Why do my spokes keep breaking? - Bike wheel science.

[rogerzilla]

The left hand spokes can't always* be tightened enough to prevent them going completely slack when you hit a bump; the lowest one may be perilously close to zero tension when just riding along, if you're fairly heavy and the rim can't take a very high average tension. The RH spokes are prone to mauling from an unshipped chain but otherwise have a better chance, since they never get anywhere near slack and the variation in tension when riding, as a proportion of initial load, is much smaller. Spokes don't break because they are too tight. The difference between leading and trailing spokes isn't massively important on a wheel with a rim brake, and builders can never agree on which side of the flange they should go (I like the "pulling" spokes to be heads-in, since they're better-braced, but some people insist they should be heads-out to pull the spokes clear of the rear mech cage on climbs!).

Derailleur wheels are a horrible compromise, really. A dishless wheel, such as you get on a fixie, singlespeed and on some IGH hubs, gives far less trouble but you just can't have one with 11 speeds (or even 7) and the usual 130mm spacing. That spacing can't be increased much further because your feet end up too far apart at the pedals.

*DT Revolution or other thinner, stretchier, spokes, are good on the LH side because they are more able to retain tension under a big shock load; they stretch by 0.5-1mm when tensioned.

A spoke (say plain gauge 2mm diameter) has a cross section of 'Pi'mm2. It's under tension of (say) 1000N so about 318N/mm2. "A simple rule of thumb calculation for the fatigue limit is one-half of the ultimate tensile strength " which for stainless steel (material of spokes) seems to be in the region of 500N/mm2. Half that is 250N/mm2. So properly tensioned spokes are above the fatigue limit of their material.

Spokes are drawn and therefore much stronger than plain stainless material. A Sapim Race spoke has a strength (they don't say what, but it's usually yield strength, not UTS which is even higher) of 1300 N/mm^2. So my 150kgf spokes aren't anywhere near that, at about 580N/mm^2.

 

[Ajax Bay]

The left hand spokes can't always* be tightened enough to prevent them going completely slack when you hit a bump; the lowest one may be perilously close to zero tension when just riding along

Roger - before I invite you to share a link to anyone else who says something along these lines, please could you confirm you've read this thread from start to finish. I think we have tried to quantify, very roughly, the sort of tension changes which might occur in the lowest spokes as they roll through.
On one of your well built rear wheels might the tensions be 1000N on the left side and 1500N on the right? A rear wheel under a heavy rider might experience a rolling load of 700N. So even if ONLY one spoke reduced in compression by that amount the spoke would stay in tension: is 300N 'perilously close'?
In practice several spokes would be taking part of the load (at least 3 even on a 20 spoke wheel) so the residual tension would be well more than 300N. A heavy rider would be well advised to choose a rear wheel with a decent number of spokes and a rim which can be built into a robust wheel.
So positing a heavy rider and a rim which "can't take a very high average tension" is a strawman - in the context of 'can't get the left hand spokes tensioned enough'. Surprised you didn't add in "low spoke count" to make it a target really easy to punch.
On the material properties of a Sapim Race spoke, I think you are agreeing with me "properly tensioned spokes are above the fatigue limit of their material" ie eventually, in normal use, they will fail through fatigue - thank you.

 

[rogerzilla]

On the material properties of a Sapim Race spoke, I think you are agreeing with me "properly tensioned spokes are above the fatigue limit of their material" ie eventually, in normal use, they will fail through fatigue - thank you.

Er...no. Even my very tight spokes aren't stressed to 50% of Sapim's quoted strength (which I believe is yield, not UTS).

Rims are usually quite stiff these days, but older rim designs (and some that you can still buy, I'm looking at you, Mavic) will elastically pringle well before the spokes are as tight as you would like. In fact, this used to be the way to determine optimal tension: get to the pringle, then back off every spoke by a half a turn to a turn. In these cases, the LH spokes can be alarmingly slack, maybe 50kgf, and prone to losing all tension very easily. The 18 LH spokes on my rear touring bike wheel (built by, allegedly, one of the UK's best wheelbuilders) are all over the place tensionwise but many are a mere 60kgf, which probably explains why it went out of true easily. The 16 LH spokes on my race bike are 100-110kgf, which is more like it.

 

[Ajax Bay]

I erred when I asserted that "properly tensioned spokes are above the fatigue limit of their material". They aren't. Thank you for picking me up on that.
Still think your strawman (heavy rider, weak/poor rim) in the context of 'can't get the left hand spokes tensioned enough' is not useful, except as an outlier.
@Yellow Saddle 's post provides detail.
"the cyclical [stress] range does have an effect on fatigue life (sometimes called Endurance Limit or even Fatigue Strength). [But] steel has a fatigue limit, which means that cyclical changes in stress have no effect on the fatigue life of the piece: if the changes are kept below a certain level, then the material effectively has an infinite life. This limit varies a bit and can be as [low] as 35% of tensile strength or as [high] as 65% of tensile strength.
"We can relate back to spokes. A [example double butted 2.0/1.8mm] spoke has a tensile strength of about 2400N. Typically, we build to no more than 1500N and the cycles will be (big approximation here) about 200N. Even with a [high fatigue] limit of 35% [of tensile strength] you can see that a wheel can easily be built with infinite spoke life.
"The important factors are: enough spokes; stiff enough rim and proper stress relieving.
"Unfortunately, rims, made from aluminium, have a fatigue limit and it is pretty low too. Rims therefore crack long before spokes break, if the wheelbuilder knows his onions.[/QUOTE]

 

[rogerzilla]

Yellow Saddle is right, and it's perfectly ok to re-use spokes when a rim wears out (I use new nipples, since they cost peanuts) since they will go on indefinitely if built properly.

Caveat: cheap unbranded spokes can break even in a good wheel. The top branded spokes (DT and Sapim) are vastly better than they were in the 1970s or earlier but OEM wheels may use nastier ones. I broke a spoke on my track bike last year; it was a no-name set of spokes that came on a 2003 Fuji Track and was on its second hub (!) and rim. I felt cheated.

 

[T4tomo]

I know nothing about wheel building so here is a question for the experts....

I have a broken spoke on a wheel from 1987, original spokes as far as I can tell. Hubs and rims are fine and the bike doesnt get a lot of use - its special occasions ride. Spoke tension appears to be be all over the place, do i just get the one spoke replaced or should i ask him to rebuild both wheels with new spokes. Apart from new spokes looking nice and shiny (which is definitely a consideration) is it generally a good idea to replace 34 year old spokes?

 

[Ajax Bay]

This non-expert would opine: replace the spoke and get the spoke tensions evened up and taught. Ride on. This assumes the current spokes are stainless steel, you can't see the spokes' threads at all (ie inside the nipples, therefore right length) and that you've had the wheel from new so you know it has done very limited mileage. Rear wheel, right side?

 

[T4tomo]

I've not had bike or wheel from new.
Broken spoke is rear non drive side.

 

[Ajax Bay]

In that case I would still replace one spoke and tighten/balance up generally (pings the same tone on the left side and higher on the right). If another one goes, that suggests that they (all xx spokes) are reaching their fatigue life limit (having done an unknown number of miles under their previous owner(s)). More will follow. Relace, after a careful check of every rim spoke hole for cracks.

 

[Punkawallah]

So, his successive failures are a result of ‘old age’?

 

[Ajax Bay]

I said: "If another one goes, that suggests that they (all xx spokes) are reaching their fatigue life limit (having done an unknown number of miles under their previous owner(s)). More will follow."

So, his successive failures are a result of ‘old age’?

@T4tomo has only had one spoke part, so these are not "successive failures"; and that's important. To adapt Goldfinger's 'happenstance, coincidence, enemy action' assessment:
Happens once: replace.
Repeat occurrence: indicator that the set of spokes are reaching their fatigue life. (NB this is not "old age"; the variable is not time but tension cycles (millions)). The set of individual spokes will have a range of cycle fatigue limits with a distribution which could be Gaussian. One may go for a reason other than fatigue (eg physical damage), but once the second goes, the risk of just replacing only that one is that a third will go shortly, to the detriment of an enjoyable ride. Bit akin to identifying a lesson, but then not learning it.
HTH

 

[Punkawallah]

Thanks for that :-) Every day’s a school day! Think I was commenting on an earlier post than the one you referenced, still getting used to this whole ‘Interweb’ thing. Serves me right for not checking :-)
Not an engineer, so happy to work with ‘old age’ and ‘fair wear & tear’ without looking to ’gotcha’ 👍

While I have you, Gaussian distribution?

 

[lane]

GT grade spokes kept breaking. I was told poor quality rims and when I googled the rims quite a few with same issues. Spa built wheels decent rims no more breakages.

It wasn't fatigue with the Grade not done many miles.

 

[Ajax Bay]

GT grade spokes kept breaking . . . it wasn't fatigue

Absolutely. Spokes sequentially breaking after a long rolling life indicates fatigue limits being hit (and a relacing required). Before that it's a wheel building issue.
Have a read of this thread: https://www.cyclechat.net/threads/gt-grade-stans-grail-spoke-problem.220652/ to which @I like Skol contributed.

 

[lane]

Absolutely. Spokes sequentially breaking after a long rolling life indicates fatigue limits being hit (and a relacing required). Before that it's a wheel building issue.
Have a read of this thread: https://www.cyclechat.net/threads/gt-grade-stans-grail-spoke-problem.220652/ to which @I like Skol contributed.

Thanks interesting thread. My GT Grade was stolen so that solved the problem in a way.