Cycling UK Forum

Re spokes; -that tallies pretty well with my experience too. Yes the 'testing' is going to take a while, normally. Years ago I built some training wheels with -pretty ordinary- new spokes in them and even though I'd built very few wheels at this point I was a bit cheesed off a months later to find that the spokes started breaking at the elbow.

I did a few cutups on the spokes, and bit of microscopy, and I found two things;

1) that the elbow bend was 'neutral' and not well supported during forming.

2) that the steel used had a few too many non-metallic inclusions in it.

1) meant that the inside of the spoke bend often had a 'compression fold' in it and

2) meant that there were often bad notch-like defects at and below the surface wherever gross compression yield had occurred.

Neither defect would be likely in a well-made spoke. However I saw the opportunity for an experiment; I used spoke washers to improve the fit of the spokes in the hub (which wasn't perfect) just to see how I'd get on. One of the wheels in question (two rims, thirty years and countless miles in all weathers later) is just coming to the end of its life now- I can't get another rim to match, and I suppose I've had my money's worth from the spokes.... I think I broke maybe two further spokes in that wheel despite the spokes being 'known bad'.

Since then I have used both better quality spokes, and spoke washers whenever I think the fit isn't good enough without them, unless I am specifically requested not to do so. I can't remember the last time I broke a spoke in one of my own wheels built such (or someone else managed it either, but then in fairness I don't build much for other people these days). By contrast (having become complacent about spoke failures in more recent years I expect) I've bought MTBs with 'factory' wheels and bitterly regretted not at least re-setting and stressing them early on, having been blighted with perfectly avoidable spoke failures subsequently.

I don't know if that qualifies as proof positive (I suppose I'd have to do cutups on the unbroken spokes to prove that they had similar defects in them...) but I was fairly convinced by it.

cheers

"All men dream: but not equally. Those who dream by night in the dusty recesses of their minds wake in the day to find that it was vanity: but the dreamers of the day are dangerous men, for they may act their dreams with open eyes, to make it possible. This I did."

At the time of his prang, he was living in deliberate near-obscurity as 'Aircraftsman Shaw', unable to bear the fame, compromise and corruption that accompanied his former life. Motorcycles are a slight interest of mine; I have come to realise that the Brough Superior SS100 was -by the standards of the time- an absurdly powerful machine. Maybe, after a past life of excitement and danger, he used it to get his 'adrenaline fix'.

I didn't know it had an SA gearbox though; I wonder if JB thinks it jumped out of top and killed him...

cheers

re the spoke set; I think the idea is/was to avoid cyclic bending on the rolled thread, which is very likely if there is a residual bending stress near the nipple. (I'm not sure the usual 'overload' setting method would have worked easily with these spokes on the rim in question as the required forces would have been... mighty.... ) If there is a slight cyclic bending stress at a small kink in a plain spoke, it seems better able to cope with it; if it were not so, any tiny bend or dink in a spoke would spell disaster. It could hardly be worse than the hub end in any event.

I reckon that most spokes fail at the hub end because they simply don't fit right and/or the wheel hasn't been stressed correctly during building. In a long second place comes failures in the threads, usually because of cycling bending where the spoke is angled where it goes into the rim and hasn't been 'set' right etc. A distant third place is reserved for spokes that fail at the transition to the butt, and although I have not carried out a metallurgial evaluation of the last of these, I suspect that the culprit might often be surface finish which can be at its poorest at this point.

I'd be interested to hear the views and experience of others on this point, maybe in another thread.

cheers

$1800? Cor blimey that is a bit 'spendy'....

cheers

Interesting comments about the wheel build, I guess some of them go for almost any chunky hub with that many crosses. Years ago I had a tandem with an 'Atom' hub brake (which eventually received a much-needed thicker axle) and there the builder copped out and used much thicker spokes with a distinct 'set' to them near the nipple so they went in the rim OK. Normally I would have rebuilt the wheel as a matter of course, but since I couldn't source stainless spokes of the correct type at the time, and we never broke any, I never bothered.

Re the gear range; everyone has different choices to make here, for sure, and I don't expect my choices (or the Alfine hub come to that) to suit everyone. For example, I used to have a gear set-up with 50,44 and a granny on my touring bike. This produces massive duplication of gears with 2T gaps at the back but they were close enough together for me (typically about 13%) and it made my life very simple, keeping double shifts to a minimum and running the chain pretty straight most of the time.

I don't see the benefit of massively high top gears on a touring bike myself, and even in the mountains on good roads I have rarely wanted more than 100" and in truth I've rarely needed more than about 90" for any length of time. At the bottom end, anytime I'm using less than 25" I'm generally thinking -quite hard- about walking anyway. I'm sure that there are a few times and places where you can argue that you would have some benefit from a wider gear range; I suspect that few people are genuinely going to need that although more will certainly want it I am sure.

If I were considering an Alfine 11 for touring I'd set the top gear to about 95" and the bottom would then be 23". This would also give me a nice 64" gear for just 'tapping along'. If I did run out of gears, so be it; I might spend more time looking at the scenery -which is the point of it, after all....

cheers

Chain suck is normally associated with a hooked chainring and a chain that isn't as stretched (usually) as the one that caused the hooking in the first place. Unusual that the extra chain tension of a shift should trigger it tho'.

New chainring or 'de-hooking' the old one with a file (if it is worth saving) usually fixes chain suck, but maybe there is something more going on here.

Most folk would go for lower gears for regular offroading, and a triple will work OK for most and avoids the double's mahooooosive double-shift.

-but I'm sure others will chip in here...

cheers

I had a thought, maybe someone has mentioned this already (in which case apologies) but what profile is the inner cage plate of the front mech? Do you think the chain hits it during a bad shift?

Many MTB front mechs have a funny profile to the inner plate and I have long suspected that the chain can fly across, hit the inner plate, and then be deflected to the right as it slithers down the inner cage plate (or something like that...)

The rise and fall may or may not be significant. There are lots of bottom/top pull, top swing MTB mechs (that work quite well) that have hardly any rise and fall to them. If it is important, it maybe interacts with some other aspect of the design?

cheers

yes (again)...

cheers

-ah, I see....

-bit worried about the sharp angle causing cable fatigue but then its pretty bad in stock form, too.

BTW isn't the Fargo just, ahem, just a 'cross bike with fatter tyres? Or am I missing something..?

cheers

JB maintained throughout this 'correspondence' that the gear would slip under hard 'continuous' load and that the cable adjustment ( which is rightly the #1 suspect normally ) was OK. Although we cannot be 100% sure I think this is most likely true, more or less; what I'm not at all sure about is that the reason for this is as he concluded.

1) I think TwoWheelGood's comment re. another, different fault in the hub remaining might be spot on. JB implied that the hub had already seen a lot of use. If previously run with the bearings loose in top gear, for example, the planet cage pawls and pawl ring could wear to give a weak rejection ramp. If (say) only one pawl engages in top gear it will certainly slip under load. Likewise if the planet cage pawls were not a perfectly matched pair, the resultant eccentricity would cause either the pawls or the clutch driver to 'walk' out of engagement, just as a bent axle might.

2) It is most certainly the case that clutch driver and planet pins will eventually wear in the pattern described and that this might, -nay, will- cause slippage. It is equally certain that this process is greatly accelerated by faulty bearing adjustment, cable adjustment, lubrication, or gear shifting technique. However JB's assertion was that the axle flexed to make the clutch driver etc axially misaligned and therefore walk out of engagement under continuous load, even with new clutch and pinion pin parts. I daresay this might be possible in theory but:

a) if an estimated chain tension (given a few givens) load of 300-400lbs is applied to the axle about 1" from the end (where it is typically reacted by the bearings), I doubt the flex would be sufficient to cause this, ( whereas the slightest bearing slack or other source of misalignment would do it right away... ) and

b) I doubt very much that the load was truly continuous; he says he was sprinting without clips and straps. By my reckoning most folk pedalling like this will produce a rather discontinuous chain load, and unless the clutch driver is notched in the wrong place through wear, or the pinion pin/clutch wear is especially bad, a good clutch driver spring should push the clutch back into engagement whenever the chain tension reaches a low level.

All in all, I think it is quite likely that the hub slipped for some other reason. Maybe he'd have figured it out better if he had persevered.

I've seen AW hubs quite happily installed on the rear wheels of tandems and apparently give little trouble. I've personally used various -mostly old- SA gears for tens of thousands of miles (including hard training for road racing) and have not encountered similar troubles; however I have seen other gears that gave a lot of trouble which had just very subtle wear or faults on them in unexpected places.

Whatever the true reason, JB's comments -which are likely to be taken seriously by the US cycling community- have convinced many people that SA gears are junk and were manufactured by a company that knew this and did everything possible to avoid admitting it. Whilst SA have had many troubles, and like many companies, have sometimes produced products that were not as good as they should have been, JB's comments smack of kicking someone when they are already down. The AW hub survived as long as it did because it was simple and reliable, and the world of cycling would have been immeasurably poorer had it not existed.

+1, 'Brevity is the soul of wit' and all that.

On track record that makes me, er, not very witty then....

Re reynolds tubes- I found this article too (scroll down to find it, nice stuff on the way there too );

http://spinwell.co.uk/tag/reynolds/

twelve men!

-blimey-

cheers

isn't the Alivio/Horizon arrangement a bottom pull/top swing? If so I'm not sure there is an alternate clamping position for the cable; if your mech is of the sort I am thinking of the pinch bolt is radial to the pivot axis of the part the cable is clamped to. This means that the cable is kind of clamped to a drum, and it doesn't matter which side of the pinch bolt it goes, it gives the same throw.

The alternate clamping position is -I think- for those mechs where the pinch bolt lies paralell -more or less- to the pivot axis of the part the cable attaches to. Normally this is a bottom pull/ bottom swing arrangement and the pinch bolt is on an arm that sticks up when the cable is slack. Here the idea is that the cable goes the other side of the pinch bolt, i.e. slightly closer to the pivot axis. This gives a longer throw.

It might be that there is another reason for your problem. Do you kow the model numbers of your shifters and front mech?

I hope my descriptions make sense...

cheers

Recently whilst trawling the web for nuggets of wisdom -as rare as the gold variety in some quarters...- I stumbled upon this (but please read on before clicking the link)
http://yarchive.net/bike/sturmey_archer_hubs.html .

This is a ten year old discussion, but it is about a product that is far older than that.

The gist of it is that Jobst Brandt, well known US cycle buff and author of a good book on wheelbuilding, amongst others, went over the bars whilst sprinting in top gear on a Sturmey Archer AW Hub which slipped. This was many years ago, when he was still a cycling neophyte. The LBS, not unreasonably, pointed out that the hub clutch driver and pinion pins were very worn and that he should change them. Despite doing this, he took another tumble in a similar fashion, then concluded that the design was flawed, and that SA's attitude was all wrong.

My question is this; was this a safe conclusion to draw? If so, fair enough, but if not, SA were perhaps done a great disservice.

Comments please.

mattsccm wrote:You will probably find that a touch of wear in your chainset will be enough to compesate for any mis matches. I di find that a 10 speed cahin was snug but working on a near new chainset that was old enough to be 8 speed I guess. When I tried again a few months later it was fine. I suppose it that 0.2mm mentiuon that worn.

-exactly- and others have had similar experiences. But you cannot rely upon wear to ensure compatibility.

IIRC shimano chainrings are 2.0mm wide for 8-speed, ~1.85mm for 9 speed and skinner again for 10 speed. Campy ones are different but also get skinnier, obviously. The chain needs a little sideways clearance to run and especially shift OK; each of the respective chains is slightly loose from side to side even on the correct, matching chainring and will therefore appear to 'fit' onto an older, wider chainring. It might even work OK if the old chainring is a little worn, but don't assume this is definitely going to be the case.

cheers