Often the first clue is that something doesn't work as the customer expects, and just as often there's no customer complaint at all, just a glowing MIL. If we're lucky, there will be fault codes. While those codes could lead down many paths, at least they give us a place to start.
Vehicles are now more complicated than mechanics in previous generations ever imagined. The cables, rods and vacuum lines that controlled them have been replaced with sensors, pulse-width-modulated valves, multiple control units and a data communications network to link them all together. Fortunately, we also have a large arsenal of sophisticated diagnostic equipment. And while the learning curve is steep, we also have the world's best trainers to show us how to use it.
Those of us who straddle both technological generations have a real advantage in that we can draw on both sets of diagnostic skills. But if the goal is to find the real source of the problem without just hanging new parts, a problem I'm helping a friend with right now has convinced me that diagnostics is much easier today.
The engine is simplicity itself. It's a motorcycle with two cylinders and two carburetors, and while it has electronic ignition (with fixed timing), the only sensor on the whole engine is for crankshaft position. It starts easily, runs smooth and feels strong up to about 40 mph. But opening the throttle further merely produces that "bwaaaaa" sound that, from past experience, leads me to believe it's not getting enough gas. Compression is dead on spec, and so is ignition timing, and there's a strong blue spark. The plugs have that patina that indicates proper air/fuel mixture, but we dove into the carburetors anyway because there was nowhere else to go.
Naturally, we cleaned everything and replaced all the gaskets. If you've ever worked on constant-velocity motorcycle carburetors, you know the diaphragms are expensive, but we replaced them anyway because they are old and the rubber felt brittle. We looked inside the tank and petcock, changed the fuel filter and measured fuel flow, even though there's no specification. But we found nothing amiss, and everything we did made no difference at all.
I'd love to put a vacuum gauge on it, but there aren't any vacuum taps on the carbs, and special tools are needed to connect a carburetor synchronizer gauge. It doesn't matter though, because there's no vacuum spec. In fact, there are precious few specifications of any kind for these carburetors.
If it was fuel injected or if it had an oxygen sensor or MAP sensor or even just a throttle position sensor, I'd have something to check with a scope. There might even be a computer with a live data stream, and maybe even fault codes. If there was a manufacturer's service information Web site, I might find reliable specifications for details such as jet size or fuel flow.
I grew up with carburetors and am equally comfortable with Quadrajets, twin SUs and a six-pack of dual Webbers. This dumb Keihin isn't going to get the best of me. But it's strange to realize how much easier it can be to troubleshoot a complex network of components simply because they're not dumb; they communicate. Kids today really have it easy.
About the Author
