DiabloScott

Continuing from the Free Ross thread:

Swaged cranks are two-piece and they somehow hammer the crankarm into the spider.

The 600 Arabesque crank came in both swaged and single-cast versions... probably different model years.







Is this something more than just a really tight press fit? Do they sort of deform the back side like a pop-rivet? Am I missing something?

I had a swaged crank on a Peugeot that came apart - essentially the arm and the spider were two pieces again, and not connected. Probably that was just a cheap crank. But Sheldon's site says that swaged cranks are especially problematic on fixed gears and tandems - how come?

Wilfred Laurier

Probably because both applications create torque on the chainring in both directions (fixies from backpedal force, tandems from the other rider trying to pedal at a different rate), essentially working the swaged joint back and forth until it breaks.

nashvillebill

Yep, swaging is essentially deforming the material, and thus making what's hoped to be a tight press fit (an interference fit). Some minor differences: in a thermal press fit, one piece is heated and/or the other piece cooled (by dry ice for example), they are then put together with relative ease (like a shaft sliding into a hole), then the materials return to room temperature so they contract/expand forming the interference fit. Makes a very stout union. Or the two pieces, both at the same temperature, are merely forced together, for example a shaft into a hole where the shaft is a few thousandths of an inch bigger than the hole, thus binding them--usually this won't be as tight as a thermal press fit. Swaging typically involves a loose or locational fit then one piece is expanded outward or contracted inward with a die to deform the piece and thus make an interference fit --great for hydraulic hoses, or low-force applications, lousy for torque IMHO.

I'm guessing they assemble the spider on the crank, then force a tapered square die into the tapered hole to force the inside out.

smontanaro 

One would think that if the crank arm is properly snugged down that the taper on the spindle would tend to keep everything together. I guess not...

DiabloScott

Thanks - that would also explain the joint's strength variability between manufacturers using different swaging techniques.

I notice that on the bottom photo of the single speed crank, there's also some kind of key between the arm and the spider to keep things from slipping.

When I broke my crank I brought it in to the bike shop and the guy said "that's what happens to swaged cranks - they suck"... I guess he should've said "cheap" swaged cranks.

thumpism 

I've had the spider on a swaged crank come loose and spin on the arm when I was pedaling, but that crank was already damaged when I installed it and I was actually only testing for fit. The swaged crank I replaced that one with has performed admirably ever since. I wouldn't be afraid or embarrassed to have one on a bike.

fietsbob

Or, different price points ..

Swaging is like a Rivet heading, its hammered together, expanding the metal out as it's forced inward.

Brompton finally stopped using a swaged chainring /crank arm , and ships now with a spider crank with a 130 bolt circle



Lower picture they punched out the same chainring for a one piece crank, thats the hole's purpose.

clasher

One can hammer most swaged cranks apart with a ball-peen held on the backside with the chainring supported on a bench vice or other sturdy thing and then striking the ball-peen hammer with a 4lb sledge hammer... highly suggest wearing safety glasses when doing this! I scrap them at the co-op all the time and it seems most of the ones I've broken apart have a lip that is squished down to help hold the steel ring on the aluminum crankarm. When the notches get rounded down the chainring will slip on the crankarm, usually it's just the cheaper ones that do this.

The keyhole on that chainring in the first post is for the notch (apparently called "drive pine hole")on one piece cranks, they probably just use the same rings on both types of cranks to keep costs down.

juvela

One often sees swaged sets where the crank arm has shifted slightly in relation to the spider or outer chainwheel. The shift is always from pedaling loads so the arm has invariably moved clockwise in relation to the spider or outer chainwheel as the case may be.

This Raleigh Super Course Mk II was purchased by Oaddjob2 at the Ann Arbor swap meet of 2014 and exhibits this symptom on its Nervar Sport Cotterless chainset.

https://dl.dropbox.com/s/y55eytn4mar..._0669.jpg?dl=0

artclone

I first learned this technique during a railroad discussion - - this is how train "tires" are put on the wheels. Interference fit has very high shear strength when done right.

noglider 

@DiabloScott, either you're very strong, or you got unlucky. It does happen, but it's rare. But with that said, it's clear enough that forged cranks are better. I'm not even sure they're still making staged cranks. Maybe just on cheap bikes.

DiabloScott

I remember it well - I was first in line at a left turn lane. When I got the green arrow I attempted my usual explosive acceleration so I could get through the intersection and into the bike lane without holding up the cars behind me. Instead, I got half a revolution of the crank and then suddenly both legs were at 6 o'clock and I had no propulsion. I had to dismount and walk through the intersection while the cars behind me laughed and sneered.

It was a Peugeot P8 like this one - clearly not up to my horsepower:

nashvillebill

Yes, a well-done interference fit with steel or iron can be quite permanent (shrink fit with heat and/or dry ice). A poorly done interference fit...like swaging aluminum crank arms...may not last.

verktyg 

There are several different processes.

Many "light rail" pubic transportation cars have aluminum alloy wheels with steel wear rings. The steel rings are shrunk fit by heating them to expand while the aluminum wheels are chilled to shrink. When the rings cool down and the wheels warm up the, joint is somewhat permanent. Worn rings are machined off so that they can be replaced.

Railroad wheels are pressed onto the axles with 500 tons of force. The bore in the wheels is smaller than the diameter of the axles so there is an interference fit. Here's a video showing this being done on an automatic machine.

https://www.youtube.com/watch?v=qd9x_6jy__Q


@nashvillebill gives some good explanations about swagging versus interference fits.


verktyg

Chas.

NJgreyhead

Good stuff, Chas and Nashvillebill. Thanks.

Surprised the Simmons company has that video on Youtube showing their machine operator wearing running shoes instead of something more industrial.

John E

I have always associated swaged spiders with cheapness and unreliability. Swaging was very common on cottered steel cranks, which is what makes the original one-piece Agrati crank/spiders that Capo used a bit special.

oneclick 

I doubt it. That square needs to be *exactly* a certain size and taper (interchangeable parts and all).

And when you look closer at the swage it very much looks as though it's a cylindrical section of the crank that fits through a round hole in the ring and then is forced outwards and over the inner edge of the ring. This makes sense, of course; even if you could push the inside outward and maintain the tolerance of the square - you would not want to force the edge of the bore *outwards*, loading the metal in tension and pruducing cracks, you want instead to squeeze the ring between two faces, one the existing shoulder on the crank, the other the inner face of the swaged part. This loads the metal in compression, and you are likely to get much more surface contact as well; rings are not very thick (even cheap ones like this).

dddd

I'm still curious as to how two crankarms can end up at 6:00, was that a cottered crank?

clasher mentioned the "notches", which I took as referring to the splines on the arm and the spider.
That's right, swaged spiders are connected via a splined interface before getting swaged onto the arm.

Interesting to see that Super Course with the yielding Nervar spider!

I've seen cottered arms where a smooth fillet braze joined the arm to the spider and I had to look at the back side to tell that it was two pieces.

Here's a swaged crankset that I've come to like, it's forged and very light!

oneclick 

I hadn't considered that, I had assumed cotterless which would only make sense with a snapped axle. A cotter falling out would do it.

Wildwood 

I have a swaged Thun (German) Coronado crank, that I like aesthetically - especially for one of several black bikes.
After reading about them I never put it on a drop handlebar bike.
I now have a city build that it is planned, but it would look better on a black bike.
Oh well, I can live with little compromises to minimize spare parts.
Only photo of Thun.

Darth Lefty 

These things are endemic on kids bikes

SurferRosa

The original owner of my Miyata mounted an SR Apex swaged crank. It's really light and I liked running a 50/39.



I guess it just wasn't good looking enough. I mounted a couple different Campy cranksets after that.



It's faster now with this 53/39. It's 3 faster.
​​​​​​​


...