Catastrophic spoke failure. While stored..

[mercalia]

I would have thought the rims would have gone rather than the spokes.

[Brucey]

FWIW I have seen some Alpina spokes that ought to have been black-finished stainless but that went rusty. I have also seen a wheel with Alpina spokes in which all the left side spokes were merrily breaking for no good reason; it turned out that they were made with bad bends, but the ones on the other side of the wheel (which were a different length and had not come from the same box) were OK. I wonder if they are not exerting the degree of control over their materials and processing that they ought to.

cheers

[531colin]

As various people have said;
if you wind up the tension on an undamaged spoke, it will eventually pull through the rim
if a couple of spokes fail, the tension in the wheel will reduce a lot.

I can't understand how a wheel which has only done hundreds of miles can break multiple spokes in storage.
If its possible for corrosion to occur to such an extent that the spokes fail in storage and under reduced tension, then the remaining similar spokes might be expected to fail with only gentle provocation?

(over the years, I have seen all sorts of things...."stainless" spokes which rusted red, and spokes so soft that you couldn't build and stress-relieve a wheel, the spokes "stretched" and failed...but nothing like this!)

[Trikeyohreilly]

No friend pranking just in house storage.

Question is, is it worth postage to and then when they say (presumably) it's nothing to do with them, postage back/repair costs to send it back to Harrogate?

I'll not ride 26inch again and still have the original wheel for the Big Dummy for resale purposes.

Last time I sent a wheel it was £12...

[Trikeyohreilly]

Original cost of wheel was £130-£140

[531colin]

If Spa built it, then Spa will warranty it. Apart from anything else, they have the balance of a box of a thousand spokes to consider, so they need to know. Phone up and speak to Andy the wheelbuilder. Be prepared for an incredulous response, because I have never heard of anything remotely like this. Customers are sometimes economical with the truth in their dealings with bike shops, and any mention of spoke failure in storage will raise the odd eyebrow. (For example, failure to mention that a spoke which failed "just riding along" was the result of a stick in the wheel; the bend in the adjacent spoke is a give-away, but its not worth antagonising a customer and the associated slagging off on the internet, better to just accept it as a warranty job.)

[SimonCelsa]

ACI spokes, I was about to buy a wheelfull from Cyclebasket.com as they are the cheapest I can find from a UK supplier.

Might have to think again!? Although it sounds like a bad batch of spokes for a 26" Dyno wheel so they are probably quite short. Might be lucky with the longer ones for a 700C build.

[Trikeyohreilly]

Thanks 531colin.

Can understand why they would be dubious. Think it's worth sending it back just because I want to know what happened. Dont really care about the wheel

[Des49]

I used some Sapim spokes a few years ago, bought from Spa. Within a couple hundred miles the spokes in the front wheel started breaking away from the ends. There is thread on the forum about this which I will try to find.

I called Spa and was put through to a wheelbuilder, who I must admit did not believe me that the spokes must be suspect. But to be fair to them they have no idea how the wheels have been built or treated. It helped that I have built wheels for 35 years or so, plus a degree in metallurgy comes in handy. Another staff member did say they had experienced a bad batch of spokes.

Spa requested some examples of the broken spokes and eventually I was sent a box of 100 replacement spokes. I hope the duff spokes found their way back to Sapim for analysis.

The experience did put me off Sapim spokes for a while, but I do use them again. Most manufacturers will suffer some quality issues at some point. Bad spokes do seem pretty rare overall.

Stress corrosion cracking does seem a likely culprit in the OP's case, but again we do not have much information to go on. It would be really interesting to know more.

[Steve O'C]

There was a similar discussion here a couple of years ago:

https://forum.cyclinguk.org/viewtopic.php?f=5&t=110887

And prior to that I recall a similar thread on a different forum

http://tinyurl.com/j7uc4vw

Previously I had assumed that this was simply a case of vandalism, not so sure now.

Seems very odd though.

[Brucey]

I think Des is correct, all spoke manufacturers have QA issues from time to time. There may be a difference in frequency though, and what they let out through the door of the factory. AIUI for many years both Sapim and DT have used Sandvik stainless steel of the same type to make their spokes. However that might be different now; DT have spoke factories all over the world and may source their material from different places now.


I don't know of any spoke manufacturer who makes spokes that are 'unbreakable' either. If you build a wheel without due care and attention to spoke fit, setting and stress-relieving etc then even with the best spokes you will make a wheel in which the spokes will break, and this is far more common than duff spokes per se, so it is inevitable that one will encounter some scepticism when trying to return genuinely duff spokes to the supplier.


Of the big brands of spoke that I have used the best has been DT. They are stupidly expensive in the UK (I can only assume that the importer applies a huge mark-up to them) and I have seen stupidly made DT spokes (with excessively long J bends), DT spokes with a poorer than normal swaged finish in the skinny part of a DB spoke, and so forth, but I have never had a duff box of DT spokes and I have not seen a wheel made with DT spokes in which you could say the spoke were duff (rather than an occasional breakage caused by bad wheelbuilding practice). Maybe I have been lucky but I have also not yet encountered a wheel I have built with DT spokes where any of the spokes has broken in service; that includes wheels which have outlasted the rest of the bike and rear wheels that have done well over 50000 miles now. However given that DT now have multiple factories across the world maybe the past performance is no predictor of future quality.

A typical good quality 14G spoke will break at something like 250-300kg load. This suggests that the mean stress in a spoke that is tensioned to (say) 125kg (which would be typical for DS rear spokes and left side front spokes in a disc brake wheel) is getting on for half yield. This is a much higher sustained stress than this material sees in most engineering applications. In DB spokes the thinner part of the spoke is work hardened during manufacture but the stress in that material is even higher; getting on for double the stress if it is genuinely 1.6mm dia (16G).

I think this can make spokes unusually sensitive to stress corrosion cracking (SCC). This is known to occur at relatively low temperatures if there are Chloride ions or Hydrogen Sulphide present, and it is thought that some of the stainless steel grades which are used for spoke manufacture are not inherently resistant to SCC. 'Stainless steel' is only corrosion resistant by virtue of the chromium oxide on the surface and this film may not form with sufficient integrity under certain conditions; if the film has formed correctly the surface is said to have been 'passivated' and it usually happens automatically, but this isn't assured under some conditions.

If the steel contains any inclusions that outcrop at the surface, this will leave a weak spot so that when the spoke is tensioned there will be local yielding, potentially cracking around the inclusion. This represents both a stress concentration and a weakness in the oxide film. Needless to say when the local stresses are very close to yield and there is a crevice, it seems likely that SCC can occur even if the conditions are not those which might normally cause it. The only defences against this are to use a better quality stainless steel (but this may not be compatible with the usual processes of spoke manufacture), steel without inclusions in, to lower the level of stress in the spoke (more spokes, with less tension where possible) and/or to use good wheelbuilding practice, i.e. to use as such stress-relief as you can.

Steel that is substandard in quality or very poor spoke manufacturing practice can result in spokes that are (literally) full of cracks and will fail regardless. This is most likely to be seen wherever the spoke has been formed in manufacture, i.e the threads, the butted section, the head, the J bends.

In any one instance of 'bad spokes' it isn't possible to say what has caused it, not for sure, not without a lot of investigation; for one thing the original tension value in the wheel is unknown, since it changes as soon as spokes start to break. However you can (crudely) assess how good the spoke material is by flexing a length back and forth and comparing the unknown spoke material with 'known good' material. if the steel is low quality (full of inclusions) it won't take many flexings before it breaks. It is quite disturbing how much easier it is to flex and break a spoke (in good material) with even tiny notches in it; this is potentially quite close to the effect of inclusions on the behaviour of the spokes.

cheers

[531colin]

I'm no engineer, but I have been repairing and building wheels for long enough to recognise some common patterns of failure.
The most common that I've seen is failure at the elbow on driveside (rear) spokes. I accept the usual explanation that this is fatigue due to fluctuations in tension as the wheel turns, superimposed on high static tension, these being the tightest spokes on the bike.
I don't have much experience of lightweight, reduced spoke count wheels, but I'm told that elbow failure is common on left side rear spokes, due to large fluctuations in tension in a flexible wheel. Seems reasonable.
Failure at the elbow (and at the nipple, less common in my experience) is down to fatigue, and you can stop it completely (at least on the driveside) by adequate stress relief of the wheel.

Theres another group of "weird failures" typified by the failure in this OP.
In my 5 or so years at Spa, we had a few duff batches of spokes, and (if memory serves) most of them* were characterised by failure in the middle of the spoke, like this current one, and unlike regular fatigue failures at the elbow or nipple.

So, WHY?
This thread is about a front wheel, so presumably butted spokes? My imperfect recollection of failures at Spa is butted spokes, eg left spokes in a rear wheel failing mid-shaft.....but that might just stick in my memory because spoke failure "should" be driveside.
So is there something in the butting process that makes the thin bit more susceptible to stress corrosion cracking? Or is it simply that tha same sized "inconsistency" is a bigger proportion of the sectional area in the thin bit?
....or is it all in my head?
Brucey and Des, could you get any useful information by examining the broken spokes?

*ignoring daft stuff like rusting "stainless" spokes and spokes too soft to build with.

[Brucey]

I think it is fair to reiterate my main point from earlier; in parts of the spokes that are close to yield, it won't take much of an imperfection (or a change in the environment) to initiate a failure.

cheers

[531colin]

See if I've got this......
The butted bit, being thinner, will be closer to yield than the elbow, being thicker. Reverse way round to fatigue, because fatigue happens where it flexes. Ok, yeah, that makes sense.

My experiences of duff batches of spokes I've seen just one or two failures in a wheel (at any one time) so that fits, I think.

I'm still staggered that the OP has had "half the spokes" on one side fail. So even if thats a bit of an exaggeration and its a 32 spoke wheel, say 6 spokes all on one side, all in half of the circumference have failed, there wasn't much tension left at the time of the last couple of failures?
Just as well its uncommon, or I'd spend even more time bewildered!

[Brucey]

it might well be 'the neighbour effect'. When a single spoke breaks the average tension in the wheel probably reduces, but the spokes immediately adjacent to the broken one (and on the same side of the wheel) usually see a tension increase, and this may be enough to tip them over the edge if they are in any way imperfect. In a dished wheel, on the tighter side, this 'neighbour' increase is not insignificant. I have never cut spokes out of a wheel to see how many spokes you need to sever before none see any tension increase, but I might do so. It probably depends on the rim and how the wheel is built.

BTW I am suspicious of corrosion having an influence here, in that new spokes are probably coated in oil before they have a chance to passivate fully. This might somehow lead to an increased sensitivity to local imperfections; as I mentioned before if the spoke is close yield anyway, it won't take much to push it over the edge.

cheers