Tuesday, January 17, 2017

Failure Mode

That long silver line on that bar is a crack. Found on MG's personal rig recently.
Failure is a topic that many of us that ride bicycles don't like to speak of. It is "negative", so you just do not speak of failure in terms of your performance. However; there is another kind of failure that we do not speak of much either. Failure of components and frames and forks. It is a very real possibility every time you ride your bicycle.

I was tipped off to a very good article covering a wide array of the topics related to parts failures on bicycles by a Twitter/blog reader contact I have. The article is by "Cyclingtips" and can be found by clicking this link. It is well written and broad based in scope. So, I feel it is a good primer on the subject for any cyclist.

I am going to take a look at this from a mechanic's/rider point of view. The above link does a good job of pointing out a little bit of the following, but like any broad ranging topic, there just wasn't room there to flesh out the subject. Not from a point of view such as mine, which includes being a mechanic for over 20 years.

In terms of parts and frame failures, we see a lot more crash related failures than we do the type of failure that happens from "life span" issues. Generally speaking, you can chalk that up to what the linked article points out where bicycles are sold then never really used. Crashes can constitute a lot of examples. The obvious is where there is something catastrophic, where bodily injury was inflicted, or there was obvious violent contact with something. (Roof mounted bike into a garage door, as an example.) However; I feel the most commonly found failure is due to a crash that did not seem to be a big deal at the time, but resulted in a failed component later.

Things will last only so long, then they die.....
This can be hard to pinpoint because most of the time people file away little biffs and dings in their mental recycle bin never to be retrieved again. However; it can take only one fall to precipitate a crack in a handle bar, as a for instance, that could lead to a "catastrophic failure" later. See MG's handle bar above. He dumped his bike in a corner, then later found this crack. Had he not seen that, the bar could have "given away all at once", or so it would have seemed. The reality was that the damage was already there waiting to fail the rest of the way.

I like to think of these things in terms of something that happened back in my youth when I was a jeweler. We worked on fine jewelry and manufactured custom designs which required lots of polishing to bring the fine metals up to their highest glow and sheen. To clean everything up, we used an ultrasonic cleaner, then a hot, high pressure steam bath. We would use these heavy stainless steel "tweezers" to hold the items as we blasted them with shots of steam. After we were finished, we would nonchalantly toss the tweezers into this heavy glass container we kept on the back drip edge of the sink. One late night, after one more of probably hundreds of thousands of tosses of tweezers into that glass, I sat down after cleaning something up to file it away in its envelope and move on to the next task. It was maybe 3-5 minutes later when the glass exploded into thousands of tiny crumbs.

I like to think of that last tweezers toss as the final crash that "broke the camel's back", as it were. The glass continued doing its job for a few more minutes, then it "just failed". In reality, every toss of those metal tweezers was building up stress in the glass which finally resulted in the glass exploding in spectacular fashion without seeming to have any "input" to make it do so.  Our bicycle parts, frames, and forks are also subject to these stresses that can eventually lead to a part failure, like the wheel I have shown here. That pulled through spoke probably had a crack next to it to begin with before it actually failed completely. Had the owner of that bike seen that crack beforehand, he could have replaced the wheel before it completely failed.

Considering that most riders almost never inspect their bicycles for cracks or loose fasteners, it is a near miracle that we don't have more catastrophic injuries related to "JRA" (Just Riding Along) incidents. Consider also that hundreds of thousands of bicycles get infused into the marketplace every year. The sheer numbers of bicycles waiting to fail must be astronomic in total. That isn't to say that efforts should not be made to improve things, but we are talking about a small number compared to what it could be, in my opinion.

Fat bikes, in particular, have had a rash of fork issues since becoming widely available.
Another thing that has been a noted issue with failure, but not a high frequency thing, is the advent of rigid fork fat bikes. The forces at work which exert stress on different parts of a fat bike are far greater, or different, (or likely both), and have caused forks to be recalled. We've seen some subtle and some not so subtle redesigns of fat bike forks which we have replaced under warranty. Obviously, this will sort itself out, but missing the mark on design is another way we've seen failures on parts.

Even less likely is the rider who refuses to repair, or replace an obviously compromised part. While this is admittedly a smaller fraction of possible failures, I see this far more often than I do the flawed part that just falls apart. At any rate, these types of situations concern me the most. In fact, I will refuse a job, rather than take the risk that something will continue to perform without causing injury or death. A famous example would be the various jobs I have refused because the steel fork was bent backward and the owner refused to replace it. Sure......it is steel, and as one guy said, "I've been riding it this way for 30 years!". I replied, "All the more reason I would not work on that. Your luck is due to run out at any time now.", and I wasn't about to take a bet on that one! He probably is still riding that bike, but you know, the minute I work on it, guess what would happen....

Finally, there is the misuse of parts and bicycles. We see this a lot! Stems and seat posts extended beyond their intended insertion points, quick release levers used incorrectly, trailers loaded with cargoes way in excess of what brake and wheel systems on the bicycles that pull them were rated to. Racks affixed to carbon fiber seat posts, bags installed incorrectly, and the list goes on. It is amazing to me how much of this people get away with without catastrophic failures. So, there are another group of potential failures just waiting to happen.

The point is, "life cycle" product limitations may be something worth pursuing, but don't for a second think that this will solve the potential issues with failures. There are far more modes of failure than that.


Phillip Cowan said...

Plastic forks belong on picnics not bike rides. I always get a few chuckles from my retrogrouchy friends with that one. But seriously I've always felt that a good bike like a good bird gun should outlast it's owner. It's hard to get worked up about the latest plastique du jour from Trekondale when you know that in 20 years it will only be safe to use as a wall hanger in some hipster coffee bar. Before you write me off as just another "steel is real" yokel let me say I'm not opposed to carbon fiber. It's wonderful stuff, and getting better, but it doesn't age well. I know this because where I work we build stuff out of it all the time. Most of our stuff goes in a high radiation environment. In a few weeks or months I get to see what would normally take years. It isn't confidence inspiring. If I had to choose between a 1949 Alex Singer or the latest carbon wonderbike I'm taking the Singer every time.

Zed F. said...

What's your take on the safety of carbon forks? Has anyone studied the actual failure rate? Is Grant Petersen's opinion correct, or based on old data? Thanks!

Guitar Ted said...

@philip Cowan- That's a great comment! Ha! Anyway....

Your experiences sound interesting concerning the carbon/radiation observances. I hadn't given that any thought, but it sounds credible and like something we all should be taking into consideration more seriously. Then again, maybe there is a "service life" designers have in mind that isn't being communicated. How would you like it if you knew that your 10K road racing rocket was only designed to be "safe" for five-ten years of consistent usage due to UV radiation concerns? I don't think that is a very marketable sales line. However; it just might have to come to that.

Also- Did you see the Singer that just came up for sale for 9K? Yikes!

@Jim Mearkle- I have no idea where Mr. Petersen got his take on the carbon forks other than to say that as an industry insider, he may know things that are not shared publicly that informed that opinion. Hard to say anything here that isn't pure speculation on my part.

Carbon forks seem to be very over-engineered, in my opinion, but the story I linked would seem to indicate that it was an aluminum steer tube that failed, and not the carbon itself. That due to a so-called "inclusion" in the metal, which I take to mean a flaw in the metal's grain structure. As the article stated, random test protocol and run of the mill visual inspections would not have uncovered that sort of issue before it reared its ugly head.

As much as we'd like to try, there is never going to be "risk free" riding when it comes to catastrophic failures. Can we try and make things better? Certainly. But to eliminate every instance in the future is not possible in my opinion. I find that, given the enormity of the numbers of bicycles in service already, and being produced now, that we're talking about something difficult to do. Even cataloging failure rates is not always a good data point, as not all failures will be reported.

So, I would hazard to say that carbon forks are actually pretty safe, given the current circumstances.

Mark Ryan said...

Ted - Check out Luescher Teknik on youtube / instagram. He specializes in repairing carbon fibre bikes and is very knowledgable about the material with some history in the aeronautics industry. He also cuts apart a lot of failed components to look inside.

Here is his youtube channel, https://www.youtube.com/channel/UCY9JUMYI54lLOHpb_zbIedQ

Unknown said...

I wash my bikes religiously. They look nice, the components last longer,and I don't have enough fingers to count the things I have found that need fixing... loose spokes, loose bottle cages, that slice in the tire you missed...

That said, I once had a seat post mounted water bottle cage snap right off during a rather bumpy portion of a double century. Obviously a fatigue crack I had missed. Stuff happens.

Unknown said...

I ordered a new bar to replace my aging Cowbell just a few days before seeing MG's pic. I was experiencing buyer's remorse until I saw it.

Part of that decision was based on Salsa's track record with aluminum recalls. How many forks, racks, rims, and anything cages died? I like Cowbells but 4ish years and 8k miles was enough.

MG said...

Good thinking, Rainier. If you think it might need to be replaced, it makes sense to just do it so you don't have the weight of the worry in the back of your mind. I wasn't bummed at all with the lifespan of my cracked Woodchipper. It was a 6-7 year old bar that had thousands of miles (and a number of minor falls) on it. I'm glad I found the crack now though, as this will be the bike I'm on at the Dirty Kanza 200 in June.

Salsa's track record of reliability/recalls isn't bad relative to the rest of the industry. And the reality is that the best riding bicycles in the world are ones that wouldn't perform particularly well in ultimate strength or stiffness tests, so there is a potential payoff for manufacturers willing to push the boundaries, so to speak. That said, I think there are some current bicycles that are both reliable and possess a fantastic ride quality. But nothing lasts forever. If you ride it hard, it'll need to be replaced at some point...

Thanks for the reminder and perspective, Guitar Ted. It's valuable and given our recent experiences (and the fact that most of us are looking forward to epic rides in 2017), it seems particularly timely.

Guitar Ted said...

@MG- Thanks Brother!

tim said...

Would you expect the 6061 salsa bars to have longer service life than the 7071 stuff?