Cycling UK Forum

"The rest of the rim remains at the same distance from the hub as when the wheel isn't loaded, and the spokes at the same tension."

So effectively the rim must be "rubbery" enough to accommodate the slackening of spokes, i.e. the circumference gets shorter? That's the only way that one bit can get closer to the hub without any other bit getting further from the hub - unless the rim pretzels of course... ok... I'd assumed that it was less likely to compress than (even several) steel spokes were to stretch.


I've often got into discussions like this where physics is debated, and nobody's physics seems complete - after all, anyone understanding the whole process completely would just go and design the unbreakable spoke and become a millionaire overnight - the point is i've seen breakages from both loose and tight wheels - the relevant factor is unevenness in tension - from what's said above it may be that my assumption that it's the tighter spokes that ping is wrong, it may be the loose ones - after all when a spoke has broken one isn't in a position to test how tight it or the others around it were before it broke, all I know from building hundreds of pairs of wheels is that when the tensions feel even, and pairs are even within the pair (checked by looking past the crossings and seeing whether they bulge in or out - I'll do a diagram to explain that if that bit comes across as chinese) - spokes don't break and largely wheels don't go out of true, give or take the odd crash. ...as long as they're not rustless spokes over ten years old that is

PS - another point about uneven spoke tensions is that, like my example with the front wheel that my mate didn't thank me for building, a duff rim means you have to compromise spoke tension evenness to get the rim straight - I remember building GP4s and G40s for a shop for a while and they were a breeze - very straight to start with - you just cranked them up evenly and there was only adjusting and de-stressing to do at the end. Cheap brands for mid-range bikes were a nightmare, and occasionally I'd have to tweak brand new factory built wheels - a double nightmare cos not only were they cheap rims but also it was a case of slackening and starting again. When a wheel has a spoke snap it sometimes seems to affect the rim permanently even if it was a good one to start with, which is why I tend to give up on a cheap wheel if it's had breakages - it'll be fine for riding to the shops but if you're going to do some serious riding I'd start again with a different wheel - preferably provided under warranty.

I would write a letter to Dawes saying that if they consider a dealer who says a spoke snapping on a front wheel in such a short time is "what you expect from a Dawes cycle" to be good judgement. Then you are regretting buying a Dawes.

Raph wrote:Take any bit of a shaft of a spoke and you can probably suspend the houses of parliament on it, but as mentioned above the usual point of breakage is at one end or other which suggests that the strength of the material as measured in the lab isn't the whole story.

you're quite right of course raph. but it's my understanding that the over 200kgs figure is for a spoke who's testing machine held it as a rim and hub would hold a spoke, if that makes sense?

andrew_s wrote:Sapim quote a strength of about 1400N/mm2, a double butted spoke is about 2.5mm2 in cross section, so that would give an expected breaking load of about 350kg.

is that from a testing machine that holds the spoke as it would be held in a wheel? its a reassuringly high figure

Raph wrote:the relevant factor is unevenness in tension - from what's said above it may be that my assumption that it's the tighter spokes that ping is wrong, it may be the loose ones - after all when a spoke has broken one isn't in a position to test how tight it or the others around it were before it broke

you're right again, you can't tell the tension of a snapped spoke but if it was tight, then it's load would vary between it's static load (the tension it's under from the nipple) and the load it's under when the wheels in use (ie. it's nipples tightness + it's contribution towards carrying the rider) neither of which are likely to be over 200 or so kgs. however, it is possible i guess to build a wheel so tight that the static load is well over 250kgs, and obviously then the extra load of the rider will snap it.

however, we've all snapped metal things by bending them back and forth, a few cycles of this and any metal will snap. as will a loose spoke at the bend/hub that gets slight movement on every rotation. this is much, much more likely than the first overtightened scenario

interesting debate nonetheless. thanks all

hubgearfreak wrote:it is possible i guess to build a wheel so tight that the static load is well over 250kgs, and obviously then the extra load of the rider will snap it.

No, because the weight of the rider only slackens the spokes

It would be a pretty meaty rim that could take 250kgf of tension in each spoke, too. An MA3 or a Rigida Nova seems to pringle at about 50kgf (I don't have a tensionometer, but 75kgf is a reasonable target for a strong rim, and those two rims pringle well before what I'd consider "normal" tension - in comparison, DT XR 4.1 rims take massive tension and the nipples will strip or round first!)

rogerzilla wrote:

hubgearfreak wrote:No, because the weight of the rider only slackens the spokes

you haven't thought it through roger old chap

if some slacken, then others must tighten. and as the wheel rotates, they change over. but hopefully none will become so slack as to have no load - ie. not under any tension, as then the bend will rattle and bend in the hub and it'll soon be curtains.(which is what's happened in the OPs wheel)

the only scenario where spokes could exclusively slacken is one in which the rim decreases in size.

"if some slacken, then others must tighten"

Yes - that's what I was thinking. To say the weight of a rider only slackens some spokes without tightening others means all the spokes are slackening on average, which is impossible unless you reckon the rim is changing its circumference. If one bit of the rim came in further than flattening (i.e. bulging inwards) then its effective circumference would be reduced. If one bit of it just flattens then the rest of it flexes outwards to accommodate the short-cut it's taking rather than carrying on the arc - that's how some spokes are shortened and others are stretched. Maybe a diagram is needed here!

I wasn't thinking that the up-down flexing of the rim was significant in the changing tensions of spokes - only sideways flexing has the leverage to pull significantly on spokes - hence common breakages while out of the saddle, also most common on rear drive side because the leverage is the greatest.

HOWEVER - it's fair enough to say that the increase in tension from adding a rider to the equation is insignificant compared to the tension in the wheelbuild - and therefore the only thing that breaks spokes is the fatigue caused by the slackening/tightening cycle - which would be worse in a slack wheel, which I admit is perfectly plausible, as a tight wheel would always keep up enough tension on all spokes so that they wouldn't flex.

Also mildly convincing is the notion that it might have been the slackest spokes that pinged on wheels whereas I always assumed it must have been the tightest. I've never even thought about that till now since the only important thing was that an even wheel has always been a strong one, and there's a slackness below which i wouldn't go when building. Thanks to this discussion I'm now just wondering...

What still doesn't add up in my experience is that I personally haven't noticed that loose wheels break more spokes. Factory wheels are usually fairly loose and break more spokes - BUT then they're almost always badly built - I've never seen a beautifully evenly tensioned but slack wheel that broke any spokes.

hubgearfreak wrote:

rogerzilla wrote:

hubgearfreak wrote:No, because the weight of the rider only slackens the spokes

you haven't thought it through roger old chap

if some slacken, then others must tighten. and as the wheel rotates, they change over. but hopefully none will become so slack as to have no load - ie. not under any tension, as then the bend will rattle and bend in the hub and it'll soon be curtains.(which is what's happened in the OPs wheel)

the only scenario where spokes could exclusively slacken is one in which the rim decreases in size.

There is indeed a miniscule increase in tension of the spokes either side of the slack ones at the bottom of the wheel, but this is so small it's usually ignored. You can't detect this increase in tension by the pitch of a plucked spoke, but you can easily detect the slack ones using the same method.

Have you read "The Bicycle Wheel"? There are calculated spoke tensions in there for a loaded wheel.

Raph - "pinging" is normally spokes untwisting as they relax at the bottom of the wheel, and indicates a crappy wheelbuild. Sadly some shops can't produce wheels with the twist removed; it's not hard.

Raph wrote:the only thing that breaks spokes is the fatigue caused by the slackening/tightening cycle

you're right again. but if the wheel's properly built, then there's no slackening, just differing positive values of tension, ie' each spoke may vary from 70kgs in it's static state to between 50-90 in it's varying loaded state. i don't know the figures, i'm sure that i'm way out, but the trick is for all spokes to be under some tension all of the time

as for them only breaking on the drive side, i wouldn't know as i've only built wheels that are fronts or with hubgears in, both of which are symmetrical. you dérailleur chaps do put up with considerable faults to have your ratios. but that's a different debate

"There is indeed a miniscule increase in tension of the spokes either side of the slack ones at the bottom of the wheel, but this is so small it's usually ignored. You can't detect this increase in tension by the pitch of a plucked spoke, but you can easily detect the slack ones using the same method."

OK so it's also partly because of numbers - few spokes slacken, many spokes tighten so the effect is less felt in the tightening - however when you're out of the saddle putting sideways stress on the wheel, the spokes getting stretched by the sideways deflection of the rim are few because the deflection is localised, so they're getting a concentrated dose of extra stretch - their counterparts on the other side are getting an equivalent amount of slackening.



"Raph - "pinging" is normally spokes untwisting as they relax at the bottom of the wheel, and indicates a crappy wheelbuild. Sadly some shops can't produce wheels with the twist removed; it's not hard."

rogerzilla - sorry, by "pinging" I meant breaking. That untwisting "ping" (from now on I don't mean breaking!) is something I make sure is resolved in a newly built wheel by standing on the rim sideways all the way round on both sides (carefully!), then re-truing. It also settles the nipples into the ferrules in the rim and settles the elbow bends into the flange so it doesn't carry on settling once the wheel's being ridden. I do that three or four times in a wheelbuild. I've only once had a rim pretzel because of it - I don't know if it means that rim was duff or my build so far was duff (most likely), or simply if it's an overkill method of "unpinging", but it certainly works. One bike shop owner I built wheels for insisted that the de-twisting thing was a waste of valuable money-making time cos what he saw was a perfectly straight wheel, then I'd go and stand on it sideways and have another ten minutes of truing. I also always lube the spoke threads to reduce friction that causes the spokes to twist when turning nipples - this is something I've never seen on a factory built wheel.



By the way, I make sure pulling spokes on the rear are laced in to out so that they have the outside of the flange to rest against, reducing the flexing at the elbow - if they go directly to the inside of the flange then the flexing is worse, and that's often where cheap back wheels give way, which adds to my skepticism about the notion that slackness per se breaks spokes, and also about the notion that the torque from the rider's pedalling is insignificant - rear pulling spokes break most often, especially if they're laced inwards.



The conclusion I get from this thread is that the range of tension on a spoke is what causes it to fail, not the peak of tension in itself - though the higher the peak the greater the range, but also if the spoke regularly slackens to the point that the twist can "ping" out of it, then that's the other extreme which also widens the range. Can't remember who's point it was but the proximity of elastic and plastic deformation in a single piece of metal is also key - the fact that a part of the spoke, e.g.t the outside of the elbow, is stretching in a way that it could do indefinitely but another bit right next to that e.g. the inside of the elbow is deforming permanently - each time the wheel goes round.

Still not sure how that affects the "tight wheel vs. loose wheel" debate! I suppose I've always tended to build tight wheels - and had only two breakages ever, so even though one of them was in an over-tight wheel, I'm struggling to stick to my original argument! I've even forgotten what it was...

I went to my dealer where I purchased it expecting them to replace the spoke as it was ...I thought ...under it's first year's guarantee. The dealer charged me £10 for labour and relacement costs ...saying that "after nearly six month's cycling this was to be expected."

Kindly tell us the name of your LBS so as nobody else has to have the misfortune of dealing with such nonsense.

[As an aside, I can't believe you actually paid them.]

PS "you dérailleur chaps do put up with considerable faults to have your ratios."

One breakage in a lifetime so far, I can live with that!

P1ssing about with internals of hub gears - not frequent but not my kind of fun. And being stuck with the gears they reckon you should have... I know cos I've had a couple.

Having said that -cleaning rear mechs? - now that bit's a total pain! I've been wondering about hub gears for a while now - but I'll ask you about that on another thread rather than hijack this one...

I admit to using the all-my-weight-on-the-rim-sideways technique to get residual twist out, but (in theory) you can build a wheel with no twist by backing off with the spoke key slightly after tightening (and vice-versa when loosening a spoke). Jobst Brandt reckons a good wheelbuilder can feel the twist; I can't, so when I built a wheel with DT Revolution spokes, which twist like nothing else because of their 1.5mm centre section, I marked one side of every spoke with a pen before tensioning. A bit obsessive, I know, but these spokes can twist through 180 degrees.

"(in theory) you can build a wheel with no twist by backing off with the spoke key slightly after tightening (and vice-versa when loosening a spoke)."

I do that as a matter of habit, but I don't trust it enough to leave the wheel as it is at the end - especially since as you mention different spokes obviously twist different amounts. I did once try to measure how much a spoke twisted so I'd know how much to back off, but I gave up on that pretty quick! The proof is the fact that I do get the odd little "ping", and even regardless of twisting, there's always a bit of settling needed at all the points of contact - threads, spoke-to-flange and nipple to ferrule. The ferrules probably settle into the rim a bit too. So a bit of "stressing" towards the end of a wheelbuild is still a necessary thing, and all the more enjoyable when it p1sses off a money-obsessed employer!

I always "settle" my wheels by riding for a day then re-trueing them.
I love the musical tinkling as I do the first 2 metres. Obviously proffesional wheelbuilders dont have this luxury.
The vast majority of spokes that have broken for me are those that snap while trying to adjust them.This is due to them being twisted by a seized up nipple. So now I put some copper grease on the threads when I do a wheel.

ALL of the other failures are when my chain or part of my luggage goes into the spokes.