On 09/06/2013 08:33 AM, AMuzi wrote:
> On 9/6/2013 9:17 AM, jim beam wrote:
>> On 09/06/2013 02:19 AM, wrote:
>>>>>>
>>>>>> unfortunately, i don't think there's any one magic
>>>>>> bullet on
>>>>>> wheel building books. jobst's book will allow you to
>>>>>> calculate spoke length from first principles [although,
>>>>>> not
>>>>>> allow for the elongation that happens as they tension],
>>>>>> lace, true and tension, but it'll also have you
>>>>>> over-tension, bend spokes unnecessarily, and he gives
>>>>>> you a
>>>>>> very iffy "stress relief" method. you'll also have no
>>>>>> initial clue on how to position a hub so the label aligns
>>>>>> with the valve hole, as per conventional practice.
>>>>>>
>>>>>> gerdt schraner's book will solve the hub label position
>>>>>> problem, and have you stress relieve better, but you'll be
>>>>>> reliant on others for spoke length. his lacing method is
>>>>>> also unnecessarily complicated.
>>>>>>
>>>>>> sheldon brown's online guide is very good on lacing, hub
>>>>>> positioning, but repeats a lot of the garbage that jobst
>>>>>> bullied into him. and his stress relief method blows
>>>>>> chunks.
>>>>>>
>>>>>> if you can find it online, the mavic wheelbuilding
>>>>>> guide is
>>>>>> pretty good. absolutely the best stress relief method. but
>>>>>> it tells you nothing about spoke length calculation,
>>>>>> presumably because if you're rebuilding a mavic wheel,
>>>>>> you're already using spokes of predetermined length.
>>>>>>
>>>>>> so, "don't read that book in isolation" is my advice. and
>>>>>> if you're thinking you want to understand wheel mechanics
>>>>>> and don't already have a strong understanding of
>>>>>> mechanical
>>>>>> principles and materials, it'll badly **** you up
>>>>>> because a
>>>>>> lot of what he says is just plain wrong.
>>>>>>
>>>>>> as to building wheels for yourself, we've discussed this
>>>>>> before. there are definitely benefits to learning in that
>>>>>> most people don't have access to a good builder.
>>>>>> statistically, that includes you. once you can build
>>>>>> yourself, and you take the time to do it right, which most
>>>>>> shops seldom invest, you can have a wheel that remains
>>>>>> true
>>>>>> and stable almost indefinitely. that's a rare beastie in
>>>>>> "other people built" wheels. my mtb wheels, complete with
>>>>>> unconventional rear spoking pattern*, have remained
>>>>>> completely true despite some fearful abuse. same for my
>>>>>> townie fixie on san fran's pot-holed, cable car tracked
>>>>>> streets.
>>>>>>
>>>>>>
>>>>>> * convention is that you go
>>>>>> trailing-leading-leading-trailing when looking at the four
>>>>>> hub sides, theory being that this allows the spokes with
>>>>>> increasing tension from drive torque to distort the
>>>>>> wheel's
>>>>>> dish less. with a disk braked rear, if you're following
>>>>>> this principle, you need to lace
>>>>>> leading-trailing-leading-trailing because braking reaction
>>>>>> opposes drive torque. it's a little more complicated to
>>>>>> lace, but it's specified in the shimano hub manual.
>>>>>>
>>>>>>
>>>>>
>>>>> Good overview.
>>>>> That rear spoking pattern for disc/drum rears is also
>>>>> the preferered
>>>>> 'old guy' way for fronts, which need no particular bias.
>>>>
>>>
>>>
>>> thanks for the comprehensive wrapup,
>>> took the time to setup thunderbird w eternal september
>>> again so I can
>>> crosspost at will ;-)
>>>
>>> My CXP-33 is laced over-4 whatever that means and the
>>> ceramic open pro
>>> is over-3
>>
>> "three cross" is conventional with an open pro. "4x" on a
>> cxp33 is not something i'd do because it's not got as much
>> socket angle available as the open pro - which is precisely
>> what "4x" requires. i'd therefore expect spokes to be more
>> likely to break at the threads.
>>
>> presumably the builder was looking to make the cxp33 a
>> slightly "softer" ride - the cxp33 is very deep and thus
>> rigid so it can ride "harsh" compared to others. ["4x"
>> spokes are longer and thus theoretically slightly more
>> elastic.] i'd have gone with a lower spoke count and "2x"
>> personally.
>>
>>
>> [another criticism of "the book" - jobst has succeeded in
>> making most people think of "spoke crossing count" as one of
>> the parameters for spoke length calculation. this is
>> somewhat misleading since you can compute a result for
>> 2.7179 "crossings", even though no such thing is possible.
>> the fact is, the "crossing" number is in fact the number of
>> hub hole offsets from the "key" position, and this doesn't
>> have to be an integer.]
>>
>>
>>> not sure how that meshes with
>>> leading-trailing-leading-trailing
>>> and trailing-leading-leading-trailing
>>
>> doesn't for "crossing" count.
>>
>>
>>>
>>> For next year cross competition I wonder if I'd reuse my
>>> roadbike
>>> to get a 26"er or 29"er. Though the latter is a bit too
>>> portly
>>> for competitions, I think
>>>
>>>
>>>> interesting.
>>>>
>>>> in that vein, old guys used to tie and solder. jobst went
>>>> out of his way
>>>> to pour scorn on that practice, but doesn't do it in what
>>>> is an actually
>>>> scientific way, just a pseudo-scientific way - i.e. the
>>>> scientific way
>>>> is "objective, methods, results, conclusions". he just
>>>> leaps from method
>>>> to conclusions, omitting actual results. yet he blithely
>>>> accepts spoke
>>>> interleaving as "stabilizing the spoke structure" with no
>>>> analysis
>>>> whatsoever.
>>>>
>>>> [i know, broken record, but i really can't understand why
>>>> other people
>>>> don't see this stuff when reading - you don't need to be
>>>> an engineer to
>>>> have logical inconsistency slap you in the face.]
>>>
>>> it's the "blond gene". you don't have to understand it,
>>> just accept "as is"
>>
>> cringe.
>>
>>
>
> Not enough information.
>
> 4x 48 spokes builds to an optimal spoke angle. 4x 40h is good, better
> than 3x.
> 4x with 32h is a clusterf**k. 4x 36 is failure prone, far from ideal,
> not recommended. And yet guys do it with 50-50 results, despite our
> admonition.

indeed. so as we know the rim is cxp33, 4x is inappropriate given that
it's either 28 or 32 hole.


--
fact check required