4 spokes failed on unridden bike!
Robert Haston
years now; so I've seen quite a bit.
I was cleaning up a couple old bikes we hadn't ridden in a few years,
preparing to donate them. My wife's road bike had a broken spoke, My old
MTB had broken two. All were rear drive-side.
The first job went fine. I hand screwed the spokes into the mountain bike's
wheel and put it in the truing stand. As I was about to start truing it, 4
spokes suddenly snapped! That made 6 spokes on a bike I hadn't ridden! My
only guess was some combination of stress-corrosion cracking (the stainless
spokes had several rusty spots) and over-tightening.
I am rebuilding the wheel with all new spokes (done this before too) but
would sure like to know what happened.
--
Robert Haston
Satellite Beach, FL
jim beam
without seeing your broken spokes, your suggestions are pretty
reasonable on both counts. another thought is that since the wheels
/had/ been ridden in the past, it's quite possible that they were
already fatigued but that your truing efforts raised tension
sufficiently to cause the stress at the fatigue crack tip to exceed
fracture. almost nothing you can do to prevent that. sometimes also
things like vibation & thermal cycling can cause sudden failure of
something that's not in service.
Jose Rizal
> without seeing your broken spokes, your suggestions are pretty
> reasonable on both counts. another thought is that since the wheels
> /had/ been ridden in the past, it's quite possible that they were
> already fatigued but that your truing efforts raised tension
> sufficiently to cause the stress at the fatigue crack tip to exceed
> fracture. almost nothing you can do to prevent that. sometimes also
> things like vibation & thermal cycling can cause sudden failure of
> something that's not in service.
>
Please explain how temperatures high enough to cause destructive thermal
cycling are reached by bicycle spokes. While you're at it, also explain
how vibration amplitudes high enough to cause significant damage (apart
from the loading on the spokes through rider weight) figure into it.
carl...@comcast.net
Dear Jose,
If a pre-tensioned spoke is already about to break, would
the additional contraction and brittleness of a drop to
sub-zero temperatures be enough to let a spoke break on a
bicycle sitting in an unheated garage?
On several cold mornings, I've found a broken rear spoke and
the rim jammed solid against one brake pad after the wheel
had spun freely while I was cleaning the chain the night
before. Typically, the riding temperature the day before was
around 34-40 F, but the overnight low was 0 to -10 F.
Jobst indicated (I think) that a failing spoke can even
break without any apparent added stress:
"Spokes can break with no additional loading once the
developing crack reduces the cross section to the yield
point. Then the crack grows plastically until failure
without change in load or temperature. I have witnessed
spoke failures more than an hour after parking the
bicycle indoors where the occurrence is conspicuous."
My layman's guess is that this means that a tiny crack can
slowly grow while the bike is just sitting there.
Come to think of it, now I'm wondering whose spokes Jobst
witnessed breaking an hour after being parked indoors.
Carl Fogel
Sheldon Brown
name to "Fogey")
> If a pre-tensioned spoke is already about to break, would
> the additional contraction and brittleness of a drop to
> sub-zero temperatures be enough to let a spoke break on a
> bicycle sitting in an unheated garage?
Dear Carl,
No, no way cold could do this, since the rim would contract too. In
fact, an aluminum rim would contract twice as much as the steel spokes,
since aluminum has twice the coefficient of expansion that steel does.
Sheldon "That's Not It" Brown
+------------------------------------------------+
| I’m currently appearing in: |
| Gilbert & Sullivan's Iolanthe at M.I.T. |
| November 12, 13, 14 and 18, 19, 20, 21 |
| http://web.mit.edu/gsp/www |
| http://sheldonbrown.com/music.html |
+------------------------------------------------+
Harris Cyclery, West Newton, Massachusetts
Phone 617-244-9772 FAX 617-244-1041
http://harriscyclery.com
Hard-to-find parts shipped Worldwide
http://captainbike.com http://sheldonbrown.com
Jose Rizal
> On Sun, 14 Nov 2004 02:55:45 GMT, Jose Rizal <_@_._> wrote:
>
> >jim beam:
> >> sometimes also
> >> things like vibation & thermal cycling can cause sudden failure of
> >> something that's not in service.
> >>
> >
> >Please explain how temperatures high enough to cause destructive thermal
> >cycling are reached by bicycle spokes. While you're at it, also explain
> >how vibration amplitudes high enough to cause significant damage (apart
> >from the loading on the spokes through rider weight) figure into it.
>
> Dear Jose,
>
> If a pre-tensioned spoke is already about to break, would
> the additional contraction and brittleness of a drop to
> sub-zero temperatures be enough to let a spoke break on a
> bicycle sitting in an unheated garage?
Thermal cycling means a repetitive load, not a load applied once to a
weakened structure that causes breakage.
A normal spoke under tension that is subjected to climatic sub-zero and
warm daytime temperatures will not experience high enough thermal cyclic
amplitudes to cause failure by itself.
If the case is that a spoke is about to break due to existing tension,
it means the load-bearing cross-section of the spoke has been reduced to
below that required to support the load. The reduced cross section will
keep reducing as the crack propagates until the spoke breaks. The
tension on a spoke is orders of magnitudes larger than any contraction
that sub-zero weather temperatures will cause. Climatic sub-zero
temperatures are also not low enough to cause "additonal brittleness" to
a stainless steel spoke.
> On several cold mornings, I've found a broken rear spoke and
> the rim jammed solid against one brake pad after the wheel
> had spun freely while I was cleaning the chain the night
> before. Typically, the riding temperature the day before was
> around 34-40 F, but the overnight low was 0 to -10 F.
It's doubtful that temperature due to weather by itself caused those
spoke failures. You need to look at the quality of the wheel build.
> Jobst indicated (I think) that a failing spoke can even
> break without any apparent added stress:
>
> "Spokes can break with no additional loading once the
> developing crack reduces the cross section to the yield
> point. Then the crack grows plastically until failure
> without change in load or temperature. I have witnessed
> spoke failures more than an hour after parking the
> bicycle indoors where the occurrence is conspicuous."
>
> http://groups.google.com/groups?q=%22spokes+can+break+with+no+additional+loading%22&hl=en&lr=&group=rec.bicycles.tech&scoring=d&selm=FDDbd.17322%2454.291551%40typhoon.sonic.net&rnum=1
Which is what I stated above. Note again the part of your quote where
it says "without change in load or temperature". This has nothing to do
with thermal cycling.
> My layman's guess is that this means that a tiny crack can
> slowly grow while the bike is just sitting there.
Yes, when it's reached the point of failure. But again, nothing to do
with thermal cycling.
carl...@comcast.net
<capt...@sheldonbrown.com> wrote:
>Quoth Carl Fogel: (For some reason my spell checker wants to change his
>name to "Fogey")
>
>> If a pre-tensioned spoke is already about to break, would
>> the additional contraction and brittleness of a drop to
>> sub-zero temperatures be enough to let a spoke break on a
>> bicycle sitting in an unheated garage?
>
>Dear Carl,
>
>No, no way cold could do this, since the rim would contract too. In
>fact, an aluminum rim would contract twice as much as the steel spokes,
>since aluminum has twice the coefficient of expansion that steel does.
>
>Sheldon "That's Not It" Brown
Dear Sheldon,
Hmmm . . . I think that you're right, but using the wrong
figures.
Unlike sluggish ordinary steels, stainless steel has roughly
the same coefficient of thermal expansion as lively
aluminum, a ratio about about 11 to 12 (not 1 to 2).
Where X is the coefficient of thermal expansion (inch/inch F
10-6) . . .
Stainless steel (as used in spokes) is around 11X:
http://www.engineeringtoolbox.com/14_283.html
You have to scroll down to see it. Nearby is a link to this
table:
http://www.engineeringtoolbox.com/24_95.html
. . . which shows the thermal expansion for aluminum (as
used in rims) is about 12X (and ordinary steel about 7x).
I'm not sure if it matters when the temperature drops that
the aluminum rim is already shrunk (in compression) by the
elongated stainless steel spokes (in tension).
But at an 11/12 ratio, I suspect that the small difference
in what would be a tiny contraction would end up negligible.
Reassuringly, "Stainless steel (type 304 or 316) does not
exhibit any ductile-to-brittle transition at low
temperatures (even down to –400°F), and is an excellent
material for frigid temperatures."
www.b-line.com/engineer/TPS/BLTPS-08.pdf
So maybe it's just coincidence that several spokes
apparently failed during unusually cold nights. I hope so,
since it's snowing lightly right now.
Carl Fogel
carl...@comcast.net
<capt...@sheldonbrown.com> wrote:
>Quoth Carl Fogel: (For some reason my spell checker wants to change his
>name to "Fogey")
>
>> If a pre-tensioned spoke is already about to break, would
>> the additional contraction and brittleness of a drop to
>> sub-zero temperatures be enough to let a spoke break on a
>> bicycle sitting in an unheated garage?
>
>Dear Carl,
>
>No, no way cold could do this, since the rim would contract too. In
>fact, an aluminum rim would contract twice as much as the steel spokes,
>since aluminum has twice the coefficient of expansion that steel does.
>
>Sheldon "That's Not It" Brown
Dear Sheldon,
Drat!
After agreeing with you (but pointing out that stainless
steel actually shrinks at almost the same rate as aluminum),
I was suddenly troubled by the vague memory that Jobst has
mentioned that cast control wheels (whose rims and spokes
have identical material) have wiggly spokes because
otherwise they'd crack when they cooled at different rates
and shrank:
"Now consider that die castings have cooling stresses caused
by sequential cooling and that the thicker part of the
spokes near the hub and the hub itself cool last causing
shrink. This puts the die cast spokes in tension. That
castings do this is known and that is why S-shaped spokes
are used on cast iron hand wheels of old to prevent cracks
that occur if they were not S-shaped. All cast railway
wheels also have such ribs on the inside."
http://groups.google.com/groups?selm=7gvpcn%24t0u%242%40hplms2.hpl.hp.com&output=gplain
Now I need to be reassured that the thin spokes with their
huge surface area don't cool and contract significantly
faster than the fat rims insulated by pneumatic tires.
This was a lot simpler with wooden ox-carts.
Befuddled,
Carl Fogel
Mike Beauchamp
somewhere, and then bring them into a different temperature environment to
start truing them?
Just a weird guess
"Robert Haston" <reha...@earthlink.net> wrote in message
news:vixld.10329$_J2....@newsread2.news.atl.earthlink.net...
Weisse Luft
Might the bikes been exposed to fumes of hypochlorite? Just a leaky
bottle of ordinary bleach would be enough over time to kill 300
stainless steel if the pH goes low. In solution with water, presence
of the Na ion keeps the pH high enough to prevent Cl induced SCC. ClO
vapors without the Na ion allows the pH to drop, attacking austinitic
SS, of which spokes are made. With the oxygen free radical, chlorine
induced SCC can happen at much lower temperatures like that of a garage
or store room during the summer.
I have seen SS spokes as brittle as uncooked spaghetti from ClO vapor
after a few months of storage near calcium hypochlorite pool
chemicals.
Just my materials experience as applied to cycling...
--
Weisse Luft
carl...@comcast.net
<capt...@sheldonbrown.com> wrote:
>Quoth Carl Fogel: (For some reason my spell checker wants to change his
>name to "Fogey")
>
>> If a pre-tensioned spoke is already about to break, would
>> the additional contraction and brittleness of a drop to
>> sub-zero temperatures be enough to let a spoke break on a
>> bicycle sitting in an unheated garage?
>
>Dear Carl,
>
>No, no way cold could do this, since the rim would contract too. In
>fact, an aluminum rim would contract twice as much as the steel spokes,
>since aluminum has twice the coefficient of expansion that steel does.
>
>Sheldon "That's Not It" Brown
Dear Sheldon,
And drat again!
I belatedly remembered specific heat, have a vague notion
that it should affect the cooling rate,and wondered if it
varies between aluminum and stainless steel.
Aluminum has a specific heat of around 0.9x, while steel has
a specific heat of only around 0.5x:
http://www.engineeringtoolbox.com/24_152.html
. . . where x = (kJ/kg.K, which means (I think) that a chunk
of aluminum absorbs about 80% more heat than a chunk of
steel, so the steel chunk would cool off a lot quicker.
Given 36 2.0 mm thick spokes about 300 mm long, my steel
spokes have a surface area of roughly 34,000 mm^2 and weigh
about 160 grams, according to the DT site (283 grams/64
pieces), so the stainless steel spokes have about 210 mm^2
of surface area to cool each gram that holds only 0.5 heat
units.
My exposed rim surface is about 1.25 inches wide and about
80 inches around, so its about 100 inches^2, or 64,000 mm^2
of surface area and weighs around 600 grams, so the rim has
about 106 mm^2 of surface area to cool each gram of aluminum
that holds 0.9 heat units.
So I think that the spokes are going to cool off about 3-4
times as fast as the rim, though they contract at almost the
same rate. Maybe it won't matter because the cooling is so
slow?
Carl Fogel
carl...@comcast.net
Dear Weisse,
Anything is possible, but the only bleach that I know of
near my garage is inside the house on the other side of
several doors in a closet.
(However, it occurs to me that I could glare suspiciously at
my neighbors, who have an indoor swimming pool, if they
weren't such nice people.)
Lke all my spokes, they broke at the elbow--I've never had
the pleasure of seeing a spoke break at the nipple, although
I gather that about one out of eight to ten spokes break
there in testing.
Despite my calculations nearby about thermal coefficients,
surface area, and latent heat, I'll probably settle for the
unsettling notion that a few spokes just happened to snap
during a few different cold snaps. Even if I could get a
thermal image camera to show the spokes cooling faster than
the rim, I suspect that any actual tension increase just
wouldn't amount to enough to be significant--though I still
want reassurance from some cold-city posters or people who
enjoy thermodynamics.
Carl Fogel
Phil, Squid-in-Training
> stress-corrosion cracking (the stainless spokes had several rusty
> spots) and over-tightening.
I have had very low quality stainless steel spokes that could be bent about
5-10 degrees before brittly *snapping* like an eggshell at certain points.
These exhibited the same "slight rust spot" corrosion and alloy
contamination that you seem to have been having with your wheel.
--
Phil, Squid-in-Training
jim beam
it's all about fracture mechanics. even a 1 degree change causes
expansion & contraction, & therefore stress. if a crack tip has a
stress concentration approaching its fracture point, then the stress
associated with a normal day/night temperature cycle can do it.
same for vibration. we don't have to be talking earthquake here.
example: i was in an historic vehicle restoration workshop a few years
back. there was a large military vehicle in a corner that had stood
unmoved for years. quite suddenly, there was a "crack" & a crunch as
one of its suspension torsion bars failed & the vehicle dropped. no one
was near and the vehicle had not moved or run in over 3 years. when we
examined the failure, it was just a normal fatigue fracture surface,
normal beach marks followed by brittle fracture. the crack had grown to
the point where tip stress was critical, and that morning was just
either one thermal cycle or one passing vehicle vibration too many.
if your question is implying that it takes significant temperature
cycling or vibration to initiate a crack, then you are correct. but
once a crack is present, it's growth accelerates as crack tip stress
concentration grows. the operation of this vehicle was extreme,
sufficient to initiate & propagate cracking to within an ace of failure.
all that needed to happen later was for it to sit about quietly
growing very very slowly to the tipping point.
"classic" example of thermal cycling fatigue: some 1980's 3-series
bmw's had head bolts with a small radius transition between the head &
the shaft. as these bolts were torqued to 98% of yield in the first
place, the normal thermal cycle of operation had the bolt operating
pretty much at yield, ie. the absolute worst point for rapid fatigue.
these bolt heads would fatigue off because of the stress concentration
at the small radius, then, if the owner was lucky, the head gasket would
go. if they were unlucky, the bolt head would float underneath the cam
shaft and get punched through the head by a cam lobe.
Weisse Luft
Forget about the thermal effects, stainless steel does not go through
brittle transition at even cryogenic temperatures.
The key is the rust spots. Stainless steel is a mix of ~18% chromium,
~8% nickel and the rest is iron. Rustable iron. Normally the chromium
"surrounds" the iron, preventing oxidation of the iron but in certain
cases, the chromium surrounding the iron crystals can become
"depleted". High temperatures associated with welding can cause growth
of chromium carbides, depleting the Cr-Fe boundaries of protection and
causing iron corrosion.
Likewise, chlorine can cause the same type of corrosion, only along the
grain boundaries instead of a heat affected zone. This causes the
frature surface to appear very rough, compared to the knife edge attack
seen by chromium carbide formation as a result of heat.
Spokes are highly cold-worked which leaves the residual stress levels
very high. Any contamination from chlorine, especially the ClO
radical, will find easy attack. It doesn't take much ClO and its
entirely possible you wouldn't find it via electron microscopy directed
x-ray diffraction elemental analysis of the fracture surface.
--
Weisse Luft
jim beam
> Forget about the thermal effects, stainless steel does not go through
> brittle transition at even cryogenic temperatures.
whoa. where do you get all this extreme temperature stuff from???
you're reading /way/ more into my words than i intend. maybe i'm not
communicating effectively, but i do /NOT/ mean anything that's either
extreme thermal cycling fatigue or phase transition [which is the
purpose of cryogenic treatments]. ok?
and what's the hard concept about thermal stress anyway? have you ever
put a bearing in fridge so it's easier to fit into an interference
housing? works a treat. that's hardly extreme.
also, where have i mentioned brittle transition? if you want to get
into that, you need to qualify your statement about "stainless steel"
and limit it to low strength austenitic stainless. brittle transition
is evident in most bcc materials - it's not unique to ferritic steel.
>
> The key is the rust spots. Stainless steel is a mix of ~18% chromium,
> ~8% nickel and the rest is iron. Rustable iron. Normally the chromium
> "surrounds" the iron, preventing oxidation of the iron but in certain
> cases, the chromium surrounding the iron crystals can become
> "depleted".
weisse, some of your engineering stuff is on target. regrettably, your
materials theory has me scratching my head. maybe i'm not understanding
you correctly, but if you are under the impression that stainless steel
comprises ferritic crystals individually coated in a rust-proof coating
of chrome & nickel, you need to do some more homework. simple pure
austenitic stainless is a single phase solid solution. essentially, the
nickel stabilizes the fcc austenitic structure and the dissolved
chromium provides a passivation effect on exposed surfaces. there's no
"surrounding" phenomenon going on here. you commonly get depletion when
the original material is inferior [in which case it's not really
"depletion"], when there is specific chemical attack, or when there's
another issue like welding.
> High temperatures associated with welding can cause growth
> of chromium carbides, depleting the Cr-Fe boundaries of protection and
> causing iron corrosion.
that's correct, but welded spokes are new to me. and that's why most
stainless is low carbon - to prevent that.
>
> Likewise, chlorine can cause the same type of corrosion, only along the
> grain boundaries instead of a heat affected zone. This causes the
> frature surface to appear very rough, compared to the knife edge attack
> seen by chromium carbide formation as a result of heat.
again, where does welding come into spoke manufacture? you're
absolutely correct that chlorides are a classic stress corrosion agent
for austenitic stainless, but i wouldn't understand that from the above
explanation.
>
> Spokes are highly cold-worked which leaves the residual stress levels
> very high. Any contamination from chlorine, especially the ClO
> radical, will find easy attack. It doesn't take much ClO and its
> entirely possible you wouldn't find it via electron microscopy directed
> x-ray diffraction elemental analysis of the fracture surface.
spokes are highly cold worked which means their dislocation densities
are high and their lattice energies are high. that does not mean the
residual stress is high. residual stress is mitigated by further low
strain working, thermal treatments, or even aging. these do /not/
reduce dislocation density significantly. chances are that if a spoke
/does/ have significant residual stress, it results from "correcting the
spoke line" rather than its condition leaving its factory.
maybe electron microscopy can't detect the scc agent, but it usually
can. depends on the specific scc system, the gear you have and whether
you know what you're doing.
Tom Sherman
> Forget about the thermal effects, stainless steel does not go through
> brittle transition at even cryogenic temperatures.
>
> The key is the rust spots. Stainless steel is a mix of ~18% chromium,
> ~8% nickel and the rest is iron....
Really? The above quantities are in the range of those for AISI Type 304
Stainless Steel but are certainly not inclusive of all stainless steels.
--
Tom Sherman
Bad Idea
> ...
> Despite my calculations nearby about thermal coefficients,
> surface area, and latent heat, I'll probably settle for the
> unsettling notion that a few spokes just happened to snap
> during a few different cold snaps. Even if I could get a
> thermal image camera to show the spokes cooling faster than
> the rim, I suspect that any actual tension increase just
> wouldn't amount to enough to be significant--though I still
> want reassurance from some cold-city posters or people who
> enjoy thermodynamics.
>
> Carl Fogel
One well designed experiment is worth more than a thousand "expert"
opinions. Someone with a tensiometer could quickly inject enough
empiricism here to bury all the cheap talk.
I know, I know...
carl...@comcast.net
wrote:
Dear BI,
I have a Park tensiometer, but am doubtful that its reading
is accurate enough to be decisive. If a spoke is already
about to fail, an extra pound or two of tension might be
enough to break the camel's back.
And lacking a walk-in freezer, I need to be poised and ready
to spend Lord knows how long one awful winter night, taking
tension readings long into the wee hours as the thermometer
drops from 35~40 F down to 0 to -10.
Once both spoke and rim are at the same temperature, I think
that Sheldon is right and that their coefficient of thermal
expansion is so close that there won't be any significant
difference.
What I'm wondering is if the stainless steel spokes that
start with about half as much specific heat and twice as
much exposed surface area will cool faster than the aluminum
rim, fast enough to make a difference.
My guess is that the cooling would be slow enough that it
would still make no difference, but I still wonder.
Carl Fogel
Robert Haston
conclude:
1. It really is weird to have 4 spokes fail suddenly (on top of two failing
while it hung on the wall).
2. The two prime suspects are long term thermal cycling of highly stressed
spokes, and some form of corrosive induced failure.
3. Corrosion looks like a better suspect. I live in Florida where the
daily temperature cycles in the garage are usually less than 20 degrees, but
the air is salt laden and humid. There are jugs and tablets of pool
chlorine about 20 feet away. That and the rust spots on the spokes were
pretty pronounced.
The good news is a couple packs of spokes and I'll be good again.
"Phil, Squid-in-Training" <phil_leeIHEA...@hotmail.com> wrote in
message news:2tEld.3312$Kk3...@news.flashnewsgroups.com...
A Muzi
-snip-
> The first job went fine. I hand screwed the spokes into the mountain bike's
> wheel and put it in the truing stand. As I was about to start truing it, 4
> spokes suddenly snapped! That made 6 spokes on a bike I hadn't ridden! My
> only guess was some combination of stress-corrosion cracking (the stainless
> spokes had several rusty spots) and over-tightening.
>
> I am rebuilding the wheel with all new spokes (done this before too) but
> would sure like to know what happened.
>
>
noticed a broken spoke. After slipping a new spoke in the
wheel and giving the whole wheel a full turn tighter I
started to true it when two spokes broke simultaneously. By
the time I finished there were several new spokes.
Here's one of the failed spokes:
http://www.yellowjersey.org/photosfromthepast/SPOKRUS2.JPG
http://www.yellowjersey.org/photosfromthepast/SPOKRUS4.JPG
I know the photo quality is poor but the outside of the
curve is cracked and that rusty area goes halfway through.
The inside of the curve is normal steel grey. That implies
that the corrosion permeated from the outside edge. We
usually think of spokes breaking from the inside of the
curve outward, not so here.
--
Andrew Muzi
www.yellowjersey.org
Open every day since 1 April, 1971
Bruce Frech
them all being close to fatigure failure and the first one to fail increased
the tension on the others enough to make them fail quickly.
I would think that if one fails you replace it, but when the second fails
you replace them all. Why use this policy?: because one might fail due to
isolated damage but if more then one fail then they all are likely to be
high risk (usually due to not being stress relieved).
Bruce
Jim Smith
> I have a Park tensiometer, but am doubtful that its reading
> is accurate enough to be decisive. If a spoke is already
> about to fail, an extra pound or two of tension might be
> enough to break the camel's back.
Measuring the frequency of a plucked spoke seems like a good way to go
for this. Use your computer and sound card. Should have better
resolution and repeatability than the park tensiometer.
jim beam
two questions:
what brand of spoke?
were they head in or head out?
Phil, Squid-in-Training
What brand of spoke were they, and how old were they?
--
Phil, Squid-in-Training
A Muzi
>>opinions.
I've many mornings brought my bike from 70F to below zeroF,
ridden it and returned it to 70F without a spoke failure.
I've also left a 70F store, ridden in 95F heat and returned
to AC environment without a spoke failure.
Most of us have similar experiences.
Temperature change might contribute but it isn't sufficient
alone.
Bad Idea
> >
> >One well designed experiment is worth more than a thousand "expert"
> >opinions. Someone with a tensiometer could quickly inject enough
> >empiricism here to bury all the cheap talk.
> >
> >I know, I know...
>
> Dear BI,
>
> ... I think ...
> ...
> My guess is ...
> ...
> Carl Fogel
Of course, long before Empiricism shows its face, Expert Opinion and
Cheap Talk gang up to kneal on the windpipe of Reasonable Suggestion.
I almost forgot that Expert Opinion and Cheap Talk were the founders
of usenet.
A Muzi
>> -snip-
>>
>>> The first job went fine. I hand screwed the spokes into the mountain
>>> bike's wheel and put it in the truing stand. As I was about to start
>>> truing it, 4 spokes suddenly snapped! That made 6 spokes on a bike I
>>> hadn't ridden! My only guess was some combination of
>>> stress-corrosion cracking (the stainless spokes had several rusty
>>> spots) and over-tightening.
>>>
>>> I am rebuilding the wheel with all new spokes (done this before too)
>>> but would sure like to know what happened.
> A Muzi wrote:
>> Today I was changing a flat for a regular customer when i noticed a
>> broken spoke. After slipping a new spoke in the wheel and giving the
>> whole wheel a full turn tighter I started to true it when two spokes
>> broke simultaneously. By the time I finished there were several new
>> spokes.
>> Here's one of the failed spokes:
>> http://www.yellowjersey.org/photosfromthepast/SPOKRUS2.JPG
>> http://www.yellowjersey.org/photosfromthepast/SPOKRUS4.JPG
>>
>> I know the photo quality is poor but the outside of the curve is
>> cracked and that rusty area goes halfway through. The inside of the
>> curve is normal steel grey. That implies that the corrosion permeated
>> from the outside edge. We usually think of spokes breaking from the
>> inside of the curve outward, not so here.
jim beam wrote:
> two questions:
> what brand of spoke?
> were they head in or head out?
It was a 1964 Steyr three speed. Steel hub flanges, 36h
3-cross and the usual Berg-Union chrome spokes.
dianne_1234
wrote:
I tried this. My setup found too many different frequencies to isolate
the spoke's tone, even though my ear could make it out okay.
I just bought a cheap microphone and plugged it in, so probably
someone with more computer or acoustic knowledge could do better.
carl...@comcast.net
<nos...@example.net> wrote:
>Weisse Luft wrote:
[snip]
>> The key is the rust spots. Stainless steel is a mix of ~18% chromium,
>> ~8% nickel and the rest is iron. Rustable iron. Normally the chromium
>> "surrounds" the iron, preventing oxidation of the iron but in certain
>> cases, the chromium surrounding the iron crystals can become
>> "depleted".
[snip]
>maybe i'm not understanding
>you correctly, but if you are under the impression that stainless steel
>comprises ferritic crystals individually coated in a rust-proof coating
>of chrome & nickel, you need to do some more homework. simple pure
>austenitic stainless is a single phase solid solution. essentially, the
>nickel stabilizes the fcc austenitic structure and the dissolved
>chromium provides a passivation effect on exposed surfaces. there's no
>"surrounding" phenomenon going on here. you commonly get depletion when
>the original material is inferior [in which case it's not really
>"depletion"], when there is specific chemical attack, or when there's
>another issue like welding.
[snip]
Dear Weisse and Jim,
At last, an experiment within my technical abilities! I've
already written the opening of what will be my first
scientific paper:
"At the end of the year 2004, I procured me a round plastic
bottle of 6% bleach, to try therewith the celebrated
phenomena of chloride's effect on stainless steel spokes."
The spoke has been sitting in the bleach bottle for about an
hour now. How long should I let it marinate before testing
for brittleness?
Isaac Fogel
Jim Smith
I just gave it a whirl with the built in microphone on a laptop:
I wouldn't try to build a wheel like this, but with a little filter
code and maybe a better mic I suspect this would work for measuring
small relative changes in tension.
Jose Rizal
> Jose Rizal wrote:
> > Please explain how temperatures high enough to cause destructive thermal
> > cycling are reached by bicycle spokes. While you're at it, also explain
> > how vibration amplitudes high enough to cause significant damage (apart
> > from the loading on the spokes through rider weight) figure into it.
>
> it's all about fracture mechanics. even a 1 degree change causes
> expansion & contraction, & therefore stress. if a crack tip has a
> stress concentration approaching its fracture point, then the stress
> associated with a normal day/night temperature cycle can do it.
>
> same for vibration. we don't have to be talking earthquake here.
Compared to the stress due to spoke tension, climatic temperature
changes contribute orders of magnitude less additional stress to the
spoke. This kind of thermal cycling will not contribute to breakage of
a cracked spoke, but one change in temperature can, provided the
fracture is large enough. Therefore thermal cycling as you imply it
isn't a significant consideration.
> example: i was in an historic vehicle restoration workshop a few years
> back. there was a large military vehicle in a corner that had stood
> unmoved for years. quite suddenly, there was a "crack" & a crunch as
> one of its suspension torsion bars failed & the vehicle dropped. no one
> was near and the vehicle had not moved or run in over 3 years. when we
> examined the failure, it was just a normal fatigue fracture surface,
> normal beach marks followed by brittle fracture. the crack had grown to
> the point where tip stress was critical, and that morning was just
> either one thermal cycle or one passing vehicle vibration too many.
Thermal cycle alone will not have caused that fracture; the load on the
torsion bar that is already fractured was the cause. When you compare
the load of the torsion bar to thermal changes, the latter contributes
insignificantly.
> if your question is implying that it takes significant temperature
> cycling or vibration to initiate a crack, then you are correct. but
> once a crack is present, it's growth accelerates as crack tip stress
> concentration grows. the operation of this vehicle was extreme,
> sufficient to initiate & propagate cracking to within an ace of failure.
> all that needed to happen later was for it to sit about quietly
> growing very very slowly to the tipping point.
It would not have required any thermal changes for the crack to
propagate through the cross-section; the load will have done that by
itself.
> "classic" example of thermal cycling fatigue: some 1980's 3-series
> bmw's had head bolts with a small radius transition between the head &
> the shaft. as these bolts were torqued to 98% of yield in the first
> place, the normal thermal cycle of operation had the bolt operating
> pretty much at yield, ie. the absolute worst point for rapid fatigue.
> these bolt heads would fatigue off because of the stress concentration
> at the small radius, then, if the owner was lucky, the head gasket would
> go. if they were unlucky, the bolt head would float underneath the cam
> shaft and get punched through the head by a cam lobe.
Irrelevant to bicycles, which do not undergo the temperature magnitudes
in an internal combustion engine.
jim beam
chrome spokes are chronic. the plating is almost entirely non-ductile,
so any bending causes cracking. once the surface layer has these cracks
in them, you have stress concentrations & fatigue propagation is under way.
i used to work in a chrome plating factory. small components were
plated by suspending them on racks with copper wire. on removal, the
wire was plated as well as the component, and the slightest bend in what
is otherwise that highly ductile material was accompanied by a creaking
cracking sound as the chrome surface fractured. not visible without a
magnifier, but it's there. "correcting the spoke line" of a plated
spoke would be the same.
jim beam
bottom line, on the large scale of things, you're absolutely correct,
there's not much to worry about and it's definitely not something most
people would encounter commonly. but in the absence of a stress
corrosion agent, and in the presence of a large pre-existing fatigue
crack in a 60 ton vehicle, to dismiss thermal expansion is somewhat
simplistic. if you /really/ want to get into small scale effects,
something metallurgists like to do, we could talk about diffusion as
well, but that would be stretching the point.
>
>
>>if your question is implying that it takes significant temperature
>>cycling or vibration to initiate a crack, then you are correct. but
>>once a crack is present, it's growth accelerates as crack tip stress
>>concentration grows. the operation of this vehicle was extreme,
>>sufficient to initiate & propagate cracking to within an ace of failure.
>> all that needed to happen later was for it to sit about quietly
>>growing very very slowly to the tipping point.
>
>
> It would not have required any thermal changes for the crack to
> propagate through the cross-section; the load will have done that by
> itself.
fracture mechanics. very small loads can result in crack propagation.
example: propagating a pre-existing crack in a piece of glass is easy.
and, come to think of it, is something that can demonstrate the
"insignificant" effects of thermal expansion too.
>
>
>>"classic" example of thermal cycling fatigue: some 1980's 3-series
>>bmw's had head bolts with a small radius transition between the head &
>>the shaft. as these bolts were torqued to 98% of yield in the first
>>place, the normal thermal cycle of operation had the bolt operating
>>pretty much at yield, ie. the absolute worst point for rapid fatigue.
>>these bolt heads would fatigue off because of the stress concentration
>>at the small radius, then, if the owner was lucky, the head gasket would
>>go. if they were unlucky, the bolt head would float underneath the cam
>>shaft and get punched through the head by a cam lobe.
>
>
> Irrelevant to bicycles, which do not undergo the temperature magnitudes
> in an internal combustion engine.
it's a relevant example of fatigue cause by relatively minor thermal
cycling. the block/head will almost never get above 240F. that's
hardly combustion temperature. much more common thermal failures are
things like spalling of brake or clutch surfaces where they're subject
to real temperature elevations or splitting of a header castings, same
reason.