Thursday, December 3, 2009
The persistent check engine light
Last week, we had a Land Rover in our shop with a check engine light. The first step with any warning light is to connect the diagnostic system to the car’s service port to see what its computers have to say. Sometimes the tester points you straight to a repair; other times the results are more ambiguous.
In this case we had a simple fault, a misfire on cylinder #3. Unfortunately, it wasn’t acting up at the moment we tested it. If the car had a misfire code and it was actually missing, diagnosis would be easy. But repairs seldom play out that way. This particular Rover was running just fine but it had that code. It also had brand new Bosch spark plugs fitted. Could it have a defective plug? Maybe. A flaky plug wire was also a possibility. We decided to change the #3 plug and the wire.
The car drove 200 miles before setting the same fault codes. At that point, we figured we had eliminated the plug and the wire, so the next step was the coil pack. We changed that, and the car went 100 miles and set the #3 misfire again.
We were now on our third try with a simple-seeming misfire. What should we do? I considered our plight. The car had a new coil, new wires, and new plugs. There seemed to be three possibilities:
1) The diagnostic equipment might be lying to us as to the nature of the fault.
2) One or more of the new parts might be defective.
3) The car might have some really bizarre problem that will defy resolution a bit longer
In my time as a service manager, I have seen all three situations. In this case, I had already tested the car with two different diagnostic systems, both of which said the same thing. Either they are colluding to trick me, or the car really has a #3 miss. I ruled out the possibility of lying test equipment for now.
I knew the car might have a one-of-a-kind problem, but I hoped it didn’t. The easiest possibility to test was the parts. I decided to start with spark plugs. I replaced the eight new Bosch plugs with a set of genuine Land Rover plugs.
The problem went away. I cringed, thinking we swapped all that other stuff and the problem was wrong plugs, all the time. We had started with plugs, but I had simply replaced the new Bosch plug with another new Bosch plug. I had not thought to try a different kind of plug.
We relied on the parts catalog, which told us what Bosch plug was right for the car. Obviously, the catalog was wrong for this particular vehicle. It would be easy to say, those people at Bosch just sell incompatible parts! But I have seen many other Land Rovers with those same plugs, and they don’t miss. I know from experience that those Bosch plugs work just fine in many Rovers. Did I get a batch of bad plugs? Or is this car particularly sensitive? We have no way of knowing. All I know is, it’s fixed.
What can we learn from that experience? Some cars are very sensitive to spark plug brand, and this model of Rover may be one of them. Next time I plug plugs in a 2003 Discovery, I will make sure they are genuine Land Rover plugs.
Parts compatibility is a big problem in this industry. Bosch is the world’s largest auto parts maker. They sell millions of spark plugs to car makers every year, and millions more to the aftermarket. Yet their plug did not work in this car.
Yet there are other situations where an aftermarket part is better. For example, Bilstein shocks outperform the stock Rover shocks every time. But how do you know which parts will be better, and which ones worse? There is no substitute for experience, and sometimes it’s hard won.