Driveshaft busted

Status
Not open for further replies.
Joined
Jun 2, 2003
Messages
23,591
The driveshaft is made of 12mm solid carbon steel, about 38 cm long. Each end has a threaded hole and a collared (?) transmission gear is bolted to the driveshaft.

Yesterday, while making quite some power uphill (~20° incline), the driveshaft sheared off without warning. I didn't feel the driveshaft "go soft," I simply lost propulsion. So far I have put about 8k miles on the bike. All transmission parts have a lifetime warranty, so getting a replacement part is no issue. I am however wondering about the cause of the driveshaft failure and the probability of it happening again. I suppose an engineer can gleam some info by looking at the images that I posted. Any comments or suggestions?

Are my legs simply creating too much torque?
laugh.gif


sheared1.jpg


sheared2.jpg
 
mori,
that outside ring of corrosion indicates that there's been a crack growing for a long time.

Looks like there's a thread in the driveshaft that may have started a crack...is there salt in the vicinity of your normal riding habits ?
grin.gif


The rest looks like the crack grew until it reached a point where plastic deformation took place (that grey area) followed by failure on that little peak.

Did it feel like a nearly half turn "slipping clutch" as it went ?

Can you degrease the pinion gear, as that looks either dirty, or suss as well.

Out of Interest, what are the gear ratios front and back, pedal radius, and wheel diameter ?
 
The air here at the coast is salty, humidity reaches often +80%, but everything is greased well. The driveshaft and transmission are enclosed in splashwater-resistant housings.

I confirm plastic deformation, because I can see how the shaft was twisted off. It happened so fast that I didn't get the slipping clutch feeling one might suspect.

The gear is just greasy (RL CV-2) and dirty from me disassembling the shaft drive.

The drive's specs:

The cranks (170 mm for increased ground clearance) are of course connected to the crank anxle. On the crank axle is the front axle bevel gear (23 teeth) which meshes with the front transmission bevel gear (pinion, 11 teeth) that is attached to the forward end of the driveshaft. At the other end of the driveshaft is the rear transmission gear (22 teeth), which meshes with the hub gear (29 teeth). The hub gear is mated to an internal Shimano Nexus 8-speed hub. The bike has 26 inch wheels.

crank/pinion = 2.09/1
rear transmission gear/hub = 0.76/1

Nexus-8 ratios unknown, though I could find a spec sheet...

5th gear on the Nexus-8 is direct drive.

In case you wonder why the collar on the gear is so long, it's because that's where the bearing goes:

sheared4.jpg
 
I don't think the rust was an indication of a crack. My high carbon steel knife will get rusty within 5 minutes here if I don't oil it. I think it was just surface rust. I could easily wipe it off:

sheared5.jpg
 
The pinion is worn (besides the cracks near the teeth's edges) and will be replaced. Imagine how tiny that gear really is and what distance it travels. The pinion is the biggest wear item in the whole drive.

Do you still think the driveshaft had a crack? As I said, I'm not sure about the rust. Do the texture and look of the piece indicate a crack? I mean, the center looks like it's been twisted off, but towards the edge the metal seems all smooth.
 
i had the same thing happen, only in a car. i was driving my pops 64 impala home from a car show at daytona beach. the rear axle snapped at the wheel bearing and the wheel, drum, and axle end went flying off. it also broke a brake line. i ended up flipping the car 3 times as i went into the woods.

thing is, the axle looks just like you're drive shaft. a relatively clean cut, with a ring of corrosion around it. we figured the drive shaft has a crack in it and was just begging to let go. finally it did.

so it seems, the same thing has happen to you. perhaps in the future, consider a less viscous grease, perhaps even consider a gear oil for less drag. of course the grease didn't cause this, but if you get another cracky drive shaft as a replacment it might last longer if theres less drag.
 
An area of stress due to the where the splines end .That is where anything made like thats usually will break. The edge is an area of stress.
 
Also insufficient radius where the splined part transitions to the rest of the shaft. I agree about the rust having formed earlier, when the crack was young. Does the shaft experience any bending?

Bikes are built to be as light as possible, so there's not much safety factor.
 
Thanks for the replies, everyone.

I find it strange that if the corrosion is not just recent corrosion, it can be wiped off with a paper towel. Looks like surface rust to me, although it's odd that there was no corrosion present in the center. The brown stuff on the pinion wasn't rust from the pinion (no rust at all), just dirt and dark red/brown grease.

I don't know about improperly heat-treated material, although I've had a quality issue with the gears previously.

Since the break occurred exactly where the pinion gear collar ends, I suspect that this is indeed a stress point. This indicates again that a crack occurred, I guess?

The driveshaft might have been subject to bending a little. The bike has a full suspension frame and the unconventionally designed bottom bracket is designed to let the front gear box swivel freely twist to avoid stress on the driveshaft. By sealing the bottom bracket with flexible RTV I have likely induced some unnecessary stress that may have contributed or even caused the driveshaft failure. After replacing the driveshaft and pinion I will not seal the gear box.

So I got underbuilt parts of not-good-enough quality in a semi-experimental drive, with which I tinkered? Awesome! I guess I can live with it, and will simply replace the driveshaft and the pinion gear from now on as maintenance items every 5k miles.

Would anybody else go with a thinner lube than the RL CV-2 that I'm using? The gearbox cannot be filled with oil.


PS: Papa Bear, how about a belt drive?
hornets_nest.gif
 
Quote:


You weren't "power shifting" were you ??
tongue.gif





Well, going up a 20° incline, standing on the pedals, I do exert a lot of force on the drive -- definitely more than for which the system is designed. It's designed for an around-town bike, but I'm using it in a mountain bike which I do not use like I would a cruiser.
wink.gif
 
mori, if I assume that you weigh 70Kgm the torsional stress on the 12mm section is around 700MPa, which while doable means that they've given you a very small safety factor (50-100%) depending on the grade of steel.

That's to be expected, as it's a bike, safety factor means weight, and the consequence of failure isn't likely to be life threatening.

And yep, I'm positive that there's a fatigue crack that started the ball rolling.

Think of the shaft, supported between the two bearings, with the pinions "climbing" their respective "ring gear". That will introduce a bending moment with the bearings as the fulcrum points...with the shaft rotating between them.

It's almost a direct parallel to the fatigue testing machines that we used at Uni.

http://www.instron.com.au/wa/products/spec_equip/rr.aspx?ref=http://www.google.com.au/search

That's the reason that the crack is so uniform.
 
Status
Not open for further replies.
Back
Top