mavic rims don't *all* suck

Discussion in 'Cycling Equipment' started by Dvt, Feb 26, 2004.

  1. Dvt

    Dvt Guest

    First, thanks a bunch to Carl Fogel for hosting the pictures and helping me with formatting this
    message. I have documented a rim failure much like one of those mentioned in the "mavic rims suck?"
    thread. This rim served for a long time without trouble except the time I stacked it into a snowbank
    off the side of a hairpin on a descent. I rebuilt the wheel after that incident and it continued to
    serve well for a few more years.

    Here are 8 *.jpg images of that rim after it failed. The links are listed here for quick reference
    and also repeated in the text.

    1. http://home.comcast.net/~carlfogel/download/rim_wide.JPG (135k)
    2. http://home.comcast.net/~carlfogel/download/rim_label.JPG ( 40k)
    3. http://home.comcast.net/~carlfogel/download/big_crack.JPG (111k)
    4. http://home.comcast.net/~carlfogel/download/small_crack.JPG ( 86k)
    5. http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG ( 20k)
    6. http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG ( 46k)
    7. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG ( 54k)
    8. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG ( 52k)

    This Mavic Open SUP CD rim from 1993 failed after 10's of thousands of miles. It failed in 2003
    while I was not on the bike. I inflated the tires in the morning, rode to work, then found the rim
    in this condition when I was ready to ride home at the end of the day. The failure was a crack in
    the sidewall. I have no complaints, as the rim served a long and useful life.

    ++++
    Photo 1: Wide view of rim, with failures at approx 11 and 12 o'clock.
    http://home.comcast.net/~carlfogel/download/rim_wide.JPG
    ++++
    Photo 2: Label of the rim showing "Open SUP CD" at the top of the orange label. This label is at
    about 8 o'clock on the previous picture. http://home.comcast.net/~carlfogel/download/rim_label.JPG
    ++++

    The failures were in the sidewall. I had originally presumed that they were caused by brake wall
    wear. But after having read some of the chatter on rec.bicycles.tech, I'm not sure anymore. Could
    this be an anodizing failure? Here are some closeup photos of the failures to help us decide.

    ++++
    Photo 3: This is the larger failure seen at 11 o'clock in photo 1.
    http://home.comcast.net/~carlfogel/download/big_crack.JPG
    ++++
    Photo 4: This is the smaller failure seen at 12 o'clock in photo 1.
    http://home.comcast.net/~carlfogel/download/small_crack.JPG
    ++++

    Notice the white powder around the spoke socket in the photo 4. Apparently the anodizing didn't
    prevent galvanic corrosion. The yellowish stuff is adhesive residue from the rim strips.

    Also note that both cracks are approximately centered on a spoke hole. The rim should bulge a little
    bit at each spoke hole, causing the brake wall to wear a little bit thinner in that area. Both of
    these failures were on the right side of a wheel, and the spoke holes nearest the failures contained
    spokes that pulled to the right. But the wheel was a Shimano 8-speed rear, so the bracing angle was
    low and the bulge was no longer noticable. Any bulge that had existed was rubbed out by brake pad
    wear. The bulge on the left side spokes are still noticeable.

    Now let's get even closer.

    ++++
    Photo 5: A blown-up view of the photo 3. This side of the crack is closest to the spoke bed. Bottom
    of photo is toward the spokes, top of photo is the braking surface.
    http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG
    ++++
    Photo 6: A blow-up of a different part of photo 3. Again, we see the side closest to the spoke bed.
    You can see a bit of the socket in the lower part of the picture. Bottom of photo is toward the
    spokes, top of photo is the braking surface.
    http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG
    ++++
    Photo 7: More magnification of a section from Photo 5. Bottom of photo is toward the spokes, top of
    photo is the braking surface. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG
    ++++
    Photo 8: More magnification of a section from Photo 6. Bottom of photo is toward the spokes, top of
    photo is the braking surface. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG
    ++++

    I don't know how to analyze the last four photos. I see that the shiny part on the top half of the
    photos has striations perpendicular to the thickness of the brake wall. I think that shiny part
    probably broke all at once while the bike was parked.

    I think the dull, rough section in the lower part of each photo came apart slowly. But it isn't
    smooth with the parallel lines that typical of a fatigue failure.

    So what was the failure mode of this rim?

    --
    Dave [email protected]
     
    Tags:


  2. Very interesting Dave, great photos by you and Carl... never seen that failure in a rim before.
    Might want to email these to Mavic, see what they say. -tom

    "dvt" <[email protected]> wrote in message news:[email protected]...
    > First, thanks a bunch to Carl Fogel for hosting the pictures and helping me with formatting this
    > message. I have documented a rim failure much like one of those mentioned in the "mavic rims
    > suck?" thread. This rim served for a long time without trouble except the time I stacked it into a
    > snowbank off the side of a hairpin on a descent. I rebuilt the wheel after that incident and it
    > continued to serve well for a few more years.
    >
    > Here are 8 *.jpg images of that rim after it failed. The links are listed here for quick reference
    > and also repeated in the text.
    >
    > 1. http://home.comcast.net/~carlfogel/download/rim_wide.JPG (135k)
    > 2. http://home.comcast.net/~carlfogel/download/rim_label.JPG ( 40k)
    > 3. http://home.comcast.net/~carlfogel/download/big_crack.JPG (111k)
    > 4. http://home.comcast.net/~carlfogel/download/small_crack.JPG ( 86k)
    > 5. http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG ( 20k)
    > 6. http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG ( 46k)
    > 7. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG ( 54k)
    > 8. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG ( 52k)
    >
    > This Mavic Open SUP CD rim from 1993 failed after 10's of thousands of miles. It failed in 2003
    > while I was not on the bike. I inflated the tires in the morning, rode to work, then found the rim
    > in this condition when I was ready to ride home at the end of the day. The failure was a crack in
    > the sidewall. I have no complaints, as the rim served a long and useful life.
    >
    > ++++
    > Photo 1: Wide view of rim, with failures at approx 11 and 12 o'clock.
    > http://home.comcast.net/~carlfogel/download/rim_wide.JPG
    > ++++
    > Photo 2: Label of the rim showing "Open SUP CD" at the top of the orange label. This label is at
    > about 8 o'clock on the previous picture. http://home.comcast.net/~carlfogel/download/rim_label.JPG
    > ++++
    >
    > The failures were in the sidewall. I had originally presumed that they were caused by brake wall
    > wear. But after having read some of the chatter on rec.bicycles.tech, I'm not sure anymore. Could
    > this be an anodizing failure? Here are some closeup photos of the failures to help us decide.
    >
    > ++++
    > Photo 3: This is the larger failure seen at 11 o'clock in photo 1.
    > http://home.comcast.net/~carlfogel/download/big_crack.JPG
    > ++++
    > Photo 4: This is the smaller failure seen at 12 o'clock in photo 1.
    > http://home.comcast.net/~carlfogel/download/small_crack.JPG
    > ++++
    >
    > Notice the white powder around the spoke socket in the photo 4. Apparently the anodizing didn't
    > prevent galvanic corrosion. The yellowish stuff is adhesive residue from the rim strips.
    >
    > Also note that both cracks are approximately centered on a spoke hole. The rim should bulge a
    > little bit at each spoke hole, causing the brake wall to wear a little bit thinner in that area.
    > Both of these failures were on the right side of a wheel, and the spoke holes nearest the failures
    > contained spokes that pulled to the right. But the wheel was a Shimano 8-speed rear, so the
    > bracing angle was low and the bulge was no longer noticable. Any bulge that had existed was rubbed
    > out by brake pad wear. The bulge on the left side spokes are still noticeable.
    >
    > Now let's get even closer.
    >
    > ++++
    > Photo 5: A blown-up view of the photo 3. This side of the crack is closest to the spoke bed.
    > Bottom of photo is toward the spokes, top of photo is the braking surface.
    > http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG
    > ++++
    > Photo 6: A blow-up of a different part of photo 3. Again, we see the side closest to the spoke
    > bed. You can see a bit of the socket in the lower part of the picture. Bottom of photo is toward
    > the spokes, top of photo is the braking surface.
    > http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG
    > ++++
    > Photo 7: More magnification of a section from Photo 5. Bottom of photo is toward the spokes, top
    > of photo is the braking surface. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG
    > ++++
    > Photo 8: More magnification of a section from Photo 6. Bottom of photo is toward the spokes, top
    > of photo is the braking surface. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG
    > ++++
    >
    > I don't know how to analyze the last four photos. I see that the shiny part on the top half of the
    > photos has striations perpendicular to the thickness of the brake wall. I think that shiny part
    > probably broke all at once while the bike was parked.
    >
    > I think the dull, rough section in the lower part of each photo came apart slowly. But it isn't
    > smooth with the parallel lines that typical of a fatigue failure.
    >
    > So what was the failure mode of this rim?
    >
    > --
    > Dave [email protected]
     
  3. Jobst Brandt

    Jobst Brandt Guest

    Dave vt? writes:

    > First, thanks a bunch to Carl Fogel for hosting the pictures and helping me with formatting this
    > message. I have documented a rim failure much like one of those mentioned in the "mavic rims
    > suck?" thread. This rim served for a long time without trouble except the time I stacked it into a
    > snowbank off the side of a hairpin on a descent. I rebuilt the wheel after that incident and it
    > continued to serve well for a few more years.

    > Here are 8 *.jpg images of that rim after it failed. The links are listed here for quick reference
    > and also repeated in the text.

    > 1. http://home.comcast.net/~carlfogel/download/rim_wide.JPG (135k)
    > 2. http://home.comcast.net/~carlfogel/download/rim_label.JPG ( 40k)
    > 3. http://home.comcast.net/~carlfogel/download/big_crack.JPG (111k)
    > 4. http://home.comcast.net/~carlfogel/download/small_crack.JPG ( 86k)
    > 5. http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG ( 20k)
    > 6. http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG ( 46k)
    > 7. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG ( 54k)
    > 8. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG ( 52k)

    > This Mavic Open SUP CD rim from 1993 failed after 10's of thousands of miles. It failed in 2003
    > while I was not on the bike. I inflated the tires in the morning, rode to work, then found the rim
    > in this condition when I was ready to ride home at the end of the day. The failure was a crack in
    > the sidewall. I have no complaints, as the rim served a long and useful life.

    That is not what I would find acceptable. How would you feel if you were descending a mountain road
    while the rim fell apart? Make that a right hand turn with opposing traffic or at the entry of a
    hairpin turn over an abyss. The rim shows no excessive wear on the braking surface so it appeared to
    have plenty of life left... but it didn't.

    > So what was the failure mode of this rim?

    Cracking is a fatigue failure. The rough fracture (formerly called "it crystallized") is a
    typical fatigue fracture. A forced rupture is smoother and bright, it having occurred all at once
    from overload.

    There is no reason to be an apologist for reckless manufacturing. Anodizing is not to our advantage!
    But thanks for documenting that it is so.

    Jobst Brandt [email protected]
     
  4. Jobst Brandt

    Jobst Brandt Guest

    Tom Nakashima writes:

    > Very interesting Dave, great photos by you and Carl... never seen that failure in a rim before.
    > Might want to email these to Mavic, see what they say.

    That's old hat and they've seen it aplenty. As I said, bicycle shops around here had rims where the
    rim cracked all the way around into the hollow section, leaving the tire on the outer half and the
    spokes on the inner, the wheel still slightly ridable. That was when the MA-40 came on the scene.

    Jobst Brandt [email protected]
     
  5. Tad Borek

    Tad Borek Guest

    [email protected] wrote:
    > Dave vt? writes:
    >>So what was the failure mode of this rim?
    >
    > Cracking is a fatigue failure. The rough fracture (formerly called "it crystallized") is a typical
    > fatigue fracture. A forced rupture is smoother and bright, it having occurred all at once from
    > overload.

    Do you know where the motion comes from that leads to this type of fatigue failure? Is it sidewall
    bending in response to, say, hard bumps that momentarily increase the pressure in the tube? Is there
    a normal cycle of sidewall bending that occurs as the wheel rotates? Is it repetitions of inflating
    & deflating the tire/tube over a long time period (akin to those old aluminum failures in aircraft,
    from pressurization cycles)?

    It seems some of this motion must be negligible - I'm wondering if there's something in particular
    that is the biggest contributor to this type of fatigue failure. Eg maybe there's an argument for
    keeping your tires topped off even if you're not riding.

    A curious thing is that it looks like two unconnected sections failed at the same time. It makes
    sense to have a fracture that creeps in both directions, even unzipping all the way around the rim,
    but unless I'm seeing wrong it seems the cracks aren't connected. Viewing fatigue failure ITO
    probabilities, it seems unlikely two unconnected sections would reach the failure point on exactly
    the same day, after all those years, especially because once it began to fail the pressure of the
    bead must have dropped a bit (hypothesis: crack one began to propogate at lower pressure than crack
    two, unless they began at the same time. Yes/no?). So what was the trigger...a way-worn brake pad,
    dragging metal on that edge? Higher than normal tire pressure that day? OP - after the snowbank
    crash did you bend out the rim in these spots?

    -Tad
     
  6. Dvt

    Dvt Guest

    Tad Borek wrote:
    > A curious thing is that it looks like two unconnected sections failed at the same time. It makes
    > sense to have a fracture that creeps in both directions, even unzipping all the way around the
    > rim, but unless I'm seeing wrong it seems the cracks aren't connected.

    The fracture is not connected. I also found that odd for all the same reaons you mention. I should
    have mentioned it in the OP.

    > Viewing fatigue failure ITO probabilities, it seems unlikely two unconnected sections would reach
    > the failure point on exactly the same day, after all those years, especially because once it began
    > to fail the pressure of the bead must have dropped a bit (hypothesis: crack one began to propogate
    > at lower pressure than crack two, unless they began at the same time. Yes/no?).

    I don't know if they began at the same time. I didn't notice the crack inceptions, only the
    final failure.

    > So what was the trigger...a way-worn brake pad, dragging metal on that edge?

    No grooves in the brake wall to indicate that. No noticeable marks on the inside or the outside of
    the brake wall. Maybe I'll get pix of that.

    > Higher than normal tire pressure that day?

    I guess it was a bit higher than "normal" since I don't pump my tires up frequently. I usually go
    weeks between inflations. I get slightly bigger tires than I need to allow this laziness to occur.

    > OP - after the snowbank crash did you bend out the rim in these spots?

    No, I just released all spoke tension and started anew. When I had removed all tension I laid the
    rim on a flat surface. The rim was still flat.

    Dave dvt at psu dot edu
     
  7. Dvt

    Dvt Guest

    [email protected] wrote:
    >>This Mavic Open SUP CD rim from 1993 failed after 10's of thousands of miles. It failed in 2003
    >>while I was not on the bike. I inflated the tires in the morning, rode to work, then found the rim
    >>in this condition when I was ready to ride home at the end of the day. The failure was a crack in
    >>the sidewall. I have no complaints, as the rim served a long and useful life.

    > That is not what I would find acceptable. How would you feel if you were descending a mountain
    > road while the rim fell apart? Make that a right hand turn with opposing traffic or at the entry
    > of a hairpin turn over an abyss. The rim shows no excessive wear on the braking surface so it
    > appeared to have plenty of life left... but it didn't.

    The braking surface is definitely concave. The braking surface *did* show significant wear. I don't
    know if it was excessive. How thin does a wall need to be in good aluminum? I don't know how thick
    the wall was when the rim was new, so I can't comment on the change in thicknes. I could measure the
    remaining wall thickness.

    In retrospect, riding this rim in that well-worn condition was probably not wise. But then we
    wouldn't be having all of this fun. :)

    >>So what was the failure mode of this rim?

    > Cracking is a fatigue failure. The rough fracture (formerly called "it crystallized") is a typical
    > fatigue fracture. A forced rupture is smoother and bright, it having occurred all at once from
    > overload.

    There is part of the surface that appears to be forced rupture, as you would expect when the crack
    propogated until the remaining wall was too weak. The recent inflation was probably the proverbial
    last straw, causing the rupture.

    > There is no reason to be an apologist for reckless manufacturing. Anodizing is not to our
    > advantage! But thanks for documenting that it is so.

    Even if, as you say, the rough surface indicates fatigue, how does this condemn the anodizing? How
    do I know this wouldn't have happened with an MA-2 under the same conditions.

    I currently have a pair of MA-2s on my commuter bike. Maybe I'll report back in a few years with
    their failure mechanism. <g>

    Dave dvt at psu dot edu
     
  8. Dianne_1234

    Dianne_1234 Guest

    On Thu, 26 Feb 2004 20:00:37 GMT, Tad Borek <[email protected]>
    wrote:

    >Do you know where the motion comes from that leads to this type of fatigue failure?

    In theory, all causes "count". (Miner's rule, etc.)

    > Is it sidewall bending in response to, say, hard bumps that momentarily increase the pressure in
    > the tube?

    The number of cycles of bump-hitting is normally many, many times less than from just rolling. I'd
    neglect them.

    > Is there a normal cycle of sidewall bending that occurs as the wheel rotates?

    This is my guess as to the largest number of cycles. The bulge at the bottom of the tire changes the
    angle at which the casing pulls on the rim. This changes the bending moment. This is cyclic with
    every wheel rotation.

    > Is it repetitions of inflating & deflating the tire/tube over a long time period (akin to those
    > old aluminum failures in aircraft, from pressurization cycles)?

    Probably (Miner's rule again). My guess is there are far fewer cycles of pumping than of rolling.

    >It seems some of this motion must be negligible - I'm wondering if there's something in particular
    >that is the biggest contributor to this type of fatigue failure. Eg maybe there's an argument for
    >keeping your tires topped off even if you're not riding.

    See above, by far the largest number of cycles comes from rolling.

    Later the OP mentions larger tires. The casing of larger tires makes a more horizontal angle with
    the rim, so increases the bending moment on the rim side wall.

    >A curious thing is that it looks like two unconnected sections failed at the same time.

    Yeah, I thought so too! What are the chances of that? Never mind -- it happened this once at
    least! :)
     
  9. Jobst Brandt

    Jobst Brandt Guest

    Tad Borek writes:

    >>> So what was the failure mode of this rim?

    >> Cracking is a fatigue failure. The rough fracture (formerly called "it crystallized") is a
    >> typical fatigue fracture. A forced rupture is smoother and bright, it having occurred all at once
    >> from overload.

    > Do you know where the motion comes from that leads to this type of fatigue failure? Is it sidewall
    > bending in response to, say, hard bumps that momentarily increase the pressure in the tube? Is
    > there a normal cycle of sidewall bending that occurs as the wheel rotates? Is it repetitions of
    > inflating & deflating the tire/tube over a long time period (akin to those old aluminum failures
    > in aircraft, from pressurization cycles)?

    I think the FAQ item explains that. There is a large cycling lateral load with each pass of the load
    affected zone:

    http://draco.acs.uci.edu/rbfaq/FAQ/8b.25.html

    > It seems some of this motion must be negligible - I'm wondering if there's something in particular
    > that is the biggest contributor to this type of fatigue failure. Eg maybe there's an argument for
    > keeping your tires topped off even if you're not riding.

    This is for most rims an insignificant load unless there are crack initiators. As I have mentioned,
    I have ridden MA-2 rims until sidewalls were worn down to 0.5mm (0.020") with no failure.

    > A curious thing is that it looks like two unconnected sections failed at the same time. It makes
    > sense to have a fracture that creeps in both directions, even unzipping all the way around the
    > rim, but unless I'm seeing wrong it seems the cracks aren't connected. Viewing fatigue failure ITO
    > probabilities, it seems unlikely two unconnected sections would reach the failure point on exactly
    > the same day, after all those years, especially because once it began to fail the pressure of the
    > bead must have dropped a bit (hypothesis: crack one began to propagate at lower pressure than
    > crack two, unless they began at the same time. Yes/no?). So what was the trigger... a way-worn
    > brake pad, dragging metal on that edge? Higher than normal tire pressure that day? OP - after the
    > snowbank crash did you bend out the rim in these spots?

    I believe the cracks opened before parking the bicycle and that they slowly peeled apart. I have
    observed this sort of failure somewhere but I don't recall the circumstance.

    Jobst Brandt [email protected]
     
  10. Jim Beam

    Jim Beam Guest

    i think you're right, the rim served a long & useful life, especially as a brake track, /by
    definition/, is subject to wear and is going to fail in this way at some point, if other failure
    modes don't get to it first.

    why did it crack when not in actual use? well, i'm becoming more & more inclined to think there's a
    stress corrosion element to this picture. the granular fracture surfaces in your pics are typical of
    such a failure mode.

    a micrograph showing this in more detail is here: http://corrosion.ksc.nasa.gov/images/scc4.jpg

    here, the crack can clearly be seen growing between the grains, not progressing through them as is
    typical of fatigue.

    your pics are not perfect, and may well show an element of fatigue under closer examination, but the
    bright crystalline intergranular nature of the bulk of the fracture surface is clear. it's not
    uncommon for failures to combine stress corrosion, corrosion and fatigue, so it could easily be a
    combination.

    further examples of stress corrosion can be seen here: http://technology.open.ac.uk/materials/mem/mem-
    corr.html

    dvt wrote:
    > First, thanks a bunch to Carl Fogel for hosting the pictures and helping me with formatting this
    > message. I have documented a rim failure much like one of those mentioned in the "mavic rims
    > suck?" thread. This rim served for a long time without trouble except the time I stacked it into a
    > snowbank off the side of a hairpin on a descent. I rebuilt the wheel after that incident and it
    > continued to serve well for a few more years.
    >
    > Here are 8 *.jpg images of that rim after it failed. The links are listed here for quick reference
    > and also repeated in the text.
    >
    > 1. http://home.comcast.net/~carlfogel/download/rim_wide.JPG (135k)
    > 2. http://home.comcast.net/~carlfogel/download/rim_label.JPG ( 40k)
    > 3. http://home.comcast.net/~carlfogel/download/big_crack.JPG (111k)
    > 4. http://home.comcast.net/~carlfogel/download/small_crack.JPG ( 86k)
    > 5. http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG ( 20k)
    > 6. http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG ( 46k)
    > 7. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG ( 54k)
    > 8. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG ( 52k)
    >
    > This Mavic Open SUP CD rim from 1993 failed after 10's of thousands of miles. It failed in 2003
    > while I was not on the bike. I inflated the tires in the morning, rode to work, then found the rim
    > in this condition when I was ready to ride home at the end of the day. The failure was a crack in
    > the sidewall. I have no complaints, as the rim served a long and useful life.
    >
    > ++++
    > Photo 1: Wide view of rim, with failures at approx 11 and 12 o'clock.
    > http://home.comcast.net/~carlfogel/download/rim_wide.JPG
    > ++++
    > Photo 2: Label of the rim showing "Open SUP CD" at the top of the orange label. This label is at
    > about 8 o'clock on the previous picture. http://home.comcast.net/~carlfogel/download/rim_label.JPG
    > ++++
    >
    > The failures were in the sidewall. I had originally presumed that they were caused by brake wall
    > wear. But after having read some of the chatter on rec.bicycles.tech, I'm not sure anymore. Could
    > this be an anodizing failure? Here are some closeup photos of the failures to help us decide.
    >
    > ++++
    > Photo 3: This is the larger failure seen at 11 o'clock in photo 1.
    > http://home.comcast.net/~carlfogel/download/big_crack.JPG
    > ++++
    > Photo 4: This is the smaller failure seen at 12 o'clock in photo 1.
    > http://home.comcast.net/~carlfogel/download/small_crack.JPG
    > ++++
    >
    > Notice the white powder around the spoke socket in the photo 4. Apparently the anodizing didn't
    > prevent galvanic corrosion. The yellowish stuff is adhesive residue from the rim strips.
    >
    > Also note that both cracks are approximately centered on a spoke hole. The rim should bulge a
    > little bit at each spoke hole, causing the brake wall to wear a little bit thinner in that area.
    > Both of these failures were on the right side of a wheel, and the spoke holes nearest the failures
    > contained spokes that pulled to the right. But the wheel was a Shimano 8-speed rear, so the
    > bracing angle was low and the bulge was no longer noticable. Any bulge that had existed was rubbed
    > out by brake pad wear. The bulge on the left side spokes are still noticeable.
    >
    > Now let's get even closer.
    >
    > ++++
    > Photo 5: A blown-up view of the photo 3. This side of the crack is closest to the spoke bed.
    > Bottom of photo is toward the spokes, top of photo is the braking surface.
    > http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG
    > ++++
    > Photo 6: A blow-up of a different part of photo 3. Again, we see the side closest to the spoke
    > bed. You can see a bit of the socket in the lower part of the picture. Bottom of photo is toward
    > the spokes, top of photo is the braking surface.
    > http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG
    > ++++
    > Photo 7: More magnification of a section from Photo 5. Bottom of photo is toward the spokes, top
    > of photo is the braking surface. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG
    > ++++
    > Photo 8: More magnification of a section from Photo 6. Bottom of photo is toward the spokes, top
    > of photo is the braking surface. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG
    > ++++
    >
    > I don't know how to analyze the last four photos. I see that the shiny part on the top half of the
    > photos has striations perpendicular to the thickness of the brake wall. I think that shiny part
    > probably broke all at once while the bike was parked.
    >
    > I think the dull, rough section in the lower part of each photo came apart slowly. But it isn't
    > smooth with the parallel lines that typical of a fatigue failure.
    >
    > So what was the failure mode of this rim?
     
  11. Werehatrack

    Werehatrack Guest

    On Thu, 26 Feb 2004 12:54:51 -0500, dvt <[email protected]> may have
    said:

    >First, thanks a bunch to Carl Fogel for hosting the pictures and helping me with formatting this
    >message. I have documented a rim failure much like one of those mentioned in the "mavic rims suck?"
    >thread. This rim served for a long time without trouble except the time I stacked it into a
    >snowbank off the side of a hairpin on a descent. I rebuilt the wheel after that incident and it
    >continued to serve well for a few more years.
    >
    >Here are 8 *.jpg images of that rim after it failed. The links are listed here for quick reference
    >and also repeated in the text.
    >
    >1. http://home.comcast.net/~carlfogel/download/rim_wide.JPG (135k)
    >2. http://home.comcast.net/~carlfogel/download/rim_label.JPG ( 40k)
    >3. http://home.comcast.net/~carlfogel/download/big_crack.JPG (111k)
    >4. http://home.comcast.net/~carlfogel/download/small_crack.JPG ( 86k)
    >5. http://home.comcast.net/~carlfogel/download/blowup_02.8_1.JPG ( 20k)
    >6. http://home.comcast.net/~carlfogel/download/blowup_02.8_2.JPG ( 46k)
    >7. http://home.comcast.net/~carlfogel/download/blowup_4_1.JPG ( 54k)
    >8. http://home.comcast.net/~carlfogel/download/blowup_4_2.JPG ( 52k)

    Looking at the photos, I'd say that this was a fatigue failure due to the way the fractured surface
    does not show a clean break. Fatigue failures, in my experience, often indicate an understimate of
    the loading that the item would receive (this often results in the manufacturer saying that the item
    was *over*loaded, even though the real problem may have been that they failed to take the probable
    real-world in-service stresses into account properly). Without a cross-section to look at, I have to
    wonder if they might not have allowed too small of a fillet, or even created what amounted to a
    sharp corner at the base of that flange. Sharp internal corners tend to become failure points, and
    the break is typical of a sharp-corner fatigue failure in my opinion.

    Despite the long period in service, I have to say that I'm not impressed. The rim's outer surface
    does not show the kind of brake wear that would have made this less of a design concern and more of
    a maintenance and inspection matter.

    --
    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.
     
  12. Werehatrack

    Werehatrack Guest

    On Thu, 26 Feb 2004 15:39:19 -0500, dvt <[email protected]> may have
    said:

    >Tad Borek wrote:
    >> So what was the trigger...a way-worn brake pad, dragging metal on that edge? Higher than normal
    >> tire pressure that day?
    >
    >I guess it was a bit higher than "normal" since I don't pump my tires up frequently. I usually go
    >weeks between inflations. I get slightly bigger tires than I need to allow this laziness to occur.

    This makes me think that the cracks had been present for a while, but had not migrated all of the
    way through the flange, and only became complete when the pressure provided the force to finally
    push the broken sections out. Have you inspected the flanges on the *inside* for signs of cracks
    around the rest of the circumference of the wheel? I'm betting that you'll find that you have more
    than just the two which precipitated the failures.

    --
    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.
     
  13. Dvt

    Dvt Guest

    jim beam wrote:
    > why did it crack when not in actual use? well, i'm becoming more & more inclined to think there's
    > a stress corrosion element to this picture. the granular fracture surfaces in your pics are
    > typical of such a failure mode.
    >
    > a micrograph showing this in more detail is here: http://corrosion.ksc.nasa.gov/images/scc4.jpg

    I don't know what that picture represents. What's the scale? Is it aluminum? What do the colors
    represent? I expect a micrograph is not optical, so the colors probably represent the reflection of
    some sort of beam other than visible wavelengths.

    > your pics are not perfect, and may well show an element of fatigue under closer examination, but
    > the bright crystalline intergranular nature of the bulk of the fracture surface is clear. it's not
    > uncommon for failures to combine stress corrosion, corrosion and fatigue, so it could easily be a
    > combination.

    Suggestions for better pics? More magnification, diffuse light source,....? I have access to a
    digital camera, optical microscopes that range from 8-40x, and some garden-variety light sources.
    That's what I used for these pictures.

    --
    Dave dvt at psu dot edu
     
  14. Dvt

    Dvt Guest

    dianne_1234 wrote:
    > Later the OP mentions larger tires.

    I should have quantified: 20-25 mm tires for the first part of the rim's life, but in later years, I
    usually ran 28mm on those rims. I had 28mm tires on at the time of failure (old Avocet Duro).

    --
    Dave dvt at psu dot edu
     
  15. Dvt

    Dvt Guest

    [email protected] wrote:
    > This is for most rims an insignificant load unless there are crack initiators. As I have
    > mentioned, I have ridden MA-2 rims until sidewalls were worn down to 0.5mm (0.020") with no
    > failure.

    These sidewalls are ~1.5-1.8 mm thick at the crack.

    How do you know when to replace a rim? Concavity in the brake wall? Do you measure the wall
    thickness? Based on your earlier post, I don't think you wait until they crack.

    --
    Dave dvt at psu dot edu
     
  16. Dvt

    Dvt Guest

    Werehatrack wrote:
    > Have you inspected the flanges on the *inside* for signs of cracks around the rest of the
    > circumference of the wheel? I'm betting that you'll find that you have more than just the two
    > which precipitated the failures.

    I don't see any cracks. The inside of the flanges is pretty gooped up with Velox adhesive, so the
    cracks may be disguised.

    --
    Dave dvt at psu dot edu
     
  17. Dvt

    Dvt Guest

    Werehatrack wrote:

    > Without a cross-section to look at, I have to wonder if they might not have allowed too small of a
    > fillet, or even created what amounted to a sharp corner at the base of that flange. Sharp internal
    > corners tend to become failure points, and the break is typical of a sharp-corner fatigue failure
    > in my opinion.

    The inside of the brake wall is pretty gradually curved. There are no sharp points near the crack.

    > Despite the long period in service, I have to say that I'm not impressed. The rim's outer surface
    > does not show the kind of brake wear that would have made this less of a design concern and more
    > of a maintenance and inspection matter.

    Same question to you: How do you tell when a brake wall is too thin for continued service?

    --
    Dave dvt at psu dot edu
     
  18. dvt <[email protected]> wrote in
    news:[email protected]:

    >
    > Same question to you: How do you tell when a brake wall is too thin for continued service?
    >
    When your brake lever begins to pulse as you apply them...and then the end is close, real close.
     
  19. Smmb

    Smmb Guest

    "dvt" <[email protected]> a √©crit dans le message de :
    news:[email protected]...
    > Werehatrack wrote:
    > > Despite the long period in service, I have to say that I'm not impressed..............

    > Same question to you: How do you tell when a brake wall is too thin for continued service?
    >

    I think all the Mavic haters really believe that a virtuous rim gives its soul to the hereafter,
    collapses into a pile of pixie-dust, having left a note of regret.

    Now that's the right failure mode ? Everything fails, somehow, sometime. Same crowd, I would bet,
    that is wistful for the days of washable oil filters for cars...
    --
    Bonne route,

    Sandy Paris FR
     
  20. Jim Beam

    Jim Beam Guest

    dvt wrote:
    > jim beam wrote:
    >
    >> why did it crack when not in actual use? well, i'm becoming more & more inclined to think there's
    >> a stress corrosion element to this picture. the granular fracture surfaces in your pics are
    >> typical of such a failure mode.
    >>
    >> a micrograph showing this in more detail is here: http://corrosion.ksc.nasa.gov/images/scc4.jpg
    >
    >
    > I don't know what that picture represents. What's the scale? Is it aluminum? What do the colors
    > represent? I expect a micrograph is not optical, so the colors probably represent the reflection
    > of some sort of beam other than visible wavelengths.

    the pic is a generic monochrome, taken with reflected light from an optical metallurgical
    microscope after a sample has been sectioned and polished. it's not to scale because it doesn't
    need to be - it simply illustrates the nature of some stress corrosion cracking following along
    grain boundary lines in the metal. [grains are individual metal crystals just like individual cells
    in biological samples.]

    this is relevant to your pics because they show a polycrystalline fracture surface where the metal
    has separated at grain boundaries.

    >
    >> your pics are not perfect, and may well show an element of fatigue under closer examination, but
    >> the bright crystalline intergranular nature of the bulk of the fracture surface is clear. it's
    >> not uncommon for failures to combine stress corrosion, corrosion and fatigue, so it could easily
    >> be a combination.
    >
    >
    > Suggestions for better pics? More magnification, diffuse light source,....? I have access to a
    > digital camera, optical microscopes that range from 8-40x, and some garden-variety light sources.
    > That's what I used for these pictures.
    >
    they need to be in focus! if you want to determine the extent of any fatigue, you need to have the
    fracture surface in focus, particularly that inside edge, which will require having the plane of the
    fracture parallel to the optical focus plane. the ideal light source is reflected from the objective
    lens itself - some biological microscopes have this feature, but most use transmitted light so
    you'll have to do the best you can with a source as close and as perpendicular as possible.

    if the cracks initiated with fatigue from the inside of the rim, there will be a comparatively
    smooth "beach mark" surface extending from that inside edge into the wall thickness. you may need
    significantly more magnification than 40x to detect this if the matreial has gone straight from the
    slow growth region to brittle fracture, but have a go and see what you can do.
     
Loading...
Loading...