Thanks, Carl. That really is interesting, and I appreciate your taking the time to explain it.
I have a first-edition copy of _The Bicycle Wheel_ that I bought over 20 years ago. I forgot about the squeezing business.
As I mentioned in another post, I wouldn't suggest that there isn't something going on there. But relieving internal stresses is not it, at least, not from the perspective of metallurgical engineering and science...at least, not that I've ever read, over 20-some years of studying and writing about the subject.
Nevertheless, the sophistication involved in wheelbuilding today is impressive and interesting.
If you buy an inexpensive tension gauge, some weights, some rope, and a ceiling hook, then you can easily duplicate my tests.
But first try this quick test:
The rim appears to deform quite noticeably in a zig-zag when four spokes A-B-C-D are squeezed together as pairs A & C and B & D.
The "basic physics" mistakenly assumed that there must be impressive tension increases to support the conclusion and the angle calculations worked only if the rim and hub stayed fixed relative to each other.
Tape a spoke at a tangent to the rim on the brake surface. Sight along the spoke, and squeeze a few spokes in that rim section.
Watch the spoke wave back and forth as the rim deforms.
As an analogy, the "basic physics" that calculated a 150-lb increase in tension for a 250-lb initial tension spoke for a 30-lb squeeze force was the "basic physics" for a hammock stretched taut between two six-foot-thick oak trees--sitting in the hammock does not draw the two trees appreciably closer.
The real physics, repeatedly measured on various wheels, suggested that the situation is much more like slinging a hammock between two saplings that bend toward each other when you sit on the hammock. Calculations based on the alarming angle at the hammock will mistakenly produce tension rises that would break the rope.
Consider that a 100-lb squeeze force bent spokes at such an impressive angle that they retained a faint bend.
Consider that if a 30-lb squeeze force actually produced a 150-pound tension increase in a 250-lb initial tension spoke, producing a 400-lb tension, then a reasonably strong poster could easily squeeze a spoke so hard that it exceeded its elastic limits. I know of no complaints that posters have permanently stretched their spokes by squeezing them and found that the spoke lost considerable tension when they let go.
Jobst's tests indicated that the popular 15-gauge stainless steel spokes start yielding permanently at around 600 pounds of tension.
Here's a diagram of the setup for my testing:
formatting link
I never found a wheel whose spokes A & C rose more than 65 lbs in tension when I squeezed them together with a 60-lb weight and clamped spokes B & D on the other side to similiar and greater angles.
Again, the calculations that mistakenly assume a fixed distance between the rim and hub predicted a 250% greater tension increase from only half as much squeeze force.
They work as handy parts for Radio Control Model linkage rods. You can put in a Z-bend on one end too. The 2x56 threading usually fits the nylon pushrod linkage too. You can also put in a Z-bend on the unthreaded end and use it as an emergency spoke replacement on those wheels with a shorter spoke too. I have bent the unthreaded end, and put on a strong little magnet on it to use to fish out parts deep inside something before.
Wow, you've really made a study of this. I'm going to re-read your message tomorrow when I have time. However, let me make one point now.
I don't know the term "scragging," but "bulldozing" a spring refers to squashing it until you've driven it as far as you can into the plastic range. Car customizers do it to lower ride height without increasing spring rates (cutting a coil spring results in a surprising increase in stiffness). Car manufacturers also sometimes do it, as in the case of the Ford springs cited in the NIST paper, in order to prevent further plastic deformation in service, which would result in premature sagging of a car's springs.
It isn't about stress relief. If I read the NIST paper correctly, their point is that the tempering (low-temp heat treatment) does the stress-relieving. What they're saying about the bulldozing is that it doesn't *increase* the residual stress, because bulldozing results in a nearly pure torsional strain on the spring material, which doesn't produce antagonistic stresses within the bulk of the material. It's all plastically deformed in one direction, which results in a strain gradient but no residual stresses result from the operation.
Anyway, that's how I interpret the NIST results, based on what they say in the paper and also on standard engineering understandings of stress and strain. I'll look more carefully at your message tomorrow. It deserves a careful reading.
Reverse the rim nut, thread partly onto spoke, fill the cavity with match head material, stuff wooden match into cavity and break off excess. Heat the head with another match. Enjoy the bang. Works best with strike-anywhere matches.
BTW: Point away from self and others until the bang happens.
You won't have to ask: "Where's the Kaboom?"
Good firecracker substitute.
We kids in my neighborhood did that a lot when about 12-13 years old.
Of course, we also experimented with homemade explosives too. It was good we had a big sand pit to play in, and understanding parents/teachers who'd supervise our efforts to make big bangs and stop us when we got into things too dangerous.
Truly!
All hail the humble wire coat hanger, temporary savior of many exhaust systems, bent doors, shaky fenders -- and a whole lot of other things as well! Its utility is surpassed only by duct tape. Maybe.
I make belt buckles and rings of three sets of three twisted coat hanger wires. Looks like a ring of rope when done.
In article , snipped-for-privacy@alphalink.com.au says..
When I broke both arms in a crash some years ago, I found an old spoke was perfect for scratching an itch way down inside the cast. Stiff, straight, long enough, flexible, and the threads would scratch an itch without breaking the skin.
I have no idea what the in-use mechanical properties of an s-bent spoke would be. I don't think existing wheelbuilding machines could handle it, but that's pretty much besides the point since wheelbuilding machines and machine-built wheels as they exist are all about making a product built to sell, rather than appropriate use of resources. Putting these bends in spokes and maintaining an accurate length, necessary for wheelbuilding, would pretty much require a fixture of some sort, and at that point it would probably make more sense just to have a little machine that squashed new heads onto the spokes, as it's done in the factory (as I understand it).
2 more issues with recycling spokes: they also should have gotten their angle adjusted near the nipple in many cases, allowing them the oppurtunity to have accumulated too much stress there to be useful in the future if they weren't. However, even if the angle was adjusted, what it gets set to is quite specific to the wheel, more so than at the hub. Think in terms of the different ERD's and cross patterns a given length of spoke might be used for. Also, stress relief probably also affects built-in stresses in the threads themselves. Between these 2 other issues and the fact that it's only a minority of trashed wheels out there with spokes you'd even want to consider any of this with, and I don't think what you propose is very practical.
Focus on the fact that wheels with reasonable quality components, including hubs that are well-designed and built for indefinite use such as Campy, Shimano (arguable but I think so), Phil, King, etc, quality spokes, and quality rims are amazingly efficient and durable when built and maintained right. Hubs and spokes can last basically forever, if the wheel is built right and the hub is maintained, and rims need only be replaced due to violence, brake track wear in rim brake wheels, and extreme fatigue in hub brake wheels. The hard part of this equation to get right today is rims, because current rims kinda suck in terms of being utiliitarian, economical, and as long-lived as possible, but fairly suitable rims are out there. This approach, wheels that are suitable for serious transportation indefinitely for a fairly low overall cost in terms of resources, makes much more sense to me than going to great lengths to try and turn wheels that were never even meant to be ridden all that much into something they're not. Most of the hubs will be damaged from not being set up and maintained right, and many of the rims will have some damage as well.
It's worse than that. Depending on whether they are left or right spokes from a rear wheel you have at least three elbow bend effects from inside to outside spokes. This is not a good idea for the reasons stated here often... and in "the Bicycle Wheel".
If a 'stress relieved' spoke is truly stress relieved would standard metallography techniques to show this difference? Or stated differently, if there is no difference between a 'stress relieved' spoke & an as manufactured one, would the non- difference shown in the metallograph be proof that there is no difference? ('Stress relieved' in this case meaning 'squeezed spokes', only)
OK, I read your post carefully tonight. I'm probably not following the whole explanation but I don't see any place where stress-relief could occur. You're changing tensions, probably, and there's probably some plastic deformation going on that changes the preloads and load angles -- in which directions, I can only guess.
But the standard engineering definition of residual stress refers to internal tensions and compressions that are due to differential expansion or contraction within the material. For example, when you heat-treat a thick piece of carbon steel, you typically have martensite on the outside and ferrite on the inside. Martensite is less dense than ferrite, which means the outside has expanded. Thus, the ferrite inside is loaded in sheer against the martensite on the outside. It may not bend the material if the heat treatment is uniform, but the skin and the core are both under stress, which is residual stress from the effects of heat treatment.
Tempering (heating to a temperature somewhat below the critical temperature) will relieve some of the stresses by a couple of mechanisms; the one that's relevant here is allowing a slight slippage at grain boundaries due to the elevated temperature (purists will also note that tempering converts a small fraction of the martensite back into ferrite, thus lowering the differential expansion/contraction). Normalizing, which is conducted at higher temperatures, relieves still more stress. "Stress-relieving," done at still higher temperatures and for longer times, relieves still more. Heating above the critical (Curie) temperature and cooling the piece slowly will anneal it, which should eliminate all internal stress if it's done properly.
That's the documented way to stress-relieve steel that has residual stresses. There is a cold method known as vibratory stress relief which remains somewhat controversial but which does seem to relieve at least the more severe stresses, as from welding.
I've not heard of any cold-working method other than sustained high-frequency vibration that relieves internal stresses. But I've been away from the field for a few years and I may have missed something anyway. Just projecting from the standard theory and practice, though, I can't see any way that just squeezing spokes as you describe could actually relieve residual stresses in a spoke.
I'm just trying to describe the theory and practice, not endorse them.
:)
As I understand it, the theory is that cold-bending the spoke elbow and rolling the threads at the nipple end leave residual internal stresses.
The spokes are then often bent a little bit more, in order to make the elbow end lie flatter against the hub flange and to make the nipple end bend at the nipple (if necessary) instead of a longer, gentler bend.
The theory is that your squeeze all the spokes, two pairs at a time, after the wheel is built and thus relieve the residual stresses.
Like you, I'm open to further education.
When I measured actual tension increases in spoke pairs squeezed with known forces on various wheels, the results were greeted with comments including "impossible," "must be cheesy wheels," and similar incredulity. My measurements can be easily checked, but so far no one has announced any different results.
The topic sometimes smacks more of faith than physics.
If you followed that NIST paper that Carl linked to, you'll see what I think is the last word in stress analysis. I don't know if it works on something as thin as a bicycle spoke.
Now I'm reaching into dim areas of memory, but I think that metallographic etching and microscopy will show only a qualitative state of stress, and only to an expert. I don't think it will tell you anything about how much stress there is, nor what effect it would have on performance of the part.
Beyond that, I know of no other methods used in production testing. Maybe there's something in lab metallography.
Again, I've never heard of it being used on something this skinny.
A call to the metallurgy department of a university that covers a lot of mechanical engineering might get you a good answer -- Purdue, Mich. State, Ohio State, etc. That's where I'd go with it now, if I had to write about it.
I didn't recall any mention of eddy current testing, so I searched the archives with hope:
"Broek p.375 refers to eddy current inspection, apparently 3 mm is the smallest detectible 50% of time... and I get the feeling these were surface flaws.... even 9mm cracks were not all reliably detected. On a handlebar, that's proof testing time!"
--Jim Papadopoulos, 1994
formatting link
Alas, a spoke elbow is only 2mm wide, so any residual stresses that take thousands or tens of thousands of miles to cause spoke failure are unlikely to show up on an eddy current crack inspector, unless the resolution has increased several orders of magnitude in the last decade.
Yes, I realize that. I didn't mean to imply that you did.
That's probably true.
I suspect -- and I'm really only guessing -- that they're seating the elbow end, thus reducing the specific stress (the stress per unit area) at that end. If the nipple itself is bent, you'd be transferring a load on the spoke to a load on the nipple, which could be better. But bending the spoke at that end, if that's what they're doing, probably does not increase strength.
I'll hypothesize one other thing: The process may actually result in a more gradual takeup and relief of tensile stress as the wheel rotates. Some process relating to spoke life (which may, or may not, be "fatigue" in the engineering sense of the word) probably benefits from that, because a lot of life-determining events in steel mechanical parts involve the strain-rate sensitivity of steel.
I'm not going to guess beyond that. I'm already over the top with that guess.
Whatever, it's interesting.
Have anybody tried boiling their spokes with wing of bat or eye of newt? You never know.
I've always been under the impression that the stress left in, say, a plate of steel or a spoke after cold bending was called residual stress. Isn't that also residual stress?
I believe that eddy current testing can be used for more that crack detection. Whether or not it can determine the state of the grain in steel, I don't know. It seems that in my memory from the dim past that it can. ET is one of the few NDT methods that I've never been certified in. I have all of ~2 days experience with it :-)
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.