The truth is, oil analysis is detective work, plain and simple. Today’s detectives are empowered with a growing bag of tricks but frankly, only a few of these tricks involve traditional “oil analysis.”
Let’s take a closer look at what’s involved in real oil detective work. But before we do, remember that the primary job of the oil analyst is not troubleshooting chronic machine problems but rather the activity of machine health management, that is, maintaining and controlling machine wellness.
I have written extensively on this subject in the past. Proactive maintenance is always where the big payoff is found. Still, even the best proactive maintenance programs can’t completely rid machines of random failures and occasional abnormal wear conditions. It is in these cases when the oil detective earns his keep.
A problem is still a problem whether it is detected early or kept out of sight. Out of sight may be of momentary convenience, but for process-critical machines, problem penalties can grow if not corrected early. Compounding and/or chain-reaction failures can cost millions of dollars or even one’s life. You’ve seen if before - the worse things get, the faster they get worse.
Trick 1
Remember machines tend to fail along old fault lines. Knowledge of the machine’s failure modes and its history is a view into the future and defines the detection strategy.
Trick 2
Hearing the “pin drop” is the key to early detection. Don’t wait until the pin drop transforms into a thunderous wrecking ball. Oil analysis is unique in its ability to detect and amplify weak signals before a machine’s performance becomes impaired.
It’s not enough to know a problem exists; you’ve got to discover its point of origin. Without accurately knowing the source of the problem, corrective action might be misguided or left to expensive trial-and-error. When it comes to localizing your problem, you probably have more tricks at your disposal than you may realize.
Trick 3
Use secondary sampling ports to surgically pinpoint the generation of wear metals and contaminant ingression. For circulating systems, pin-drop concentrations of wear metals can quickly get lost unless samples near the point of origin are obtained.
Trick 4
Elemental families and sequential wear trends can also effectively isolate the source of a machine fault. This requires baseline knowledge of machine metallurgy, including the layer sequence of metals of journal bearing claddings. Additionally, the increasing availability of SEM/EDX spectroscopy by commercial and in-house labs offers the unique ability to characterize metallurgy of individual particles.
Trick 5
When it comes to localizing problems, the use of companion technologies can be powerful. Consider vibration analysis, proximity probes, bearing metal temperature probes, thermal imaging, acoustic emissions, motor current analysis and magnetic chip detectors.
Trick 6
Don’t forget about the debris on the system filter. This is a repository of the machine’s wear history. There are many modern practices for analyzing and interpreting filter debris.
By the time a problem has been detected and localized, the cause of the problem is often discovered as well, but not always. A suspect cause (misalignment, degraded oil, etc.) may need further confirmation or there may be two or more causes working in concert. Knowing the true root cause is vital to prescribing a remedy that works. Slowing the rate of progress may, in many cases, be the best response, enabling complete correction at the next scheduled outage.
Trick 7
For most impending problems, evidence identifying the cause will reside in the oil analysis data. Look at rate-of-change trends of the fluid’s physical and chemical properties. Consider running a few ASTM performance tests (copper corrosion, four-ball, etc.) to see if noncomplying conditions exist. Also look for oil contamination (dirt, water, glycol, etc.).
Trick 8
Defining the wear mode is where the real strength in microscopic wear particle analysis (analytical ferrography) lies. Properly sampled lubricants often contain particles of unique shape and size that characterize how they were created. The skillful eye of a well-trained wear particle microscopist can be valuable.
Trick 9
Investigate if operating conditions changed: Loads, speeds, pressures, duty cycle and operating environment. Did these changes coincide with an increase in wear metal production? Ask questions and study the operating history. A careful machine inspection is also important.
Trick 10
Use pattern recognition (unique data patterns/trends) to confirm a cause by combining oil analysis data with companion technologies, such as those mentioned in
Trick 5.
Once the location and cause are determined, the severity of the condition may still not be clear. Can the unit survive until to the next shutdown or outage? Simple answers to such complex questions are rare. Even analysts with considerable experience can occasionally miss the telltale signs that distinguish an incipient failure condition (early stage) from one more precipitous and severe. Yet the availability and skillful interpretation of good data can usually avert the need to play Russian roulette with machine reliability.
Trick 11
Wear particle size, shape and concentration are important clues to failure severity and component residual life. With few exceptions, the progressive stages of machine wear will correlate to increasing particle size and concentration.
Trick 12
Wear metal production sequence helps identify not only the source of wear, but severity as well. As lower journal bearing layers are reached (tin, then lead, then copper, for instance) the progress of wear is revealed.
Trick 13
Go where the wear particles go, such as the filter, magnetic plugs and sediment. Often these collections are virgin particles that haven’t been pulverized or mechanically reworked by the moving frictional surfaces of the machine.
Trick 14
Companion tests can play an important role. How do the vibration overalls compare to baseline levels? How much has bearing metal temperature changed? Has a thrust bearing pad’s temperature increased several degrees? Was there a shift in gap-voltage from the proximity probe?
As evidenced by the list of tricks, today’s oil detective needs a complement of skills - both art form and science. While training is important, critical skills and knowledge can be acquired in a variety of ways. Often it is not a matter of having knowledge that counts, but rather knowing where and how to find knowledge. This includes knowing how to ask questions and who to ask.
Good oil detectives know where to find subject-matter experts they can go to for help. Keep their resumes and areas of specialty organized in a file. There is also a wealth of information in books, magazine and journal articles. Build your oil analysis and tribology library, including electronic references such as bookmarked Web sites where information can be found.
The best oil detectives develop penetrating sleuth-like skills. Do you have what it takes to uncover clues that expose both problem and root cause? If so, these are hot career skills that are in increasing demand in today’s reliability-conscious maintenance world. Do you know what occurs when there is something of limited supply for which there is increasing demand? It gets more valuable . . . $.