Re-lace wheel the wrong way?

I recently bought a new bicycle for my wife that was built with sew-ups. It's a NOS bike that's
about 15 years old, and I want to keep the hubs but lace them to clinchers so I unlaced the sew-up
rims. Last night I laced the hubs with fresh spokes and Mavic MA-3s. I laced the rear according to
the steps in Jobst Brandt's book, like I always do.

The problem is the elongation of the spoke holes caused by the first build has the spokes going in
the wrong direction.

Following the directions, I have the hub drive-side down. I keep the in-bound (spoke head facing
outside) spokes in the holes as it was originally laced, and the first spoke goes to the higher hole
on the rim, left of the valve stem. I repeat on the opposite side. Then when it comes time to wrap
the hub so that the first spoke pulls away from valve stem hole, I see that the elongation marks on
the hub are opposite from the way the spokes are oriented. In other words I would have to wrap the
hub the opposite direction, so that the first spoke would angle across the imaginary line between
valve stem hole and rim, if I were to match the existing spoke-hole elongation marks.

Make sense?

What to do?

Not good at these sorts of conceptual puzzles,

Morgan Fletcher Oakland, CA

 

Morgan Fletcher writes:

> I recently bought a new bicycle for my wife that was built with sew-ups. It's a NOS bike that's
> about 15 years old, and I want to keep the hubs but lace them to clinchers so I unlaced the sew-up
> rims. Last night I laced the hubs with fresh spokes and Mavic MA-3s. I laced the rear according to
> the steps in Jobst Brandt's book, as I always do.

> The problem is the elongation of the spoke holes caused by the first build has the spokes going in
> the wrong direction.

In another part of the book the reason for the preferred spoking pattern is explained and not
claimed to be better than the one that your wheel had, at least not that one could measure. If you
have not tightened the spokes yet, you should re-lace it the way the spokes were to avoid risking
flange failure.

If you have tightened the spokes, then new dimples have already been formed and cannot be undone, so
just finish the job and hope for the best. If you are not extra heavy, and don't ride much rough
stuff, the hub should last a long time. Before people thought about failure analysis, wheels were
often re-spoked in a different pattern and no one thought about it when, once in a long while, a hub
failed. It was usually attributed to some other effect. Besides, spoke failures were so common they
masked most other failings.

Jobst Brandt [email protected] Palo Alto CA

 

In article <[email protected]>, Morgan Fletcher <[email protected]> wrote:
>I recently bought a new bicycle for my wife that was built with sew-ups. It's a NOS bike that's
>about 15 years old, and I want to keep the hubs but lace them to clinchers so I unlaced the sew-up
>rims. Last night I laced the hubs with fresh spokes and Mavic MA-3s. I laced the rear according to
>the steps in Jobst Brandt's book, like I always do.
>
>The problem is the elongation of the spoke holes caused by the first build has the spokes going in
>the wrong direction.

Relace it the way it was done before. This will increase the hub's odds of survival. Even
laced "wrong" it won't have a significant effect on the finished product (esp. compared to a
broken flange).

--Paul

 

lace it the opposite way. if you lace it the way it was originally, you'll keep moving the stress
cycle graph towards the point of failure like this:

http://technology.open.ac.uk/materials/mem/mem-ccf1.html

if you lace it the other way, you effectively set the clock back to near zero because the stress is
now oriented at 90 degrees to any cracks that may have nucleated. you should be able to
conceptialize that if you look at how the cracks in this picure have grown in relation to the
applied stress from each spoke and their relative orientations for each hole.

ford.

Morgan Fletcher wrote:
> I recently bought a new bicycle for my wife that was built with sew-ups. It's a NOS bike that's
> about 15 years old, and I want to keep the hubs but lace them to clinchers so I unlaced the sew-up
> rims. Last night I laced the hubs with fresh spokes and Mavic MA-3s. I laced the rear according to
> the steps in Jobst Brandt's book, like I always do.
>
> The problem is the elongation of the spoke holes caused by the first build has the spokes going in
> the wrong direction.
>
> Following the directions, I have the hub drive-side down. I keep the in-bound (spoke head facing
> outside) spokes in the holes as it was originally laced, and the first spoke goes to the higher
> hole on the rim, left of the valve stem. I repeat on the opposite side. Then when it comes time to
> wrap the hub so that the first spoke pulls away from valve stem hole, I see that the elongation
> marks on the hub are opposite from the way the spokes are oriented. In other words I would have to
> wrap the hub the opposite direction, so that the first spoke would angle across the imaginary line
> between valve stem hole and rim, if I were to match the existing spoke-hole elongation marks.
>
> Make sense?
>
> What to do?
>
> Not good at these sorts of conceptual puzzles,
>
> Morgan Fletcher Oakland, CA

 

anonymous snipes:

> Lace it the opposite way. If you lace it the way it was originally, you'll keep moving the stress
> cycle graph towards the point of failure like this:

http://technology.open.ac.uk/materials/mem/mem-ccf1.html

> If you lace it the other way, you effectively set the clock back to near zero because the stress
> is now oriented at 90 degrees to any cracks that may have nucleated. You should be able to
> conceptialize that if you look at how the cracks in this picure have grown in relation to the
> applied stress from each spoke and their relative orientations for each hole.

The dimples caused by spokes occur when the wheel is first tensioned, trued and stress relieved.
Failure comes later from metal fatigue. Lacing the spokes in the opposite direction will cause
additional plastic flow that will increase the number of micro cracks from which separation can
occur. Initial deformation and subsequent crack initiation, if any, does not go away by lacing
spokes in the opposite direction, nor does the likelihood of failure. The 36-hole hub shown was
laced cross-3. This should interest people who think fewer spokes, especially with radial spoking is
advisable.

I prefer to spoke as tangential as possible so that the forces are as far from radial as possible.
This lessens the probability of flange failure and in my case, a Campagnolo hub like the on shown in
the above URL has survived substantially more than 200,000 miles over the years. I am neither a
lightweight nor do I restrict my riding to smooth paved roads. I think this is adequate proof that
the failure shown is not a necessary outcome of long use.

I saved two samples of the same kind of hub that look the same as the one in the picture. I saved
these for people who want to know why they should build a wheel conservatively. It probably doesn't
make much difference in my experience because people interested in unconventional wheels usually
don't ride much or at least keep changing equipment to keep abreast of the latest fad.

Jobst Brandt [email protected] Palo Alto CA

 

bs writes-<< lace it the opposite way. if you lace it the way it was originally, you'll keep moving
the stress cycle graph towards the point of failure like this:

Wouldn't do this-lacing opposite asks for flange failures..

Peter Chisholm Vecchio's Bicicletteria 1833 Pearl St. Boulder, CO, 80302
(303)440-3535 http://www.vecchios.com "Ruote convenzionali costruite eccezionalmente bene"

 

> Lacing the spokes in the opposite direction will cause additional plastic flow that will increase
> the number of micro cracks from which separation can occur. Initial deformation and subsequent
> crack initiation, if any, does not go away by lacing spokes in the opposite direction, nor does
> the likelihood of failure.

are you seriously trying to contend that crack growth is not affected by stress orientation?

if so, you should publish a paper and share this spectaular phenomenon and turn the engineering
world on it's head.

ford

 

the 10 spd ridden here began reconditioning with a m.o. belgian rim and a 'three triangle' tiawan
hub of soft aluminum. not knowing what i was doing( aphenomena that continues) i ran the spokes in a
low torque mode and under heavy loads-the flange wore under spoke flex with elliptical spoke holes.
almost through the hub. so, i 'laced', using the word loosely, backwards over fresh aluminum and
continued on the road. no problems. frankly, the hub material seems so malleable or ductile, sorry
jb, that fracture seeems possible only with several solid hits from a 11lb. hammer. what did reduce
hub wear(besides more spoke torque) was cleaning and lubricating the spoke holes/spoke bends with
wax/teflon lube-instead of cracks in the spokes and hub wear, the hub wear reducted 50% and spoke
breakage(now with DT not generic)stopped! with wear becoming burnishing-the spokes moved rather than
galled and cracked.

 

ford exploder <[email protected]> writes:

> > Lacing the spokes in the opposite direction will cause additional plastic flow that will
> > increase the number of micro cracks from which separation can occur. Initial deformation and
> > subsequent crack initiation, if any, does not go away by lacing spokes in the opposite
> > direction, nor does the likelihood of failure.
>
> are you seriously trying to contend that crack growth is not affected by stress orientation?
>

You are proposing that fatigue crack are definitely initiated after the hub is initially laced.
What Jobst is saying (and the empirical evidence discussed in this forum seems to support) is that
you are much more likely to generate cracks in the first place by relacing the opposite way.
Apparently, hubs rarely fail this way except for radial lacing and relacing as described in this
thread. Most structures never initiate fatigue cracks in the first place, and I suspect the same is
true for hubs.

Dave Korzekwa

 

in article [email protected], [email protected] wrote:

> The dimples caused by spokes occur when the wheel is first tensioned, trued and stress relieved.
> Failure comes later from metal fatigue. Lacing the spokes in the opposite direction will cause
> additional plastic flow that will increase the number of micro cracks from which separation can
> occur. Initial deformation and subsequent crack initiation, if any, does not go away by lacing
> spokes in the opposite direction, nor does the likelihood of failure.

I have a relatively new bike (about 2500km) that came with a radially spoked front wheel (on a
Shimano 105 hub). Would rebuilding it cross-3 at this stage be more likely to cause problems
than cure any?

-- Shane Stanley

 

Shane Stanley writes:

>> The dimples caused by spokes occur when the wheel is first tensioned, trued and stress relieved.
>> Failure comes later from metal fatigue. Lacing the spokes in the opposite direction will cause
>> additional plastic flow that will increase the number of micro cracks from which separation can
>> occur. Initial deformation and subsequent crack initiation, if any, does not go away by lacing
>> spokes in the opposite direction, nor does the likelihood of failure.

> I have a relatively new bike (about 2500km) that came with a radially spoked front wheel (on a
> Shimano 105 hub). Would rebuilding it cross-3 at this stage be more likely to cause problems than
> cure any?

As I mentioned in an earlier response to this subject, lighter riders that do not blast down rough
roads most likely won't have any failures. Some hubs are specifically designed to withstand radial
spoking so that is also a consideration. No one blanket statement applies to all cases or even a
few, but it is something to consider.

Jobst Brandt [email protected] Palo Alto CA

 

<[email protected]> wrote:

>
> > Lace it the opposite way. If you lace it the way it was originally, you'll keep moving the
> > stress cycle graph towards the point of failure like this:
>
> http://technology.open.ac.uk/materials/mem/mem-ccf1.html ... The dimples caused by spokes occur
> when the wheel is first tensioned, trued and stress relieved. Failure comes later from metal
> fatigue. ... The 36-hole hub shown was laced cross-3. This should interest people who think fewer
> spokes, especially with radial spoking is advisable.
>
> I prefer to spoke as tangential as possible so that the forces are as far from radial as possible.
> This lessens the probability of flange failure and in my case, a Campagnolo hub like the on shown
> in the above URL has survived substantially more than 200,000 miles over the years.

Michel Gagnon asks: One technical question. What if we were to push the design a little further?
Would a 36-hole wheel laced cross-4 would be even stronger than a cross-3? I know it would be "past"
the tangential alignment and that there might be problems in trying to get the spokes in the holes,
but if it's possible to solve the problem (with a large flange, 1975-style, for example), what would
be the results?

Regards,

Michel Gagnon

 

Michel Gagnon writes:

>>> Lace it the opposite way. If you lace it the way it was originally, you'll keep moving the
>>> stress cycle graph towards the point of failure like this:

http://technology.open.ac.uk/materials/mem/mem-ccf1.html

>> The dimples caused by spokes occur when the wheel is first tensioned, trued and stress relieved.
>> Failure comes later from metal fatigue. ...The 36-hole hub shown was laced cross-3. This should
>> interest people who think fewer spokes, especially with radial spoking is advisable.

>> I prefer to spoke as tangential as possible so that the forces are as far from radial as
>> possible. This lessens the probability of flange failure and in my case, a Campagnolo hub like
>> the on shown in the above URL has survived substantially more than 200,000 miles over the years.

> One technical question. What if we were to push the design a little further? Would a 36-hole wheel
> laced cross-4 would be even stronger than a cross-3? I know it would be "past" the tangential
> alignment and that there might be problems in trying to get the spokes in the holes, but if it's
> possible to solve the problem (with a large flange, 1975-style, for example), what would be the
> results?

The maximum cross number is roughly INT(N/9) or x4 for a 36 spoke wheel. This doesn't work for the
smallest front wheel flanges but it works well for the hub shown in the URL above. The result is
that pairs of spokes in adjacent holes pull toward each other and that they do not pull outward on a
bridge-like periphery of the flange. The periphery of the flange is not stressed in bending.

Jobst Brandt [email protected] Palo Alto CA