front wheel flexes whilst climbing? cause

[martinn]

Spoke pattern is i think is this

https://www.sheldonbrown.com/wheelbuild.html the muliti coloured pattern about halfway down

[531colin]

what is the spoke pattern on this wheel?

Spoke pattern is i think is this
https://www.sheldonbrown.com/wheelbuild.html the muliti coloured pattern about halfway down

All you need to do is to count how many other spokes every spoke crosses, including the cross that's right next to the hub flange. (Its probably 3 cross). Number of crossings has very little impact on (lateral) stiffness.

[531colin]

With higher spoke tensions a given external load makes for a proportionally smaller deflection, no? At least that's what I get when I try and visualise the vector triangles.

No. Elongation is linearly proportional to load increase.....Hooke's Law/Young's Modulus/whatever.

[Brucey]

Hi Brucey,

If I have interpreted your post correctly, the current wheel set up is less stiff than an equivalent one which had a wider hub/ flange.

But the flexing may be due to the tension being either too much or not enough?

yup.

so how do you tell what tension it should be? I don't have any tension meters and I am tone deaf in case you can hear a tone difference.

hence I suggested go to a good wheelbuilder and ask his opinion.

Out of interest Brucey would you build a wheel set up as mine for an individual weighing 80Kg? I ask as it seems to be that if you are over 80-85Kg your considered a heavy rider and need "more robust" solutions?

quite a lot of wheelbuilders won't build that hub at all, regardless of the drilling or the weight of the rider. Not every rider the same weight rides out of the saddle in the same way. I'd err on the side of caution myself, and build that wheel either 32 or 36 spokes depending on the way the rider rides, and I certainly wouldn't use a rim any more flimsy than an archetype. In choosing a 28h rim you have made the wheel about 14% or 28% weaker than it might have been (i.e. 32h or 36h) and for what.....? saving about 30g or 60g.

I spoke to a good wheelbuilder this afternoon and he reckoned that it might be impossible to self-pretzel a wheel with an archetype rim; at least not without great risk of stripping threads on the spokes or breaking something first. This would suggest that it is likely that the spokes in your wheel are too slack rather than too tight.

If you want an ad-hoc test, squeeze the spoke crossings and compare the tension with a rear wheel. If the SP wheel compares with the NDS rear tension, it is too slack.

cheers

[531colin]

...........
BTW I've had to partially rethink my idea of fatigue in spokes; on the one hand it might usually be worse in the spokes which have higher tension in them. On the other hand the spokes that see the largest tension variations can have the worst fatigue damage and spokes that go completely slack may also be flexed in bending, which may be extremely harmful. It turns out (according to some FEA models) that the pushing spokes on the NDS may see the largest tension variations and they are also the ones that are most likely to momentarily go completely slack...

Does practical experience (of which spokes in a wheelset most often fail) support or refute this re-think?
There are other situations where the spoke elbow flexes, for example if the elbow is too long to be supported by a thin spoke flange, or if the elbow bend is wrong for the spokes' situation (and the wheel hasn't been adequately stress-relieved).
Feeling mischievous, I might suggest that NDS spokes should be first to fatigue in 130mm, 11-speed, adequately stress-relieved wheels with a good match between spoke elbow length and flange thickness.
Therefore I expect the Driveside spokes to fail first in the great majority of wheelsets.

[531colin]

Hi all,

I have a SP dynamo hub with a 32 hole H-plus son rim (Handbuilt wheel from a reputable supplier). I weigh about 80Kg. The forks are carbon

When I climb,and get out of the saddle and put some effort in, ............

[Brucey]

Does practical experience (of which spokes in a wheelset most often fail) support or refute this re-think?
.

There's a significant minority of wheels in which the first spokes to break are NDS ones not DS ones.

Opinion is divided about whether (or how) a small cyclic stress at high mean stress compares with larger cyclic stress at a lower mean stress.

In addition it may be that there is some variation of wheel building technique (like lack of setting and stress-relieving) that causes spokes on the NDS to break prematurely. It isn't quite clear what is the most important thing.

cheers

[Samuel D]

What gauge are the spokes in this flexy wheel?

[JakobW]

With higher spoke tensions a given external load makes for a proportionally smaller deflection, no? At least that's what I get when I try and visualise the vector triangles.

No. Elongation is linearly proportional to load increase.....Hooke's Law/Young's Modulus/whatever.

Imagine a wire strung between two points with a weight hung off the middle so it deflects somewhat. Now double the tension on the wire; the deflection will become less. The forces diagram for a rim segment is more complex, but the same principle applies, doesn't it?

Edited to add - I meant a load normal to the rim side walls, so as to cause the wheel to flex side-to-side.

[Brucey]

I agree with Jakob.

The lateral restoring force that arises when the rim is displaced laterally (considering only lateral forces and displacements) arises from two sources

1) the spokes each side may lengthen and shorten respectively and therefore have different tensions in them

and

2) the angles at which the spokes are pulling the rim changes

The first of these will provide a restoring force that is almost invariant of the preload (if you don't make life complicated for yourself by considering spoke crossings, rim stiffness etc) because as Colin points out the spokes are linearly elastic.

However the second of these provides a restoring force that is proportional to the preload, more or less.

I will work through a practical example to show the relative magnitudes of the two effects, but the bottom line is that this suggests that the rim does get harder to displace laterally when the spoke tension is increased. In practical terms this is limited by either the maximum tension the rim will accept before it is liable to crack in service, or the tendency of the wheel to collapse via Euler buckling ( mostly older single-wall and shallow double-wall rims).

cheers

[Samuel D]

I’d be interested to see the worked out example, because it all depends on the relative magnitude of source (2). My gut instinct – often wrong! – says it must be tiny over the necessarily small deflections that may be permitted when using rim brakes, and that even if it isn’t, there’s not much wheel-to-wheel variability within the range of preloads available to us in functional wheels.

Here’s a thought arising from the small deflection that I observe with my XR2 rims, even with a 32-spoke SONdelux wheel (though it must be said that I’m barely 65 kg). The lateral force on the wheel occurs at the contact patch. For that to displace the rim in the other direction at the brakes implies the rim is transmitting this force. So paradoxically, stiffer rims may show more deflection at the brake, given a few givens (as Brucey likes to say). They may do this even as they deflect less at the contact patch (not that you’ll notice that).

Perhaps I’ve overlooked something obvious.

[Brucey]

if you consider the rim to be very stiff and the spoking to be (laterally at least) somewhat flexible then you can see how the wheel can flex so as to cause brake rub. As Samuel says rims that are relatively flexible (with stiffer spoking) may move at the contact patch but not so much in the brake caliper.

But in consideration of the forces required to laterally displace a section of rim, one can examine two simple situations that may or may not represent reality (for a normal wheel) in a meaningful fashion.

1) Tension variations caused by length changes in spokes; Supposing that the rim is only displaced laterally and the spokes change tension in a linear elastic fashion. For sake of argument lets use flanges that are 35mm off the centreline connected to the rim by spokes that are 296mm length. Supposing that the rim is displaced laterally by 1mm then the spoke change in length so that they become lengthened by 0.12mm and shortened by 0.117mm respectively.

What tension change does this create in the spokes? Well using 209GPa as the modulus of steel it works out that a 0.5mm extension of a ~300mm long 14g spoke causes a tension variation of ~112Kgf. Which implies that in this case the spokes get 26.9kg tighter and 26.2kg slacker respectively. This (once the bracing angle of the spokes is allowed for) generates a net lateral force of about 6kg on the rim. Note that if the flanges were spaced half as far apart, the net force would also be about half.

2) Changes in bracing angle. With the same geometry and a 1mm displacement the lateral component of the spoke tension force T is initially given by

(35/296) x T.

Once the rim is displaced the lateral components become (36/296) x T and (34/296) x T respectively giving a net lateral force of (2/296)xT (provided the average T value between pairs of spokes doesn't change much).

Suppose that the tension is 120 Kgf then this generates a lateral force of 0.81kg.

So on the face of it the bracing angle changes might have a much smaller effect vs the changes in spoke length. But the reality is that

a) lateral forces might not cause an extension of the spokes as initially suggested. In extremis one spoke may slacken and the other one might not increase in tension by the amount assumed.

b) at the contact patch there is a near-vertical thrust (due to dead weight + dynamic loading) onto a wheel that is not vertical that may cause both spokes to slacken, but by different amounts.

c) double-butted spokes are a lot more flexible so length changes are accompanied by smaller (about 2/3rds) tension changes

I suspect that in some cases (eg when a strong rider is climbing vigorously out of the saddle) the spokes at the contact patch don't just reduce in tension, they can briefly go completely slack on one side. At this stage they make zero contribution to the stiffness of the wheel. The point at which this occurs varies with the initial tension in the spokes. So wheels that are built tight may be 'stiffer' than those with spokes that are less tight.

It is also worth noting that once some spokes go completely slack, the wheel is liable to (elastically initiated) collapse; only the rim stiffness can save you.

If the rim is displaced by (say) 3mm then the contribution of the angle changes becomes increased and the lateral force on the rim (given that the spoke on one side is all but completely slack) is again proportional to the tension load in the (tight side) spokes, i.e. it may again vary with preload (spoke tension).

If the flanges are closer together then the contribution of angle changes alone to the lateral thrust is approximately the same. But the lateral component of the tension force is smaller anyway, which means that the wheel might be half as stiff if length changes in the spokes mainly cause the initial lateral reaction load.

Note that if the spokes are in fact crossed and braced, tension changes may not occur in the normal way; they may be 'shared' with the braced spokes such that the spokes appear to be more elastic than they are in straight tension.

So it looks as if there are several mechanisms by which the preload on the structure (higher tensions) might contribute to the stiffness of the wheel. The relative contributions are difficult to be exact about, because none of the (simplified) situations examined represents reality perfectly. It might be for very small movements, the number and type of spokes, the stiffness of the rim, and the flange spacing are the primary determinants, but at larger applied loads/ deflections, the preload becomes more important.

One thing is for sure; if the rim is moving +/- 3mm between the brake blocks, and other wheels in the same bike don't, that wheel is very flexible, relatively speaking! I have very little doubt that

a) a similar wheel built with wider spaced flanges (and/or more spokes) would be stiffer and
b) that if the spoke tension is low, the wheel may deflect more than normal, by an unknown amount, that will vary with the build of the wheel.

cheers

[[XAP]Bob]

I’d be interested to see the worked out example, because it all depends on the relative magnitude of source (2). My gut instinct – often wrong! – says it must be tiny over the necessarily small deflections that may be permitted when using rim brakes, and that even if it isn’t, there’s not much wheel-to-wheel variability within the range of preloads available to us in functional wheels.

Here’s a thought arising from the small deflection that I observe with my XR2 rims, even with a 32-spoke SONdelux wheel (though it must be said that I’m barely 65 kg). The lateral force on the wheel occurs at the contact patch. For that to displace the rim in the other direction at the brakes implies the rim is transmitting this force. So paradoxically, stiffer rims may show more deflection at the brake, given a few givens (as Brucey likes to say). They may do this even as they deflect less at the contact patch (not that you’ll notice that).

Perhaps I’ve overlooked something obvious.

The rim is a continuous arch, kept in compression at all times by the spokes.
It rotates around the axle in all three axes - but the rim isn't going to deform sideways significantly unless it is pushed from top and bottom in the same direction.

[NATURAL ANKLING]

Hi,

I’d be interested to see the worked out example, because it all depends on the relative magnitude of source (2). My gut instinct – often wrong! – says it must be tiny over the necessarily small deflections that may be permitted when using rim brakes, and that even if it isn’t, there’s not much wheel-to-wheel variability within the range of preloads available to us in functional wheels.

Here’s a thought arising from the small deflection that I observe with my XR2 rims, even with a 32-spoke SONdelux wheel (though it must be said that I’m barely 65 kg). The lateral force on the wheel occurs at the contact patch. For that to displace the rim in the other direction at the brakes implies the rim is transmitting this force. So paradoxically, stiffer rims may show more deflection at the brake, given a few givens (as Brucey likes to say). They may do this even as they deflect less at the contact patch (not that you’ll notice that).

Perhaps I’ve overlooked something obvious.

The rim is a continuous arch, kept in compression at all times by the spokes.
It rotates around the axle in all three axes - but the rim isn't going to deform sideways significantly unless it is pushed from top and bottom in the same direction.

Well....maybe if we consider that this rim deflection is not wholly to the rim and its component parts fixing to the hub.
That leaves the bearing, they are not that truly rigid like motorised vehicles, and their design (excluding cartridge) allows to displace the balls (riding up) on the cone and cup.
If we consider the force at bearing which is many times that of the lateral force at rim. (vector / moments etc)

As the rider pushes on pedal and turns the steering which is a reaction to the body sway, the rim deflects and unloads the downward weight of rider and bike (later is smaller) on the bearing, pedal side of bike.
Similarly the opposite side (bike) bearing is seeing / sharing the unloading force of pedal side.
This could easily be mimicking rim and spoke deflection.

Though I discounted the forks they might be adding some percentage however slight.

At a point in increased tension (to practicable gain) in spoke when building, it might not add any discernable rigidity, masking the deflection at bearings.

I don't have the tools you would need to show this, a simple dial test indicator clamped to fork will show bearing deflection.

Simple thumb pressure deflects the rim at brake easily.
The force at road end on rim will be considerably more? But whether this force is transferred equally or not to the brake side is questionable as said further up posts.
My gut says it is not wholly transferred but I cant say what percentage.

Of course I think we agree? That 5 mm is still excessive, this might be largely due to the taboo wheel build component's as said up posts, for me 1.5 mm is best I can get approaching pretzel on a standard touring front wheel.

[Samuel D]

It might be for very small movements, the number and type of spokes, the stiffness of the rim, and the flange spacing are the primary determinants, but at larger applied loads/ deflections, the preload becomes more important.

I’d agree with that, now that I’ve read your analysis and thought about the points raised.

The rim is a continuous arch, kept in compression at all times by the spokes.
It rotates around the axle in all three axes - but the rim isn't going to deform sideways significantly unless it is pushed from top and bottom in the same direction.

As NATURAL ANKLING pointed out, light thumb pressure pushes the rim visibly off centre, so your point is lost on me. Too subtle!

I’m still curious which spokes are in martinn’s problematic wheel, just in case they’re something particularly light.