    delta) is spread over 5 spokes.

    On an 18-spoke wheel, the same tension loss is spread over only 3
    spokes, so the total tension loss _may_ be greater.

    Unfortunately, we have only FEA's for 36-spoke wheels and don't know
    what fewer spokes do. Here's an FEA for 36-spokes:

    http://www.astounding.org.uk/ian/wheel/index.html

    Ian's FEA suggests that the 5 spokes lose tension at roughly these
    relative rates (67~69, 244~250, ~350):

    / / | \ \ 36-spoke wheel
    -7 -25 -35 -25 -7 total ~99

    But what happens to an 18-spoke wheel with the same load?

    It could produce a much greater change in the middle spoke:

    / | \ 18 spokes?
    -7 -85 -7 total ~99 much greater change

    Or the same change:

    / | \ 18 spokes?
    -32 -35 -32 total ~99 same change

    Or even slight less change:

    / | \ 18 spokes?
    -33 -33 -33 total ~99 slightly less change

    The assumption is that the correct model tends toward the greater
    change example, meaning that a low spoke-count wheel's spokes see a
    greater delta or change in tension, from o to say 85 rather than just
    from 0 to 35.

    The greater the range of loading delta (spoke tension change), the
    sooner the flange will fatigue and fail.

    Cheers,

    Carl Fogel- Hide quoted text -

    - Show quoted text -

    Thanks Carl, this weekend I'll probably go over the details in that
    link. So delta just pertains to change in loading, and with lower
    spokes, there's a higher change? And in the life of a low spoked
    wheel, the greater loading per each cycling reduces its time to
    failure at some spot? Correct?

  11. bicycle_disciple said:
    Quoted message said:

    On Fri, 23 Nov 2007 15:40:53 -0800 (PST), bicycle_disciple

    Quoted message said:

    On Nov 23, 2:48 pm, Werehatrack <[email hidden]> wrote:
    > On Fri, 23 Nov 2007 12:46:30 -0600, Werehatrack
    > <[email hidden]> said:

    Quoted message said:

    > >I would add to the observations above that the low-spoke-count wheels
    > >have loading delta during rotation,

    Quoted message said:

    > make that "higher loading delta during rotation relative to that
    > typical of a high-spoke-count wheel."

    Quoted message said:

    > --
    > My email address is antispammed; pull WEEDS if replying via e-mail.
    > Typoes are not a bug, they're a feature.
    > Words processed in a facility that contains nuts.

    Quoted message said:

    I do not understand. Please explain loading delta.

    Quoted message said:

    B.D

    Quoted message said:

    http://cozybeehvie.blogspot.com

    Dear B.D.,

    The greek delta letter is often used to indicate change in equations.

    On a low spoke count wheel, the change in load may be greater.

    That is, on a 36-spoke wheel, the bottom 5 spokes lose considerable
    tension and then regain it. The total tension loss (or change or
    delta) is spread over 5 spokes.

    On an 18-spoke wheel, the same tension loss is spread over only 3
    spokes, so the total tension loss _may_ be greater.

    Unfortunately, we have only FEA's for 36-spoke wheels and don't know
    what fewer spokes do. Here's an FEA for 36-spokes:

    http://www.astounding.org.uk/ian/wheel/index.html

    Ian's FEA suggests that the 5 spokes lose tension at roughly these
    relative rates (67~69, 244~250, ~350):

    / / | \ \ 36-spoke wheel
    -7 -25 -35 -25 -7 total ~99

    But what happens to an 18-spoke wheel with the same load?

    It could produce a much greater change in the middle spoke:

    / | \ 18 spokes?
    -7 -85 -7 total ~99 much greater change

    Or the same change:

    / | \ 18 spokes?
    -32 -35 -32 total ~99 same change

    Or even slight less change:

    / | \ 18 spokes?
    -33 -33 -33 total ~99 slightly less change

    The assumption is that the correct model tends toward the greater
    change example, meaning that a low spoke-count wheel's spokes see a
    greater delta or change in tension, from o to say 85 rather than just
    from 0 to 35.

    The greater the range of loading delta (spoke tension change), the
    sooner the flange will fatigue and fail.

    Cheers,

    Carl Fogel- Hide quoted text -

    - Show quoted text -

    Thanks Carl, this weekend I'll probably go over the details in that
    link. So delta just pertains to change in loading, and with lower
    spokes, there's a higher change? And in the life of a low spoked
    wheel, the greater loading per each cycling reduces its time to
    failure at some spot? Correct?

    Dear BD,

    The greater the change in tension, the sooner the failure, as long as
    the tension doesn't reach the yield point for anything involved.

    But whether a wheel with fewer spokes behaves that way is the
    question.

    Someone with a tension gauge and an 18-spoke wheel could probably
    determine what's actually going on more quickly and more convincingly
    than anyone can do an FEA for such wheels.

    Part of the problem is that low-spoke-count wheels tend to be much
    stiffer, so it's not clear what happens.

    Another part of the problem is that there's disagreement about whether
    stiffer wheels with fewer spokes need higher tension.

    I suspect that most of the low-spoke-count wheels have been developed
    in just the same way as the original 36-spoke wheels--trial and error
    come first, then maybe some theorizing, and maybe not.

    The spoke holes in the flange in question actually resemble an old
    highwheeler trick, but the idea was to allow cross-7 lacing that would
    normally have had the shafts of spokes covering the heads of other
    spokes:

    http://i3.tinypic.com/5ymzs40.jpg

    Marten has pointed out elsewhere in this thread that spokes in this
    arrangement look like a bad idea because the two spoke heads close
    together are trying to pull the hub apart.

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    The modern version is even worse than the highwheeler version, since
    the two spoke holes are now a single elongated hole, with no
    reinforcing material between them.

    Cheers,

    Carl Fogel

  12. Quoted message said:

    On Fri, 23 Nov 2007 17:28:35 -0800 (PST), bicycle_disciple

    Quoted message said:
    Quoted message said:

    On Fri, 23 Nov 2007 15:40:53 -0800 (PST), bicycle_disciple

    Quoted message said:
    Quoted message said:

    <[email hidden]> wrote:
    >On Nov 23, 2:48 pm, Werehatrack <[email hidden]> wrote:
    >> On Fri, 23 Nov 2007 12:46:30 -0600, Werehatrack
    >> <[email hidden]> said:

    Quoted message said:
    Quoted message said:

    >> >I would add to the observations above that the low-spoke-count wheels
    >> >have loading delta during rotation,

    Quoted message said:
    Quoted message said:

    >> make that "higher loading delta during rotation relative to that
    >> typical of a high-spoke-count wheel."

    Quoted message said:
    Quoted message said:

    >> --
    >> My email address is antispammed; pull WEEDS if replying via e-mail.
    >> Typoes are not a bug, they're a feature.
    >> Words processed in a facility that contains nuts.

    Quoted message said:
    Quoted message said:

    >I do not understand. Please explain loading delta.

    Quoted message said:
    Quoted message said:

    >B.D

    Quoted message said:
    Quoted message said:

    >http://cozybeehvie.blogspot.com

    Quoted message said:
    Quoted message said:

    Dear B.D.,

    Quoted message said:
    Quoted message said:

    The greek delta letter is often used to indicate change in equations.

    Quoted message said:
    Quoted message said:

    On a low spoke count wheel, the change in load may be greater.

    Quoted message said:
    Quoted message said:

    That is, on a 36-spoke wheel, the bottom 5 spokes lose considerable
    tension and then regain it. The total tension loss (or change or
    delta) is spread over 5 spokes.

    Quoted message said:
    Quoted message said:

    On an 18-spoke wheel, the same tension loss is spread over only 3
    spokes, so the total tension loss _may_ be greater.

    Quoted message said:
    Quoted message said:

    Unfortunately, we have only FEA's for 36-spoke wheels and don't know
    what fewer spokes do. Here's an FEA for 36-spokes:

    Quoted message said:
    Quoted message said:

    http://www.astounding.org.uk/ian/wheel/index.html

    Quoted message said:
    Quoted message said:

    Ian's FEA suggests that the 5 spokes lose tension at roughly these
    relative rates (67~69, 244~250, ~350):

    Quoted message said:
    Quoted message said:

    / / | \ \ 36-spoke wheel
    -7 -25 -35 -25 -7 total ~99

    Quoted message said:
    Quoted message said:

    But what happens to an 18-spoke wheel with the same load?

    Quoted message said:
    Quoted message said:

    It could produce a much greater change in the middle spoke:

    Quoted message said:
    Quoted message said:

    / | \ 18 spokes?
    -7 -85 -7 total ~99 much greater change

    Quoted message said:
    Quoted message said:

    Or the same change:

    Quoted message said:
    Quoted message said:

    / | \ 18 spokes?
    -32 -35 -32 total ~99 same change

    Quoted message said:
    Quoted message said:

    Or even slight less change:

    Quoted message said:
    Quoted message said:

    / | \ 18 spokes?
    -33 -33 -33 total ~99 slightly less change

    Quoted message said:
    Quoted message said:

    The assumption is that the correct model tends toward the greater
    change example, meaning that a low spoke-count wheel's spokes see a
    greater delta or change in tension, from o to say 85 rather than just
    from 0 to 35.

    Quoted message said:
    Quoted message said:

    The greater the range of loading delta (spoke tension change), the
    sooner the flange will fatigue and fail.

    Quoted message said:
    Quoted message said:

    Cheers,

    Quoted message said:
    Quoted message said:

    Carl Fogel- Hide quoted text -

    Quoted message said:
    Quoted message said:

    - Show quoted text -

    Quoted message said:

    Thanks Carl, this weekend I'll probably go over the details in that
    link. So delta just pertains to change in loading, and with lower
    spokes, there's a higher change? And in the life of a low spoked
    wheel, the greater loading per each cycling reduces its time to
    failure at some spot? Correct?

    Dear BD,

    The greater the change in tension, the sooner the failure, as long as
    the tension doesn't reach the yield point for anything involved.

    But whether a wheel with fewer spokes behaves that way is the
    question.

    Someone with a tension gauge and an 18-spoke wheel could probably
    determine what's actually going on more quickly and more convincingly
    than anyone can do an FEA for such wheels.

    Part of the problem is that low-spoke-count wheels tend to be much
    stiffer, so it's not clear what happens.

    Another part of the problem is that there's disagreement about whether
    stiffer wheels with fewer spokes need higher tension.

    I suspect that most of the low-spoke-count wheels have been developed
    in just the same way as the original 36-spoke wheels--trial and error
    come first, then maybe some theorizing, and maybe not.

    The spoke holes in the flange in question actually resemble an old
    highwheeler trick, but the idea was to allow cross-7 lacing that would
    normally have had the shafts of spokes covering the heads of other
    spokes:

    http://i3.tinypic.com/5ymzs40.jpg

    Marten has pointed out elsewhere in this thread that spokes in this
    arrangement look like a bad idea because the two spoke heads close
    together are trying to pull the hub apart.

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    The modern version is even worse than the highwheeler version, since
    the two spoke holes are now a single elongated hole, with no
    reinforcing material between them.

    Cheers,

    Carl Fogel- Hide quoted text -

    - Show quoted text -

    Yes I noticed how the spoke head are in one elongated hole, trying to
    pull that area apart. Actually, I have the wheel in question with me
    in my basement, I was surprised I hardly noticed this before!

    B.D

    http://cozybeehive.blogspot.com

  13. Arthur Shapiro said:

    I was riding with a fellow club member on a Thursday club ride, and
    heard some light pinging noises coming from the rear of his bicycle.
    They got a little bit worse as we were on one climb. It sounded as
    if his rear derailleur might be just barely touching one spoke, but
    visually that didn't appear to be the case.

    Quoted message said:

    The noise got louder and more frequent as we started up a fairly
    healthy climb. Then suddenly something broke. It appeared to be
    one of the spokes on his Bontrager rear wheel.

    Quoted message said:

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    Quoted message said:

    and the subsequent "Next Picture" link to picture #02. (Others in
    the set are of no interest beyond the club. The pictures will be up
    for one week.)

    Quoted message said:

    Is there enough visual evidence to speculate on the cause / nature
    of this flange failure?

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes, as
    shown in this picture, this part of the flange is in tension and fails
    similarly to radial spoking where stress in spokes arise primarily
    from tension rather than wheel loads that in this wheel cause these
    high tension loads to vary with each wheel revolution.

    Jobst Brandt

  14. Arthur Shapiro said:

    I was riding with a fellow club member on a Thursday club ride, and
    heard some light pinging noises coming from the rear of his bicycle.
    They got a little bit worse as we were on one climb. It sounded as
    if his rear derailleur might be just barely touching one spoke, but
    visually that didn't appear to be the case.

    Quoted message said:

    The noise got louder and more frequent as we started up a fairly
    healthy climb. Then suddenly something broke. It appeared to be
    one of the spokes on his Bontrager rear wheel.

    Quoted message said:

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    Quoted message said:

    and the subsequent "Next Picture" link to picture #02. (Others in
    the set are of no interest beyond the club. The pictures will be up
    for one week.)

    Quoted message said:

    Is there enough visual evidence to speculate on the cause / nature
    of this flange failure?

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes, as
    shown in this picture, this flange section is in tension and fails
    similarly as with radial spoking where stress in spokes arise
    primarily from static tension rather than wheel loads that in this
    wheel cause spoke tension loads to vary with wheel revolutions.

    Jobst Brandt

  15. Quoted message said:

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes, as
    shown in this picture, this part of the flange is in tension and fails
    similarly to radial spoking where stress in spokes arise primarily
    from tension rather than wheel loads that in this wheel cause these
    high tension loads to vary with each wheel revolution.

    I'm not following you here. Don't conventional 36-spoke wheels
    have flange sections that alternate being in compression and
    in tension?

    Tom Ace

  16. Tom Ace said:
    Quoted message said:

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes,
    as shown in this picture, this part of the flange is in tension and
    fails similarly to radial spoking where stress in spokes arise
    primarily from tension rather than wheel loads that in this wheel
    cause these high tension loads to vary with each wheel revolution.

    Quoted message said:

    I'm not following you here. Don't conventional 36-spoke wheels have
    flange sections that alternate being in compression and in tension?

    Not only do these spoke pairs pulling in opposite directions not
    overlap and compress the material between, there is no material
    between. These spokes appear to be laced into slots that contain two
    spokes. The bridge between has got to fail. Most hubs are deigned to
    have opposing spokes pull against each other when properly laced.
    This wheel was not and the hub was designed to not do that, judging
    from the slots containing two spokes. Being black that isn't easy to
    see but the failed section shows no material between the ends of the
    slot.

    Look at response #1 to see the effect phrased slightly differently.

    Jobst Brandt

  17. Quoted message said:
    Tom Ace said:
    Quoted message said:

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes,
    as shown in this picture, this part of the flange is in tension and
    fails similarly to radial spoking where stress in spokes arise
    primarily from tension rather than wheel loads that in this wheel
    cause these high tension loads to vary with each wheel revolution.


    I'm not following you here. Don't conventional 36-spoke wheels have
    flange sections that alternate being in compression and in tension?

    Not only do these spoke pairs pulling in opposite directions not
    overlap and compress the material between, there is no material
    between. These spokes appear to be laced into slots that contain two
    spokes. The bridge between has got to fail. Most hubs are deigned to
    have opposing spokes pull against each other when properly laced.
    This wheel was not and the hub was designed to not do that, judging
    from the slots containing two spokes. Being black that isn't easy to
    see but the failed section shows no material between the ends of the
    slot.

    Look at response #1 to see the effect phrased slightly differently.

    Jobst Brandt

    Hi there Jobst.

    I just lightened and cropped the image of the hub and posted it on
    Flickr @:

    http://www.flickr.com/photos/73832500@N00/2058503237/

    You can clearly see that the 2 spokes are in a slot and not 2
    individual spoke holes.

    Great design huh?

    Cheers from Peter

  18. Joe Riel said:

    bicycle_disciple <[email hidden]> writes:


    [...]

    Quoted message said:
    Quoted message said:

    Why is it poor? How off is one potentially going to be? I thought the
    100 swings sort of evens things out, and you get a good figure, but I
    understand this is not a foolproof method.

    You are computing the rotation inertia via the formula

    Irot = m*R^2*(g*(T/2/pi)^2/R - 1)

    where
    R = length of the pendulum
    m = mass of wheel
    T = period of oscillation

    Let's assume most of the error is in the measurement of R
    (that is almost certainly the case with the setup you show).
    The sensitivity of Irot with respect to R is

    (dIrot/dR)*(R/Irot) = 1 - R^2/R0^2,

    where Irot = m*R0^2. We know that R0 < Rw (wheel radius).
    From the the pictures it looks like R ~ 3*Rw, so the
    magnitude of the sensitivity is greater than 3^2-1 = 8.

    If you use the technique I suggest (use the inside top edge
    of the rim as a pivot point on a fixed knife edge), the
    sensitivity is reduced more than eight times.

    It might be harder to keep the oscillation in the plane of the wheel
    with that method though. At least if you use string, you can make a sort
    of triangular swing that's relatively stable.

  19. Quoted message said:
    Arthur Shapiro said:

    I was riding with a fellow club member on a Thursday club ride, and
    heard some light pinging noises coming from the rear of his bicycle.
    They got a little bit worse as we were on one climb. It sounded as
    if his rear derailleur might be just barely touching one spoke, but
    visually that didn't appear to be the case.

    Quoted message said:

    The noise got louder and more frequent as we started up a fairly
    healthy climb. Then suddenly something broke. It appeared to be
    one of the spokes on his Bontrager rear wheel.

    Quoted message said:

    Then we looked closer:

    http://www.ocrebels.com/images/GelsonsNBNovember2007/pic01.htm

    Quoted message said:

    and the subsequent "Next Picture" link to picture #02. (Others in
    the set are of no interest beyond the club. The pictures will be up
    for one week.)

    Quoted message said:

    Is there enough visual evidence to speculate on the cause / nature
    of this flange failure?

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes, as
    shown in this picture, this flange section is in tension

    Isn't the situation just the same with 32H or 36H? The basic pattern is
    the same as with this failed hub-- if you number the spokes A, B, C, D,
    etc. then if A and B are crossing over and compressing the flange, then
    B and C will be pulling the section of flange between them apart. C and
    D compressing again, D and E pulling.

    The unusual thing about this Bontrager hub is that the holes aren't
    evenly spaced. The sections of flange that are being pulled apart by
    spokes are shorter-- just the length of those slots. The sections in
    compression are longer. I don't know why or whether that made any
    difference to why it failed.

  20. "Ben C" <[email hidden]> wrote in message
    news:[email hidden]...

    Quoted message said:

    On 2007-11-24, [email hidden]
    <[email hidden]> wrote:


    ....

    Quoted message said:
    Quoted message said:

    Cross laced spokes in hubs with a larger number of spokes have the
    flange section between two spokes in compression. With few spokes, as
    shown in this picture, this flange section is in tension

    The section under tension didn't fail - the failures are both in sections
    that were under compression.

    Quoted message said:

    Isn't the situation just the same with 32H or 36H? The basic pattern is
    the same as with this failed hub-- if you number the spokes A, B, C, D,
    etc. then if A and B are crossing over and compressing the flange, then
    B and C will be pulling the section of flange between them apart. C and
    D compressing again, D and E pulling.

    Yes, but see below.

    Quoted message said:

    The unusual thing about this Bontrager hub is that the holes aren't
    evenly spaced. The sections of flange that are being pulled apart by
    spokes are shorter-- just the length of those slots. The sections in
    compression are longer. I don't know why or whether that made any
    difference to why it failed.

    If you look closely at the photos, the crack on the left is in line with the
    spoke that was attached in that slot. I think the crack on the right is
    unrelated to the primary failure, and happened after the crack on the left.
    The left crack looks very much like the result of the spoke effectively
    cleaving the flange. The spoke tension is partially translated into tension
    in the flange that is orthogonal to the direction of the spoke, effectively
    trying to split the flange with a pressure that increases with the spoke
    tension. Imagine the spoke end as the tip of a chisel, applying pressure
    against the flange.

    Mike

Active in the last 60 minutes

Active in this thread

0 users · 0 guests ·0 bots ·0 total

No signed-in users are active right now.

No known search crawlers active right now.