A broken connecting rod poking through the side of the block is a sure sign that something's not quite right inside an engine. So are unusual sounds from within, lower than normal oil pressure, and smoke or strange smells, to name a few. While perfectly valid indicators by themselves, they all have one thing in common: It's too late. Much like relying on so-called idiot lights, by the time the alarm goes off, some damage is already done. Keeping an eye on good gauges is a big step in the right direction, but those still only tell what's going on now, not what's about to happen-and certainly not why.
Having saved an expensive motor with oil analysis, we are firm believers in this early warning system. It may sound high-zoot, expensive, and complicated, but using oil analysis is really quite simple and affordable. You basically grab a bit of the old lube from the engine-or any component that's lubricated by oil that you wish to sample-and mail it to a laboratory. About two weeks later you get a report back that shows what levels of contaminants and evil things were in there. It's a lot like taking a blood test at the doctor's office, except that you'd normally take the oil sample when changing the oil. If you need to replace your blood, you have problems well beyond the scope of this story.
Should the doctor find something unusual with your blood test results you'd get a phone call right away, and that's how a good oil analysis lab operates. Both are aimed at catching a problem early and thereby minimizing the damage, hopefully avoiding catastrophic failure altogether.
Oil analysis isn't meant for finding nuts and bolts in the oil pan-that's what magnets are for. Rather, through physical tests and spectrochemical analysis, the oil sample is searched for over two dozen contaminants, from metals to water. The two services we used for this story, Shell Care and Staveley Services Fluid Analysis, both have a listing of the possible sources for these contaminants on the back of the lab reports. Boron, for example, is used in most antifreezes, but in some cases also as a motor oil additive. That's one reason the laboratory must know what oil and grade was used in what engine, and for how long, in order to make a more meaningful report.
When asked what would be the last thing you'd want to see on the analysis report, both Staveley's Stan Light and Shell Care's Gene Wagenseller agreed that coolant in the oil is the worst for the motor. Coolant attacks tri-metal bearings aggressively; a chemical reaction thickens the oil and the bearings go out quickly. Since the water in the coolant mixture normally boils off, there's often no real clue to be had by checking the dipstick-unless the oil level is suddenly higher and the coolant level is down. It's when finding coolant in the oil that you get a phone call, e-mail, or fax from the lab, to let you know now. That's what saved one of our motors from becoming expensive junk. Instead we had a usable core.
Both also feel that dirt is the second worst. "It's just plain abrasive," says Wagenseller, and can enter the engine through loose connections or bad filters. As four-wheelers, we should probably all take note on that one. After studying our reports more carefully, we ordered up a True Flow filter for our pickup to replace the paper stocker, having seen a very convincing demonstration at a trade show.
Third on the list is raw fuel. Gasoline in particular isn't a good lubricant, and diesel isn't exactly great, either. Wagenseller suggests using heavy stationery to test for fuel dilution. Pour some oil on the paper and if it runs out from the center quickly, it's likely diluted. Obviously, an analysis is a far more precise testing method. We've all seen the resulting black smoke from poorly adjusted carburetors on the trail, and surely some gas made it into the crankcase, too, if the fuel mixture was that rich.
Each of the contaminants listed on the report will say something about the state of the sampled component, and by comparing the results to the listed sources, you can get a darn good idea of what's going on in there.
Motor oil is definitely the most common lubricant to be analyzed, but just about anything with lubricant inside could be. Automatic transmissions are prime candidates, being heat-sensitive and expensive to have rebuilt. We certainly wouldn't bother with analyzing the two quarts of dinosaur oil in a Dana 44, but the $100-plus worth of synthetic gear lube in our tow vehicle's rear axle would be a very good candidate for analysis.
Before thinking that oil analysis may be a good thing, but mostly for old stuff, think again. We sampled an '05 Chevy at its first oil change with 1,385 miles on the odometer. At the dealership, they didn't understand why we wanted to use the free oil change coupon so early-we thought it was a bit late. Far from unexpectedly, the oil wasn't exactly clean, and neither was the Shell Care report card. There were quite high levels of iron, aluminum, lead, copper, silicon (dirt), and sodium, among other things. Shell Care recommended a re-sampling in another 1,000 miles and also wrote that, "Analysis indicates typical conditions for break-in." Our point, exactly. Nobody in their right mind would build a fresh motor and not change the oil after a dyno run or initial break-in before installing it in a vehicle. Why then would it be OK to leave the factory fill in a new vehicle for upwards of 7,000 miles with all the new-motor crud in there?
Today's motor oil is superior to that of years past, and any brand-name oil of the correct API grade for your engine should do a very good job. However, while the new oil in itself is generally clean, the containers it comes in can be a problem, says Shell Care's Gene Wagenseller. Although it's rare that something made it into the system, he suggests using a funnel with a strainer, which is definitely simple and cheap enough insurance. Mixing brands isn't usually a problem, but different viscosities could be, he adds. From an analysis standpoint, it could be trickier to figure out what's really going on since different oils have different additive packages, we'd imagine.
Ironically, the first sample is the least informative since there's no history to compare it to. Engines are all different, so it takes a few samples to get an idea of what is normal and have a trend to study for anomalies. Staveley's reports have graphs that show the past five levels of certain contaminants, making it very easy to spot if something unusual is happening.
Helping to avoid costly problems is the main benefit of oil analysis, but to (hopefully) catch a potential problem early is not the only advantage of taking oil samples. It's not uncommon for the lab results to show that the oil could've stayed in service for quite a bit longer than what you may have used for a change interval. Especially with the more expensive synthetic oils, it definitely makes sense to get the most out of them. Of course, oil analysis could also prove that you're running your oil too long, in which case using oil analysis would still save you money in the long run. After all, oil and filters cost much less than the parts they protect.
Whether using oil analysis to extend oil-change intervals, shorten them, or to help prevent the engine from experiencing catastrophic failure, it usually seems to more than pay for itself.
If nothing else, we found it interesting that even our beater Blazer had only five times more aluminum in its very dirty oil than what the nearest local water district lists for its drinking water. Fortunately, perhaps, we don't know how our well water compares.
Samples are normally taken at a scheduled oil change, directly from the stream when draining the oil, but it can be done though the dipstick tube if sampling between changes is needed. Taking a sample should be done shortly after shutdown when the oil is still plenty warm and well mixed, says Stan Light. It should also be taken about "midstream," so there is usually sufficient time to somewhat leisurely set the drain plug aside and grab the bottle, and maybe even to wait a few seconds before filling it up. Don't worry about the bottle getting oily-just screw the cap on to keep the sample inside from getting contaminated and finish the oil change in normal fashion. Later, wipe the outside of the bottle clean and fill out the paperwork-thoroughly and correctly, or the lab report will not be all it could've been. Next, stick the sample and paperwork into the larger bottle and mail it. Ideally, in the next day or so.
Meant for small- and medium-sized diesel truck fleets rather than us four-wheelers, Shell Lubricants' VideoCheck is an interesting concept nevertheless. By removing an injector nozzle and inserting a lighted fiber optic cable into the cylinder, a technician can view valves, cylinder bores, piston crowns, and such. Everything is photographed with a high-resolution digital camera and can be examined in detail later.
It's obviously a lot quicker and less expensive to use the Shell VideoCheck than to tear the motor apart for a look-see, and in this case a video can literally prevent a blockbuster.
We changed the oil and took a sample from a new pickup at just 1,385 miles. Sure glad we did, because the oil probably should've been changed 1,000 miles earlier. Shell Care attributed the high levels of contaminants to break-in and recommended another sample in 1,000 miles (we never did). As the report shows, there was a lot of debris in there. The truly scary part was that our Jeep-based tractor (which usually gets used oil, has no oil filter, often doesn't get fully warmed up before shutdown, and has lots of oil in the coolant) had better numbers on all but lead, phosphorous, and magnesium. Silicon, for example, was 23 times higher in the pickup's fresh motor-and that's dirt. Yikes.
After a few samplings, you get a much better idea of what the norm is for your vehicle and the trend is a very important part of the analysis. The graphic portion of this report from Staveley shows four samples (from 23,775 to 35,126 miles), and based on iron contents, at 35,000 miles, this 6.2 diesel is still not fully broken in. Or, the rust from sitting for a decade is finally coming out. Judging by the amount of silicon (dirt) in the oil, we should either get a better air filter or stay on the pavement.