Just how long can a lubricating oil last? This question is often asked not only in training seminars, but also by end users who utilize oil analysis for condition-based fluid changes.
While it is nearly impossible to answer this question concretely, we can certainly consider various operating conditions and monitor the lubricant properties to help us determine the optimum time for changing out a lubricant before it reaches the point of condemnation.
Lubrication Properties
Standard test slates for routine oil analysis generally include lubricant properties testing. For the most part, these tests include viscosity, additive metals, oxidation, neutralization number and RULER. The methodology used to determine these values can vary between labs; however, the basic monitoring techniques remain the same.
Lubricant properties testing employs the use of aging alarms. These types of alarm limits help determine when a lubricant is nearing the end of its useful life. Establishing aging limits does not require a set number of samples prior to commencement, as required in statistical alarms used in wear debris analysis.
For most industrial lubricating oils and hydraulic fluids, assigning an age-based alarm value to properties is relatively easy. Knowing what the fluid was like at the beginning of its life is important when assigning an age-based alarm. This requires a valid new lubricant sample in order to establish a relevant baseline.
Once baseline values are established, alarm setting and trending can take place. General guidelines for these alarms are shown in Figures 1a and 1b.
1a. Alarm Guidelines
Lubricant Monitoring
As a lubricant ages, we can expect to see changes in the physical characteristic of the oil as compared to those same characteristics in the baseline sample. For example, oil has a natural tendency to oxidize over time. As the oil oxidizes, the acid number tends to increase in a near lock-step fashion. Along with these two telltale identifiers of lubricant age, viscosity also begins to increase over time.
While identify aging lubricant does not require all three of the factors mentioned above, the occurrence of two of these factors carries a high probability of lubricant failure.
This scenario can be seen in Figure 2. Both the acid number and oxidation values have been steadily increasing for two and a half years. Given the alarm value recommendations in Figure 1, we can tell that this fluid is now in a caution state.
1b. Alarm Percentages
Lubricant Change
As mentioned previously, a goal for proper proactive maintenance is to replace oil in a component before it reaches a level of condemnation. The reason for this is simple: Once you wait until the fluid is condemned before scheduling an oil change, it is likely that you are running your machines with fluid that is causing damage.
As time passes and a lubricant begins to degrade, the actual rate of degradation increases. This trend can be spotted in Figure 2. While the fluid in Figure 2 is still usable, albeit in a caution state, we know this is the prime time to put a work order into the CMMS for a fluid change. Doing so at this point will allow ample time for planning and scheduling of this task so that only good oil remains in the machine.
Keep in mind that just because a fluid condition breaks an initial alarm, there is no need to make immediate plans for a changeout. It is important to utilize confirmation tests such as those mentioned here. In oil analysis, a single test result violating that initial caution alarm should always be confirmed with other testing. It is also good practice that if a single test result violates a critical alarm with no confirming data, an additional sample should be taken for lab testing. This helps rule out any unwanted circumstances during the previous testing.
Oil analysis is a powerful tool for condition monitoring. With the right experience and appropriate trending, we can use oil analysis to optimize our lubricant change intervals, thereby optimizing our machine life by keeping "live" lubricant in our systems.
About the Author
