Frame durability -- A great study shines light on retro frames
#1
Senior Member
Thread Starter
Frame durability -- A great study shines light on retro frames
Probably a lot of you have run across this article, which was translated from a 1997 German magazine article and posted on Sheldon Brown, but it still fascinates me and I think it's useful for anyone who likes to build up something special on the cheap.
The synopsis: Testers put a dozen high-performance, light racing frames into a stress machine that flexed them until they broke or two days passed. Among the 12 frames, there was a good cross section of materials -- lugged steel (Columbus SLX), welded steel, aluminum, carbon, titanium and carbon/alloy. All but three frames broke. Of the three that didn't break, two were aluminum and one was carbon: Trek OCLV, Cannondale CAAD3 and Principia RSL.
There's some lessons about materials -- mainly that the stereotypes about aluminum alloy and carbon lacking durability don't have to be true.
But I think the bigger point is about craftsmanship -- Trek and Cannondale were making great frames here in the U.S. and they are still pretty easily available for ridiculously low prices. I don't know anything about Principia except for what I've just read by googling.
Trek came to campus at the University of Wisconsin and participating in engineering days and showed (this was late 90s) their carbon manufacturing and testing. At that time, and in diminishing capacity all the way to 2017, Trek engineers and manufacturers worked at the same facility surrounded by corn fields 20-odd miles east of Madison. Now, sadly, none of their frames are built in the U.S.
The "common wisdom" I hear about carbon or aluminum frames from the late 90s or early 2000s is that they're suspect; yet people will snap up vintage steel frames and pay $1,000 or more for a Merlin Extralight frame.
Meanwhile, lighter and more durable Trek 5900s are going for a few hundred bucks.
My point is, I always look first for Cannondales and Treks from that era if I'm looking to build the highest performing, most durable bike for the least amount of money.
Trek wasn't "figuring out" carbon fiber in the late 90s or early 2000s. They were extremely adept, using the highest quality stuff they could buy, and supporting domestic workers who painstakingly laid it up.
I've seen plenty of Trek carbon crack due to fatigue -- recent mountain bikes most often. But never anything from the 97-2006 era, and these are the frames that practically get given away.
The synopsis: Testers put a dozen high-performance, light racing frames into a stress machine that flexed them until they broke or two days passed. Among the 12 frames, there was a good cross section of materials -- lugged steel (Columbus SLX), welded steel, aluminum, carbon, titanium and carbon/alloy. All but three frames broke. Of the three that didn't break, two were aluminum and one was carbon: Trek OCLV, Cannondale CAAD3 and Principia RSL.
There's some lessons about materials -- mainly that the stereotypes about aluminum alloy and carbon lacking durability don't have to be true.
But I think the bigger point is about craftsmanship -- Trek and Cannondale were making great frames here in the U.S. and they are still pretty easily available for ridiculously low prices. I don't know anything about Principia except for what I've just read by googling.
Trek came to campus at the University of Wisconsin and participating in engineering days and showed (this was late 90s) their carbon manufacturing and testing. At that time, and in diminishing capacity all the way to 2017, Trek engineers and manufacturers worked at the same facility surrounded by corn fields 20-odd miles east of Madison. Now, sadly, none of their frames are built in the U.S.
The "common wisdom" I hear about carbon or aluminum frames from the late 90s or early 2000s is that they're suspect; yet people will snap up vintage steel frames and pay $1,000 or more for a Merlin Extralight frame.
Meanwhile, lighter and more durable Trek 5900s are going for a few hundred bucks.
My point is, I always look first for Cannondales and Treks from that era if I'm looking to build the highest performing, most durable bike for the least amount of money.
Trek wasn't "figuring out" carbon fiber in the late 90s or early 2000s. They were extremely adept, using the highest quality stuff they could buy, and supporting domestic workers who painstakingly laid it up.
I've seen plenty of Trek carbon crack due to fatigue -- recent mountain bikes most often. But never anything from the 97-2006 era, and these are the frames that practically get given away.
#2
Senior Member
Join Date: Oct 2015
Posts: 15,929
Bikes: 2015 Workswell 066, 2017 Workswell 093, 2014 Dawes Sheila, 1983 Cannondale 500, 1984 Raleigh Olympian, 2007 Cannondale Rize 4, 2017 Fuji Sportif 1 LE
Liked 3,934 Times
in
2,053 Posts
Not to be negative but (go with your strengths, they tell me ... ...) a 1997 study is pretty irrelevant today.
Trek OCLV are widely acknowledged to be tanks .... all the weight of steel but made from carbon! As far as where the frames were made .... pretty sure pretty much no frames don't come from across the ocean anymore ... and those frames might be stronger.
Also, I am not sure how to best test a frame ... to actually simulate the loads and stresses of riding.
Your point abut people's prejudices about frame materials is well taken ... but as with most prejudices, facts here don't matter. They Know they are wrong, but they prefer their prejudices .... There is so much evidence that CF is a good material, used properly, and perfectly safe for a bike, but still people, after three decades, claim only steel is real ...
The Flat Earth Society has members all around the globe .... so why bother?
Trek OCLV are widely acknowledged to be tanks .... all the weight of steel but made from carbon! As far as where the frames were made .... pretty sure pretty much no frames don't come from across the ocean anymore ... and those frames might be stronger.
Also, I am not sure how to best test a frame ... to actually simulate the loads and stresses of riding.
Your point abut people's prejudices about frame materials is well taken ... but as with most prejudices, facts here don't matter. They Know they are wrong, but they prefer their prejudices .... There is so much evidence that CF is a good material, used properly, and perfectly safe for a bike, but still people, after three decades, claim only steel is real ...
The Flat Earth Society has members all around the globe .... so why bother?
Likes For Maelochs:
#3
climber has-been
Join Date: Dec 2004
Location: Palo Alto, CA
Posts: 7,572
Bikes: Scott Addict R1, Felt Z1
Liked 4,136 Times
in
2,038 Posts
There's no way to be sure why some random carbon frame failed. And given that carbon fiber has essentially infinite fatigue life, the reason is probably not fatigue.
#4
Senior Member
Thread Starter
Not to be negative but (go with your strengths, they tell me ... ...) a 1997 study is pretty irrelevant today.
Trek OCLV are widely acknowledged to be tanks .... all the weight of steel but made from carbon! As far as where the frames were made .... pretty sure pretty much no frames don't come from across the ocean anymore ... and those frames might be stronger.
Also, I am not sure how to best test a frame ... to actually simulate the loads and stresses of riding.
Trek OCLV are widely acknowledged to be tanks .... all the weight of steel but made from carbon! As far as where the frames were made .... pretty sure pretty much no frames don't come from across the ocean anymore ... and those frames might be stronger.
Also, I am not sure how to best test a frame ... to actually simulate the loads and stresses of riding.
And the fact is that a 58 cm OCLV weighed 1,200 grams vs the 2080 grams of the 58 cm Columbus SLX frame.
The article explains how the engineers did the stress testing to simulate the forces of riding.
#5
Senior Member
Join Date: Jul 2009
Posts: 6,514
Bikes: Colnago, Van Dessel, Factor, Cervelo, Ritchey
Liked 8,364 Times
in
3,317 Posts
Likes For tomato coupe:
#6
Senior Member
Join Date: Oct 2015
Posts: 15,929
Bikes: 2015 Workswell 066, 2017 Workswell 093, 2014 Dawes Sheila, 1983 Cannondale 500, 1984 Raleigh Olympian, 2007 Cannondale Rize 4, 2017 Fuji Sportif 1 LE
Liked 3,934 Times
in
2,053 Posts
#7
Senior Member
Not to be negative but (go with your strengths, they tell me ... ...) a 1997 study is pretty irrelevant today.
Trek OCLV are widely acknowledged to be tanks .... all the weight of steel but made from carbon! As far as where the frames were made .... pretty sure pretty much no frames don't come from across the ocean anymore ... and those frames might be stronger.
Also, I am not sure how to best test a frame ... to actually simulate the loads and stresses of riding.
Your point abut people's prejudices about frame materials is well taken ... but as with most prejudices, facts here don't matter. They Know they are wrong, but they prefer their prejudices .... There is so much evidence that CF is a good material, used properly, and perfectly safe for a bike, but still people, after three decades, claim only steel is real ...
The Flat Earth Society has members all around the globe .... so why bother?
Trek OCLV are widely acknowledged to be tanks .... all the weight of steel but made from carbon! As far as where the frames were made .... pretty sure pretty much no frames don't come from across the ocean anymore ... and those frames might be stronger.
Also, I am not sure how to best test a frame ... to actually simulate the loads and stresses of riding.
Your point abut people's prejudices about frame materials is well taken ... but as with most prejudices, facts here don't matter. They Know they are wrong, but they prefer their prejudices .... There is so much evidence that CF is a good material, used properly, and perfectly safe for a bike, but still people, after three decades, claim only steel is real ...
The Flat Earth Society has members all around the globe .... so why bother?
Likes For GhostRider62:
#8
Senior Member
Join Date: Oct 2015
Posts: 15,929
Bikes: 2015 Workswell 066, 2017 Workswell 093, 2014 Dawes Sheila, 1983 Cannondale 500, 1984 Raleigh Olympian, 2007 Cannondale Rize 4, 2017 Fuji Sportif 1 LE
Liked 3,934 Times
in
2,053 Posts
#10
Klaatu..Verata..Necktie?
Join Date: May 2007
Location: SF Bay Area
Posts: 19,454
Bikes: Litespeed Ultimate, Ultegra; Canyon Endurace, 105; Battaglin MAX, Chorus; Bianchi 928 Veloce; Ritchey Road Logic, Dura Ace; Cannondale R500 RX100; Schwinn Circuit, Sante; Lotus Supreme, Dura Ace
Liked 13,274 Times
in
6,826 Posts
Likes For genejockey:
#11
Sunshine
Join Date: Aug 2014
Location: Des Moines, IA
Posts: 17,093
Bikes: '18 class built steel roadbike, '19 Fairlight Secan, '88 Schwinn Premis , Black Mountain Cycles Monstercross V4, '89 Novara Trionfo
Liked 8,133 Times
in
4,503 Posts
The study is relevant because it showed that a 1997 Trek OCLV withstood more stress cycles than a 1997 lugged Columbus SLX frame, among several others. So if you're shopping for a 1997 frame, this sheds a lot of light on some potential choices.
And the fact is that a 58 cm OCLV weighed 1,200 grams vs the 2080 grams of the 58 cm Columbus SLX frame.
The article explains how the engineers did the stress testing to simulate the forces of riding.
And the fact is that a 58 cm OCLV weighed 1,200 grams vs the 2080 grams of the 58 cm Columbus SLX frame.
The article explains how the engineers did the stress testing to simulate the forces of riding.
Likes For mstateglfr:
#12
Senior Member
Join Date: Jan 2023
Location: Eastern Shore MD
Posts: 1,233
Bikes: Lemond Zurich/Trek ALR/Giant TCX/Sette CX1
Liked 1,161 Times
in
572 Posts
Its funny, when I bought my CAD3 in 1997+/- - the "bike people" gave me crap for buying a Crackendale. I guess some of the previous versions had issues... To date, that was the stiffest and fastest frame I've owned, and I regret passing it along to my nephew... who never rode it and gave it away.
I think people who are weary of carbon fiber, like myself, are sometimes misunderstood. It's not that I think they are more prone to assploding from normal use than any other bike - it's that I think they are not durable in terms of minor impacts, and when they do fail - it's often catastrophic.
Over torquing (you hope the bike shop kid knows what he is doing/uses a torqe wrench), hidden cracks from minor impacts that won't show until they do - often after the crash.
I think people who are weary of carbon fiber, like myself, are sometimes misunderstood. It's not that I think they are more prone to assploding from normal use than any other bike - it's that I think they are not durable in terms of minor impacts, and when they do fail - it's often catastrophic.
Over torquing (you hope the bike shop kid knows what he is doing/uses a torqe wrench), hidden cracks from minor impacts that won't show until they do - often after the crash.
#13
Senior Member
Join Date: May 2022
Location: USA - Southwest PA
Posts: 3,969
Bikes: Cannondale - Gary Fisher - Giant - Litespeed - Schwinn Paramount - Schwinn (lugged steel) - Trek OCLV
Liked 2,821 Times
in
1,496 Posts
Likes For t2p:
#14
I don't think carbon fails catastrophically. It's made up of many individual strands, If some strands break, it will flex, but other strands will still be holding the part together. It's unlikely that all the strands would break at the same time.
I have a CF handlebar that's been visibly crushed in the clamp area. And yet you cannot flex that bar in any way, shape or form. It is a myth that they just assplode out of the blue.
I have a CF handlebar that's been visibly crushed in the clamp area. And yet you cannot flex that bar in any way, shape or form. It is a myth that they just assplode out of the blue.
#15
Senior Member
Join Date: Oct 2015
Posts: 15,929
Bikes: 2015 Workswell 066, 2017 Workswell 093, 2014 Dawes Sheila, 1983 Cannondale 500, 1984 Raleigh Olympian, 2007 Cannondale Rize 4, 2017 Fuji Sportif 1 LE
Liked 3,934 Times
in
2,053 Posts
Yup .... there is a laundry list of stuff people can do wrong .... but oddly so very few people every seem to do those things or have them happen, when so many people ride CF bikes.
But ... as I said above, prejudices are immune to fact. And I am fine wit other people's prejudices .... but the phrase "weary of carbon fiber' sort of mystifies me. I don't want a unicycle, nor plan to ever buy one, and the reason is because they are hard to ride and easy to wreck ... actual facts ... yet I am not "weary" of them ... more "wary" of them.
(lol ... I have actually considered unicycles, but as I age and crash damage takes longer to heal ... )
But ... as I said above, prejudices are immune to fact. And I am fine wit other people's prejudices .... but the phrase "weary of carbon fiber' sort of mystifies me. I don't want a unicycle, nor plan to ever buy one, and the reason is because they are hard to ride and easy to wreck ... actual facts ... yet I am not "weary" of them ... more "wary" of them.
(lol ... I have actually considered unicycles, but as I age and crash damage takes longer to heal ... )
#16
Senior Member
Join Date: Jan 2023
Location: Eastern Shore MD
Posts: 1,233
Bikes: Lemond Zurich/Trek ALR/Giant TCX/Sette CX1
Liked 1,161 Times
in
572 Posts
I don't think carbon fails catastrophically. It's made up of many individual strands, If some strands break, it will flex, but other strands will still be holding the part together. It's unlikely that all the strands would break at the same time.
I have a CF handlebar that's been visibly crushed in the clamp area. And yet you cannot flex that bar in any way, shape or form. It is a myth that they just assplode out of the blue.
I have a CF handlebar that's been visibly crushed in the clamp area. And yet you cannot flex that bar in any way, shape or form. It is a myth that they just assplode out of the blue.
I've was in the hospital with 4 broken bones, concussion and a knee injury from a set of CF handlebars that assploded out of the blue. The cause - upon removal they looked to be crushed a little bit at the clamp area.
#17
Senior Member
Thread Starter
I don't think carbon fails catastrophically. It's made up of many individual strands, If some strands break, it will flex, but other strands will still be holding the part together. It's unlikely that all the strands would break at the same time.
I have a CF handlebar that's been visibly crushed in the clamp area. And yet you cannot flex that bar in any way, shape or form. It is a myth that they just assplode out of the blue.
I have a CF handlebar that's been visibly crushed in the clamp area. And yet you cannot flex that bar in any way, shape or form. It is a myth that they just assplode out of the blue.
Of course, this depends on a good layup and design.
#18
Randomhead
Join Date: Aug 2008
Location: Happy Valley, Pennsylvania
Posts: 24,909
Liked 4,067 Times
in
2,769 Posts
Steel bikes are definitely more marginal in a lab. In real life they aren't nearly as bad.
Just from listening to the people at the LBS, Trek had a number of bikes with bonded dropouts that fell out due to a glue failure. I ain't buying an old Trek carbon bike.
Just from listening to the people at the LBS, Trek had a number of bikes with bonded dropouts that fell out due to a glue failure. I ain't buying an old Trek carbon bike.
#19
Senior Member
Thread Starter
It's not that challenging to find one a few years newer, after the switch to 1 1/8" headsets, with full 7700 Dura Ace for significantly less than a thousand dollars.
What's a better deal than that?
Likes For ljsense:
#20
Senior Member
A couple of the guys I worked with were what I came to think of as steel apologists. To them, any aluminum or carbon frame failure happened because of the material---and any steel or titanium frame failure happened despite the material.
Likes For Trakhak:
#21
Senior Member
Join Date: May 2022
Location: USA - Southwest PA
Posts: 3,969
Bikes: Cannondale - Gary Fisher - Giant - Litespeed - Schwinn Paramount - Schwinn (lugged steel) - Trek OCLV
Liked 2,821 Times
in
1,496 Posts
trek almost immediately began to include / ship / install (?) a large steel ring / ‘washer’ to be installed on the drive side bottom bracket cup on OCLV frames so a thrown chain would not saw through the area around the drive side chain stay
also heard some internal bottom bracket sleeves / shells worked loose on early frames
also heard some internal bottom bracket sleeves / shells worked loose on early frames
Last edited by t2p; 05-31-23 at 12:01 PM.
#22
Senior Member
#23
El Rayo X
Re: Malox in post #2
The Flat Earth Society has members all around the globe .... so why bother?[/QUOTE]
You're a funny guy! I can't be the only one who gets it.
The Flat Earth Society has members all around the globe .... so why bother?[/QUOTE]
You're a funny guy! I can't be the only one who gets it.
Last edited by Litho dbh; 05-31-23 at 12:34 PM.
Likes For Litho dbh:
#24
Senior Member
Why buy an old bike in the first place? A 5-year-old mid-range premium brand carbon 105 bike such as will outperform a so-called old bike any day of the week. A 2018 Specialized Roubaix 105 bike blue book is around $1,000, how can you go wrong with that?
Likes For Atlas Shrugged:
#25
aged to perfection
Join Date: Jun 2009
Location: PacNW
Posts: 1,928
Bikes: Dinucci Allez 2.0, Richard Sachs, Alex Singer, Serotta, Masi GC, Raleigh Pro Mk.1, Hetchins, etc
Liked 1,349 Times
in
717 Posts
A bike frame - steel ones at least - are fairly complex structures and it's really hard to simulate the loads imposed by a rider on real roads.
However for a properly built frame, where the front forks and rear dropouts are firmly held and deflection loads imposed at the bottom bracket, the structure should still behave according to Hooke's law which is shown below. "Stress" is the load applied to the frame, and "strain" is the deflection measured in response to that load.
Looking at the stress / strain curve below, if the loads imposed are in the range of A and B, the frame is being loaded well within the elastic range, and will spring back with no permanent deformation. Fatigue effects for loads in this range will not occur for millions - 10s of millions - of load cycles.
HOWEVER once you go beyond "C" on the stress / stain curve, then the frame may be permanently distorted (not return to original unloaded position) and fatigue cycles build up very quickly.
the trick in load testing is to go as close to "C" as possible, still staying in the elastic region (where the frame springs back) and then applying hundreds of thousands of load cycles at that level until a crack appears.
However for a properly built frame, where the front forks and rear dropouts are firmly held and deflection loads imposed at the bottom bracket, the structure should still behave according to Hooke's law which is shown below. "Stress" is the load applied to the frame, and "strain" is the deflection measured in response to that load.
Looking at the stress / strain curve below, if the loads imposed are in the range of A and B, the frame is being loaded well within the elastic range, and will spring back with no permanent deformation. Fatigue effects for loads in this range will not occur for millions - 10s of millions - of load cycles.
HOWEVER once you go beyond "C" on the stress / stain curve, then the frame may be permanently distorted (not return to original unloaded position) and fatigue cycles build up very quickly.
the trick in load testing is to go as close to "C" as possible, still staying in the elastic region (where the frame springs back) and then applying hundreds of thousands of load cycles at that level until a crack appears.
Likes For mpetry912: