Exploding Spokes

The season of urgent repairs has arrived. People come to town for the weekend. Something goes wrong with the bike. They plead for emergency care.

Last Thursday it was two guys carrying a Specialized Secteur. It was 4:30 p.m. The shop would close in one hour.

"The bike was hanging in storage all winter," said the owner. "When I brought it out, a bunch of spokes were broken in the rear wheel."

I asked about the usual things: could a car have hit it, could a heavy object have fallen against it, do any people have access to the storage area who might be pissed off at him? He dismissed each of those possibilities. That left us with four broken spokes on the drive side, cause unknown.

The owner said that the bike had been hanging by the front wheel from a storage hook, in an unheated space during last winter's phenomenal cold spell in Boston. I had never encountered spoke failure due to cold, but there's a lot I haven't encountered. Still, I would have expected the rim to fail before the spokes. Or, more likely, the rim would contract more than the spokes, relieving tension on them.

I said I could not adequately investigate this mystery in what remained of the day. I would need a few hours to see if this thing would be safe to fly, even with the broken spokes replaced.

The spokes were all in a row. The rim is DT. The spokes are DT. The hub is one of those Specialized-labeled ones that look like they say SIXE if you read them upside down. It's "Axis." They've fixed the font now, but I liked it the old way. I can't remember what I read about those hubs, whether they're DT or Formula or some other generic maker. No matter.

The spokes had not broken at the bend or the threads. They had snapped somewhere in the middle. The edges of the breaks showed no pinching, as they would if a cutter had been used.

A cruise through the Internet turned up some forum posts about bad OEM spokes of Asian origin a few years ago. I had seen failure of various OEM spokes myself, but never DTs. Another forum thread mentioned corrosive environments that weakened spokes. The spokes on this bike did show signs of that. They had a black finish. On close examination you could see cracks in that, and some crusty residue. I sanded away the black around one, carefully and gradually, to see if the crack extended into the metal beneath. In my test area it did not appear to.

I determined the proper length for replacement spokes and installed four. Put the wheel on the stand and tensioned the new spokes. Straight, smooth, round, the wheel looked good. I stepped away to work on something else before I put the bike back together.

BANG! A black spoke snapped, sending the nipple-end across the room.

That's annoying. Glad I didn't have my eyeball down there, sighting along the truing stand calipers.

To me, that was the end of any stop-gap repair. This wheel needs to be rebuilt. To complicate the problem further, the non-drive-side spoke holes in the rim all showed little stress bulges. Nothing had cracked yet, but it made the rim a questionable candidate for the investment of parts and labor to respoke it. I don't know why the non-drive side would show stress that the drive side did not, but this wheel was clearly weird.

BANG! Another spoke fired off. That's it. This thing is dead. I took our spokes back out of it and set the remains by the rest of the bike, with some cloth draped over it as blast matting in case any other spokes decided to blow. It was less likely now that the tension had been reduced. Still, I didn't want any of us to have to dig a spoke fragment out of our leg.

While the front wheel had not misbehaved at all, its spokes had been in the same environment as the rear ones, and there are only 24 of them between the rider and a screaming face plant. Low spoke count wheels are technically strong enough for the stresses of performance riding, but they sacrifice most of their margin of safety by putting more and more load on fewer and fewer parts. When one part fails, the structure deviates much more than it would if more parts were carrying the load to begin with. Are you really going to rip down some exhilarating descent, congratulating yourself on the money you saved by not rebuilding or replacing that front wheel?

Not my decision. I did advise the owner to think about getting both wheels redone or replaced. Because he claimed -- as do they all -- that the bike barely had any miles on it, he was going to take up the matter with the shop where he bought it.

Barely any miles. Not that old. We hear that all the time. Your average ten-year-old bike is between fifteen and twenty years old. A guy brought us his 15-year-old Cannondales on Sunday, and they are 23 years old. Bikes we "just tuned up" show up in the repair records from two years ago.

Sunday started with barely enough work for the number of people on the schedule. We did not get a wave of check-ins, but the couple we did get kept us turning wrenches and looking useful until closing time. Summer has shrunk from the cheerful chaos of the 1990s and early 2000s to a brittle, somewhat desperate couple of months at best, starting on the Fourth of July weekend and not quite making it to Labor Day. But the warm weather can bring us little waves of business any time from now until the fall foliage fades.