Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

stress relieving stainless steel bicycle spokes? 4

Status
Not open for further replies.

carlfogel

Computer
Feb 29, 2004
4
Hello,

Here's a stainless-steel stress-relief question.

Briefly, can stainless-steel bicycle spokes be usefully stress-relieved by grabbing a pair of spokes already tensioned to 200-400 pounds in a wheel and squeezing them together with your hands?

This procedure is one of the sacred beliefs of bicycle wheel-building, but I can't find any actual testing or data connected with it.

Now for the details.

Typical bicycle spokes are 314 (8/18) stainless steel, about 290mm to 300mm long with an short elbow and flattened button at the hub end and a threaded end for the nipple.

Spokes vary from 1.6mm to 2.0 mm in thickness and are often double-butted. The theory is that thicker ends resist fatigue, while a thinner mid-section is more elastic. (Pre-tensioned spoked wheels are a system that supports loads by elastically losing tension in the lower spokes and effectively standing on them.)

Except for rare crash damage, spokes fail by fatigue at either the nipple/thread junction or at their hub elbow. In well-built wheels, spokes often outlast several rims over 50,000 miles and ten years. (The rims slowly wear out under caliper-pad braking.)

Two major spoke manufacturers are Sapim and DTSwiss:

During wheel-building, the spoke elbow may be bent slightly (less than five degrees) to make it lie flat along the flange of the hub. Here's the prevailing theory underlying stress relief for stainless steel spokes:

"Metals such as steel and aluminum are elastic materials that spring back if they are deformed or bent. If they are bent far enough they take a set, and do not return entirely to their original shape. The stress level at which a metal takes a set is called its elastic limit or yield point. Below the elastic limit the metal works in its elastic stress zone. Above the elastic limit it is forced into its plastic stress zone. Beyond the plastic zone is the
failure stress at which the metal breaks. Brittle metals have little or no plastic region and break shortly beyond the yield point. Bending causes tension and compression on opposite sides of a piece of metal. Both forces cause the
same kind of failure."
--"The Bicycle Wheel," Jobst Brandt, 1981, p. 31-32

[an idealized stress vs. strain graph follows:]

S o o o breaking point
t o
r o
e o
s elastic o
s limit o
o o
o o o
o o o
o
o
o

Strain (deflection)
[figure 15; stress here is tension and strain is elongation]

It's been pointed out that stainless steel does not show this drop-off after any elastic limit, but instead shows a steadily climbing stress-strain curve more like aluminum. Indeed, figure 69 in the same book shows the results of
actual testing of butted and unbutted spokes in 2.0mm and 1.8mm diameters from DT and WheelSmith, which all show this kind of stress-strain curve:

o o o o o
o
o
o
o
o
o
o
o
o
o
o

[figure 69's actual spoke tests show no dip after elastic limit]

The long passage following my signature details the prevailing theory and procedure for relieving residual stresses in spokes. It can be found in the FAQ
section of rec.bicycles.tech at:

Unfortunately, I can't find any author addressing this spoke-squeezing stress-relief procedure other than Jobst Brandt. I'm pig-ignorant and just trying to find out whether non-bicycle metals engineers find it plausible.

Naturally, I'll do my best to answer any questions and scan and provide links to any graphs if it will help. Sorry if this is covered in Engineering for Dummies or is considered inappropriate by any members. (I ended up here because I found a 304 wire stress-relief thread.)

Thanks,

Carl Fogel
carlfogel@comcast.net

Subject: 8c.1 Stress Relieving Spokes
From: Jobst Brandt <jobst.brandt@stanfordalumni.org>
Date: Mon, 29 Nov 1999 17:13:28 PST

> I wonder if "stress-relieving" is entirely correct? I > see it as a yielding/hardening process, in which the > yield load is increased by embedding the spoke elbow in > the hub, bending the elbow to a different angle, etc. > When unloaded from a high load, this area of the spoke > should be more or less elastic.
> So I think the term should be "overloading" > or "hardening" -- any thoughts?

Yes. It appears that the process of stress relieving is obscure to many if not most people, because after seeming to have made it clear, comments like the above surface. Spokes are cold formed from wire that is (at least DT) as hard and work hardened as it can become.

Tensioning does not further harden spokes, there being no plastic deformation. Besides, wire ductility is important in both forming spokes and in use.

The coiled wire from which spokes are made is straightened by running it first between rollers staggered in X and then in Y, the wire moving in the Z direction. Reverse bending acts as a degausser, having ever diminishing excursions that affect ever shallower depths of the wire.

This stress relieves the wire while removing the curl of being shipped in a coil. If it had no curl, releasing its free end on the spool would allow it to uncoil explosively into a huge birds nest.

Wire is cut into suitable lengths, the first operation being to cold form a spoke head onto one end with one axial blow of a die, after which the spoke is cut to a specific length before rolling the thread and bending a 100 degree elbow.

Threads, head, and elbow, contain metal that was elastically deformed (beyond yield) as well as metal that was elastically deformed, each having elastic memory. In these transitions, parts that yielded and ones that did not conflict, each wanting to return to or stay in a different shape. This is why a spoke bent by hand springs back only
partially when released.

On lacing spokes into a wheel, elbows are often additionally bent (brought to yield), thus remaining at or exceeding yield stress during tensioning. Threads also have internal tensile stress besides local compressive stress at the threads. The thread core is already in
tension from the lengthening effect of thread rolling and its stress only increases with tensioning.

Therefore, spokes in a newly built wheel have locations where stress is near yield, some more so than others. Because fatigue endurance of a metal at or near the yield stress is short, cyclic loads in such spokes will cause failures at high stress points. In normal use, a wheel only unloads spokes, but with spokes near yield, even these
stress cycles readily cause fatigue failures. Only the lightest riders on smooth roads might be spared failures with a wheel whose spokes have not been stress relieved.

Stress relieving to relax these high stress points is accomplished by over-stressing them in order to erase their memory. It is not done to bed the spokes into the hub, as is often stated. Bedding-in occurs sufficiently from tension. However, stretching spoke pairs with a strong grasp at midspan, can momentarily increased tension by 50% to 100%. Because spokes are usually tensioned no higher than 1/3 their yield stress, this operation has no effect on the spoke as a whole, affecting only the small high stress zones where spokes are near yield. By stretching them, these zones relax below yield by as much as the overload.

Stress relieving with a light grasp of spoke pairs is worthless, as is bouncing the wheel or bending it in a partially opened drawer. Pressing axially on the hub, while supporting the rim, requires a force larger than is manually possible but is effective for spoking machines (except the left side rear spokes that would collapse the
rim). Another not recommend method, is laying the wheel on the floor and walking on it with tennis shoes, carefully stepping on each pair of crossed spokes. The method works but bends the rim and is difficult to control.

It is STRESS RELIEVING! Even though people insist on calling it pre-stressing or seating-in. The wheel is already prestressed when tensioned.

Jobst Brandt <jbrandt@hplabs.hp.com>

[end of FAQ section]
 
Replies continue below

Recommended for you

"stress relieving" in the metallurgical sense is normally used to refer to releaving of internal stress (by mechanical or thermal means) in the material, that occur because of welding, forming, or heat treating. The stress relieving of spokes as described would just be relieving the external mechanical pre-stress load. The same thing could probably be accomplished by unthreading the threaded end of the spoke a bit after the spokes are seated, and the result would be more predictable.
 
Agreed. What the book means by stress-relieving is 'untwist' I think. I had a go on my own bicycle at squeezing the spokes. Some very satisfying 'pinging' noises as they shifted themselves, and while it's relieving stress in the spokes, I wouldn't call it stress-reliving in an engineering sense. Bit of a word play going on there.

For example, if you bend a spare spoke in your hands and then let go so it returns to it's shape, you're relieving the stress in the spoke you've put in by bending it, so arguably you are stress-relieving it to the dictionary definition. However, you're not 'stress-relieving' it in the way it's normally meant in an engineering sense. I suppose 'unload' would be a better word.


Excessive accuaracy is a sign of poor breeding. -Socrates.
 
Dear GRoberts and Knobhead,

I appreciate the time and trouble that you both took to wade through my question. If I follow you, you both think that the spoke-squeezing serves to release mechanical torison from twisting the spoke nipples by allowing them to rotate slightly.

Such "pinging" is indeed often heard when spokes are squeezed, and spokes do suffer surprisingly from what's called wind-up, turning into faint imitations of barber poles.

Jobst Brandt, however, indicates emphatically that this is not what he means at all and that internal metallic stress-relief is what he believes occurs.

Like both of you, I'm doubtful. I can't find any stress-strain measurements that show any change after such squeezing, nor can I find any formal spoke testing that shows improved fatigue resistance. But spoke-squeezing has become an article of faith among many wheel-builders, so I'm pestering helpful folks like you in hopes of finding out what actually happens.

Thanks again,

Carl Fogel
 
From 'Machinery's Handbook' (the engineer's bible):

'Stress-relieveing: This process reduces internal stresses, caused by machining, cold working or welding, by heating the steel to a temperature below the critical range and holding it there long enough to equalise the temperature throughout the piece.'

By 'critical temperature' it means, roughly speaking, the temperature at which the internal stucture of the material begins to change.




Excessive accuaracy is a sign of poor breeding. -Socrates.
 
There IS a mechanical stress relieving technique known as VSR or Vibration Stress Relief. A company called Meta-lax do it (read about it at but I don't see how squeezing a spoke performs any kind of stress-relieving. VSR involves vibrating the material for quite a long period and requires some fine tuning to match whatever it is you're stress-relieving.

I don't see what Brandt is saying has any effect at all.


Excessive accuaracy is a sign of poor breeding. -Socrates.
 
Paraphrasing Brandt: stretching spoke pairs with a strong grasp at midspan momentarily increases the tension, BUT the tension added only affects the highly stressed zones near yield.

From what I can make of it, what he's calling "erasing the memory" is plastic deformation. Essentially you're exceeding the yield strength of the material in these localized high-stress zones, causing very slight plastic deformation. Due to the short time, and relatively low load applied (even though it's a "strong grasp"), you won't exceed the ulimate strength of the material, which would cause the spoke to break. It is this extremely slight PLASTIC deformation that effectively "relieves" the stress. The rest of the spoke is still elastically deformed (from tensioning), although now that you've plastically deformed the high stress zone, the rest of the spoke isn't as elastically deformed as it was before.

Now, metallurgically, a REAL, bonafide STRESS RELIEF requires imparting some form of energy, be it in the form of thermal energy or vibration, in order to give the atoms enough energy to move slightly more than they normally do. The effect of this increased atomic energy is evenly distributed dislocations. It is dislocations in metal crystals that are the "stress". Cold forming, work hardening, etc. all introduce dislocations. By performing a stress relief, you're evening out the dislocations, and more evenly distributing the high stress imparted from some other process. In contrast, an anneal provides enough energy for atoms within the crystal to realign themselves and eliminate dislocations, effectively softening the material.

I hope this is clear, and if anyone wants to amend or correct me, feel free.
 
Dear RD400guy,

I'm delighted--this is the first time that I've seen something that might explain things.

If I follow you, some of the confusion might arise from your distinction between real stress relief versus what you see as a plastic deformation that would sorta-kinda achieve the same effect in localized areas of high stress that are near yield.

Now I have to scurry off and peer at plastic versus elastic and yield and so forth, but your explanation with the distinction between true stress relief and plastic deformation gives me hope that there's something to be found. I may even be able to see what you're saying what the spoke-squeeze people have written.

Thanks,

Carl Fogel

 
Dear RD400guy,

Here's part of a private email from a materials engineer
about your explanation that I thought you might like to see.

Thanks again,

Carl Fogel

>Carl,
>
>To me, the pertinent bit is:
>
>"It is dislocations in metal crystals that are the
>'stress.' Cold forming, work hardening, etc. all introduce
>dislocations." The spoke-squeezing theory doesn't address
>dislocations, so I lose sight of how the theory is derived.
>
>Anyhow, I concur with RD400guy on the theory. The
>practical questions that I have about spoke-squeezing are
>these:
>
>1. Is any local yielding sufficient to bring a very high
>dislocation density material such as spoke wire to a state
>of metallurgical"relief." Bear in mind that these have
>been drawn from say 500N/mm^2 yield to 1,400N/mm^2 yield.
>Materials theory intimates that further yielding in these
>circumstances is more likely to activate slip bands in the
>material than achieve stress relief.
>
>2. It ignores the other practical sources of spoke
>failure, such as material cleanliness [metallurgical
>cleanliness, not external grime] or surface quality. No
>amount of stress relief, real or imagined, can compensate
>for poor material or poor surface finish.
>
>3. It conflicts with another fatigue-reduction effort
>popular with spoke-squeeze theorists, who advocate the
>plastic deformation [bending] of spoke elbows during wheel-
>building. This flies in the face of the theoretical
>desire to eliminate internal stress--it serves to increase
>dislocation density and also to introduce residual stress
>in the very region in which it is supposed to be critical.
>
>4. Per my copy of "Mechanical Metallurgy" by George
>Ellwood Dieter [isbn 0-07-016893-8--my bible--which you
>should be able to find in a good library, or at least have
>them order it] the chapter on fatigue mentions many
>aspects of this phenomenon, including fatigue under
>combined stresses, cumulative damage, and sequence
>effects, etc. There is only one fleeting reference [page
>415] to the potential to enhance fatigue life in notched
>specimens by overload. But that same paragraph is
>prefaced by the words "the fatigue strength of smooth
>specimens is reduced more than would be predicted by
>Miner's linear damage rule if a few cycles of high stress
>are applied before testing at lower stress"; this is in
>the "cumulative damage" section.
>
>Ultimately, what Brandt calls "stress relief" really isn't
>stress-relief in the metallurgical sense. RD400guy seems
>to say the same thing.
 
I work in the spring making industry and what you are describing is what we call a scrag. When applied to a spring the material is placed in a state of higher stress than normally applied (but lower than yield - so long as they don't mess up), and this allows the material to be mechanically stress relieved. What this causes in the material is that those areas that still have some residual stress have the stress relieved by localised yielding (dislocation movement).

This process is applied for different types of spring material, from stainless steel to Hard drawn spring wire to oil tempered wire. Differnet materials will show different amounts of yield in the same application, some may not even yield at all. This then comes down to the effectiveness of the stress relief that cold coiled springs are put through.

I hope my perspective has helped. CHeers
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor