Thursday, February 26, 2009

The (Un)Reliability of Bikes

In an earlier post I tallied up the maintenance cost of my bicycle, and concluded that it came to an average of $0.067/mile over the lifetime of the bike. In this post I'd like to address the question as to whether this is reasonable.

How much should one spend for maintenance? It turns out that in the engineering world there is an answer to this question. When I worked as a consulting engineer for the pulp & paper industry we used to estimate annual maintenance costs for any piece of major equipment at between two and five percent of the installed cost. So say you put in a shiny new oxygen delignification system for $40 million; you can expect to pay $800,000 - $2 million annually to keep that baby in top running form.

But much of the installed cost of major equipment is stuff that just sits there: big tanks with tons of expensive steel. What about something with moving parts, like a pump? One estimate for pump maintenance costs is $13-18 per horsepower per year. A typical installed cost for an industrial-scale 100 horsepower pump is around $60,000, based on EPA cost curves. Put these together and you have an annual maintenance cost factor of ... 2-3%. Hmm, pretty much the same.
OK, well maybe vehicles are just more complicated. Let's see what maintenance costs are for a car. In my earlier post I had determined that the cost of maintaining my 2003 Honda Accord was $0.042/mile. This was based on total maintenance cost of $2,086 over 5 years and total mileage of 50,000 miles. With a purchase price of $17,500, the annual maintenance as a percent of the purchase cost is 2.4%. I'm noticing a trend here!

OK, I won't keep you in suspense any longer. When I run the same calculation on my bike, I get 40.5%.

WHAT??  Can that be right? Let's double-check: $975 total maintenance costs in 3.4 years of operation, purchase cost of $707:

     (975/3.4)/707 = 0.405

Yes, it's true: the maintenance cost of a bike is TEN TIMES as high as other moving machinery. This is such a stunning statistic that it's worth digging a bit into the details. Here is my complete list of bicycle maintenance costs, rounded to the nearest dollar:
8-29-05   Chain, fender   47
3-2-06Chain, tuneup86
5-26-06Tuneup, new rear wheel224
9-21-06Chain, tuneup98
3-15-07Chain, disk brake, tuneup128
4-30-07Tube, tire39
4-29-08Chain, tuneup, rear cassette224
variousChains & tubes85

What can we conclude from this? Simply, that even supposedly "quality" bikes are the equivalent of Yugos.

This is actually a fairly apt comparison, not just another cheap shot at the Yugo. The Yugo was a perfectly functional car, so long as you rigorously maintained it. According to Wikipedia,
One critical issue specific to the Yugo was the need for regular replacement of the interference engine's timing belt — every 40,000 miles (64,000 km). In a non-interference engine, timing belt failure does not cause further damage to the engine. In an interference engine, however, timing belt failure disrupts synchronization between pistons and valves, causing them to smash into each other (hence the name interference engine), thus destroying the engine. Though this requirement was stressed in owners' manuals, it was too frequently overlooked by owners.

In the same way, a bicycle will slowly become non-functional (though not in such a catastrophic fashion as the Yugo) if not given regular maintenance. Where a car can go months between lubes, a commuter bike has to be lubed every few weeks. A car's tires last five years; a bike's last one or two. And on, and on, and on. This despite the fact that a bike is much simpler, mechanically, than a car.

But maybe that last point is part of the problem. There are many people who actually enjoy working on their bikes. A bike is simple enough that it doesn't take a shop computer to tune it up. I have several friends with a bicycle stand and a full complement of tools who do all their own maintenance.

I'm not like that.

Every hour I spend working on my bike is an hour lost from reading Cryptonomicon. This weekend I replaced the chain on my bike, only to discover that I had threaded it incorrectly and had to redo it. When I finally managed to unhook the special gold link, a piece of it fell off and disappeared. The next day I found it, rethreaded the chain, found I had done it incorrectly again, unhooked it a second time, and finally got it right on the third try. Say what you want about my mechanical skills — I will happily admit to it all — but I want an industrial-strength bike that just works.

Fortunately, there's a new movement afoot with just this aim in mind. As more and more of us take up bike commuting, our voices are beginning to be heard above the roar of the racers and mountain bikers. Bikes are being produced with chain guards and enclosed hubs that better keep road grit away from the chain. Some even do away with the chain altogether. Disc brakes are an incremental improvement, though they still require too-frequent adjustment. Self-sealing tubes and puncture-resistant tires are easy to find, so flat tires are quite a rarity for me now.

I'm looking forward to the day, maybe not far away, when I plop down good money for a truly industrial-strength commuter bike. Every six months or so I'll get a quick tune-up for $50, and that will be it for maintenance. Let's say that through superior design, materials, and workmanship the maintenance cost of a commuter bike is reduced to 5% of its purchase price. That implies my dream bike would cost $2,000, three times what I paid for my current bike. But it would be worth it. Simple economics shows that I'd get my money back after seven years. If a seven-year payback is good enough for the paper industry (and it was, when I worked in it), then it's good enough for me. And besides, I'll get more reading done.

Wednesday, February 4, 2009

Unwritten Rules of the Road

My son has been learning to drive. It has been an education both for him and for me. For him, of course, everything is new, and he has been soaking up the rules of the road like a sponge. He has surprised me with a few things that you'd think I'd know by now. (For example, did you know that yield signs are not yellow, but are actually red and white? Blew my mind.).

But, as in any learning situation, there are rules that can be written down, and then there are rules that defy codification and can only be learned through experience. The example that came up with us recently occurred at a four-way stop. Here the written rules are clear:
  1. First one to the stop sign gets to go first.
  2. If it's a tie, the driver on the right gets to go first.
Three cars were approaching the stop. The first to arrive was the car directly across from us; we were second; and the car to our right was last (see diagram). So according to the
 rules, we should have proceeded in the same order in which we arrived. But there was a twist:  the first car, shown in green, was turning left in front of us, and the third car (yellow) was signaling a right turn.

So instead of waiting his turn, the yellow car started to go while the green car was still in front of us. My son got steaming mad (he is 16) and honked, incensed that the yellow car had gone out of turn. After he cooled down a bit I explained that it made more sense in this situation for the yellow car to go before us because he had freedom of movement while we were still blocked by the green car.

In other words, this case presented an exception to the written rules. Although the case of a four-way stop is probably small enough in scope that you could, if you wanted, write a comprehensive set of rules that covered every possible arrival sequence and turning pattern, it's more efficient to just lay out the main rules as guidelines, and learn the specifics in practice.

In biking, there are even fewer written rules than there are in driving. One rule is, "bicycles should follow the same traffic laws as cars." But in Seattle there is an explicit exception to this rule that gives bicyclists permission to ride on sidewalks, where cars are prohibited. And there are other exceptions that prohibit bikes (and pedestrians) from going onto freeways. These rules recognize the fact that bicycles are not exactly like cars, and they're not exactly like pedestrians. They are something in between.

Unfortunately, no one has ever really tried to write down what the rules should be, exactly, for bikes. The law, as far as I know (correct me if I'm wrong), has very few bike-specific rules. So once in awhile some misguided cop decides to start enforcing vehicle traffic laws on bikes, which most of the time make no sense. (A full defense of this statement would take another entire column. But anyone who has regularly ridden the infamous"many-driveways" portion of the Burke-Gilman trail in Lake Forest Park will understand how absurd it is to try to enforce the many stop-signs along that stretch). 

I've been riding 4,000 miles a year for many years now, and through experience I have learned many of these unwritten rules. For example, when coming to a legitimate stop-sign such as the one on the Burke-Gilman trail at 65th (unlike the illegitimate ones in Lake Forest Park), the most reasonable thing to do is slow down enough that you could stop if you had to, and then either proceed, if there is no traffic, or, if there is approaching traffic, wait for a clear signal from drivers to either proceed or stop. Any variation from this rule only causes more problems.

Let me take one paragraph to defend this statement. First, let's say you, as a newbie biker, decide to follow traffic laws rigorously and come to a full stop every time. Now if a car happens to be approaching the crosswalk while you are slowing down, chances are it's going to slow down and wait for you; 75% of cars do just that. The driver's expectation is that you will proceed across with dispatch, causing him minimal delay. When you act contrary to his expectations, you increase his delay (and probably his aggravation). Furthermore, you will make him less likely to slow down next time he approaches the crosswalk, surprising the next biker to come along. When it comes to traffic flow, causing surprise is the worst thing you can do. Behaving in a predictable way is the best thing you can do. Only experience can teach you what the "expected" behavior is.

I'm completely comfortable with the ambiguity inherent in the rule, "do what's expected." But there are certain personality types that can't stand such fuzzy logic. (I think traffic cops are disproportionately of that type). All I can say is, get over it. This is life, and life is complicated. Stop acting like a sixteen-year-old.