A friend of mine, Jason Cammisa, just sent me this very interesting writeup on an experiment of his. If you want to do your own analysis, grab a kit from Blackstone Labs and send it in. It is about $20 or so.
Hi everyone, I just got the results of a very interesting experiment I conducted. As you all know, I'm a proponent of extended oil change intervals, but only if I'm sure it's not going to damage my precious machinery. The whole "oil change at 3,000 miles" movement is a marketing farce, and I just proved it. My 2003 525i has a flexible service program. A computer monitors your driving and decides when it's necessary to change oil. There has been a lot of criticism of these systems, because they often recommend oil changes after 12,000 or more miles. BMW's system monitors gallons of fuel consumed. After a pre-programmed number of gallons used, it wants an oil change. The logic here is that all the things that increase fuel consumption (cold starts, high revs, hard acceleration, etc) also reduce oil life. I think all of you know how I drive... especially in a relatively underpowered, heavy car with a small, rev-happy engine. In other words, I beat the shit out of the 5-series. It was always floored, and as soon as it was warm, practically every shift was at 6,000rpm. And yet, the oil change computer refused to acknowledge the need for an oil change after nearly 16,000 miles. Even the service manager at the dealership agreed that it was unusual to go that long, but he commended me as I must be an "economical driver." Um, right. Since it had been more than a year, the dealer did an oil change when I returned the car at lease-end. But not before I whisked away a sample of the oil and sent it to the laboratory for testing. The report is attached. The verdict, for those who don't care to look at all those numbers: After 15,600 of my abuse, there was nothing wrong with the oil. The average oil sample in the "Universal Average" column had 6,400 miles on it... so the wear items (Aluminum, Iron, etc.) should be higher in proportion to the mileage difference. The big wear item on this engine is Iron. Adjusting the universal average to my mileage, one would have expected 27ppm in my oil. With a result of 31, we're not looking at a significantly higher amount of wear. Especially when you consider how I drive. The TBN (or Total Base Number) mentioned in the comments is a measure of how much detergent is left in the oil. With 0.5, the detergents are almost used up. But not completely! The bottom line here? This oil is fine, and so is a 15,000 mile oil change -- on this car, with this synthetic oil, with me beating the hell out of it. Your mileage may vary, of course, but I never, ever expected this oil to look this good. I run Mobil One Synthetic on my personal cars, and I have been changing oil at 9,000 miles or one year, whichever comes first. I'll be sending out samples of those oils, too, and I'm guessing that extending the interval to 12,000 miles would be just fine. Perhaps you all should do the same?