Problem: American Classic CR-420's

[buckybux]

I purchased the American Classic CR-420's with bladed spokes on Ebay in May 2005, cost about $500 (retail is about $900). In September I busted my first spoke, took it to the LBS where they informed me the spokes were not DT's as stated in Ebay. The seller agreed to pay for new DT spokes (about $100). The LBS re-built the wheels using DT Aero Speed Spokes. Now about 3000 miles later, I am busting spokes again. Three in the last 2.5 months, two at the nipple and one at the bend, all on the drive side of the rear wheel. In doing some background research, I learned that that Aero Speed spokes are thinner and lighter weight, then the original equipment New Aero spokes. Rear hub has the serial number xxxx'd out, which tells me it is possibly a factory second. To rebuild the wheels with the New Aero will cost me about $60.

I have talked with American Classic, and my LBS, bottom line is that no one knows for sure what is the problem, it could be the spokes, hub or rim.

My question is should I try the $60 rebuild, or toss the wheels and start over with a new set. My front wheels have the Aero Speed, but none of the spokes have broken on that wheel. Should I be concerned about breaking a front spoke (which would be a disaster if it broke while descending at 40mph)?

 

[JTE83]

How heavy are you? Maybe your too heavy for these wheels?

 

[buckybux]

How heavy are you? Maybe your too heavy for these wheels?

I am 158 lbs. I am within range. I am also not unusually hard on the wheels either.

 

[hd reynolds]

Spokes usually break at the bend where the stress is the greatest. But in your case you have encountered more breakage on the nipple side. Spokes that break not from the bend means you are exceeding their maximum tensile strength which point to too much tension. I would suggest for your wheel builder to stay within the recommended tension limits for the spokes - your wheelbuilder can check this by using a spoke tensiometer.

 

[buckybux]

Spokes usually break at the bend where the stress is the greatest. But in your case you have encountered more breakage on the nipple side. Spokes that break not from the bend means you are exceeding their maximum tensile strength which point to too much tension. I would suggest for your wheel builder to stay within the recommended tension limits for the spokes - your wheelbuilder can check this by using a spoke tensiometer.

Spoke tension has been checked three times. At the time of rebuild, about a month later, and then again in May (when the first one broke). Tension has been within American Classic Standards, which is fairly broad, which is 90-120 kgf.

 

[ScienceIsCool]

Actually, spokes don't break where the stress is highest (since the stress is same all through the spoke), but at areas where the spoke is full of residual stress from manufacturing. I.e., at the threads and bends. Additional stress cycles (i.e., riding) can then push these areas to more rapid fatigue and hence, quick failure.

Build quality can affect this too. During the building process, more experienced builders will grab the spokes and stress the hell out of them. What this does is yield the spokes at the bend (i.e., bend them a bit more) and remove those residual stresses. This allows a more natural fatigue life and fewer failures. I don't know of a way to relieve stress from the threads other than annealing... Typically, spokes will fail there but not after many, many miles.

The rim will often fail way before you reach the tensile strength of your spoke, so no worries there. My recommendation would be to rebuild the wheel with new spokes after checking the hub flange for damage and the rim for straightness and cracks at the holes.

Spokes usually break at the bend where the stress is the greatest. But in your case you have encountered more breakage on the nipple side. Spokes that break not from the bend means you are exceeding their maximum tensile strength which point to too much tension. I would suggest for your wheel builder to stay within the recommended tension limits for the spokes - your wheelbuilder can check this by using a spoke tensiometer.

 

[buckybux]

Actually, spokes don't break where the stress is highest (since the stress is same all through the spoke), but at areas where the spoke is full of residual stress from manufacturing. I.e., at the threads and bends. Additional stress cycles (i.e., riding) can then push these areas to more rapid fatigue and hence, quick failure.

Build quality can affect this too. During the building process, more experienced builders will grab the spokes and stress the hell out of them. What this does is yield the spokes at the bend (i.e., bend them a bit more) and remove those residual stresses. This allows a more natural fatigue life and fewer failures. I don't know of a way to relieve stress from the threads other than annealing... Typically, spokes will fail there but not after many, many miles.

The rim will often fail way before you reach the tensile strength of your spoke, so no worries there. My recommendation would be to rebuild the wheel with new spokes after checking the hub flange for damage and the rim for straightness and cracks at the holes.

Thanks for the input. The wheel was true at the time of the spoke break, there are no cracks on the rim, and the hub looks good. So it looks like I will try re-building with stronger spokes. The cost is only $60 and my labor, while the cost for a new wheel is about $400.

 

[ScienceIsCool]

I would still check the rim for straightness once you've removed the spokes. It's possible to have a true wheel with a rim that isn't straight, but the tension will be a bit wonky. This will also lead to rapid fatigue failures and frequent truing.

John Swanson

Thanks for the input. The wheel was true at the time of the spoke break, there are no cracks on the rim, and the hub looks good. So it looks like I will try re-building with stronger spokes. The cost is only $60 and my labor, while the cost for a new wheel is about $400.

 

[buckybux]

I would still check the rim for straightness once you've removed the spokes. It's possible to have a true wheel with a rim that isn't straight, but the tension will be a bit wonky. This will also lead to rapid fatigue failures and frequent truing.

John Swanson

I will check it again for straigtness. Again, I appreciate your help. At least I feel a little more comfortable about rebuilding them.

 

[alfeng]

My question is should I try the $60 rebuild, or toss the wheels and start over with a new set.

Presuming the rim & hub are sound, have you considered the less exotic option of having the wheels laced with standard, 14g spokes ... or, at the most, "traditional" double butted spokes? While it lacks the appeal of the exotic spokes, potential reliability does count as a consideration ...

BTW. Did the nipple "strip" or did the spoke actually break?

The spoke failure which you describe suggests to me (certainly, others may disagree) that the wheel has too much lateral flex and that you should consider rebuilding the wheel with straight gauge spokes on both the drive & non-drive sides.

 

[ScienceIsCool]

Why would that suggest lateral flex? Failure due to fatigue is caused by repeated stress cycles, which cause stress risers at micro inclusions or flaws in the base material. These flaws grow until a crack propagates and the material fails.

This effect can be exaggerated if the stress cycle passes near or through a zero tension point, or near the yield point. Too little tension is bad. Too much tension is bad, though it will usually cause the rim and/or hub flange to crack or rupture first.

The shape and gauge of the spoke has little to do with fatigue failures, since it's the bend of the spoke and the threads that have high residual stresses which lead to failure. I.e., they are pre-stressed to a point that's too close to yield for good fatigue life. I've never seen or heard of a spoke that has failed due to fatigue near the middle of the spoke or at the swage where it goes from low gauge to high gauge.

Oh yeah. The spoke nipple (aluminum) and rim will fail long before the threads of the spoke begin to strip.
John Swanson

Presuming the rim & hub are sound, have you considered the less exotic option of having the wheels laced with standard, 14g spokes ... or, at the most, "traditional" double butted spokes? While it lacks the appeal of the exotic spokes, potential reliability does count as a consideration ...

BTW. Did the nipple "strip" or did the spoke actually break?

The spoke failure which you describe suggests to me (certainly, others may disagree) that the wheel has too much lateral flex and that you should consider rebuilding the wheel with straight gauge spokes on both the drive & non-drive sides.

 

[alfeng]

Why would that suggest lateral flex? Failure due to fatigue is caused by repeated stress cycles, which cause stress risers at micro inclusions or flaws in the base material. These flaws grow until a crack propagates and the material fails ...

So, YOU are suggesting the flaw is in the spoke, itself? The build? The maintenance?

Okay, considering 3000 miles were logged before the failure, I reckon there was rather lax maintenance ...

OR, are you suggesting that a double-butted or bladed spoke somehow has the same lateral flex as a straight gauge spoke AND/OR that a wheel built with double-butted or bladed spokes will be as laterally stiff?

Are YOU suggesting that repeated lateral flex doesn't influence the stress on the spoke?

Well, maybe repeated flexing doesn't have any effect ... heck, you bend a coat hanger enough times and it will ... wait, it will break!

Presumng that the reference to breaks at the nipples were stripped nipples, then what do YOU (make that a generic 'you') suppose caused the micro-piston motion (for want of a better description) which resulting in the stripping?

Well, not only do I think the particular "problem" wheel should be rebuilt with straight 14 gauge spokes, but probably with a x3 lacing on the drive side instead of x2 (which I believe is how it is laced).

Clearly, periodically checking for uniform spoke tension would assist in ensuring the spoke breakage is kept to a minimum ...

But, as I said, "others may disagree" ... it was just my opinion as to what the problem is and a possible way to avoid it in the future. Certainly, other approaches may be better ...

 

[ScienceIsCool]

Awesome discussion. I'll try to go point by point below:

So, YOU are suggesting the flaw is in the spoke, itself? The build? The maintenance?

Okay, considering 3000 miles were logged before the failure, I reckon there was rather lax maintenance ...

<I don't think maintenance is an issue. What would you maintain? 5000 km is also not a ridiculously low number for fatigue failures depending on tension and load. Bad, but not ridiculous.>

OR, are you suggesting that a double-butted or bladed spoke somehow has the same lateral flex as a straight gauge spoke AND/OR that a wheel built with double-butted or bladed spokes will be as laterally stiff?

<Yes, that is what I'm saying. The lateral flex is almost exclusively a function of spoke tension. The spoke does not achieve any kind of bending moment (i.e., spoke would have to be completely slack before it did this) therefore the material stiffness does not come into play.>

Are YOU suggesting that repeated lateral flex doesn't influence the stress on the spoke?

<You're right. Lateral loading, like radial loading will induce cyclical stress which contributes to fatigue life.>

Well, maybe repeated flexing doesn't have any effect ... heck, you bend a coat hanger enough times and it will ... wait, it will break!

<Of course, your demonstration involves bending the wire hanger past its yield point. Not good for fatigue... That clearly doesn't happen in a wheel.>

Presumng that the reference to breaks at the nipples were stripped nipples, then what do YOU (make that a generic 'you') suppose caused the micro-piston motion (for want of a better description) which resulting in the stripping?

<I figure the threads didn't strip. I figure the spoke cracked at the threads due to poor fatigue characteristics. Those characteristics are a function of how the cold rolled threads were made. Induced stresses in the metal and all that.>

Well, not only do I think the particular "problem" wheel should be rebuilt with straight 14 gauge spokes, but probably with a x3 lacing on the drive side instead of x2 (which I believe is how it is laced).

<Why?>

Clearly, periodically checking for uniform spoke tension would assist in ensuring the spoke breakage is kept to a minimum ...

But, as I said, "others may disagree" ... it was just my opinion as to what the problem is and a possible way to avoid it in the future. Certainly, other approaches may be better ...

 

[alfeng]

I don't think maintenance is an issue. What would you maintain? 5000 km is also not a ridiculously low number for fatigue failures depending on tension and load. Bad, but not ridiculous.

Agreed, it isn't a lot of miles ...

The spoke tension on the "problem" wheel could/should have been checked for uniformity ...

OR, are you suggesting that a double-butted or bladed spoke somehow has the same lateral flex as a straight gauge spoke AND/OR that a wheel built with double-butted or bladed spokes will be as laterally stiff?

Yes, that is what I'm saying. The lateral flex is almost exclusively a function of spoke tension. The spoke does not achieve any kind of bending moment (i.e., spoke would have to be completely slack before it did this) therefore the material stiffness does not come into play.

Well, don't most wheelbuilders tension their spokes by kgf? Doesn't that mean that a wheel laced with double-butted spokes will be laterally less stiff (i.e., be more easily "deflected") than one with thicker spokes?

Those super-duper spokes (2.0/1.5mm) have to be tensioned to ~175kgf to have approximately the same deflection as a 14g spoke tensioned at ~100kgf ... a "regular" double butted spoke would need to be tensioned to ~131kgf to achieive the same deflection level.

OR, to put it another way, a super-duper spoke would only need significantly less force to deflect it if it were tensioned to 100kgf ...

Of course, your demonstration involves bending the wire hanger past its yield point. Not good for fatigue... That clearly doesn't happen in a wheel.

Except at the J-bend which is essentially at a fulcrum point which will, in the course of 3000 miles, have gone through approximately 190,000 stress cycles based on a 24 spoke wheel ...

Presumng that the reference to breaks at the nipples were stripped nipples, then what do YOU (make that a generic 'you') suppose caused the micro-piston motion (for want of a better description) which resulting in the stripping?

I figure the threads didn't strip. I figure the spoke cracked at the threads due to poor fatigue characteristics. Those characteristics are a function of how the cold rolled threads were made. Induced stresses in the metal and all that.

Well, we don't know which it is ... I guess I've never seen a spoke-itself actually break at the nipple; but, I have seen stripped nipples. I guess I saw ONE spoke break due to unrelieved wind-up stress(es) ... but, that wasn't at the nipple.

Why?

Just my opinion ...

There are certainly wheelbuilders with MORE experience than I have ... for what it's worth, the only wheels I have had problems with were laced & tensioned by someone else.

Again, if it were MY wheel, the way I would "salvage" it would be to lace it with 14g spokes, x3 on the drive side & x2 on the non-drive side -- it's just MY opinion ... no one has to agree with it -- as I said, another spec on the rebuild may be better ... I just wouldn't do any "experimenting" at this point if I wanted to get more mileage out of the rim & hub.

 

[ScienceIsCool]

Sounds like we agree on most things. Just a few points though. Stress is typically measured in Pa, pressure, which is force per cross-sectional area. That's why it takes more force to achieve the same stress for a thick wire. (the kgf variations you mentioned below) That said, the stress you can put on a spoke is typically far away from yield regardless of spoke thickness. The rim just can't take that much force and will rupture. If it doesn't then the hub will.

The thick and thin spokes are well away from their yield point in a wheel. Ergo, a thin spoke and a thick spoke have roughly the same fatigue life at mid-span. That's why you don't see fatigue failure in the middle of a spoke. At least I've never heard of one.
My last point is that fatigue is a probability function based on the number and depth of stress cycles. 200,000 cycles is near the low end of expected failure, but well within a normal range. Coincidentally, this is why I won't ride aluminum spoked wheels like Mavic. 250,000 cycles is near the 50% likely point that a spoke will fail. That's why you hear of so many cyclists grumbling about their Ksyrium's after 6000 km. BTW, I have a paper on my website www.bikephysics.com which contains a link or two that might interest you. One of them discusses modelling of spoke fatigue failure. Big-time math but a good read. Some of the other links are regarding lateral stiffness. Also interesting. Thanks for the thought-provoking discussion!

John Swanson

Agreed, it isn't a lot of miles ...

The spoke tension on the "problem" wheel could/should have been checked for uniformity ...

OR, are you suggesting that a double-butted or bladed spoke somehow has the same lateral flex as a straight gauge spoke AND/OR that a wheel built with double-butted or bladed spokes will be as laterally stiff?

Well, don't most wheelbuilders tension their spokes by kgf? Doesn't that mean that a wheel laced with double-butted spokes will be laterally less stiff (i.e., be more easily "deflected") than one with thicker spokes?

Those super-duper spokes (2.0/1.5mm) have to be tensioned to ~175kgf to have approximately the same deflection as a 14g spoke tensioned at ~100kgf ... a "regular" double butted spoke would need to be tensioned to ~131kgf to achieive the same deflection level.

OR, to put it another way, a super-duper spoke would only need significantly less force to deflect it if it were tensioned to 100kgf ...

Except at the J-bend which is essentially at a fulcrum point which will, in the course of 3000 miles, have gone through approximately 190,000 stress cycles based on a 24 spoke wheel ...

Presumng that the reference to breaks at the nipples were stripped nipples, then what do YOU (make that a generic 'you') suppose caused the micro-piston motion (for want of a better description) which resulting in the stripping?

Well, we don't know which it is ... I guess I've never seen a spoke-itself actually break at the nipple; but, I have seen stripped nipples. I guess I saw ONE spoke break due to unrelieved wind-up stress(es) ... but, that wasn't at the nipple.

Just my opinion ...

There are certainly wheelbuilders with MORE experience than I have ... for what it's worth, the only wheels I have had problems with were laced & tensioned by someone else.

Again, if it were MY wheel, the way I would "salvage" it would be to lace it with 14g spokes, x3 on the drive side & x2 on the non-drive side -- it's just MY opinion ... no one has to agree with it -- as I said, another spec on the rebuild may be better ... I just wouldn't do any "experimenting" at this point if I wanted to get more mileage out of the rim & hub.

 

[buckybux]

Just for point of information:

1. What maintainance is needed? I keep them clean, and check that they are true weekly. I also randomly pluck the spokes for tension.

2. The spokes that broke at the nipple did exactly that, they did not strip. The break was at the top thread.
I have considered going with traditional 14 gauge or butted spokes, but I would prefer bladed. I think it is worth it to try the heavier Bladed New Aero spokes. Bottom line, is that no one is quite sure what is wrong with the wheels.

 

[ScienceIsCool]

Oops. I also wanted to say that I totally agree with you that thinner spoked wheels will flex more laterally. You are right. I'll have to reconsider the concept of bending moment when under tension. I just don't think it significantly affects the total fatigue life of the spoke. Again, the spoke will break where residual stresses are making a mess of things. This can also be mitigated during the build process.

Stripped nipples are also no good. Aluminum just isn't up to the job. Brass is standard for a reason...

John Swanson

Agreed, it isn't a lot of miles ...

The spoke tension on the "problem" wheel could/should have been checked for uniformity ...

OR, are you suggesting that a double-butted or bladed spoke somehow has the same lateral flex as a straight gauge spoke AND/OR that a wheel built with double-butted or bladed spokes will be as laterally stiff?

Well, don't most wheelbuilders tension their spokes by kgf? Doesn't that mean that a wheel laced with double-butted spokes will be laterally less stiff (i.e., be more easily "deflected") than one with thicker spokes?

Those super-duper spokes (2.0/1.5mm) have to be tensioned to ~175kgf to have approximately the same deflection as a 14g spoke tensioned at ~100kgf ... a "regular" double butted spoke would need to be tensioned to ~131kgf to achieive the same deflection level.

OR, to put it another way, a super-duper spoke would only need significantly less force to deflect it if it were tensioned to 100kgf ...

Except at the J-bend which is essentially at a fulcrum point which will, in the course of 3000 miles, have gone through approximately 190,000 stress cycles based on a 24 spoke wheel ...

Presumng that the reference to breaks at the nipples were stripped nipples, then what do YOU (make that a generic 'you') suppose caused the micro-piston motion (for want of a better description) which resulting in the stripping?

Well, we don't know which it is ... I guess I've never seen a spoke-itself actually break at the nipple; but, I have seen stripped nipples. I guess I saw ONE spoke break due to unrelieved wind-up stress(es) ... but, that wasn't at the nipple.

Just my opinion ...

There are certainly wheelbuilders with MORE experience than I have ... for what it's worth, the only wheels I have had problems with were laced & tensioned by someone else.

Again, if it were MY wheel, the way I would "salvage" it would be to lace it with 14g spokes, x3 on the drive side & x2 on the non-drive side -- it's just MY opinion ... no one has to agree with it -- as I said, another spec on the rebuild may be better ... I just wouldn't do any "experimenting" at this point if I wanted to get more mileage out of the rim & hub.

 

[ScienceIsCool]

I think you did all the right things. Check the tension every so often to make sure things are cool. Though poor tension will show itself as a wobble in the rim. Just remember that if one spoke breaks due to fatigue, it's likely that the rest are near their fatigue limit too. In other words, when one breaks, many more will follow in short order. The answer is simply to rebuild as you are doing.

Personally, I wouldn't get too worked up over the beefiness of the spokes, but your plan sounds liek a good one. Just remember that Sapim CX-Rays have some of the best fatigue characteristics of any spoke in the world and they're skinny little suckers.

John Swanson

Just for point of information:

1. What maintainance is needed? I keep them clean, and check that they are true weekly. I also randomly pluck the spokes for tension.

2. The spokes that broke at the nipple did exactly that, they did not strip. The break was at the top thread.
I have considered going with traditional 14 gauge or butted spokes, but I would prefer bladed. I think it is worth it to try the heavier Bladed New Aero spokes. Bottom line, is that no one is quite sure what is wrong with the wheels.

 

[alfeng]

1. What maintainance is needed? I keep them clean, and check that they are true weekly. I also randomly pluck the spokes for tension.

I guess you are ready to buy a tensiometer ... it will allow you to quantify that the tension is actually close-to-the-same on all the spokes.

A problem CAN arise when you (that's a generic "you" ... as in "anyone") simply tweaks a wheel without regard for spoke tension ... to state the obvious, spokes are generally paired AND a wheel can be trued by errantly making one of the two in a pair (much) more taut than the other rather than having them as-close-as-possible to the same tension ...

Although I have a Wheelsmith (ugh!), the Park is significantly CHEAPER (a good thing!) AND easier to use, IMO ... so, I would recommend you spring for a Park tensiometer at some point in the near future.

I think the DT tensiometer has a dial gauge (i.e., easy-to-use), but is really expensive ...

 

[hd reynolds]

Actually, spokes don't break where the stress is highest (since the stress is same all through the spoke), but at areas where the spoke is full of residual stress from manufacturing. I.e., at the threads and bends. Additional stress cycles (i.e., riding) can then push these areas to more rapid fatigue and hence, quick failure.

I am around competitive cyclists on a daily basis and I have observed that spokes rarely break at the threads. If they dont break at the bend, it is usually below the threads where they do.