Update on frame failure

Chalo says...

Quoted message said:

On close examination, I was also unable to find a tire mark, paint
scuff, or any other marking at the site of the crease in the downtube.

The bike didn't hit anything except me, after I had hit the ground .
It came to rest with my legs all tangled up in it, and it took me
probably half a minute to extract myself.

Quoted message said:

Given your weight, and the suspension forks, and the fact that you were
intentionally conducting a "hard braking test" with newly mounted disk
brakes, it seems possible that the forks took a sudden downward dive to the
bottom of their travel, sending your weight forward and over the bars (it's
a fairly common "over the bars" scenario in mountain biking accidents).

I've bottomed out a lot of suspension forks,and none of them make the
"crack" sound that preceded me getting tossed. These forks, Marzocchi
MX Comp 29", I never managed to bottom out at all, despite their
insufficient spring rate.

I was trying to wear in the new pads, plus see if the improved braking
would make anything bend from the added force, so I was stopping from
30-35mph on a steep downhill. I didn't make a sudden jerk at the
brakes, so there was no sudden plunge to the bottom of the fork's
travel. I had been hard on the brakes for about a full second before
the "crack", so the fork had already sunk as low as it was going to
go.

Chalo Colina

I also heard a resounding crack when I broke an aluminum frame last
year. It was surprisingly loud. I think suspension forks are much
flexier than rigid forks, especially the ones with 30mm or less
stanchion diameter. I think I see MX comp on these forks, which I
believe are 30mm, but even my 32mm Fox isn't nearly as stiff as a rigid
fork. I know you said the fork was compressed and this should take away
the excess flex, but I strongly doubt that would make it equivalent to a
rigid fork. Maybe you didn't find any tire marks on the frame because
they didn't rub together. When the tire and frame met, it instantly
locked the wheel and the frame and tire moved as one until the weight
was off.

"Steven M. Scharf" <[email hidden]> wrote

Quoted message said:

"Chalo" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

I have been advised in earnest by the Cannondale dealer that I should
use the original steel unicrown fork, with a rim brake, when I resume
the use of this bike.

How about forget the use of this bike and move to a frame that is less
likely to break, i.e. a steel frame.

I have had a lot better luck over the last 17 years with Cannondale
frames than with any other manufaturer's frame. I've had steel frames
by Raleigh, Schwinn, Maruishi, Specialized, Ross, Surly, Dyno, and
Nishiki. Some have held up pretty well; others have failed in as
little as one ride-- but none have offered me the combination of
stiffness, light weight, and durability that Cannondale frames have.

When I bought a made-to measure custom, I commissioned a steel one
from David Bohm of Tucson AZ. In order to make it suitable for my
size and weight, he didn't use any conventional frame tube its
intended location, except for the seat stays-- for the top tube he
used a downtube; for the downtube he used a tandem boom tube; for
chainstays he used fork blades. I think that is as good an
illustration as any why a commercial steel frame is unlikely to be a
good choice for me.

Chalo Colina

Dan Burkhart <[email hidden]>

Quoted message said:

Quoted message said:

http://www.uploadimages.net/images/154121bohmside.JPG

A Rohloff hub in a 48 spoke wheel? Rohloff tells me they only do 32
spoke drilling. This must be a custom job.
Dan Burkhart

I used a CNC vertical mill to add 16 extra spoke holes to the right
flange and 8 to the left. I then laced a modified crow's-foot pattern
on a 48 hole rim.

Chalo Colina

"Dan" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

"jim beam" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Peter Cole said:

"Steven M. Scharf" <[email hidden]> wrote in message
news:[email hidden]...

>"Chalo" <[email hidden]> wrote in message
>news:[email hidden]...
>
>
>>I have been advised in earnest by the Cannondale dealer that I

should

Quoted message said:

Quoted message said:

Quoted message said:

>>use the original steel unicrown fork, with a rim brake, when I

resume

Quoted message said:

Quoted message said:

Quoted message said:

>>the use of this bike.
>
>How about forget the use of this bike and move to a frame that is

less

Quoted message said:

Quoted message said:

Quoted message said:

>likely to break, i.e. a steel frame.

Actually, steel frames are *more* likely to fail from overload. Since
aluminum has no fatigue limit, it has to be somewhat overbuilt to

have

Quoted message said:

Quoted message said:

Quoted message said:

adequate fatigue life. This means it can take higher peak loads.

good! go to the top of the class.

I don't agree. Assuming that "overload" refers to any non-fatigue

structural

Quoted message said:

failure...

Aluminum has about 1/3 the stiffness (modulus of elasticity) of steel and
about the same strength (yield). To compensate for its low stiffness,
aluminum is usually formed into larger diameter thin tubes, which are

more

Quoted message said:

likely to fail in local buckling because of the low modulus of elasticity
and the thinner tube wall in relation to the tube diameter. I don't

believe

Quoted message said:

that in this case the failure was a material strength or fatigue

failure,

Quoted message said:

it was a stiffness or buckling failure. Recall that Euler's solution to
elastic columns is a function of material and section stiffness - not
material strength or fatigue properties. I also believe that a sudden
buckling of the tube wall could produce a "crack" or "snap" sound.

You're presuming Euler buckling, that's conjecture on your part. It remains
to proven if that's how this frame failed. In any case, I wasn't
specifically referring to that mode of failure in my comments about
aluminum being overbuilt, just to the general case of bending loads. To
avoid early fatigue, working stress must be kept lower in aluminum than
steel, the inescapable consequence of this is that yield points are higher.

Chalo said:

Tom Sherman said:

Ryan Cousineau said:

Is there a particular reason for choosing Al as the material of choice?
I mean, for that price there's a certain Ti builder who might do the
deed, and I think he does custom work too...

http://www.habcycles.com/mtb.html

Very funny, Mr. Cousineau!

I have an economic embargo on businesses whose owners give support to
terrorists. ;^)

We could also start a flame-war over the subject of buying products made
in totalitarian countries.

--
Tom Sherman

"Jose Rizal" <_@_._> wrote in message
news:[email hidden]...

Quoted message said:

jim beam:

Quoted message said:

Peter Cole said:

"Steven M. Scharf" <[email hidden]> wrote in message
news:[email hidden]...

>"Chalo" <[email hidden]> wrote in message
>news:[email hidden]...
>
>
>>I have been advised in earnest by the Cannondale dealer that I

should

Quoted message said:

Quoted message said:

Quoted message said:

>>use the original steel unicrown fork, with a rim brake, when I

resume

Quoted message said:

Quoted message said:

Quoted message said:

>>the use of this bike.
>
>How about forget the use of this bike and move to a frame that is

less

Quoted message said:

Quoted message said:

Quoted message said:

>likely to break, i.e. a steel frame.

Actually, steel frames are *more* likely to fail from overload. Since
aluminum has no fatigue limit, it has to be somewhat overbuilt to

have

Quoted message said:

Quoted message said:

Quoted message said:

adequate fatigue life. This means it can take higher peak loads.

good! go to the top of the class.

Which class is that?

Unless you have dimensions to compare structures built with both
materials, you're both in lala land

I think you missed the point, care to expand?

jim beam said:

[email hidden] wrote:

Quoted message said:

Quoted message said:

Now that you mention it, this could have been caused by a shoe sole or
plastic cleat, although I don't see that cleats were in use. However,
the "tangled up in the bicycle" scenario is an important clue. In any
case, I am certain that the damage on the downtube is not from
overload but rather from an impact of a foreign object.

Jobst Brandt
[email hidden]

translation: "don't know so i'll make something up that can only be
verbally contradicted".

it's buckling. get that empty beer can out and prove it to yourself.

If Chalo was doing a maximal braking test (which he was), the front
wheel was very nearly locked up. Some event prompted the fork to bend
backward, and this would have allowed the front wheel to come to a
complete stop for the split second it took the tire to reach the down
tube, where it would have initiated the buckling of the frame (which
wouldn't need any more force from the tire to proceed).

This could explain how the front tire could have been the "foreign
object" without leaving a telltale mark on the down tube.

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

Mark Hickey said:

jim beam said:

[email hidden] wrote:

Quoted message said:

Quoted message said:

Now that you mention it, this could have been caused by a shoe sole or
plastic cleat, although I don't see that cleats were in use. However,
the "tangled up in the bicycle" scenario is an important clue. In any
case, I am certain that the damage on the downtube is not from
overload but rather from an impact of a foreign object.

Jobst Brandt
[email hidden]

translation: "don't know so i'll make something up that can only be
verbally contradicted".

it's buckling. get that empty beer can out and prove it to yourself.

If Chalo was doing a maximal braking test (which he was), the front
wheel was very nearly locked up. Some event prompted the fork to bend
backward, and this would have allowed the front wheel to come to a
complete stop for the split second it took the tire to reach the down
tube, where it would have initiated the buckling of the frame (which
wouldn't need any more force from the tire to proceed).

This could explain how the front tire could have been the "foreign
object" without leaving a telltale mark on the down tube.

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

it's possible, but i doubt it. my view is that described by ed cory.
the suspension suddenly bottoming out causes an impulse in excess of
other braking forces. that with a 300lb rider? the reason crumple
zones on cars save lives is because they reduce the magnitude of the
crash impulse. it's still the same crash so the same total energy is
expended, but the duration over which the crash impulse ocurrs is what
affects the force experienced.

Dan said:

"jim beam" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Peter Cole said:

"Steven M. Scharf" <[email hidden]> wrote in message
news:[email hidden]...

>"Chalo" <[email hidden]> wrote in message
>news:[email hidden]...
>
>
>
>>I have been advised in earnest by the Cannondale dealer that I should
>>use the original steel unicrown fork, with a rim brake, when I resume
>>the use of this bike.
>
>How about forget the use of this bike and move to a frame that is less
>likely to break, i.e. a steel frame.

Actually, steel frames are *more* likely to fail from overload. Since
aluminum has no fatigue limit, it has to be somewhat overbuilt to have
adequate fatigue life. This means it can take higher peak loads.

good! go to the top of the class.

I don't agree. Assuming that "overload" refers to any non-fatigue structural
failure...

Aluminum has about 1/3 the stiffness (modulus of elasticity) of steel and
about the same strength (yield).

that must be a typo - you can't say that a frame tube steel with a yield
of 1000N/mm2 has "about the same strength" as a 7005 aluminum frame tube
with a yield of 400N/mm2. even the exotic lithium-aluminum alloys have
trouble getting to 700/mm2. steel can easily get to 1300N/mm2 or more
if it's air hardening.

Quoted message said:

To compensate for its low stiffness,
aluminum is usually formed into larger diameter thin tubes, which are more
likely to fail in local buckling because of the low modulus of elasticity
and the thinner tube wall in relation to the tube diameter. I don't believe
that in this case the failure was a material strength or fatigue failure,
it was a stiffness or buckling failure.

kinda. read my posts - we both agree that it's buckling. i never said
fatigue.

Quoted message said:

Recall that Euler's solution to
elastic columns is a function of material and section stiffness - not
material strength or fatigue properties.

to paraphrase: "...solution ... is a function of material... - ...not
material strength..." you need to clarify.

Quoted message said:

I also believe that a sudden
buckling of the tube wall could produce a "crack" or "snap" sound.

again, we agree on that.

"Peter Cole" <[email hidden]> wrote in message
news:h1Jld.95858$R05.65305@attbi_s53...

Quoted message said:

You're presuming Euler buckling, that's conjecture on your part. It

remains

Quoted message said:

to proven if that's how this frame failed. In any case, I wasn't
specifically referring to that mode of failure in my comments about
aluminum being overbuilt, just to the general case of bending loads. To
avoid early fatigue, working stress must be kept lower in aluminum than
steel, the inescapable consequence of this is that yield points are

higher.

Quoted message said:

No I wouldn't assume an Euler model but Euler does demonstrate the
importance of E. If you look up buckling of thin cylindrical columns you
will find that there are no good theoretical models for predicting failure*.
Empirical models for critical stress of thin cylindrical tubes usually have
the form KE where E is the Modulus of elasticity and K is dependant on the
dimensionless ratio of radius to wall thickness (e.g. Roark & Young).
Anyway, the capacity is proportional to E. I referenced Euler simply to stay
away from a discussion of advanced mechanics of materials.

I still suspect that the combined stress (bending and compression) exceeded
the critical buckling stress.

*This is generally true for non-Euler columns as well. For non-Euler
columns, Johnson's equation is the most practical predictor but it is
empirical. In actual practice, very few columns are slender enough to be
modeled by Euler's equation.

"Mark Hickey" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

If Chalo was doing a maximal braking test (which he was), the front
wheel was very nearly locked up. Some event prompted the fork to bend
backward, and this would have allowed the front wheel to come to a
complete stop for the split second it took the tire to reach the down
tube, where it would have initiated the buckling of the frame (which
wouldn't need any more force from the tire to proceed).

This could explain how the front tire could have been the "foreign
object" without leaving a telltale mark on the down tube.

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

I still doubt that the tire contacted the tube. When I load the profile
photo into AutoCAD and scale things, I find that the tire is about 1.5
inches from the tube at full fork compression and that the tire will not
contact at the crease even if the fork and head tube bend far enough back
for the tire to contact the down tube. Also, if the head tube and fork flex
backward, the portion of down tube above the crease rotates up and farther
from the tire - the gap remains fairly constant.

The simple answer is often the correct answer - the tube buckled locally
under combined stress. The bike is fairly new and in good shape so I doubt
pre-existing damage at the crease. A large enough object getting caught
between the tire and down tube seems unlikely to go unnoticed. The sudden
buckling made a sharp sound - just like a buckling beer can.

jim beam says...

Quoted message said:

it's possible, but i doubt it. my view is that described by ed cory.
the suspension suddenly bottoming out causes an impulse in excess of
other braking forces. that with a 300lb rider? the reason crumple
zones on cars save lives is because they reduce the magnitude of the
crash impulse. it's still the same crash so the same total energy is
expended, but the duration over which the crash impulse ocurrs is what
affects the force experienced.

I think it happened just like Chalo said it did. The frame buckled under
the load and the wheel came back and hit the frame. For the instant that
the frame was buckling, the tire wasn't rotating. Then it hit the frame.
That's why there aren't any tire marks on the frame, because there
wasn't any significant rubbing.

"jim beam" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

that must be a typo - you can't say that a frame tube steel with a yield
of 1000N/mm2 has "about the same strength" as a 7005 aluminum frame tube
with a yield of 400N/mm2. even the exotic lithium-aluminum alloys have
trouble getting to 700/mm2. steel can easily get to 1300N/mm2 or more
if it's air hardening.

I stand corrected, I hastily stated a generality based on the popular
assumption I encounter that steel is "stronger" than aluminum. Both
materials are available in a wide range of overlapping yield and ultimate
strengths. What doesn't vary much is the modulus of elasticity. Generalities
are generally wrong when applied to a specific case (logic joke).

Quoted message said:

Quoted message said:

To compensate for its low stiffness,
aluminum is usually formed into larger diameter thin tubes, which are

more

Quoted message said:

Quoted message said:

likely to fail in local buckling because of the low modulus of

elasticity

Quoted message said:

Quoted message said:

and the thinner tube wall in relation to the tube diameter. I don't

believe

Quoted message said:

Quoted message said:

that in this case the failure was a material strength or fatigue

failure,

Quoted message said:

Quoted message said:

it was a stiffness or buckling failure.

kinda. read my posts - we both agree that it's buckling. i never said
fatigue.

We agree and I don't think you said it was fatigue.

Quoted message said:

Quoted message said:

Recall that Euler's solution to
elastic columns is a function of material and section stiffness - not
material strength or fatigue properties.

to paraphrase: "...solution ... is a function of material... - ...not
material strength..." you need to clarify.

I said "Recall that Euler's solution to elastic columns is a function of
material and section stiffness..." I should have phrased it: material
stiffness and section stiffness. Material stiffness is E the modulus of
elasticity. Section stiffness in this case is a function of tube radius and
wall thickness.

Quoted message said:

Quoted message said:

I also believe that a sudden
buckling of the tube wall could produce a "crack" or "snap" sound.

again, we agree on that.

It seems that we agree on all of it.

Mark Hickey said:

jim beam said:

[email hidden] wrote:

Quoted message said:

Quoted message said:

Now that you mention it, this could have been caused by a shoe sole
or plastic cleat, although I don't see that cleats were in use.
However, the "tangled up in the bicycle" scenario is an important
clue. In any case, I am certain that the damage on the downtube is
not from overload but rather from an impact of a foreign object.

Jobst Brandt
[email hidden]

translation: "don't know so i'll make something up that can only be
verbally contradicted".

it's buckling. get that empty beer can out and prove it to yourself.

If Chalo was doing a maximal braking test (which he was), the front
wheel was very nearly locked up. Some event prompted the fork to bend
backward, and this would have allowed the front wheel to come to a
complete stop for the split second it took the tire to reach the down
tube, where it would have initiated the buckling of the frame (which
wouldn't need any more force from the tire to proceed).

This could explain how the front tire could have been the "foreign
object" without leaving a telltale mark on the down tube.

The tire would have been rotating somewhat, even if going at a slow rate.
Any relative movement of the tire and downtube would have left some kind of
mark. Tire marks on clearcoated paint usually leave a rubber streak that is
easily rubbed off with some elbow grease. Chalo hasn't mentioned anything
about marks directly, but he has responded to a comment concerning the lack
of marks without mentioning the marks themselves.

If the tire wasn't moving and Chalo was, then the coefficient of friction
would have gone from static to kinetic, and the wheel would have slid
forward, away from the downtube.

I think a close-up of the downtube exterior and interior, and/or a
cross-section of the downtube, would be really helpful.

--
Phil, Squid-in-Training

Quoted message said:

Just as an observation, quite a few years ago I buckled a steel frame
by running into a rider who had fallen right in front of me. It was at
fairly slow speed (going uphill) and my front wheel struck him
squarely, bringing my bike to an nearly instant stop. I went over the
bars, but neither I nor the other rider suffered any injuries. It was
not until some time after I had picked myself up and continued on that
I noticed the peculiar handling of my bike.

Was this a road bike or a mountain bike with an un-suspension-corrected
fork? Suspension forks put a lot of distance between the tire and downtube.

--
Phil, Squid-in-Training

Quoted message said:

Quoted message said:

Quoted message said:

Ryan Cousineau wrote:
>Is there a particular reason for choosing Al as the material of choice?
>I mean, for that price there's a certain Ti builder who might do the
>deed, and I think he does custom work too...
>http://www.habcycles.com/mtb.html

Quoted message said:

Quoted message said:

Tom Sherman said:

Very funny, Mr. Cousineau!

Quoted message said:

(Chalo) said:

I have an economic embargo on businesses whose owners give support to
terrorists. ;^)

Ryan Cousineau wrote:
-snip-

Quoted message said:

Mark has shown a general (and reasonable) reluctance to enter into most
discussions which refer directly to his company in this way, and I think
this is a generally sensible policy, since it avoids excessive
self-plugging, and allows him to productively discuss bike tech here
without unseemly self-promotion.

As I once pointed out in defending Jobst, the alternative to the prickly
Jobst we all know and love is probably not a kinder, gentler Jobst, it's
that Jobst will stop posting in this newsgroup. Similarly, I don't think
the alternative to the Mark Hickey we all know and love is a Mark who
suddenly sees the light and whips out a Nader-themed bicycle ("designed
according to safety principles laid down by Ralph himself!"😉, it's a
Mark who stops posting here because it's more trouble than joy.

I don't think that's a positive outcome.

A well considered piece, Ryan.

I find myself frequently pointing out in my store that
Michelin doesn't formulate policy for the French State
Department. Mr Bush doesn't consult Mark. Or me.

--
Andrew Muzi
www.yellowjersey.org
Open every day since 1 April, 1971

Chalo said:

Blair P. Houghton said:

Dan said:

I agree. The fork isn't the problem.

It is. It's too long until it bottoms out.

By the time the fork is seeing a large braking load, it _is_ almost
bottomed, and no longer than a rigid fork. (Suspension forks are
usually shorter than rigid unicrowns when bottomed out.)

Are you saying the weight alone has it bottomed-out?

--Blair
"Then what's it for?"

"Phil, Squid-in-Training" <[email hidden]>

Quoted message said:

The tire would have been rotating somewhat, even if going at a slow rate.
Any relative movement of the tire and downtube would have left some kind of
mark. Tire marks on clearcoated paint usually leave a rubber streak that is
easily rubbed off with some elbow grease. Chalo hasn't mentioned anything
about marks directly, but he has responded to a comment concerning the lack
of marks without mentioning the marks themselves.

I think I mentioned in an earlier post that there were no marks I
could see. I thought there may have been a little darkening on first
inspection, but a closer examination revealed nothing. That is the
chief reason I suspect that contact might not have occurred, because
the way in which I was pitched over definitely led me to believe the
tire had jammed against the frame.

Quoted message said:

I think a close-up of the downtube exterior and interior, and/or a
cross-section of the downtube, would be really helpful.

Cannondale took the frame back for analysis. Pardo's photos are all I
can offer.

I suppose I could try to contact the company and ask them what they
think happened. I will look into it.

Chalo Colina

"Phil, Squid-in-Training" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

Quoted message said:

Just as an observation, quite a few years ago I buckled a steel frame
by running into a rider who had fallen right in front of me. It was at
fairly slow speed (going uphill) and my front wheel struck him
squarely, bringing my bike to an nearly instant stop. I went over the
bars, but neither I nor the other rider suffered any injuries. It was
not until some time after I had picked myself up and continued on that
I noticed the peculiar handling of my bike.

Was this a road bike or a mountain bike with an un-suspension-corrected
fork? Suspension forks put a lot of distance between the tire and downtube.

Road

DR

Phil, Squid-in-Training says...

Quoted message said:

The tire would have been rotating somewhat, even if going at a slow rate.
Any relative movement of the tire and downtube would have left some kind of
mark. Tire marks on clearcoated paint usually leave a rubber streak that is
easily rubbed off with some elbow grease. Chalo hasn't mentioned anything
about marks directly, but he has responded to a comment concerning the lack
of marks without mentioning the marks themselves.

If the tire wasn't moving and Chalo was, then the coefficient of friction
would have gone from static to kinetic, and the wheel would have slid
forward, away from the downtube.

I think a close-up of the downtube exterior and interior, and/or a
cross-section of the downtube, would be really helpful.

I don't know why you would say that. It seems logical to me that in the
instant after the frame buckled, the brake would lock and tire would
have almost no rotation. The brake would instantly lock because when the
frame let go, there would be no counter force to balance braking force.
I tend to think there were some marks made, but they may be so faint
that only glare from light at the correct angle would show them. The
marks might be no more visible than a fingerprint, probably no more than
a smudge.