# Bicycle spoke tension increase?

1. May 8, 2004

### carlfogel

Hello,

Typical bicycle wheels have 36 spokes that are
trued to a final tension of roughly 200 pounds.

If a pair of adjacent spokes are squeezed together
with one hand with a force of roughly 100 pounds,
what does the spoke tension rise to?

I'm asking because this healthy-squeeze operation
is popular among some wheel-building engineers,
who believe that it relieves stresses and is crucial
to preventing fatigue-broken spokes. They may be
right, but data, details, and testing are not forthcoming.

So far, I'm just trying to figure out how high the
tension rises. That is, does the existing 200 pound
tension rise to 250 pounds, 300 pounds, 400 pounds,
or what?

This spoke question may be different from the classic
example of tension in a previously slack horizontal rope
after a weight is added, but I'm not sure whether the
pre-existing 200 pound tension matters.

If it's of any importance, the stainless steel spokes
are about 295 mm long, 2 mm thick, and stretch about
1 mm under 200 pounds of tension. My best estimate
of the angle of deflection under a 100 pound sideways
force is 4 to 5 degrees from the ends of the spoke to
the center.

Thanks,

Carl Fogel

2. May 9, 2004

### Gokul43201

Staff Emeritus
When you squeeze the spokes together, do they get permanently deformed or are you within the elastic limit ?

Purely from the info you have given, and assuming 4-5 deg is correct, there is a rough calculation that can be done. Assuming 295 mm is the length under strain, and calculating the new length after a 5 deg deflection as 295/cos(5) = 296.13 mm, gives an additional increase of about 1.13 mm. That should roughly be another 200 lbs of tension.

Total = about 400 lbs.

I still don't see the point of this.

3. May 9, 2004

### carlfogel

Gokul43201 wrote:

When you squeeze the spokes together, do they get permanently deformed or are you within the elastic limit ?

Purely from the info you have given, and assuming 4-5 deg is correct, there is a rough calculation that can be done. Assuming 295 mm is the length under strain, and calculating the new length after a 5 deg deflection as 295/cos(5) = 296.13 mm, gives an additional increase of about 1.13 mm. That should roughly be another 200 lbs of tension.

Total = about 400 lbs.

I still don't see the point of this.

Dear Gokul,

Thanks for the suggestion of looking at the increase in length.

As for the elastic limit and the broader point, both are murky.

As I understand things, stainless steel exhibits no clear yield
point on a stress-strain curve where the ascending curve
abruptly dips, flattens out, and then resumes ascending
until it flattens out and then begins to fail.

Instead, the stainless steel used in spokes just climbs until
the flatten-out-and-fail phase is reached, much like aluminum.

The spoke-squeezing theory, as far as I can follow it, claims
to relieve internal microscopic stresses that would later encourage
early fatigue in the spoke. Presumably, this involves microscopic
yielding and the metal taking a set beyond the elastic limit,
a permanent deformation.

However, there is no side-to-side setting, no permanent
elongation is claimed, and the people who believe most
deeply in this are remarkably touchy when asked about
details, data, or testing.

Unfortunately, irritability has no effect on spokes one way
or the other, so I can't judge the theory by how nice or
nasty they are.

The original theory was put forward with a stress-strain
graph showing an ordinary steel's sharply defined yield
point and dip. But actual testing of stainless-steel spokes
shows, of course, no such dip, just steady elongation
under steady tension increase.

The amount of increased tension has never, as far as
I know, been calculated openly--it's just been declared
to be just about the right amount by happy coincidence.

Like irritability, coincidence is no grounds for belief or
disagreement. The moon just happens to look about the
same size as the sun.

Some engineers have argued that the spoke-squeezing
is nonsense and that improved spoke life is due to improved
materials that appeared around the same time as the
spoke-squeezing scheme, and to subtle improvements in the
manufacturing and shape of spokes. But they didn't write
the FAQ's on the technical newsgroup where spoke-squeezing
is considered an article of faith.

At this point, I'm agnostic, but inquiring. The widespread
spoke-squeezing trick may improve fatigue resistance,
perhaps because of stress-relief, perhaps for other reasons.

Or it may do nothing at all. The chief expounder of the
spoke-squeezing theory spent considerable time and
testing debunking a previous theory, namely that
wrapping wires around where the spokes cross and
soldering the wire strengthened wheels. The same
kind of data and testing is missing from the spoke-squeeze
theory.

Figuring out the actual increase in tension seems like a
good place to start. I have a spoke tensiometer, but
figuring out a clean experiment turns out to be harder
than it looks.

Thanks again,

Carl Fogel

4. May 9, 2004

### krab

A 100 lb. sideways force is 50 lb. at each end. But the ratio of side force to total force is sin(4 to 5 degrees). This means the total force is 50lb/sin(4 to 5 deg.)= 570 to 720 lbs. So the additional is 370 to 520 lbs.

5. May 9, 2004

### carlfogel

Krab wrote: A 100 lb. sideways force is 50 lb. at each end. But the ratio of side force to total force is sin(4 to 5 degrees). This means the total force is 50lb/sin(4 to 5 deg.)= 570 to 720 lbs. So the additional is 370 to 520 lbs.

Dear Krab,

Thanks--I take it that you'd apply the classic
weight-hanging-from-a-horizontal-rope
angle-equation, with the pre-existing 200-pound
tension being irrelevant?

This makes sense. Unfortunately, it means
a lot of finicky angle measuring. (I was
hoping that someone would pat me on
the head and explain a method that doesn't
rely on measuring such small angles.)

I think that your approach will also square
with Gokul's approach of measuring the
temporary elastic increase in length of the
spoke and then calculating from the ~1mm
elongation per 295 mm length per 200 pounds
original tension.

Your two answers vary, his being around 200
pounds, yours around 370 to 520, but I think
that this variation is entirely due to the rough
figures that I gave you to work with--4 or 5
degrees versus about a millimeter per 295 mm
per 200 pounds.

Time to ponder how to measure small angles or
lengths with greater accuracy.

Thanks again,

Carl Fogel

6. May 11, 2004

### Chi Meson

There is much evidence that the "Parallel spoke squeeze" is the wrong way to "detwist " the spokes. The additional force is indeed as calculated by Krab, and with todays doub;e-butted spokes, it is not a good idea since elongation can occur. There are other methods for detwisting spokes. THe best is to use bladed spokes, this way the twist is obvious.

7. May 11, 2004

### carlfogel

Chi Meson wrote:There is much evidence that the "Parallel spoke squeeze" is the wrong way to "detwist " the spokes. The additional force is indeed as calculated by Krab, and with todays doub;e-butted spokes, it is not a good idea since elongation can occur. There are other methods for detwisting spokes. THe best is to use bladed spokes, this way the twist is obvious.

Dear Chi,

I need to clarify the point that you raise. Spokes do indeed twist as their nipples are turned to increase tension and this twist (or wind-up) is often mechanically released by squeezing the spokes together, riding around the block, or by special wheel-building tools that press the rim in slightly for this purpose. A pinging noise is often noticed. Bladed spokes, as you mention, make any twist painfully obvious.

However, the proponents of spoke-squeezing emphatically and repeatedly state that the squeezing is not to allow the spokes to untwist (a mechanical rotating motion), but to accomplish some kind of metallurgical stress-relief that involves the yield point of the metal of the spoke.

Stainless steel shows no clearly defined yield point and is usually treated as reaching yield when it elongates some tiny percentage of its original length. As far as I can tell, the spoke-squeeze theory claims some microscopic relief of internal stresses that significantly reduces fatigue failure, but which does not signficantly lengthen the spoke.

Here's a link to a typical explanation:

http://yarchive.net/bike/stress_relieve.html

Unfortunately, I can't find any evidence at all concerning this spoke-squeeze scheme in the form of measurements or fatigue-testing.

Thanks for taking the trouble to reply. So far, no one here or in an engineering forum has found the spoke-squeezing scheme of stress relief convincing, but I'd love to find tests or a clear explanation for or against it.

Carl Fogel

8. May 11, 2004

### Chi Meson

I guess I misread the post initially. During the wheel-building process, the squeezing of the parallel spokes, along with the spreading of the crossed spokes (as was explained to me by my "master tech" about 12 years ago) was to slightly bend the spoke at the nipple end and at the hub end.
(he spreading of the crossed spoke was done by jamming a screwdriver into the cross toward the hub). It was emphasized that the spokes should not be under full tension when this is done.

It was explained to me that by creating a slight bend during the building process (before the spoke was under much stress) would cause less strain on the spoke than would have been the case if the spoke was simply tightened. This is because the nipple never really "aims" at the hub hole, (well, with the nice wheels they make these days, they do) so a straight spoke would be subject to a sharper "corner" (right at the tip of the nipple) if the spoke wasn't initially given a proper gentle bend before full tension. The sharper bend would create greater sheer stress and the spoke would be more likely to fail. Ditto usually at the hub end.

My master tech made it abundantly clear that a common mistake was to continue the practice after the wheel was fully tensioned. THis is when the squeezing produces falure inducing stress at the nipple and hub. (We used a homemade rig that put lateral pressure on the rims to detwist, as you said)

9. May 12, 2004

### carlfogel

Dear Chi,

Sorry, but I'm afraid that it's easy to mis-read my long-winded posts.

Using a smooth screwdriver shaft to twist the spokes tighter where they cross is another common method of stress relief (both methods increase spoke tension). It also can bend the spokes into slightly better alignment, as your tech suggested.

But I suspect that your tech is in the minority. The spoke squeezers and their kindred screwdriver twisters both emphasize their belief that this is some kind of yield-point stress-relief and that it is to be performed after each round of spoke-tightening.

Here's a link to an awfully nice bicycle web site, where twisting spokes with a screwdriver is explained as an alternative for this stress-relief:

http://sheldonbrown.com/wheelbuild.html#seating

Near the end is the section on this stress-relieving notion. Actually, Sheldon uses a smooth, rounded bicycle crank arm instead of a screwdriver, but the idea is the same. He quotes Jobst Brandt on the theory.

Again, I'm wondering whether the metal is under stress at a microscopic level and is indeed stress-relieved by the briefly increased and then released extra tension.

Carl Fogel

10. May 13, 2004

### Chi Meson

The people to ask are at Barnett's Bicycle Institute:
www.bbinstitute.com

They have engineers there who are constantly learning the best methods and reasons for any bicycle adjustment.

11. May 16, 2004

### carlfogel

Dear Chi,

Regrettably, mechanic sites like Barnett and Park Tool and spoke companies like Sapim and Swiss DT prefer to say nothing about spoke-squeezing, pro, con, or explanatory, even when asked.

They do address the visible and audible mechanical unwinding of spoke twist, but not the invisible metallurgical stress relief from brief extra tension claimed by spoke-squeezers:

http://www.bbinstitute.com/BM5%20chap%2017.pdf [Broken]

Given how unpleasant some spoke-squeezing proponents can be when their claims are questioned, the policy of silence seems quite sensible. But I'm still curious about the physics of what spoke-squeezing actually does or doesn't do in terms of relieving stress and making spokes more fatigue-resistant.

I do appreciate your suggestion to try Barnett, so thanks again.

Carl Fogel

Last edited by a moderator: May 1, 2017
12. May 17, 2004

### Chi Meson

Well, I've been a load of help, haven't I? :uhh:

I still am under the personal belief that the squeezing of the parallel spokes while under full tension is a bad idea. I've built my own wheels for over ten years now, and the only time I've had a broken spoke was when I reused a set of bladed spokes for a new rim. I knew it was a bad idea, but even these lasted for quite a while before...ping!

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook