What happened to non-boutique inexpensive MO wheels?

[Peter Cole]

jim beam wrote:
> Peter Cole wrote:
>> jim beam wrote:
>>> Peter Cole wrote:
>>>> jim beam wrote:
>>>>> Peter Cole wrote:
>>>>>> jim beam wrote:
>>>>>>> [email protected] wrote:
>>>>>>>> In article <[email protected]>,
>>>>>>>> [email protected] says...
>>>>>>>>
>>>>>>>>> since we have yet to be privileged with sight of the jobstian
>>>>>>>>> research that will revolutionize the world of materials by
>>>>>>>>> showing how fatigue can be eliminated in a material with no
>>>>>>>>> endurance limit
>>>>>>>>
>>>>>>>> I thought most spokes were 304 stainless steel?
>>>>>>>>
>>>>>>> there's various grades used for spokes, but none have an endurance
>>>>>>> limit. they have what's called a "fatigue limit" which is an
>>>>>>> attempt to
>>>>>>> quantify the stress they can endure for a given number of fatigue
>>>>>>> cycles, but that's not the same thing. an endurance limit is the
>>>>>>> stress
>>>>>>> at which the material can endure an infinite number of cycles,
>>>>>>> and is
>>>>>>> characterized by a "knee" in the s-n graph where the line goes
>>>>>>> horizontal. mild steel is the classic material with this
>>>>>>> property, but
>>>>>>> some titanium alloys also have it. no stainless alloys that i
>>>>>>> know of
>>>>>>> have it.
>>>>>>
>>>>>> <users.wpi.edu/~cfurlong/me3320/lect13/Lect13.pdf>
>>>>>> <bama.ua.edu/~mweaver/courses/MTE455/MTE455_2006_L26.pdf>
>>>>>>
>>>>>> That's not what they're teaching today.
>>>>>
>>>>> how so?
>>>>>
>>>>> regarding endurance limit, both seem to corroborate the same
>>>>> definition as i, although the second uses fatigue limit and
>>>>> endurance limit interchangeably between diagrams and text.
>>>>>
>>>>> regarding which alloys evidence an endurance limit, the first one
>>>>> says some types of stainless can show it, but i saw no examples.
>>>>> or if they did, and/or you know any, please share. i'm interested
>>>>> because fwiu, endurance limits originate with the same mechanism
>>>>> that causes strain aging, i.e. interstitial elements [carbon in
>>>>> mild steel and oxygen in titanium] locking dislocations. the thing
>>>>> with chromium passivated stainless is that there's very little
>>>>> carbon possible without having chromium carbide precipitate at
>>>>> grain boundaries, a big no-no for fatigue, among other things.
>>>>> there are other types of "stainless", but they're not something you
>>>>> or i would buy off the shelf and i've not heard of their use for
>>>>> fatigue resistance. i definitely don't see them being used in
>>>>> bicycle spokes.
>>>>
>>>> As you say, the first notes claim some stainless steels have a flat
>>>> S/N curve, some don't.
>>>
>>> i'm not saying that. i've not seen a stainless steel that evidences
>>> an endurance limit, as defined by a "knee" in the graph. some
>>> stainless steels appear to flatten, but there is no knee, hence no
>>> endurance limit.
>>>
>>>> The second reference claims that all stainless are flat (pg, 4)
>>>> "Ferrous metals and other strain aging materials -- Examples: low
>>>> carbon steel, stainless steel, titanium, etc."
>>>
>>> cite the alloys and explain the mechanism. endurance limit is cause
>>> by the mechanism i explained, carbon in ferrous metals, and is
>>> problematic in chromium passivated stainless, again for the reasons i
>>> explained.
>>>
>>>>
>>>> Another source is:
>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>
>>>>
>>>> Sec. 5.1
>>>>
>>>> "The S-N curves for ferrous and titanium alloys exhibit a limiting
>>>> stress below which failure does not occur; this is called the
>>>> fatigue or the endurance limit. The branch point or "knee" of the
>>>> curve lies normally in the 105 to 107 cycle range. In aluminium and
>>>> other nonferrous alloys there is no stress asymptote and a finite
>>>> fatigue life exists at any stress level. All materials, however,
>>>> exhibit a relatively flat curve in the high-cycle region, ie. at
>>>> lives longer than about 105 cycles."
>>>
>>> which supports what i said before.
>>
>> I cited 3 sources, 2 of which said all stainless steels have an
>> endurance limit, the other said some did, some didn't.
>
> with respect, one of those two uses "endurance limit" and "fatigue
> limit" interchangeably. where i come from, the words "endurance limit"
> are reserved for graphs showing the knee, "fatigue limit" is for those
> that don't. engineers don't seem to regard materials definitions as
> important as materials people, hence the dearth of engineering papers
> perpetuating the confusion.
>
>>
>> From:
>> <www.stainless-steel-world.net/pdf/11021.pdf>
>>
>> Page 30 shows an S/N graph for 2 types of stainless, one of which is
>> common 316. There is clearly a fatigue limit for both alloys.
>
> ok, that's a good link, citing specific alloys. i have two comments:
>
> 1. endurance limit at 20-25% of yield is not really a reliable benefit.

From:
<http://www.azom.com/details.asp?ArticleID=863>

Tensile strength of 316 is given as 515MPa = 75ksi.

From S/N referenced above, the asymptote is about 33ksi.

Ratio is 44%, typical for steel.



> 2. fatigue is statistical. i'm not saying the publishers of this paper
> don't know what they're doing, but be honest peter, there are not a lot
> of other materials out there supporting this position - most to the
> contrary in fact.


Cite 'em.

 

[Peter Cole]

jim beam wrote:
> Peter Cole wrote:
>> jim beam wrote:
>>> Peter Cole wrote:
>>>> Another interesting quote from
>>>>
>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>
>>>> Sec. 6.6
>>>>
>>>> "Residual stresses or internal stresses are produced when a region
>>>> of a part is strained beyond the elastic limit while other regions
>>>> are elastically deformed. When the force or deformation causing the
>>>> deformation are removed, the elastically deformed material springs
>>>> back and impose residual stresses in the plastically deformed
>>>> material. Yielding can be caused by thermal expansion as well as by
>>>> external force. The residual stresses are of the opposite sign to
>>>> the initially applied stress. Therefore, if a notched member is
>>>> loaded in tension until yielding occurs, the notch root will
>>>> experience a compressive stress after unloading. Welding stresses
>>>> which are locked in when the weld metal contracts during cooling are
>>>> an example of highly damaging stresses that cannot be avoided during
>>>> fabrication. These stresses are of yield stress magnitude and
>>>> tensile and compressive stresses must always balance each other, as
>>>> indicated in Figure 26. The high tensile welding stresses contribute
>>>> to a large extent to the poor fatigue performance of welded joints."
>>>>
>>>> "Residual stresses have a similar influence on fatigue life as
>>>> externally imposed mean stresses, ie. a tensile stress reduces
>>>> fatigue life while a compressive stress increases life. There is,
>>>> however, an important difference which relates to the stability of
>>>> residual stresses. While an externally imposed mean stress, eg.
>>>> stress caused by dead weight always acts (as long as the load is
>>>> present), residual stress may relax with time, especially if there
>>>> are high peaks in the load spectrum that cause local yielding at
>>>> stress concentrations."
>>>
>>> dude, with respect, residual stress has NOTHING to do with endurance
>>> limits!!!
>>
>> No, it doesn't, but you were the one who put them together upthread:
>>
>> "Since we have yet to be privileged with sight of the jobstian
>> research that will revolutionize the world of materials by showing how
>> fatigue can be eliminated in a material with no endurance limit and,
>> we're somewhat thin on evidence for this supposition that spokes are
>> sent from the factory containing residual stress"
>>
>> My cite above was intended as a rebuttal to your (frequent) denial
>> that mechanical stress relieving doesn't reduce residual stress.
>
> context: i was contrasting the credibility of someone that claims they
> can eliminate fatigue in a situation where they can't, with the fact
> that they're making unsupported suppositions.
>
> i've never said you cannot mechanically relieve residual stress. i
> /have/ however questioned whether spokes leave the factory with
> sufficient residual stress to initiate fatigue /and/ i've questioned the
> efficacy of "stress relief" applied after spoke manufacture. mechanical
> stress relief is typically only effective a short time after initial
> deformation. waiting more than a few hours [as will be necessary to
> ship from the spoke manufacturer to the wheel builder] will render this
> method useless. and then we can get into the debate of spoke squeezing
> itself and whether a process that produces no detectable deformation is
> actually yielding the material sufficiently to work as is being
> supposed, [timing issues aside of course].

These claims (time dependency & lack of overall plastic yield) seem to
be not supported by the link above and others I have provided. I will
wait for you to provide any evidence to support your claims.

 

[Qui si parla Campagnolo]

monkeyboy wrote:
> damyth wrote:
> > OK, so I'm in the market for some nice but inexpensive pre-built
> > traditional (32 spoke) 700c non-boutique wheels. I go over to the
> > usual mail order places like Nashbar and Performance, but there are
> > none to be found! All they have now are boutique wheels and
> > non-boutique 650c.
> ....
>
> take a look at cbike.com, they have campy scrirroco wheels for $299,
> great wheels at a great price...I guess they could be considered
> boutique...?

Except a handbuilt, using Veloce hubs(the base hubs for above mentioned
wheels), Velocity rims and 32 or 36 spokes would cost about the same $
but could be 'designed' for any rider.

 

[jim beam]

Peter Cole wrote:
> jim beam wrote:
>> Peter Cole wrote:
>>> jim beam wrote:
>>>> Peter Cole wrote:
>>>>> Another interesting quote from
>>>>>
>>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>>
>>>>> Sec. 6.6
>>>>>
>>>>> "Residual stresses or internal stresses are produced when a region
>>>>> of a part is strained beyond the elastic limit while other regions
>>>>> are elastically deformed. When the force or deformation causing the
>>>>> deformation are removed, the elastically deformed material springs
>>>>> back and impose residual stresses in the plastically deformed
>>>>> material. Yielding can be caused by thermal expansion as well as by
>>>>> external force. The residual stresses are of the opposite sign to
>>>>> the initially applied stress. Therefore, if a notched member is
>>>>> loaded in tension until yielding occurs, the notch root will
>>>>> experience a compressive stress after unloading. Welding stresses
>>>>> which are locked in when the weld metal contracts during cooling
>>>>> are an example of highly damaging stresses that cannot be avoided
>>>>> during fabrication. These stresses are of yield stress magnitude
>>>>> and tensile and compressive stresses must always balance each
>>>>> other, as indicated in Figure 26. The high tensile welding stresses
>>>>> contribute to a large extent to the poor fatigue performance of
>>>>> welded joints."
>>>>>
>>>>> "Residual stresses have a similar influence on fatigue life as
>>>>> externally imposed mean stresses, ie. a tensile stress reduces
>>>>> fatigue life while a compressive stress increases life. There is,
>>>>> however, an important difference which relates to the stability of
>>>>> residual stresses. While an externally imposed mean stress, eg.
>>>>> stress caused by dead weight always acts (as long as the load is
>>>>> present), residual stress may relax with time, especially if there
>>>>> are high peaks in the load spectrum that cause local yielding at
>>>>> stress concentrations."
>>>>
>>>> dude, with respect, residual stress has NOTHING to do with endurance
>>>> limits!!!
>>>
>>> No, it doesn't, but you were the one who put them together upthread:
>>>
>>> "Since we have yet to be privileged with sight of the jobstian
>>> research that will revolutionize the world of materials by showing
>>> how fatigue can be eliminated in a material with no endurance limit
>>> and, we're somewhat thin on evidence for this supposition that spokes
>>> are sent from the factory containing residual stress"
>>>
>>> My cite above was intended as a rebuttal to your (frequent) denial
>>> that mechanical stress relieving doesn't reduce residual stress.
>>
>> context: i was contrasting the credibility of someone that claims
>> they can eliminate fatigue in a situation where they can't, with the
>> fact that they're making unsupported suppositions.
>>
>> i've never said you cannot mechanically relieve residual stress. i
>> /have/ however questioned whether spokes leave the factory with
>> sufficient residual stress to initiate fatigue /and/ i've questioned
>> the efficacy of "stress relief" applied after spoke manufacture.
>> mechanical stress relief is typically only effective a short time
>> after initial deformation. waiting more than a few hours [as will be
>> necessary to ship from the spoke manufacturer to the wheel builder]
>> will render this method useless. and then we can get into the debate
>> of spoke squeezing itself and whether a process that produces no
>> detectable deformation is actually yielding the material sufficiently
>> to work as is being supposed, [timing issues aside of course].
>
> These claims (time dependency & lack of overall plastic yield) seem to
> be not supported by the link above and others I have provided. I will
> wait for you to provide any evidence to support your claims.

dieter, mechanical metallurgy, isbn 0-07-016893-8, chapter 19.6,
residual stresses in rod, wire & tubes. [forgive me if i don't dig out
my old lecture notes - one of my old profs was mr. cold drawn steel.]

typical mechanical relief strain is 1%-2%. for a 294mm spoke, where
does that leave us? arguing as jobst does that the relief strain is at
levels microscopic and can't be detected is suppositional straw clutching.

 

[jim beam]

Peter Cole wrote:
> jim beam wrote:
>> Peter Cole wrote:
>>> jim beam wrote:
>>>> Peter Cole wrote:
>>>>> jim beam wrote:
>>>>>> Peter Cole wrote:
>>>>>>> jim beam wrote:
>>>>>>>> [email protected] wrote:
>>>>>>>>> In article <[email protected]>,
>>>>>>>>> [email protected] says...
>>>>>>>>>
>>>>>>>>>> since we have yet to be privileged with sight of the jobstian
>>>>>>>>>> research that will revolutionize the world of materials by
>>>>>>>>>> showing how fatigue can be eliminated in a material with no
>>>>>>>>>> endurance limit
>>>>>>>>>
>>>>>>>>> I thought most spokes were 304 stainless steel?
>>>>>>>>>
>>>>>>>> there's various grades used for spokes, but none have an endurance
>>>>>>>> limit. they have what's called a "fatigue limit" which is an
>>>>>>>> attempt to
>>>>>>>> quantify the stress they can endure for a given number of fatigue
>>>>>>>> cycles, but that's not the same thing. an endurance limit is
>>>>>>>> the stress
>>>>>>>> at which the material can endure an infinite number of cycles,
>>>>>>>> and is
>>>>>>>> characterized by a "knee" in the s-n graph where the line goes
>>>>>>>> horizontal. mild steel is the classic material with this
>>>>>>>> property, but
>>>>>>>> some titanium alloys also have it. no stainless alloys that i
>>>>>>>> know of
>>>>>>>> have it.
>>>>>>>
>>>>>>> <users.wpi.edu/~cfurlong/me3320/lect13/Lect13.pdf>
>>>>>>> <bama.ua.edu/~mweaver/courses/MTE455/MTE455_2006_L26.pdf>
>>>>>>>
>>>>>>> That's not what they're teaching today.
>>>>>>
>>>>>> how so?
>>>>>>
>>>>>> regarding endurance limit, both seem to corroborate the same
>>>>>> definition as i, although the second uses fatigue limit and
>>>>>> endurance limit interchangeably between diagrams and text.
>>>>>>
>>>>>> regarding which alloys evidence an endurance limit, the first one
>>>>>> says some types of stainless can show it, but i saw no examples.
>>>>>> or if they did, and/or you know any, please share. i'm interested
>>>>>> because fwiu, endurance limits originate with the same mechanism
>>>>>> that causes strain aging, i.e. interstitial elements [carbon in
>>>>>> mild steel and oxygen in titanium] locking dislocations. the
>>>>>> thing with chromium passivated stainless is that there's very
>>>>>> little carbon possible without having chromium carbide precipitate
>>>>>> at grain boundaries, a big no-no for fatigue, among other things.
>>>>>> there are other types of "stainless", but they're not something
>>>>>> you or i would buy off the shelf and i've not heard of their use
>>>>>> for fatigue resistance. i definitely don't see them being used in
>>>>>> bicycle spokes.
>>>>>
>>>>> As you say, the first notes claim some stainless steels have a flat
>>>>> S/N curve, some don't.
>>>>
>>>> i'm not saying that. i've not seen a stainless steel that evidences
>>>> an endurance limit, as defined by a "knee" in the graph. some
>>>> stainless steels appear to flatten, but there is no knee, hence no
>>>> endurance limit.
>>>>
>>>>> The second reference claims that all stainless are flat (pg, 4)
>>>>> "Ferrous metals and other strain aging materials -- Examples: low
>>>>> carbon steel, stainless steel, titanium, etc."
>>>>
>>>> cite the alloys and explain the mechanism. endurance limit is cause
>>>> by the mechanism i explained, carbon in ferrous metals, and is
>>>> problematic in chromium passivated stainless, again for the reasons
>>>> i explained.
>>>>
>>>>>
>>>>> Another source is:
>>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>>
>>>>>
>>>>> Sec. 5.1
>>>>>
>>>>> "The S-N curves for ferrous and titanium alloys exhibit a limiting
>>>>> stress below which failure does not occur; this is called the
>>>>> fatigue or the endurance limit. The branch point or "knee" of the
>>>>> curve lies normally in the 105 to 107 cycle range. In aluminium and
>>>>> other nonferrous alloys there is no stress asymptote and a finite
>>>>> fatigue life exists at any stress level. All materials, however,
>>>>> exhibit a relatively flat curve in the high-cycle region, ie. at
>>>>> lives longer than about 105 cycles."
>>>>
>>>> which supports what i said before.
>>>
>>> I cited 3 sources, 2 of which said all stainless steels have an
>>> endurance limit, the other said some did, some didn't.
>>
>> with respect, one of those two uses "endurance limit" and "fatigue
>> limit" interchangeably. where i come from, the words "endurance
>> limit" are reserved for graphs showing the knee, "fatigue limit" is
>> for those that don't. engineers don't seem to regard materials
>> definitions as important as materials people, hence the dearth of
>> engineering papers perpetuating the confusion.
>>
>>>
>>> From:
>>> <www.stainless-steel-world.net/pdf/11021.pdf>
>>>
>>> Page 30 shows an S/N graph for 2 types of stainless, one of which is
>>> common 316. There is clearly a fatigue limit for both alloys.
>>
>> ok, that's a good link, citing specific alloys. i have two comments:
>>
>> 1. endurance limit at 20-25% of yield is not really a reliable benefit.
>
> From:
> <http://www.azom.com/details.asp?ArticleID=863>
>
> Tensile strength of 316 is given as 515MPa = 75ksi.
>
> From S/N referenced above, the asymptote is about 33ksi.
>
> Ratio is 44%, typical for steel.

there's no asymptote referenced in the azom link. the one showing ~35%
from the stainless-steel-world link is for DP3, whatever that is. and
fyi, ~35% is indistinguishable from the fatigue limit for a lot of
structural aluminum alloys, so i don't think you'll find many people
relying on an endurance limit for DP3 in their fatigue design.

>
>
>
>> 2. fatigue is statistical. i'm not saying the publishers of this
>> paper don't know what they're doing, but be honest peter, there are
>> not a lot of other materials out there supporting this position - most
>> to the contrary in fact.
>
>
> Cite 'em.

dude, google 'stainless steel endurance limit'. there are 170,000
returns, and while i haven't read all of them, the first 50 or so don't
reference an asymptote - or at least, not without also confusing fatigue
and endurance limits. it's been years since you and i first started
this kind of debate, and so far you've cited just one link with a
graphical asymptote. everything else shows a more typical flattening
curve, no asymptote. done any curve fitting exercises lately?

 

[dvt]

jim beam wrote:
> Peter Cole wrote:
>> jim beam wrote:
>>> where i come from, the words "endurance
>>> limit" are reserved for graphs showing the knee, "fatigue limit" is
>>> for those that don't.

Amen to that. Google for "define:endurance limit" and you'll find that
most definitions say endurance = fatigue. Ugh.

I do think you "knee" and "asymptote" terms are not quite correct,
though. You use asymptote as a synonym for horizontal. Most S/N curves
that I've seen have a knee (a change in slope). They have two
asymptotes; one above the knee, one below. But materials showing an
endurance limit have a *horizontal* asymptote above the knee.

>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>> paper don't know what they're doing, but be honest peter, there are
>>> not a lot of other materials out there supporting this position -
>>> most to the contrary in fact.
>>
>>
>> Cite 'em.
>
> dude, google 'stainless steel endurance limit'. there are 170,000
> returns, and while i haven't read all of them, the first 50 or so don't
> reference an asymptote - or at least, not without also confusing fatigue
> and endurance limits.

I just did the Google search that you suggested. Since I was looking for
pdfs in the hopes of finding graphics, the first link I clicked was the
5th one on the list:
<http://www.aksteel.com/pdf/markets_products/stainless/austenitic/AK_Nit30PDB_1104_2.pdf>

Look at pages 9 and 10. They show a horizontal portion in the S/N curve.

--
Dave
dvt at psu dot edu

Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)

 

[[email protected]]

On Thu, 02 Nov 2006 09:18:44 -0500, dvt <[email protected]> wrote:

[snip]

>How would you design a test for residual stress in spoke elbows? I have
>access to some equipment, but not "neutron/x-ray inferometry" [sic] that
>you mentioned in the above-mentioned thread.

[snip]

Dear Dave,

Here's a stab at it:

http://groups.google.com/group/rec.bicycles.tech/msg/34e3b90b360992f4

After tedious setup explanations, it all boils down to before and
after pictures of four stainless steel spokes whose U-bends have been
tensioned to 0, 76, 121, and 179 kgf, then heated to an orange glow:

http://i12.tinypic.com/2zgtfs5.jpg

http://i11.tinypic.com/2qsc507.jpg

Only one spoke shows any change when stress-relieved by heating, the
untensioned one on the left. This suggests that the other three spokes
had all significant stress removed by tension before they reached
normal bicycle tension levels.

Cheers,

Carl Fogel

 

[dvt]

jim beam wrote:
> dvt wrote:
>> How would you design a test for residual stress in spoke elbows? I
>> have access to some equipment, but not "neutron/x-ray inferometry"
>> [sic] that you mentioned in the above-mentioned thread.
>
> chlorides are well known for their stress corrosion in stainless steel.
> google will help you identify other agents as well. simply leave a
> spoke in a jar of the requisite solution and wait.

Thanks for the response, jim.

I see how that tests for corrosion. As I have said in the past, my
metallurgy knowledge is not enough to make the leap from a corrosion
test to a test for residual stress. Maybe you can help me make that
connection.

>>> sure, if residual stress is sufficient to be a factor, it can help to
>>> reduce it. but the necessary supposition that spoke manufacturers
>>> are incompetent in order to take that position is something of a
>>> stretch.
>>
>> I don't think it takes a supposition of incompetence.
>
> i disagree. according to jobstian theory, the supposition is that
> spokes leave the factory with residual stress, ergo the manufacturer is
> incompetent for not relieving it.

I don't think it requires incompetence on the spoke manufacturer's part.
The wheel building process can change the angle of the elbow, producing
some residual stress. I know you have debated that point (with Benjamin
Lewis, if memory serves). His photographs show that it's at least
possible to change the elbow angle by building a wheel, and it agrees
with my experience, so I'm sticking with that. Since we've bent the
spoke elbow without any form of stress relief, there are quite likely
some residual stresses.

So if the manufacturer supplied spokes without residual stress, residual
stresses would be induced during the wheel build. Supplying spokes free
of residual stress would be rather pointless, in that case.

> perhaps, but we'd then have to show that residual stress was a factor
> and that material quality was not.

I think this is where we disagree. The first part of your sentence makes
sense to me, but the second part doesn't. Why does one have to be
exclusive of the other? Why couldn't some failures be caused by a
*combination* of material quality and residual stress?

> seriously, if spoke squeezing was
> able to eliminate fatigue regardless, don't you think manufacturers
> would just do that and use cheaper materials and cheaper processing
> rather than what they do now?

Perhaps if they were able to assume no residual stresses in a spoke in a
built wheel, they *could* go to cheaper materials and processing.

--
Dave
dvt at psu dot edu

 

[dvt]

[email protected] wrote:
> On Thu, 02 Nov 2006 09:18:44 -0500, dvt <[email protected]> wrote:
>
> [snip]
>
>> How would you design a test for residual stress in spoke elbows? I have
>> access to some equipment, but not "neutron/x-ray inferometry" [sic] that
>> you mentioned in the above-mentioned thread.

> Here's a stab at it:
>
> http://groups.google.com/group/rec.bicycles.tech/msg/34e3b90b360992f4

I read that. Is it possible that the modern stainless spokes are
undergoing a phase change (or some other effect) that would affect your
results? For example, could there be a phase change that acts in the
opposite direction of stress relief, resulting in no net change in the
case of higher tension?

I do not know the answer to my questions. So I don't know if your test
is showing *only* stress relief, or if something else is going on as well.

By the way, nice work with the experiment. It was a good idea.

--
Dave
dvt at psu dot edu

Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)

 

[dvt]

dvt wrote:
> jim beam wrote:
>> Peter Cole wrote:
>>> jim beam wrote:
>>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>>> paper don't know what they're doing, but be honest peter, there are
>>>> not a lot of other materials out there supporting this position -
>>>> most to the contrary in fact.

>>> Cite 'em.

>> dude, google 'stainless steel endurance limit'. there are 170,000
>> returns, and while i haven't read all of them, the first 50 or so
>> don't reference an asymptote - or at least, not without also confusing
>> fatigue and endurance limits.

> I just did the Google search that you suggested. Since I was looking for
> pdfs in the hopes of finding graphics, the first link I clicked was the
> 5th one on the list:
> <http://www.aksteel.com/pdf/markets_products/stainless/austenitic/AK_Nit30PDB_1104_2.pdf>
>
> Look at pages 9 and 10. They show a horizontal portion in the S/N curve.

Here's another one that shows a S/N curve for 304 SS with an endurance
limit, found via the Google search you suggested:

"Fatigue Life Improvements of the AISI 304 Stainless Steel Ground
Surfaces by Wire Brushing"
Nabil Ben Fredj, Mohamed Ben Nasr, Amir Ben Rhouma, Chedly Braham, and
Habib Sidhom
Journal of Materials Engineering and Performance, Volume 13(5), October
2004, page 565.

Sorry, I can't directly link that one.

--
Dave
dvt at psu dot edu

Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)

 

[[email protected]]

On Fri, 03 Nov 2006 13:06:22 -0500, dvt <[email protected]> wrote:

>[email protected] wrote:
>> On Thu, 02 Nov 2006 09:18:44 -0500, dvt <[email protected]> wrote:
>>
>> [snip]
>>
>>> How would you design a test for residual stress in spoke elbows? I have
>>> access to some equipment, but not "neutron/x-ray inferometry" [sic] that
>>> you mentioned in the above-mentioned thread.
>
>> Here's a stab at it:
>>
>> http://groups.google.com/group/rec.bicycles.tech/msg/34e3b90b360992f4
>
>I read that. Is it possible that the modern stainless spokes are
>undergoing a phase change (or some other effect) that would affect your
>results? For example, could there be a phase change that acts in the
>opposite direction of stress relief, resulting in no net change in the
>case of higher tension?
>
>I do not know the answer to my questions. So I don't know if your test
>is showing *only* stress relief, or if something else is going on as well.
>
>By the way, nice work with the experiment. It was a good idea.

Dear Dave,

If something else is involved, it's darned clever.

The U-bends of all four spokes were heated to an orange glow.

The U-bend that hadn't been tensioned moved quite noticeably as the
heat relieved internal stresses.

The other three U-bends didn't move, having been tensioned to around
76, 121, and 179 kgf.

Either their stresses had already been relieved by tension below 76
kgf, or some unknown force was holding all three steady, despite twice
as much stress-relieving tension being applied to one spoke.

I'm still trying to poke holes in the experiment.

Two things seem worth checking.

First, I'm going to pre-straighten the elbows on another batch by
bending them instead of pulling on them. Watching the elbows slither
into the holes drilled in the bolts is probably the best part of the
experiment, but it's possible that the smooth straightening of the
elbows as the vise jaws are opened somehow applies lots and lots of
tension briefly, so that there's a hidden stress relief tension.

That is, the 76kgf spoke was just one whose elbow had straightened.
Maybe in straightening it briefly reached 169 kgf?

Second, the pre-straightened elbows may let me test lower tensions,
though the Park tension gauge and the bending wire make that tricky.
If stress is indeed being relieved below 76 kgf, then there should be
a tension level at which the spoke bend-angle changes, but not as much
as large change of the untensioned spoke-bend.

The logic would seem to be that the greatest residual stresses are
going to be relieved first, by the least tension, so I expect a graph
of bend-change versus tension that drops off sharply.

Cheers,

Carl Fogel

 

[Peter Cole]

jim beam wrote:
> Peter Cole wrote:
>> jim beam wrote:
>>> Peter Cole wrote:
>>>> jim beam wrote:
>>>>> Peter Cole wrote:
>>>>>> jim beam wrote:
>>>>>>> Peter Cole wrote:
>>>>>>>> jim beam wrote:
>>>>>>>>> [email protected] wrote:
>>>>>>>>>> In article <[email protected]>,
>>>>>>>>>> [email protected] says...
>>>>>>>>>>
>>>>>>>>>>> since we have yet to be privileged with sight of the jobstian
>>>>>>>>>>> research that will revolutionize the world of materials by
>>>>>>>>>>> showing how fatigue can be eliminated in a material with no
>>>>>>>>>>> endurance limit
>>>>>>>>>>
>>>>>>>>>> I thought most spokes were 304 stainless steel?
>>>>>>>>>>
>>>>>>>>> there's various grades used for spokes, but none have an endurance
>>>>>>>>> limit. they have what's called a "fatigue limit" which is an
>>>>>>>>> attempt to
>>>>>>>>> quantify the stress they can endure for a given number of fatigue
>>>>>>>>> cycles, but that's not the same thing. an endurance limit is
>>>>>>>>> the stress
>>>>>>>>> at which the material can endure an infinite number of cycles,
>>>>>>>>> and is
>>>>>>>>> characterized by a "knee" in the s-n graph where the line goes
>>>>>>>>> horizontal. mild steel is the classic material with this
>>>>>>>>> property, but
>>>>>>>>> some titanium alloys also have it. no stainless alloys that i
>>>>>>>>> know of
>>>>>>>>> have it.
>>>>>>>>
>>>>>>>> <users.wpi.edu/~cfurlong/me3320/lect13/Lect13.pdf>
>>>>>>>> <bama.ua.edu/~mweaver/courses/MTE455/MTE455_2006_L26.pdf>
>>>>>>>>
>>>>>>>> That's not what they're teaching today.
>>>>>>>
>>>>>>> how so?
>>>>>>>
>>>>>>> regarding endurance limit, both seem to corroborate the same
>>>>>>> definition as i, although the second uses fatigue limit and
>>>>>>> endurance limit interchangeably between diagrams and text.
>>>>>>>
>>>>>>> regarding which alloys evidence an endurance limit, the first one
>>>>>>> says some types of stainless can show it, but i saw no examples.
>>>>>>> or if they did, and/or you know any, please share. i'm
>>>>>>> interested because fwiu, endurance limits originate with the same
>>>>>>> mechanism that causes strain aging, i.e. interstitial elements
>>>>>>> [carbon in mild steel and oxygen in titanium] locking
>>>>>>> dislocations. the thing with chromium passivated stainless is
>>>>>>> that there's very little carbon possible without having chromium
>>>>>>> carbide precipitate at grain boundaries, a big no-no for fatigue,
>>>>>>> among other things. there are other types of "stainless", but
>>>>>>> they're not something you or i would buy off the shelf and i've
>>>>>>> not heard of their use for fatigue resistance. i definitely
>>>>>>> don't see them being used in bicycle spokes.
>>>>>>
>>>>>> As you say, the first notes claim some stainless steels have a
>>>>>> flat S/N curve, some don't.
>>>>>
>>>>> i'm not saying that. i've not seen a stainless steel that
>>>>> evidences an endurance limit, as defined by a "knee" in the graph.
>>>>> some stainless steels appear to flatten, but there is no knee,
>>>>> hence no endurance limit.
>>>>>
>>>>>> The second reference claims that all stainless are flat (pg, 4)
>>>>>> "Ferrous metals and other strain aging materials -- Examples: low
>>>>>> carbon steel, stainless steel, titanium, etc."
>>>>>
>>>>> cite the alloys and explain the mechanism. endurance limit is
>>>>> cause by the mechanism i explained, carbon in ferrous metals, and
>>>>> is problematic in chromium passivated stainless, again for the
>>>>> reasons i explained.
>>>>>
>>>>>>
>>>>>> Another source is:
>>>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>>>
>>>>>>
>>>>>> Sec. 5.1
>>>>>>
>>>>>> "The S-N curves for ferrous and titanium alloys exhibit a limiting
>>>>>> stress below which failure does not occur; this is called the
>>>>>> fatigue or the endurance limit. The branch point or "knee" of the
>>>>>> curve lies normally in the 105 to 107 cycle range. In aluminium
>>>>>> and other nonferrous alloys there is no stress asymptote and a
>>>>>> finite fatigue life exists at any stress level. All materials,
>>>>>> however, exhibit a relatively flat curve in the high-cycle region,
>>>>>> ie. at lives longer than about 105 cycles."
>>>>>
>>>>> which supports what i said before.
>>>>
>>>> I cited 3 sources, 2 of which said all stainless steels have an
>>>> endurance limit, the other said some did, some didn't.
>>>
>>> with respect, one of those two uses "endurance limit" and "fatigue
>>> limit" interchangeably. where i come from, the words "endurance
>>> limit" are reserved for graphs showing the knee, "fatigue limit" is
>>> for those that don't. engineers don't seem to regard materials
>>> definitions as important as materials people, hence the dearth of
>>> engineering papers perpetuating the confusion.
>>>
>>>>
>>>> From:
>>>> <www.stainless-steel-world.net/pdf/11021.pdf>
>>>>
>>>> Page 30 shows an S/N graph for 2 types of stainless, one of which is
>>>> common 316. There is clearly a fatigue limit for both alloys.
>>>
>>> ok, that's a good link, citing specific alloys. i have two comments:
>>>
>>> 1. endurance limit at 20-25% of yield is not really a reliable benefit.
>>
>> From:
>> <http://www.azom.com/details.asp?ArticleID=863>
>>
>> Tensile strength of 316 is given as 515MPa = 75ksi.
>>
>> From S/N referenced above, the asymptote is about 33ksi.
>>
>> Ratio is 44%, typical for steel.
>
> there's no asymptote referenced in the azom link.


No, the graph (with asymptote) is in:
www.stainless-steel-world.net/pdf/11021.pdf.



> the one showing ~35%
> from the stainless-steel-world link is for DP3, whatever that is.

http://www.duplexss.com/


> and
> fyi, ~35% is indistinguishable from the fatigue limit for a lot of
> structural aluminum alloys, so i don't think you'll find many people
> relying on an endurance limit for DP3 in their fatigue design.

The endurance limit I referred to was for 316. The ratio of the
endurance limit to tensile strength for 316 according to the sources
above is 44%.


>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>> paper don't know what they're doing, but be honest peter, there are
>>> not a lot of other materials out there supporting this position -
>>> most to the contrary in fact.
>>
>>
>> Cite 'em.
>
> dude, google 'stainless steel endurance limit'. there are 170,000
> returns, and while i haven't read all of them, the first 50 or so don't
> reference an asymptote - or at least, not without also confusing fatigue
> and endurance limits. it's been years since you and i first started
> this kind of debate, and so far you've cited just one link with a
> graphical asymptote. everything else shows a more typical flattening
> curve, no asymptote. done any curve fitting exercises lately?

Quibbling about terminology aside, all 3 of the sources I cited:

http://users.wpi.edu/~cfurlong/me3320/lect13/Lect13.pdf
bama.ua.edu/~mweaver/courses/MTE455/MTE455_2006_L26.pdf
http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1

unambiguously discussed S/N curves where the materials had infinite
stress life (horizontal asymptote). There was no confusion about the
topic. 2 of the authors contended that all stainless steels have this
characteristic, the 3rd said some.

The last source I cited: www.stainless-steel-world.net/pdf/11021.pdf,
showed a S/N curve for a specific alloy (which you requested), that of
the very common 316.

Do you have any examples of S/N curves that dispute this? Say one for 316?

And I'd appreciate you stop calling me "dude".

 

[Peter Cole]

jim beam wrote:
> Peter Cole wrote:
>> jim beam wrote:
>>> Peter Cole wrote:
>>>> jim beam wrote:
>>>>> Peter Cole wrote:
>>>>>> Another interesting quote from
>>>>>>
>>>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>>>
>>>>>> Sec. 6.6
>>>>>>
>>>>>> "Residual stresses or internal stresses are produced when a region
>>>>>> of a part is strained beyond the elastic limit while other regions
>>>>>> are elastically deformed. When the force or deformation causing
>>>>>> the deformation are removed, the elastically deformed material
>>>>>> springs back and impose residual stresses in the plastically
>>>>>> deformed material. Yielding can be caused by thermal expansion as
>>>>>> well as by external force. The residual stresses are of the
>>>>>> opposite sign to the initially applied stress. Therefore, if a
>>>>>> notched member is loaded in tension until yielding occurs, the
>>>>>> notch root will experience a compressive stress after unloading.
>>>>>> Welding stresses which are locked in when the weld metal contracts
>>>>>> during cooling are an example of highly damaging stresses that
>>>>>> cannot be avoided during fabrication. These stresses are of yield
>>>>>> stress magnitude and tensile and compressive stresses must always
>>>>>> balance each other, as indicated in Figure 26. The high tensile
>>>>>> welding stresses contribute to a large extent to the poor fatigue
>>>>>> performance of welded joints."
>>>>>>
>>>>>> "Residual stresses have a similar influence on fatigue life as
>>>>>> externally imposed mean stresses, ie. a tensile stress reduces
>>>>>> fatigue life while a compressive stress increases life. There is,
>>>>>> however, an important difference which relates to the stability of
>>>>>> residual stresses. While an externally imposed mean stress, eg.
>>>>>> stress caused by dead weight always acts (as long as the load is
>>>>>> present), residual stress may relax with time, especially if there
>>>>>> are high peaks in the load spectrum that cause local yielding at
>>>>>> stress concentrations."
>>>>>
>>>>> dude, with respect, residual stress has NOTHING to do with
>>>>> endurance limits!!!
>>>>
>>>> No, it doesn't, but you were the one who put them together upthread:
>>>>
>>>> "Since we have yet to be privileged with sight of the jobstian
>>>> research that will revolutionize the world of materials by showing
>>>> how fatigue can be eliminated in a material with no endurance limit
>>>> and, we're somewhat thin on evidence for this supposition that
>>>> spokes are sent from the factory containing residual stress"
>>>>
>>>> My cite above was intended as a rebuttal to your (frequent) denial
>>>> that mechanical stress relieving doesn't reduce residual stress.
>>>
>>> context: i was contrasting the credibility of someone that claims
>>> they can eliminate fatigue in a situation where they can't, with the
>>> fact that they're making unsupported suppositions.
>>>
>>> i've never said you cannot mechanically relieve residual stress. i
>>> /have/ however questioned whether spokes leave the factory with
>>> sufficient residual stress to initiate fatigue /and/ i've questioned
>>> the efficacy of "stress relief" applied after spoke manufacture.
>>> mechanical stress relief is typically only effective a short time
>>> after initial deformation. waiting more than a few hours [as will be
>>> necessary to ship from the spoke manufacturer to the wheel builder]
>>> will render this method useless. and then we can get into the debate
>>> of spoke squeezing itself and whether a process that produces no
>>> detectable deformation is actually yielding the material sufficiently
>>> to work as is being supposed, [timing issues aside of course].
>>
>> These claims (time dependency & lack of overall plastic yield) seem to
>> be not supported by the link above and others I have provided. I will
>> wait for you to provide any evidence to support your claims.
>
> dieter, mechanical metallurgy, isbn 0-07-016893-8, chapter 19.6,
> residual stresses in rod, wire & tubes. [forgive me if i don't dig out
> my old lecture notes - one of my old profs was mr. cold drawn steel.]
>
> typical mechanical relief strain is 1%-2%. for a 294mm spoke, where
> does that leave us? arguing as jobst does that the relief strain is at
> levels microscopic and can't be detected is suppositional straw clutching.

I think you are confusing techniques for completely removing (bulk)
residual stresses from those designed only to remove very localized
residual stresses. An example of the latter is given below. The
technique is identical (except for scale) as that for stress relieving
spokes.

From:
http://a257.g.akamaitech.net/7/257/....gpo.gov/cfr_2004/octqtr/pdf/46cfr54.30-5.pdf

http://tinyurl.com/y2co8y

"Subpart 54.30—Mechanical Stress
Relief
§ 54.30–1 Scope.
(a) Certain pressure vessels may be
mechanically stress relieved in accordance
with the requirements in this subpart.
(b) [Reserved]
§ 54.30–3 Introduction.
(a) Large conventional pressure vessels
used to transport liquefied petroleum
and natural gases, at ‘‘low temperatures’’
may often be difficult to
thermally stress relieve. Where no
other problem, such as corrosion exists,
mechanical stress relief will be
permitted for Class II–L pressure vessels.
(b) Mechanical stress relief serves to
cause small flaws, particularly in the
weld zone, to yield plastically at the
flaw tip resulting in a local relief of
stress and a blunting of the crack tip.
To achieve the maximum benefit from
mechanical stress relief, it is necessary
that the stresses so imposed be more
severe than those expected in normal
service life. At the same time, it is necessary
that the stresses which are imposed
are not so high as to result in appreciable
deformation or general yielding."

"§ 54.30–10 Method of performing mechanical
stress relief.
(a) The mechanical stress relief shall
be carried out in accordance with the
following stipulations using water as
the pressurizing medium:
(1) At a hydrostatic pressure (measured
at the tank top) of 11⁄2 times the
design pressure. (See UA–60(e) of the
ASME Code.)
(2) At a temperature of 70 °F. or the
service temperature plus 50 °F., whichever
is higher. Where the ambient temperature
is below 70 °F., and use of
water at that temperature is not practical,
the minimum temperature for
mechanical stress relief may be below
70 °F. but shall not be less than 50 °F.
above service temperature.
(3) The stress relief shall be at the required
temperature and pressure and
held for a period not less than 2 hours
per inch of metal thickness, but in no
case less than 2 hours."

 

[[email protected]]

On Fri, 03 Nov 2006 20:41:16 -0500, Peter Cole
<[email protected]> wrote:

[snip]

>(3) The stress relief shall be at the required
>temperature and pressure and
>held for a period not less than 2 hours
>per inch of metal thickness, but in no
>case less than 2 hours."

Dear Peter,

This is the first time that I've seen a time versus metal thickness
aspect of mechanical stress relief.

Two hours per inch of metal thickness works out to 7200 seconds per
25.4 mm, or 283 seconds per millimeter.

That would be about 10 to 12 minutes for a 2 mm thick spoke, but I
assume that the shape, absolute pressure, and absolute thickness would
modify that naive calculation considerably.

Still, it seems to be a few orders of magnitude longer than most wheel
builders squeeze their spokes.

Do you know of anything that might shed light on the effect of the
length of time that spokes are squeezed together?

Right now, I'm using a vise-rig to stretch spokes, but they're being
left on the rack, so to speak, for much longer than I can hope to
squeeze with my bare hand.

Cheers,

Carl Fogel

 

[jim beam]

dvt wrote:
> jim beam wrote:
>> dvt wrote:
>>> How would you design a test for residual stress in spoke elbows? I
>>> have access to some equipment, but not "neutron/x-ray inferometry"
>>> [sic] that you mentioned in the above-mentioned thread.
>>
>> chlorides are well known for their stress corrosion in stainless
>> steel. google will help you identify other agents as well. simply
>> leave a spoke in a jar of the requisite solution and wait.
>
> Thanks for the response, jim.
>
> I see how that tests for corrosion. As I have said in the past, my
> metallurgy knowledge is not enough to make the leap from a corrosion
> test to a test for residual stress. Maybe you can help me make that
> connection.

long, long and highly specialized subject, but basically, there's a
synergy/susceptibility between certain materials and certain
environmental factors that lead to a specific form of corrosion.
materials that may be passive in an unstressed state, can become highly
susceptible to localized corrosion when stressed. certain
material/reagent pairs exhibit dramatic crack propensity. examples
include http://www.corrosion-doctors.org/Forms/scc-environments.htm

>
>>>> sure, if residual stress is sufficient to be a factor, it can help
>>>> to reduce it. but the necessary supposition that spoke
>>>> manufacturers are incompetent in order to take that position is
>>>> something of a stretch.
>>>
>>> I don't think it takes a supposition of incompetence.
>>
>> i disagree. according to jobstian theory, the supposition is that
>> spokes leave the factory with residual stress, ergo the manufacturer
>> is incompetent for not relieving it.
>
> I don't think it requires incompetence on the spoke manufacturer's part.
> The wheel building process can change the angle of the elbow, producing
> some residual stress.

well, i don't agree with the "correcting the spoke line" practice for
exactly this reason. most hub flanges are canted and drilled so that
spoke angles are optimum for their exit geometry.

http://www.flickr.com/photos/38636024@N00/104463818/

to "correct the spoke line" /prior/ to raising spoke tension
sufficiently to bed the spokes fully into the hub not only leaves the
spoke elbow with a non-factory angle, it can indeed introduce residual
stress into an otherwise optimized component.

> I know you have debated that point (with Benjamin
> Lewis, if memory serves). His photographs show that it's at least
> possible to change the elbow angle by building a wheel, and it agrees
> with my experience, so I'm sticking with that. Since we've bent the
> spoke elbow without any form of stress relief, there are quite likely
> some residual stresses.

i've built wheels and have "stress relieved" them using the "mavic
method" [block of wood and leaning on the rim] and have subsequently
disassembled them. there may have been some elbow angle change, but
it's not one i have been able to see in comparison with new spokes.
granted, you need to be careful to achieve this, but it seems possible.

>
> So if the manufacturer supplied spokes without residual stress, residual
> stresses would be induced during the wheel build. Supplying spokes free
> of residual stress would be rather pointless, in that case.

i disagree. no manufacturer that cares about their reputation enough to
put their name on their product is going to risk damage to sales by
shipping something they know to be a problem. otherwise why bother with
the considerable expense of vacuum degassed steels as well?

>
>> perhaps, but we'd then have to show that residual stress was a factor
>> and that material quality was not.
>
> I think this is where we disagree. The first part of your sentence makes
> sense to me, but the second part doesn't. Why does one have to be
> exclusive of the other? Why couldn't some failures be caused by a
> *combination* of material quality and residual stress?

they could, but why spend your time fixing the hubcap if the fan belt's
broken?

>
>> seriously, if spoke squeezing was able to eliminate fatigue
>> regardless, don't you think manufacturers would just do that and use
>> cheaper materials and cheaper processing rather than what they do now?
>
> Perhaps if they were able to assume no residual stresses in a spoke in a
> built wheel, they *could* go to cheaper materials and processing.

doesn't work that way in practice. real world usage is that material
quality /does/ affect failure rate. spoke squeezing has no quantified
efficacy that i've ever seen.

 

[jim beam]

dvt wrote:
> jim beam wrote:
>> Peter Cole wrote:
>>> jim beam wrote:
>>>> where i come from, the words "endurance limit" are reserved for
>>>> graphs showing the knee, "fatigue limit" is for those that don't.
>
> Amen to that. Google for "define:endurance limit" and you'll find that
> most definitions say endurance = fatigue. Ugh.
>
> I do think you "knee" and "asymptote" terms are not quite correct,
> though. You use asymptote as a synonym for horizontal. Most S/N curves
> that I've seen have a knee (a change in slope). They have two
> asymptotes; one above the knee, one below. But materials showing an
> endurance limit have a *horizontal* asymptote above the knee.
>
>>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>>> paper don't know what they're doing, but be honest peter, there are
>>>> not a lot of other materials out there supporting this position -
>>>> most to the contrary in fact.
>>>
>>>
>>> Cite 'em.
>>
>> dude, google 'stainless steel endurance limit'. there are 170,000
>> returns, and while i haven't read all of them, the first 50 or so
>> don't reference an asymptote - or at least, not without also confusing
>> fatigue and endurance limits.
>
> I just did the Google search that you suggested. Since I was looking for
> pdfs in the hopes of finding graphics, the first link I clicked was the
> 5th one on the list:
> <http://www.aksteel.com/pdf/markets_products/stainless/austenitic/AK_Nit30PDB_1104_2.pdf>
>
>
> Look at pages 9 and 10. They show a horizontal portion in the S/N curve.
>
yes, but dude, there are not enough data points on those graphs to
support their curve fit!

context: this is a marketing piece aimed at engineers - engineers, for
some reason that is still unclear to me, seem to be fixated with
endurance limits. fact is, it's /highly/ dangerous to design with
endurance limits in mind - they are /not/ reliable.

i'm willing and able to accept new data if it's available, but back in
the day when i used to do this stuff, stainless wasn't attributed with
an endurance limit because there was no interstitial diffusion mechanism
to support it. maybe things have changed, but i doubt it.

 

[jim beam]

Peter Cole wrote:
> jim beam wrote:
>> Peter Cole wrote:
>>> jim beam wrote:
>>>> Peter Cole wrote:
>>>>> jim beam wrote:
>>>>>> Peter Cole wrote:
>>>>>>> jim beam wrote:
>>>>>>>> Peter Cole wrote:
>>>>>>>>> jim beam wrote:
>>>>>>>>>> [email protected] wrote:
>>>>>>>>>>> In article <[email protected]>,
>>>>>>>>>>> [email protected] says...
>>>>>>>>>>>
>>>>>>>>>>>> since we have yet to be privileged with sight of the
>>>>>>>>>>>> jobstian research that will revolutionize the world of
>>>>>>>>>>>> materials by showing how fatigue can be eliminated in a
>>>>>>>>>>>> material with no endurance limit
>>>>>>>>>>>
>>>>>>>>>>> I thought most spokes were 304 stainless steel?
>>>>>>>>>>>
>>>>>>>>>> there's various grades used for spokes, but none have an
>>>>>>>>>> endurance
>>>>>>>>>> limit. they have what's called a "fatigue limit" which is an
>>>>>>>>>> attempt to
>>>>>>>>>> quantify the stress they can endure for a given number of fatigue
>>>>>>>>>> cycles, but that's not the same thing. an endurance limit is
>>>>>>>>>> the stress
>>>>>>>>>> at which the material can endure an infinite number of cycles,
>>>>>>>>>> and is
>>>>>>>>>> characterized by a "knee" in the s-n graph where the line goes
>>>>>>>>>> horizontal. mild steel is the classic material with this
>>>>>>>>>> property, but
>>>>>>>>>> some titanium alloys also have it. no stainless alloys that i
>>>>>>>>>> know of
>>>>>>>>>> have it.
>>>>>>>>>
>>>>>>>>> <users.wpi.edu/~cfurlong/me3320/lect13/Lect13.pdf>
>>>>>>>>> <bama.ua.edu/~mweaver/courses/MTE455/MTE455_2006_L26.pdf>
>>>>>>>>>
>>>>>>>>> That's not what they're teaching today.
>>>>>>>>
>>>>>>>> how so?
>>>>>>>>
>>>>>>>> regarding endurance limit, both seem to corroborate the same
>>>>>>>> definition as i, although the second uses fatigue limit and
>>>>>>>> endurance limit interchangeably between diagrams and text.
>>>>>>>>
>>>>>>>> regarding which alloys evidence an endurance limit, the first
>>>>>>>> one says some types of stainless can show it, but i saw no
>>>>>>>> examples. or if they did, and/or you know any, please share.
>>>>>>>> i'm interested because fwiu, endurance limits originate with the
>>>>>>>> same mechanism that causes strain aging, i.e. interstitial
>>>>>>>> elements [carbon in mild steel and oxygen in titanium] locking
>>>>>>>> dislocations. the thing with chromium passivated stainless is
>>>>>>>> that there's very little carbon possible without having chromium
>>>>>>>> carbide precipitate at grain boundaries, a big no-no for
>>>>>>>> fatigue, among other things. there are other types of
>>>>>>>> "stainless", but they're not something you or i would buy off
>>>>>>>> the shelf and i've not heard of their use for fatigue
>>>>>>>> resistance. i definitely don't see them being used in bicycle
>>>>>>>> spokes.
>>>>>>>
>>>>>>> As you say, the first notes claim some stainless steels have a
>>>>>>> flat S/N curve, some don't.
>>>>>>
>>>>>> i'm not saying that. i've not seen a stainless steel that
>>>>>> evidences an endurance limit, as defined by a "knee" in the
>>>>>> graph. some stainless steels appear to flatten, but there is no
>>>>>> knee, hence no endurance limit.
>>>>>>
>>>>>>> The second reference claims that all stainless are flat (pg, 4)
>>>>>>> "Ferrous metals and other strain aging materials -- Examples: low
>>>>>>> carbon steel, stainless steel, titanium, etc."
>>>>>>
>>>>>> cite the alloys and explain the mechanism. endurance limit is
>>>>>> cause by the mechanism i explained, carbon in ferrous metals, and
>>>>>> is problematic in chromium passivated stainless, again for the
>>>>>> reasons i explained.
>>>>>>
>>>>>>>
>>>>>>> Another source is:
>>>>>>> <http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1>
>>>>>>>
>>>>>>>
>>>>>>> Sec. 5.1
>>>>>>>
>>>>>>> "The S-N curves for ferrous and titanium alloys exhibit a
>>>>>>> limiting stress below which failure does not occur; this is
>>>>>>> called the fatigue or the endurance limit. The branch point or
>>>>>>> "knee" of the curve lies normally in the 105 to 107 cycle range.
>>>>>>> In aluminium and other nonferrous alloys there is no stress
>>>>>>> asymptote and a finite fatigue life exists at any stress level.
>>>>>>> All materials, however, exhibit a relatively flat curve in the
>>>>>>> high-cycle region, ie. at lives longer than about 105 cycles."
>>>>>>
>>>>>> which supports what i said before.
>>>>>
>>>>> I cited 3 sources, 2 of which said all stainless steels have an
>>>>> endurance limit, the other said some did, some didn't.
>>>>
>>>> with respect, one of those two uses "endurance limit" and "fatigue
>>>> limit" interchangeably. where i come from, the words "endurance
>>>> limit" are reserved for graphs showing the knee, "fatigue limit" is
>>>> for those that don't. engineers don't seem to regard materials
>>>> definitions as important as materials people, hence the dearth of
>>>> engineering papers perpetuating the confusion.
>>>>
>>>>>
>>>>> From:
>>>>> <www.stainless-steel-world.net/pdf/11021.pdf>
>>>>>
>>>>> Page 30 shows an S/N graph for 2 types of stainless, one of which
>>>>> is common 316. There is clearly a fatigue limit for both alloys.
>>>>
>>>> ok, that's a good link, citing specific alloys. i have two comments:
>>>>
>>>> 1. endurance limit at 20-25% of yield is not really a reliable benefit.
>>>
>>> From:
>>> <http://www.azom.com/details.asp?ArticleID=863>
>>>
>>> Tensile strength of 316 is given as 515MPa = 75ksi.
>>>
>>> From S/N referenced above, the asymptote is about 33ksi.
>>>
>>> Ratio is 44%, typical for steel.
>>
>> there's no asymptote referenced in the azom link.
>
>
> No, the graph (with asymptote) is in:
> www.stainless-steel-world.net/pdf/11021.pdf.
>
>
>
>> the one showing ~35% from the stainless-steel-world link is for DP3,
>> whatever that is.
>
> http://www.duplexss.com/
>
>
>> and fyi, ~35% is indistinguishable from the fatigue limit for a lot of
>> structural aluminum alloys, so i don't think you'll find many people
>> relying on an endurance limit for DP3 in their fatigue design.
>
> The endurance limit I referred to was for 316. The ratio of the
> endurance limit to tensile strength for 316 according to the sources
> above is 44%.
>
>
>>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>>> paper don't know what they're doing, but be honest peter, there are
>>>> not a lot of other materials out there supporting this position -
>>>> most to the contrary in fact.
>>>
>>>
>>> Cite 'em.
>>
>> dude, google 'stainless steel endurance limit'. there are 170,000
>> returns, and while i haven't read all of them, the first 50 or so
>> don't reference an asymptote - or at least, not without also confusing
>> fatigue and endurance limits. it's been years since you and i first
>> started this kind of debate, and so far you've cited just one link
>> with a graphical asymptote. everything else shows a more typical
>> flattening curve, no asymptote. done any curve fitting exercises lately?
>
> Quibbling about terminology aside, all 3 of the sources I cited:
>
> http://users.wpi.edu/~cfurlong/me3320/lect13/Lect13.pdf
> bama.ua.edu/~mweaver/courses/MTE455/MTE455_2006_L26.pdf
> http://www.kuleuven.ac.be/bwk/materials/Teaching/master/wg12/l0200.htm#SEC_6_1

that last one is great in that slide 4 shows crack initiation at an
inclusion - the reason why we degas steels for maximum fatigue resistance.

but that's tangential. otherwise i see repeated confusion between
endurance limit and fatigue limit. there's not much chance for an
engineer to get this stuff right if they're not being taught correctly
in the first place.

>
>
> unambiguously discussed S/N curves where the materials had infinite
> stress life (horizontal asymptote). There was no confusion about the
> topic. 2 of the authors contended that all stainless steels have this
> characteristic, the 3rd said some.

there is no confusion about fatigue peter - don't say there is. what i
am questioning is the mechanism by which stainless steels are alleged to
have endurance limits. in mild steels, it's diffusion of carbon - a
well known and researched mechanism. in stainless, what is it? just
making a curve fit assumption doesn't do it for me.

>
> The last source I cited: www.stainless-steel-world.net/pdf/11021.pdf,
> showed a S/N curve for a specific alloy (which you requested), that of
> the very common 316.
>
> Do you have any examples of S/N curves that dispute this? Say one for 316?

http://www.nrc.gov/reading-rm/doc-collections/nuregs/contract/cr6878/cr-6878-est-fatigue-end.pdf

>
> And I'd appreciate you stop calling me "dude".

du^H^, oh, never mind.

 

[Peter Cole]

jim beam wrote:
> dvt wrote:
>> jim beam wrote:
>>> Peter Cole wrote:
>>>> jim beam wrote:
>>>>> where i come from, the words "endurance limit" are reserved for
>>>>> graphs showing the knee, "fatigue limit" is for those that don't.
>>
>> Amen to that. Google for "define:endurance limit" and you'll find that
>> most definitions say endurance = fatigue. Ugh.
>>
>> I do think you "knee" and "asymptote" terms are not quite correct,
>> though. You use asymptote as a synonym for horizontal. Most S/N curves
>> that I've seen have a knee (a change in slope). They have two
>> asymptotes; one above the knee, one below. But materials showing an
>> endurance limit have a *horizontal* asymptote above the knee.
>>
>>>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>>>> paper don't know what they're doing, but be honest peter, there are
>>>>> not a lot of other materials out there supporting this position -
>>>>> most to the contrary in fact.
>>>>
>>>>
>>>> Cite 'em.
>>>
>>> dude, google 'stainless steel endurance limit'. there are 170,000
>>> returns, and while i haven't read all of them, the first 50 or so
>>> don't reference an asymptote - or at least, not without also
>>> confusing fatigue and endurance limits.
>>
>> I just did the Google search that you suggested. Since I was looking
>> for pdfs in the hopes of finding graphics, the first link I clicked
>> was the 5th one on the list:
>> <http://www.aksteel.com/pdf/markets_products/stainless/austenitic/AK_Nit30PDB_1104_2.pdf>
>>
>>
>> Look at pages 9 and 10. They show a horizontal portion in the S/N curve.
>>
> yes, but dude, there are not enough data points on those graphs to
> support their curve fit!
>
> context: this is a marketing piece aimed at engineers - engineers, for
> some reason that is still unclear to me, seem to be fixated with
> endurance limits. fact is, it's /highly/ dangerous to design with
> endurance limits in mind - they are /not/ reliable.
>
> i'm willing and able to accept new data if it's available, but back in
> the day when i used to do this stuff, stainless wasn't attributed with
> an endurance limit because there was no interstitial diffusion mechanism
> to support it. maybe things have changed, but i doubt it.

Here's an almost 50 year old NASA paper that shows endurance limit for
stainless.

ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19640015138_1964015138.pdf

 

[Peter Cole]

jim beam wrote:
> Peter Cole wrote:
>>
>> The last source I cited: www.stainless-steel-world.net/pdf/11021.pdf,
>> showed a S/N curve for a specific alloy (which you requested), that of
>> the very common 316.
>>
>> Do you have any examples of S/N curves that dispute this? Say one for
>> 316?
>
> http://www.nrc.gov/reading-rm/doc-collections/nuregs/contract/cr6878/cr-6878-est-fatigue-end.pdf

I see nothing in that paper to support your position, or even anything
relevant to this thread. Perhaps you could point out the sections.

 

[jim beam]

Peter Cole wrote:
> jim beam wrote:
>> dvt wrote:
>>> jim beam wrote:
>>>> Peter Cole wrote:
>>>>> jim beam wrote:
>>>>>> where i come from, the words "endurance limit" are reserved for
>>>>>> graphs showing the knee, "fatigue limit" is for those that don't.
>>>
>>> Amen to that. Google for "define:endurance limit" and you'll find
>>> that most definitions say endurance = fatigue. Ugh.
>>>
>>> I do think you "knee" and "asymptote" terms are not quite correct,
>>> though. You use asymptote as a synonym for horizontal. Most S/N
>>> curves that I've seen have a knee (a change in slope). They have two
>>> asymptotes; one above the knee, one below. But materials showing an
>>> endurance limit have a *horizontal* asymptote above the knee.
>>>
>>>>>> 2. fatigue is statistical. i'm not saying the publishers of this
>>>>>> paper don't know what they're doing, but be honest peter, there
>>>>>> are not a lot of other materials out there supporting this
>>>>>> position - most to the contrary in fact.
>>>>>
>>>>>
>>>>> Cite 'em.
>>>>
>>>> dude, google 'stainless steel endurance limit'. there are 170,000
>>>> returns, and while i haven't read all of them, the first 50 or so
>>>> don't reference an asymptote - or at least, not without also
>>>> confusing fatigue and endurance limits.
>>>
>>> I just did the Google search that you suggested. Since I was looking
>>> for pdfs in the hopes of finding graphics, the first link I clicked
>>> was the 5th one on the list:
>>> <http://www.aksteel.com/pdf/markets_products/stainless/austenitic/AK_Nit30PDB_1104_2.pdf>
>>>
>>>
>>> Look at pages 9 and 10. They show a horizontal portion in the S/N curve.
>>>
>> yes, but dude, there are not enough data points on those graphs to
>> support their curve fit!
>>
>> context: this is a marketing piece aimed at engineers - engineers, for
>> some reason that is still unclear to me, seem to be fixated with
>> endurance limits. fact is, it's /highly/ dangerous to design with
>> endurance limits in mind - they are /not/ reliable.
>>
>> i'm willing and able to accept new data if it's available, but back in
>> the day when i used to do this stuff, stainless wasn't attributed with
>> an endurance limit because there was no interstitial diffusion
>> mechanism to support it. maybe things have changed, but i doubt it.
>
> Here's an almost 50 year old NASA paper that shows endurance limit for
> stainless.
>
> ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19640015138_1964015138.pdf

at elevated temperatures!!!

which leads me back to the question i originally asked - what is the
mechanism for endurance limit in stainless? you get endurance in mild
steel at ambient because the diffusion coefficient of carbon is
sufficiently high to allow dislocation locking. what's going on here?
this paper is just an engineering test - there is no material or
mechanism investigation whatsoever.