Re: I fixed a broken spoke! - Cycling Mob: Road Cycling, Mountain Biking and More

12-23-2006, 02:10 PM

Peter Cole wrote:
> jim beam wrote:
>> Peter Cole wrote:
>>> jim beam wrote:
>>>> Peter Cole wrote:
>>>>> jim beam wrote:
>>>>>> Peter Cole wrote:
>>>>>>> jim beam wrote:
>>>
>>>>>>>> but again, "correcting" prior to hub hole deformation has the
>>>>>>>> spoke out of the load line too! do you dispute the geometry i
>>>>>>>> illustrated?
>>>>>>>
>>>>>>> I saw nothing to dispute, just simple trig.
>>>>>>
>>>>>> right, and that trig supports the spoke angle at which spokes are
>>>>>> manufactured, not arbitrarily bent.
>>>>>
>>>>> I don't draw that conclusion.
>>>>
>>>> what's not clear about it? what conclusion /do/ you draw?
>>>
>>> None.
>>
>> yes you do. you disagree. that means taking a contrary position. i
>> want you to state that and analyze it.
>
> You are conjecturing that the combination of cant (in a less than
> interference fit) and flange deformation will combine to make the spoke
> "factory" angle perfect. You also claim that some flange holes are
> drilled canted, while I measured at least one (common) one that wasn't.
> How could both yield a perfect match to factory angle?

if the flange is canted /and/ drilled correctly, factory spoke angle is
perfect. if the flange is either uncanted or drilled parallel to the
hub axis, then factory spoke angle may indeed not be perfect. solution:
buy better hubs.

>
> Theory, in this case, really is academic. If the spoke, after
> tensioning, has a noticeable bow at the flange, the angle is not perfect
> and should be adjusted. That's all Jobst's book says.

and i'm saying that it's advice that is premature and doesn't consider
the obvious effects that result from hub hole deformation.

>
>
>>>> majority of deformation, yes, but not all. as you say, load/area
>>>> equilibrium depends on load applied. load when "stress relieving"
>>>> therefore needs to exceed that experienced in service or else it
>>>> will occur when the wheel's in use and so send the wheel out of true.
>>>
>>> Yes, perhaps, but we're discussing fatigue.
>>
>> cop out. /you/ were discussing spoke seating. i responded. not
>> you're avoiding the question.
>
> Change in tension after stress relieving may come from plastic
> deformation of the flange or spoke. I think analysis shows it to be more
> likely to be predominated by spoke yield. That's not to say that there's
> no additional flange yield, just that it's not the important factor that
> you make it out to be.

so how are you going to judge which is which? you're using a
supposition about spoke yield to suit your argument, which is unproven,
whereas hub hole yield, which is obvious and easily observed, you're
trying to trivialize.

now, are you going to say that this pic is evidence of misrepresentation?
[Only registered and activated users can see links. ]

>
>
>
>>>>> If the inside radius of the elbow is completely supported by the
>>>>> flange, how can it flex? It is not able to move.
>>>
>>>> now, here we have a spoke whose load is offset from the spoke axis
>>>> by the length of the elbow shank. the elbow will therefore bend on
>>>> loading. period. that bending will "amplify" the stress
>>>> experienced in the skin of the spoke by the leverage exerted. real
>>>> simple. now add cyclic loading, surface quality irregularities,
>>>> microstructure irregularities, and we have fatigue. end of story.
>>>
>>> If the spoke is constrained from moving, it can't move, hence it
>>> can't bend. End of story.
>>
>> that is a serious misunderstanding. where do you think fatigue comes
>> from? by definition, if it fatigues, it's experiencing strain, in
>> this case, bending.
>
> But you haven't explained how a fully supported spoke can bend.

peter, with respect, you're either conceptually blocked /or/ you're
avoiding the question. if there is not a completely straight line
between two points, whatever is joining them will bend. the load point
of a spoke is not axial to the spoke, it's offset by the elbow. if you
don't see that, i'm not sure we can debate it further because you're not
getting it. if you're denying it, that's a whole other ball of wax.

>
>
>>>>>>> Furthermore, if the load brings the spoke near yield, it will
>>>>>>> operate at that point, drastically reducing fatigue life. That's
>>>>>>> why most of your spokes failed at the outside of the bend.
>>>>>>
>>>>>> but "residual stress theory" has /compressive/ residual stress on
>>>>>> the outside of the bend...
>>>>>
>>>>> I'm talking about obtuse angles. If the spoke angle is too obtuse,
>>>>> load will make the angle more acute, causing tensile stress in the
>>>>> outside (elbow) skin. If the (initial) spoke angle was obtuse
>>>>> enough to cause slight yield (bend) during tensioning, then the
>>>>> stress on the skin will remain just below yield during the (short)
>>>>> operational life of the spoke. If the spoke bows out from the
>>>>> flange after tensioning, the angle is too obtuse, and the outside
>>>>> (elbow) with remain under (additional) tension, perhaps near yield.
>>>>
>>>> but how does hub hole distortion on yielding affect that angle?
>>>> [rhetorical]
>>>
>>> It doesn't matter if, at the end of tensioning, the angle is still
>>> too obtuse.
>>
>> no peter, you're dodging again. i demonstrated the effect of spoke
>> seating on the need to bend the spoke. address the question.
>
> No, you conjectured. The proof is in the pudding. If the spoke is too
> obtuse, it won't lay flat on the flange, and should be corrected. If the
> spoke does lay flat, it's still possible that it was over corrected or
> the factory angle was too acute.

so at which point does it require correction? hopefully we now agree
that bending prior to hub hole deformation is premature. and if the hub
hole has deformed and the spoke is full tensioned, you're not going to
see anything other than a correctly aligned spoke.

>
>
>
>>>>> I don't know what you mean by "significant". The Mavic method has
>>>>> you apply about half the final tension, then stress relieve, then
>>>>> fully tension and stress relieve again.
>>>>
>>>> ok, /you/ can apply the leaning method long before mavic say to do so.
>>>
>>> Read it yourself:
>>> <http://home.comcast.net/~carlfogel/download/SpokeTension.pdf>
>>
>> /i/ mailed that to carl. and i have read it. you're avoiding my
>> point. again.
>
> Well, then how am I "applying the Mavic method long before they say to"?
> Perhaps you should reread it to refresh your memory.

no dude, you're twisting my words again. i've said repeatedly that you
can use the "mavic method" [as in pressing on opposite sides of the rim]
even when spokes are slack. and you can. now you're trying to say
that's b.s. because mavic say to start the process after the wheel is
partially tensioned. hence you're avoiding my point.

>
>
>>>>>> i agree with what you say, but jobst says to bend before the
>>>>>> spokes are tensioned, and that's premature.
>>>>>
>>>>> No, he doesn't -- at least not in his book. I don't know why you
>>>>> keep repeating this.
>>>>
>>>> maybe you read a different version of the book to me? it clearly
>>>> has a section on "correcting the spoke line" that's spoke bending.
>>>> and he says to do it after lacing, but before tensioning. and
>>>> because that precedes hub hole deformation, it can't account for
>>>> final spoke angle and is therefore premature. i don't understand
>>>> what's hard to understand about that.
>>>
>>> I have the third edition. In Part II "Building and Repairing Wheels",
>>> under the section "Tensioning the Wheel" (pg. 95), there are six
>>> sub-sections: "Warning", "Taking Out the Slack", "Making Them Tight",
>>> "Spoke Twist", "Tension by Tone", and finally "Correcting the Spoke
>>> Line".
>>>
>>> Does your copy have these sections in a different order?
>>
>> so you /do/ have "correcting the spoke line"! and what does it say
>> there?
>
> I quoted this verbatim in an earlier post, and I have paraphrased it
> several times since (including in this post). You're simply
> misrepresenting what he said.

i stand corrected on sequence. but that's maybe why i asked you for the
quote. and maybe you'll talk with sheldon about that. but now i want
to see how the line is supposed to be corrected on fully tensioned spoke
wire! actually, i don't - i don't think it should be done at all.

>
>
>>
>>>
>>>
>>>>> If the amount of stress relief applied at that step more or less
>>>>> equals final tension, then any additional seating would be
>>>>> identically achieved just by bringing the overall tension up.
>>>>
>>>> except that spoke tension in static loading is not the highest the
>>>> wheel will ever experience in service - as proved by my "round the
>>>> block" experiment.
>>>
>>> When you ride an un-stress relieved wheel "around the block",
>>> "honking up hills", as I said before, you can put lateral loads on
>>> the wheel, which will nominally increase (some) spoke tension.
>>
>> ok, we agree.
>>
>>> Bear in mind that the contact patch is also loaded vertically, so you
>>> are first reducing static tension, then superimposing a (potentially
>>> large) lateral load.
>>
>> ok.
>>
>>> What this does, more than force additional flange seating, is to
>>> yield the spoke line to the load line, essentially "correcting the
>>> spoke line" somewhat.
>>
>> elastically, not plastically.
>
> Not necessarily. Obviously not so if the spokes are close to yield at
> the start (as they might well be if the angle was too acute/obtuse).

elastically.
[Only registered and activated users can see links. ]

>
>
>
>>> That will shorten the spoke path and reduce static tension.
>>
>> i don't buy that. the rim bends too - that pulls the spokes further
>> from the flange. tension increase is related to 1/sin theta.
>
> You don't buy what? That riding loads don't deflect the rim and lower
> spoke tension?

in axial loading, yes. that's why we were talking about lateral loading
- hill honking, remember?

>
>
>>> Your own experiences seem to confirm the need to correct the spoke
>>> line. The outcomes you describe are entirely consistent with this.
>>
>> no, you're trying to twist my words to suit your prejudice.
>>
>>>
>>>
>>>>> The real reason must be correction of the spoke line.
>>>>
>>>> you're trying to twist the facts to fit your preconception.
>>>
>>> I don't think so.
>>
>> but you've just done it again on almost every point in this response!
>
> You're the one who is misrepresenting both Jobst and Mavic. I don't
> think I'm misrepresenting your experiences with slacking and breaking
> spokes -- just trying to explain them to you.

hardly. waving your hands about spokes plastically deforming in
preference to a soft aluminum hub flange, without benefit of evidence,
isn't what i'd call explanation.