How strong actually is carbon fibre?
#51
Senior Member
Since CF doesn't appear to have a fatigue limit, "correctly used and designed" really needs to include specifying the intended safe lifespan - including a suitably large safety factor. Because since CF doesn't have a fatigue limit, it will eventually fail in routine use. How long will that take? Good question; and it depends in great part on the design.
And since AL similarly lacks a fatigue limit, the same is true for AL as well.
In practice, it probably doesn't matter for the next decade or two (or three or . . . dunno). But when frames made from both are old enough to be considered "old vintage" or "antique", if I'm still around the design lifespan of a CF or AL frame or fork would be something I'd really like to know before riding one made today that far in future.
Last edited by Hondo6; 09-17-21 at 11:06 AM. Reason: Correct typo.
#52
Habitual User
Join Date: Jan 2020
Location: Altadena, CA
Posts: 7,986
Bikes: 2023 Niner RLT 9 RDO, 2018 Trek Procaliber 9.9 RSL, 2018 Storck Fascenario.3 Platinum, 2003 Time VX Special Pro, 2001 Colnago VIP, 1999 Trek 9900 singlespeed, 1977 Nishiki ONP
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 4946 Post(s)
Liked 8,087 Times
in
3,826 Posts
Bingo - with one caveat.
Since CF doesn't appear to have a fatigue limit, "correctly used and designed" really needs to include specifying the intended safe lifespan - including a suitably large safety factor. Because since CF doesn't have a fatigue limit, it will eventually fail in routine use. How long will that take? Good question; and it depends in great part on the design.
And since AL similarly lacks a fatigue limit, the same is true for AL as well.
In practice, it probably doesn't matter for the next decade or two (or three or . . . dunno). But when frames made from both are old enough to be considered "old vintage" or "antique", if I'm still around the design lifespan of a CF or AL frame or fork would be something I'd really like to know before riding one made today that far in future.
Since CF doesn't appear to have a fatigue limit, "correctly used and designed" really needs to include specifying the intended safe lifespan - including a suitably large safety factor. Because since CF doesn't have a fatigue limit, it will eventually fail in routine use. How long will that take? Good question; and it depends in great part on the design.
And since AL similarly lacks a fatigue limit, the same is true for AL as well.
In practice, it probably doesn't matter for the next decade or two (or three or . . . dunno). But when frames made from both are old enough to be considered "old vintage" or "antique", if I'm still around the design lifespan of a CF or AL frame or fork would be something I'd really like to know before riding one made today that far in future.
__________________
"Swedish fish. They're protein shaped." - livedarklions
"Swedish fish. They're protein shaped." - livedarklions
#53
I'm good to go!
Join Date: Jul 2017
Location: Mississippi
Posts: 14,988
Bikes: Tarmac Disc Comp Di2 - 2020
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 6193 Post(s)
Liked 4,810 Times
in
3,318 Posts
Do you fly anywhere via an aircraft? Aircraft designers and engineers don't really know either. They make an educated suggestion based on testing and analysis. Then as real world evidence comes in from inspections done while aircraft are in service they adjust them when necessary. And there are changes to inspection and maintenance required for a particular aircraft component all the time. Even for privately owned aircraft.
Essentially it just boils down to someone saw something started to look bad and then something was determined to be a solution. Just as they do you can look at your bike, and if you see a problem, find out how to solve it. It's not like it is about to explode underneath you.
Though those head tubes that separated from the top and down tubes of some aluminum bikes being used in BMX and other events going down extremely steep ravines might have left their riders in about the same shape as if they had exploded.
Essentially it just boils down to someone saw something started to look bad and then something was determined to be a solution. Just as they do you can look at your bike, and if you see a problem, find out how to solve it. It's not like it is about to explode underneath you.
Though those head tubes that separated from the top and down tubes of some aluminum bikes being used in BMX and other events going down extremely steep ravines might have left their riders in about the same shape as if they had exploded.
Last edited by Iride01; 09-17-21 at 01:16 PM.
#54
Senior Member
Your materials knowledge depth is clearly way beyond mine, but I am a bit puzzled that multiple CF bike manufacturers have stated that the fatigue life of CF is infinite. This seems to conflict with what you're saying. Or, maybe I'm just not understanding the different terms clearly, which is certainly possible. Are you referring to the fibers themselves, or to the composite, which includes both fibers and resin? If it's the composite, I would expect that to vary a lot with the properties of the resin material.
Fatigue life is the number of stress cycles using a specified amount of stress at which a material can be expected to fail due to material fatigue.
Fatigue limit (some sources instead use the term an endurance limit) refers to the amount of stress below which a material will never fail due to material fatigue, regardless of how many times that stress is applied and removed.
Every material has a fatigue life for large stresses. Not every material has a fatigue limit.
Steel and titanium are generally believed to have a fatigue limit. AL and CF composites are not.
CF fatigues. Due to its nature, it behaves and fatigues differently than metals - but it does fatigue. IMO anyone claiming otherwise is either (1) incorrect or (2) is deliberately choosing misleading wording that implies that without actually stating it outright (e.g., "Will last a lifetime!"). I suppose I could be wrong, but I don't believe I am.
Lennard Zinn wrote an article for VeloNews that gave a pretty decent, nontechnical explanation of the issue some time back. If you're interested, here's the link.
https://www.velonews.com/gear/techni...fiber-fatigue/
As I've said before: CF technology is fairly mature now; ditto AL. Frames and forks can be made from either to last as long as desired using proper design, and doubtless today are designed to last a long time - probably longer than you or I will be around. Absent damage, I wouldn't sweat their durability.
But I would check either periodically for signs of damage. And if you notice anything suspicious, I'd have it checked out pronto by a competent pro.
Last edited by Hondo6; 09-17-21 at 01:38 PM. Reason: Wordsmithing for clarity.
Likes For Hondo6:
#55
Habitual User
Join Date: Jan 2020
Location: Altadena, CA
Posts: 7,986
Bikes: 2023 Niner RLT 9 RDO, 2018 Trek Procaliber 9.9 RSL, 2018 Storck Fascenario.3 Platinum, 2003 Time VX Special Pro, 2001 Colnago VIP, 1999 Trek 9900 singlespeed, 1977 Nishiki ONP
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 4946 Post(s)
Liked 8,087 Times
in
3,826 Posts
"Fatigue life" and "fatigue limit" do not refer to the same concept.
Fatigue life is the number of stress cycles using a specified amount of stress at which a material can be expected to fail due to material fatigue.
Fatigue limit (some sources instead use the term an endurance limit) refers to the amount of stress below which a material will never fail due to material fatigue, regardless of how many times that stress is applied and removed.
Every material has a fatigue life for large stresses. Not every material has a fatigue limit.
Steel and titanium are generally believed to have a fatigue limit. AL and CF composites are not.
CF fatigues. Due to its nature, it behaves and fatigues differently than metals - but it does fatigue. IMO anyone claiming otherwise is incorrect or is deliberately choosing misleading wording that implies that without actually stating it outright (e.g., "Will last a lifetime"). I suppose I could be wrong, but I don't believe I am.
Lennard Zinn wrote an article for VeloNews that gave a pretty decent, nontechnical explanation of the issue at some time back. If you're interested, here's the link.
https://www.velonews.com/gear/techni...fiber-fatigue/
As I've said before: CF technology is fairly mature now; ditto AL. Frames and forks can be made from either to last as long as desired using proper design, and doubtless today are designed to last a long time - probably longer than you or I will be around. Absent damage, I wouldn't sweat their durability.
But I would check either periodically for signs of damage. And if you notice anything suspicious, I'd have it checked out pronto by a competent pro.
Fatigue life is the number of stress cycles using a specified amount of stress at which a material can be expected to fail due to material fatigue.
Fatigue limit (some sources instead use the term an endurance limit) refers to the amount of stress below which a material will never fail due to material fatigue, regardless of how many times that stress is applied and removed.
Every material has a fatigue life for large stresses. Not every material has a fatigue limit.
Steel and titanium are generally believed to have a fatigue limit. AL and CF composites are not.
CF fatigues. Due to its nature, it behaves and fatigues differently than metals - but it does fatigue. IMO anyone claiming otherwise is incorrect or is deliberately choosing misleading wording that implies that without actually stating it outright (e.g., "Will last a lifetime"). I suppose I could be wrong, but I don't believe I am.
Lennard Zinn wrote an article for VeloNews that gave a pretty decent, nontechnical explanation of the issue at some time back. If you're interested, here's the link.
https://www.velonews.com/gear/techni...fiber-fatigue/
As I've said before: CF technology is fairly mature now; ditto AL. Frames and forks can be made from either to last as long as desired using proper design, and doubtless today are designed to last a long time - probably longer than you or I will be around. Absent damage, I wouldn't sweat their durability.
But I would check either periodically for signs of damage. And if you notice anything suspicious, I'd have it checked out pronto by a competent pro.
__________________
"Swedish fish. They're protein shaped." - livedarklions
"Swedish fish. They're protein shaped." - livedarklions
#56
Senior Member
Thanks for the info. I've done a little other reading, too. I think I'm starting to get it now. A properly designed CF structure (like a bike frame) will have a fatigue life end point eventually, but it may be so many cycles that it will realistically outlast the lifetime of the user, assuming "normal" operating conditions. Am I close?
Assuming the designer did a good design, absent damage (crash or other impact) and significant material or fabrication flaws, a long lifetime is IMO very likely. Precisely how long depends on the designer and how "hard" the structure is used, but good design should take stresses from expected use into account and add a decent safety factor when designing the structure.
I'd still inspect both AL and CF regularly for evidence of failure. And do take note of the fact that metals show evidence of failure first on the outside, while CF may not - it has several failure modes, some of which are internal and externally not visible before actual failure (though they can sometimes be detected via other means - see Zinn's article for further info on this).
FWIW: one of my two current bikes is has an AL frame with a CF fork; the other is steel. Absent a crash, I regard both as very likely to be perfectly safe for as long as I expect to be still riding.
Last edited by Hondo6; 09-17-21 at 02:04 PM. Reason: Wording changes for clarity.
Likes For Hondo6:
#58
Habitual User
Join Date: Jan 2020
Location: Altadena, CA
Posts: 7,986
Bikes: 2023 Niner RLT 9 RDO, 2018 Trek Procaliber 9.9 RSL, 2018 Storck Fascenario.3 Platinum, 2003 Time VX Special Pro, 2001 Colnago VIP, 1999 Trek 9900 singlespeed, 1977 Nishiki ONP
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 4946 Post(s)
Liked 8,087 Times
in
3,826 Posts
Lotto-Jumbo rider, Steven Kruijswijk, was involved in a heavy crash in one of the early stages of the TdF this year that completely broke out his seat stay. He got back on the bike and finished the stage, not even realizing it was missing.
__________________
"Swedish fish. They're protein shaped." - livedarklions
"Swedish fish. They're protein shaped." - livedarklions
Likes For Eric F:
#59
Senior Member
They indeed are. When properly designed, they're far stronger than is absolutely necessary.
But after seeing this photo, there's no way in hell I'd ride that frame afterwards. Or re-use the handlebars, stem, or fork on another bike. Internal damage you can't see in any of them is a possibility. And since CF often fails without warning or external sign if it is damaged, undetected internal damage can be a Very Bad Thing - especially in a safety-critical component like handlebars, stem, or fork.
But after seeing this photo, there's no way in hell I'd ride that frame afterwards. Or re-use the handlebars, stem, or fork on another bike. Internal damage you can't see in any of them is a possibility. And since CF often fails without warning or external sign if it is damaged, undetected internal damage can be a Very Bad Thing - especially in a safety-critical component like handlebars, stem, or fork.
Last edited by Hondo6; 09-17-21 at 03:37 PM. Reason: Poor initial wording.
#60
Senior Member
Join Date: Dec 2009
Location: Melbourne, Oz
Posts: 9,547
Bikes: https://weightweenies.starbike.com/forum/viewtopic.php?f=10&t=152015&p=1404231
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 1529 Post(s)
Liked 718 Times
in
510 Posts
If you're bending carbon to the point it shatters, any comparable metal component would have yielded too.
#61
Banned
Join Date: Apr 2010
Location: Lincoln Ne
Posts: 9,924
Bikes: RANS Stratus TerraTrike Tour II
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Quoted: 3352 Post(s)
Liked 1,056 Times
in
635 Posts
It appears that the NTSB disagrees. Quoted from the Executive Summary of the NTSB's report on the accident, which is readily available in PDF format on the Internet:
The "Advanced Aircraft Maneuvering Program" referenced here appears to be at least part of AA's pilot training curriculum. And in another part of the report, it goes on to indicate that the first officer had a history of being aggressive (a previous aircraft Captain who had flown with the individual as First Officer had referred to it as being "very aggressive") in the use of rudder controls during a similar previous incident while flying another type of aircraft. The Captain of that flight also indicated he regarded the Flight Officer's actions as excessive in that case; however, also indicated he felt that this behavior was "our of character". (This previous behavior on the part of AA 587's late Flight Officer is discussed on p. 12-13 of the report.) The Advanced Aircraft Maneuvering Program and its issues are discussed in detail later in the report.
I have no axe to grind here, and I'm not a pilot. But yeah - per the NTSB, there apparently were issues with the Flight Officer's behavior that contributed to the crash. There also appear to have been issues with the training he received and with the aircraft's rudder system design as well.
The "Advanced Aircraft Maneuvering Program" referenced here appears to be at least part of AA's pilot training curriculum. And in another part of the report, it goes on to indicate that the first officer had a history of being aggressive (a previous aircraft Captain who had flown with the individual as First Officer had referred to it as being "very aggressive") in the use of rudder controls during a similar previous incident while flying another type of aircraft. The Captain of that flight also indicated he regarded the Flight Officer's actions as excessive in that case; however, also indicated he felt that this behavior was "our of character". (This previous behavior on the part of AA 587's late Flight Officer is discussed on p. 12-13 of the report.) The Advanced Aircraft Maneuvering Program and its issues are discussed in detail later in the report.
I have no axe to grind here, and I'm not a pilot. But yeah - per the NTSB, there apparently were issues with the Flight Officer's behavior that contributed to the crash. There also appear to have been issues with the training he received and with the aircraft's rudder system design as well.
#62
Senior Member
Join Date: Sep 2017
Posts: 7,880
Mentioned: 38 Post(s)
Tagged: 0 Thread(s)
Quoted: 6963 Post(s)
Liked 10,964 Times
in
4,689 Posts
Of course that is BS. A pilot will move the controls to get the response he or she wants. Do you really think that a pilot will say to himself ----gee that is all the rudder I dare to put in------------In the case of the tail breaking off the Airbus airplanes is a combination of CF design not being strong enough, and the fly by wire being programed wrong. On the very light weight Cessna 150s I learned to fly in, in cross wind landings often I used full rudder to keep the plane straight on landing.
I have an old friend who flew F4 Phantoms in Vietnam - this was back when the F4 was the most advanced fighter jet in the world. My friend once described how his jet could pull enough G-forces to damage itself, even enough to cause the pilot to momentarily lose consciousness. I expressed surprise, and asked how he handled it. His response was, "You have to fly it carefully."
I believe Jim flew 134 missions and obviously lived to tell about it, so I suspect that he knew what he was talking about.
#63
Senior Member
Yes - if it isn't obvious, the above was 100% sarcasm.
From what I've seen, the NTSB actually has a very good reputation for both integrity and objective analysis when it comes to accident investigation. And given the Flight Officer's documented past history, I tend to think that maybe they got this one right. Pilots sometimes do goof. Sometimes that's the sole cause of an incident; sometimes it's not a factor at all. Here, I think we're somewhere in between - e.g., overreaction coupled with and/or aggravated by sub-optimal training and a design that also wasn't optimal. IMO all three contributed to that particular crash.
But hey - believe whatever you like.
Likes For Hondo6:
#65
Senior Member
Join Date: Dec 2009
Location: Melbourne, Oz
Posts: 9,547
Bikes: https://weightweenies.starbike.com/forum/viewtopic.php?f=10&t=152015&p=1404231
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 1529 Post(s)
Liked 718 Times
in
510 Posts
Not so much in practice, since if you're shattering your carbon, you no doubt have bigger problems than the shattered carbon. Consider the unsuccessfully garaged Look above; the rim is bent pretty far without shattering, and the rest has held its shape. If you're on that bike when you break that frame... It doesn't matter what the bike was made of at that point.
And if, instead of carbon, you had highly optimised aluminium, that's likely to asplode as well. Besides, carbon doesn't shatter like pottery, does it. Some of the fibres break, the matrix actually shatters, and usually there's remaining fibres holding the mess together.
And if, instead of carbon, you had highly optimised aluminium, that's likely to asplode as well. Besides, carbon doesn't shatter like pottery, does it. Some of the fibres break, the matrix actually shatters, and usually there's remaining fibres holding the mess together.
Last edited by Kimmo; 09-18-21 at 05:54 PM.
#66
Senior Member
Join Date: Mar 2010
Posts: 1,659
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 1248 Post(s)
Liked 1,323 Times
in
674 Posts
Your desperation gets a bit pathetic at times. Both statements made have no basis in reality and were proven false. You would be better served arguing that carbon fibre is highly conductive and prone to lightning strikes increasing risk to riders.
#67
Banned
Join Date: Apr 2010
Location: Lincoln Ne
Posts: 9,924
Bikes: RANS Stratus TerraTrike Tour II
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Quoted: 3352 Post(s)
Liked 1,056 Times
in
635 Posts
The thing about CF is the fact that while as far as any one knows, the CF itself doesnt lose strength with age, the resin that holds the CF into shape DOES. The resin is a form of plastic, and plastics do age. One hundred year old steel bikes can still be ridden, but I suspect that when some of todays CF bikes are 100 years old, they would easily shatter because the resin has aged and turned brittle.
BTW if I have unlimited funds, I would buy a titanium frame.
BTW if I have unlimited funds, I would buy a titanium frame.
#68
Habitual User
Join Date: Jan 2020
Location: Altadena, CA
Posts: 7,986
Bikes: 2023 Niner RLT 9 RDO, 2018 Trek Procaliber 9.9 RSL, 2018 Storck Fascenario.3 Platinum, 2003 Time VX Special Pro, 2001 Colnago VIP, 1999 Trek 9900 singlespeed, 1977 Nishiki ONP
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 4946 Post(s)
Liked 8,087 Times
in
3,826 Posts
Was the pilots' reason specifically the use of CF composite materials in those planes, and not at all related to design issues, control difficulties, or the failure of other components? How is this related to CF bike frames and components?
__________________
"Swedish fish. They're protein shaped." - livedarklions
"Swedish fish. They're protein shaped." - livedarklions
#69
Senior Member
Join Date: Feb 2018
Posts: 1,410
Bikes: 2017 Specialized Allez Sprint Comp
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 850 Post(s)
Liked 344 Times
in
247 Posts
Every material has a fatigue life for large stresses. Not every material has a fatigue limit.
Steel and titanium are generally believed to have a fatigue limit. AL and CF composites are not.
CF fatigues. Due to its nature, it behaves and fatigues differently than metals - but it does fatigue. IMO anyone claiming otherwise is either (1) incorrect or (2) is deliberately choosing misleading wording that implies that without actually stating it outright (e.g., "Will last a lifetime!"). I suppose I could be wrong, but I don't believe I am.
Steel and titanium are generally believed to have a fatigue limit. AL and CF composites are not.
CF fatigues. Due to its nature, it behaves and fatigues differently than metals - but it does fatigue. IMO anyone claiming otherwise is either (1) incorrect or (2) is deliberately choosing misleading wording that implies that without actually stating it outright (e.g., "Will last a lifetime!"). I suppose I could be wrong, but I don't believe I am.
Likes For smashndash:
#70
Habitual User
Join Date: Jan 2020
Location: Altadena, CA
Posts: 7,986
Bikes: 2023 Niner RLT 9 RDO, 2018 Trek Procaliber 9.9 RSL, 2018 Storck Fascenario.3 Platinum, 2003 Time VX Special Pro, 2001 Colnago VIP, 1999 Trek 9900 singlespeed, 1977 Nishiki ONP
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 4946 Post(s)
Liked 8,087 Times
in
3,826 Posts
This is illustrated in the video in Post #4 of this thread.
__________________
"Swedish fish. They're protein shaped." - livedarklions
"Swedish fish. They're protein shaped." - livedarklions
#71
Duke Ulysses
Join Date: Jul 2013
Location: Southern California
Posts: 800
Bikes: An old orange one for dirt, and for the other stuff: a white one, a kinda mint green one, and a black one.
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked 175 Times
in
86 Posts
yes
*EDIT* actually, yes.
*EDIT* actually, yes.
Last edited by growlerdinky; 09-22-21 at 11:59 AM. Reason: this will blow over. don't worry.
#72
Senior Member
I think the confusion stems from the fact that "carbon fiber" doesn't refer to carbon fiber. It refers to CFRP, which is, materially, VERY different from carbon fiber. As Zinn mentions, there are many ways that CFRP can fatigue but very rarely is it due to the breakage of fibers (caused by fatigue. Obviously impacts are different). I believe you can say that a raw carbon fiber loaded in tension has a very high fatigue limit and a very long fatigue life. But the story changes when you look at CFRP and the properties of the structure will vary VASTLY depending on the layup. So it's impossible to come up with a number for CFRP has a monolith. Some "carbon fiber" frames are just fancy papier-mâché
"Fatigue limit" refers to an amount of stress below which a material apparently can be stressed an infinite number of times and never fail due to material fatigue. "Fatigue life" refers to the number of stress cycles of a given magnitude that can be expected to cause material failure due to material fatigue.
The two concepts refer to completely different things. However, the two terms are unfortunately similar enough that they're easy to confuse.
Any material without a fatigue limit will eventually fail if repeatedly stressed enough times. How long will it take? Depends on the design and the magnitude of the stresses. For a material without a fatigue limit, larger stresses cause failure more quickly than smaller ones - but any stress level will eventually cause failure if repeated often enough. And of course, ANY material can fail if stressed enough.
In contrast, a material with a fatigue limit behaves very differently. Barring material flaws, damage, or fabrication error material with a fatigue limit will in theory never fail from material fatigue if all stresses are kept below the material's fatigue limit. I say "in theory" because testing to confirm something has an infinite lifetime can be a bit problematic. (smile) But that's what test data appears to show for titanium and steel.
CF (and AL) do not appear to have a fatigue limit; steel and titanium do. The former two materials will eventually fail due to material fatigue if stressed enough times, regardless of how large or small the stresses are; the question is when, not if. Absent damage or fabrication flaws, the latter two in theory will not - provided all stresses are kept below their respective fatigue limits.
Last edited by Hondo6; 09-23-21 at 07:16 AM. Reason: Add info and clarity; correct grammatical error and typo.
#73
Senior Member
Join Date: Feb 2018
Posts: 1,410
Bikes: 2017 Specialized Allez Sprint Comp
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 850 Post(s)
Liked 344 Times
in
247 Posts
CF is indeed strong in tension. However, your conceptually mixing apples and stones here.
"Fatigue limit" refers to an amount of stress below which a material apparently can be stressed an infinite number of times and never fail due to material fatigue. "Fatigue life" refers to the number of stress cycles of a given magnitude that can be expected to cause material failure due to material fatigue.
The two concepts refer to completely different things. However, the two terms are unfortunately similar enough that they're easy to confuse.
Any material without a fatigue limit will eventually fail if repeatedly stressed enough times. How long will it take? Depends on the design and the magnitude of the stresses. For a material without a fatigue limit, larger stresses cause failure more quickly than smaller ones - but any stress level will eventually cause failure if repeated often enough. And of course, ANY material can fail if stressed enough.
In contrast, a material with a fatigue limit behaves very differently. Barring material flaws, damage, or fabrication error material with a fatigue limit will in theory never fail from material fatigue if all stresses are kept below the material's fatigue limit. I say "in theory" because testing to confirm something has an infinite lifetime can be a bit problematic. (smile) But that's what test data appears to show for titanium and steel.
CF (and AL) do not appear to have a fatigue limit; steel and titanium do. The former two materials will eventually fail due to material fatigue if stressed enough times, regardless of how large or small the stresses are; the question is when, not if. Absent damage or fabrication flaws, the latter two in theory will not - provided all stresses are kept below their respective fatigue limits.
"Fatigue limit" refers to an amount of stress below which a material apparently can be stressed an infinite number of times and never fail due to material fatigue. "Fatigue life" refers to the number of stress cycles of a given magnitude that can be expected to cause material failure due to material fatigue.
The two concepts refer to completely different things. However, the two terms are unfortunately similar enough that they're easy to confuse.
Any material without a fatigue limit will eventually fail if repeatedly stressed enough times. How long will it take? Depends on the design and the magnitude of the stresses. For a material without a fatigue limit, larger stresses cause failure more quickly than smaller ones - but any stress level will eventually cause failure if repeated often enough. And of course, ANY material can fail if stressed enough.
In contrast, a material with a fatigue limit behaves very differently. Barring material flaws, damage, or fabrication error material with a fatigue limit will in theory never fail from material fatigue if all stresses are kept below the material's fatigue limit. I say "in theory" because testing to confirm something has an infinite lifetime can be a bit problematic. (smile) But that's what test data appears to show for titanium and steel.
CF (and AL) do not appear to have a fatigue limit; steel and titanium do. The former two materials will eventually fail due to material fatigue if stressed enough times, regardless of how large or small the stresses are; the question is when, not if. Absent damage or fabrication flaws, the latter two in theory will not - provided all stresses are kept below their respective fatigue limits.
I tried looking this up, but couldn't find anything. Do you have a source for the fatigue behavior of raw carbon fiber?
#74
Senior Member
Right, I'm personally unsure whether raw CF has a fatigue limit. I was sorta saying that because even if raw CF has a high fatigue limit, it's irrelevant because raw CF is pretty much never used. It's much more useful to look at CFRP, which, as you say, has no fatigue limit.
I tried looking this up, but couldn't find anything. Do you have a source for the fatigue behavior of raw carbon fiber?
I tried looking this up, but couldn't find anything. Do you have a source for the fatigue behavior of raw carbon fiber?
Abstract: https://www.sciencedirect.com/scienc...42112305001015
Diagram: https://ars.els-cdn.com/content/imag...01015-gr14.gif
The full article must be purchased (or access through an institutional subscription), but the abstract and diagram are enough. Besides the blanket statement in the abstract, the diagram doesn't show any evidence of approaching a limit (e.g., the curve never "goes horizontal"). Instead, it shows clear evidence of a continued linear decrease with no sign of limiting behavior as the magnitude of stress cycles decrease and the number of applications increase.
Last edited by Hondo6; 09-23-21 at 03:17 PM. Reason: Add info and delete material at end inadvertently left during editing.
#75
Senior Member
Join Date: Feb 2018
Posts: 1,410
Bikes: 2017 Specialized Allez Sprint Comp
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 850 Post(s)
Liked 344 Times
in
247 Posts
See the Zinn article I linked above from VeloNews. If you want something more technical, here's a diagram from an article from the International Journal of Fatigue titled "Fatigue strength of carbon fibre composites up to the gigacycle regime (gigacycle-composites)". The abstract clearly states that ". . . a fatigue limit (endurance limit) was not found." (emphasis added).
Abstract: https://www.sciencedirect.com/scienc...42112305001015
Diagram: https://ars.els-cdn.com/content/imag...01015-gr14.gif
The full article must be purchased (or access through an institutional subscription), but the abstract and diagram are enough. Besides the blanket statement in the abstract, the diagram doesn't show any evidence of approaching a limit (e.g., the curve never "goes horizontal"). Instead, it shows clear evidence of a continued linear decrease with no sign of limiting behavior as the magnitude of stress cycles decrease and the number of applications increase.
Abstract: https://www.sciencedirect.com/scienc...42112305001015
Diagram: https://ars.els-cdn.com/content/imag...01015-gr14.gif
The full article must be purchased (or access through an institutional subscription), but the abstract and diagram are enough. Besides the blanket statement in the abstract, the diagram doesn't show any evidence of approaching a limit (e.g., the curve never "goes horizontal"). Instead, it shows clear evidence of a continued linear decrease with no sign of limiting behavior as the magnitude of stress cycles decrease and the number of applications increase.
https://drive.google.com/file/d/1_QX...w?usp=drivesdk
I'm gonna be a pirate and share this for a few days for anyone who's interested. I'll read this in the next couple days and come back if I find anything interesting.