How to Effectively Use Oil Analysis

Few fleet managers dispute the value of oil analysis, but there are some who take a sample, send it in to the lab, receive a report, and think they're done. That's a mistake, says Ken Hill, vice president of sales and marketing, Wearcheck.

"The reason you do an oil analysis in the first place is to reduce your owning and operating costs," he says. "But one of the problems we see is that a lot of fleet managers don't integrate the data with their preventive-maintenance programs."

There are reasons for that, according to Drew Nagle, national accounts manager, BP Lubricants. "We are finding it difficult and costly for maintenance managers to go to oil-analysis labs and obtain the data from them in a format they can use," he says. "For example, many oil-analysis labs have a software program or web-site that doesn't necessarily integrate, or interface, with a customer's PM program that is on their mainframe. Their IT departments may not be very comfortable allowing integration, as well."

Fleet managers who used to obtain paper reports are now using websites or software where they can download data and view it in that fashion. "Obviously," says Nagle, "we have evolved from a reactive maintenance environment, which is what we had many years ago (when it broke, we fixed it) to a planned, more preventive maintenance domain or environment."

Titan Laboratories sends e-mail alerts directing customers online when new data is available, says Marina Radoumis, vice president, technical services. "They can go in and actually look at graphics of an engine in addition to just looking at numbers," she says.

Hill says Wearcheck recommends managers use its web tools to periodically review overall data. For instance, if a company submits 100 samples a month and suddenly it sees that 20 of them are abnormal or severe, they can use the oil-analysis information as a road map that leads to the problem.

"All oil analysis starts out with an OEM threshold number, wear, for instance," he says. "The oil-analysis company can do trend analysis. You may have two pieces of equipment that are identical, but they will wear differently based on how they are operated. The OEM may say that iron in an engine is abnormal if it is 100 ppm. The statistical data for that individual fleet on that make and model might say abnormal is anything above 85. This is conditioned based monitoring, which is strictly for individual pieces of equipment."

There are three types of analysis, says Hill. OEM threshold analysis has a set, or guide, limit; trend analysis of equipment or individual component shows spikes that exceed the norm; and statistical analysis, used when an engine manufacturer says one number is abnormal and fleet statistics show a different number as abnormal.

The type of oil-analysis reports that help fleet professionals effectively maintain their vehicles depends on the manager himself, says Walt Huysman, director of field services, Polaris Labs.

"We have a program called Compass that allows fleet managers to drill down and filter through the information they need, for instance, look at specific types of engines by manufacturer and model," he says. "They can look at specific engines, differentials; and if they plan to purchase a piece of equipment, they can track the maintenance history of a particular brand or type of product. In the long run, it saves them money because there is less maintenance and downtime costs." Also, says Huysman, oil-analysis reports can show fleet managers how to monitor lubricants' performance, which could influence their buying decisions.

Any number of reports can be generated, but Huysman has identified three that most fleet managers find valuable. The first, and most important, he says, is the Severity Summary Report, which tells fleet managers if any machines are showing abnormal wear or contamination. These can be done monthly or during whatever time period the fleet manager wants. "The Summary Report tells you what things you need to look at first," Huysman says. "These are the things that are going to cause problems."

The second report is the Volume Report, which involves the samples submitted each month. They are particularly helpful for customers who have several locations, because the data identifies which locations are submitting samples regularly and which locations are not. Third is the Turn-Around-Time Report that tells the fleet manager how long it took for a sample to be submitted, analyzed and returned.

Although Polaris has three laboratory locations, all data analysis and reporting is done in Indianapolis. This allows Polaris to tie everything together in one data base. No matter where across the country a fleet manager has equipment, and no matter where the oil sample comes from, he receives the same data analysis and reporting format that he is used to seeing.

Titan's Radoumis says reports also assist fleet managers with replacement parts. "When they bring in the equipment for PM they know they are going to need certain parts, say a head gasket, for the repairs, so they can schedule that as well," she says. "They also use the data to monitor the wear particles and the viscosity changes, along with soot levels. They actually have to monitor all three areas to be effective."

Innovative fleet managers, such as Michael Porter, fleet maintenance coordinator, BOC Linde, have fine-tuned these reports to zero in on data that's critical to the fleet.

"We have about 40 locations around the country, and the reports can spot the locations that are allowing their PMs to run too long," Porter says. "I do this in two ways. First, by the mileage they record and, second, by the elevation of contaminants reflected in the reports."

Such information allowed him, in 2002, to test extended oil-drain intervals for his fleet before deciding which interval was best. "When we started out, drain intervals were due every 15,000 miles and overdue at 18,000," he says. "We moved that so that drain intervals were due at 30,000 miles and overdue at 32,000 miles." But this created fuel-economy problems. He found the right combination at 28,000 miles and over due at 30,000.

The reports he receives let him sort data by location, by unit number or model year. "I can put in our own contaminant levels, and then highlight the data with different colors," Porter says. "Everything shows a certain criteria. Chromium, for instance, is highlighted in blue, copper in red and so forth. Once I get that all set up by color, I go through a major sort by unit number or location, or whatever I want. I can pick out those units that we have to take a look at or that we want to trade early before we see engine problems."

But no matter how well-laid the plans, stuff happens. Porter recalls one incident. He installed a new maintenance software package that was highly customized. After it had operated awhile, weird things popped up on the oil-analysis reports. He began receiving comments on the lab reports that the miles on the units didn't match previous samples. This simple comment sent up a red flag, he says, and he started making phone calls that led to a deeper investigation into what was happening.

"The problem wound up being a combination of training issues and a software glitch," he says. "The system wasn't generating the PMs accurately. It wasn't the fault of our mechanics or our locations, but a software glitch."

Nagle says one of the most important elements of oil analysis is trending. "We can't apply standardized wear limits, for example, to a particular machine," he says. "All machines are not standardized on what they do on a day-to-day basis. Maybe there are two identical wheel loaders in a fleet purchased on the same date and only one serial number apart. One may be down in the quarry loading shot rock, which is a very, very harsh environment, and the other may be up top loading aggregate onto light-duty trucks. One machine may use 16 gallons of fuel an hour; the other seven gallons an hour.

"It's not fair to apply the same wear limits to each machine because they have different operating environments."

The next step in the evolution of oil analysis and how it's used by fleet professionals will be the development of what Nagle calls"a thinking maintenance program." Instead of warning fleet managers of a potential problem through an oil-analysis report, BP wants to identify chronic conditions that have been lingering "just below the radar" for a long period of time, Nagle says.

"These are the things that are ticking time bombs," he says. "They are the conditions that are moderately out of normal range and have been for years, but because they are not considered elevated issues, they basically go unscrutinized."

Many failures that occur today are sudden and catastrophic, but they had tell-tale signs long in advance, says Nagle. Moderate levels of lead, copper or iron wear in an engine might come back on a lab report, but the levels aren't significant so they don't necessarily require corrective action. "When that occurs over a long period of time, however, it weakens the component and makes components susceptible to sudden changes in the work-load environment," he says. "That could lead to unexpected failure."

The program that BP developed about a year and a half ago summarizes, or calculates, such issues over a period of time and determines whether the component on the machine is susceptible to failure or has reached a higher level of average repair or maintenance.

"I think this is the next step for the industry," says Nagle, "what I call heuristic (discovering knowledge) maintenance programs."

In the meantime, the message from oil labs, oil companies and innovative fleet managers alike is clear: Use the data you have. As Michael Porter at BOC summed it up, "if there's any negative, it's finding time to use oil analysis to its fullest capability."