Monitoring a lubricant’s condition is essential to keep machinery healthy and prevent costly failures. The use of tribology to gather such data could be more effective than traditional methods. Ameneh Schneider and Mathias Woydt share their research into tribological condition monitoring for the first time with Lubes’n’Greases.

Many changes in oil chemistry can affect the appropriate ability or tribological function of a lubricant – for example, if crucial additives have been depleted and protection against wear and scuffing is lost. 

 What follows is a presentation of methodologies to include tribological quantities to establish a fully functional profile of an operating lubricant. The aim is to guarantee safe and reliable operation and visualize the point of safe retention of tribological performance. 

 In addition to the development of lubricant formulations, it is desirable to determine the lowest effective additive concentration, or the concentration at which failure is initiated or excessive wear starts, and consequently the rate of additive depletion during service.

Setting Up

The operating conditions of industrial oils and even engine oils are broad, but generally are subject to fluctuating temperatures, loads, velocities, environmental gases and contamination. Used oil samples contain impurities such as wear particles, water and degraded molecules from additives and fuels. Consequently, the tribological stability of used oil is different from fresh samples. 

There are two approaches for analyzing the condition of used oil. The first is to gather chemical and viscometric data, which can indicate a stable evolution of the selected quantities but can lead to no actionable requirements. The second, and where the focus of this article lies, is adding tribological test data to the analysis program to produce a broader decision base, where, in defined time intervals, chemical and physical values are determined. Incorporating tribological data in condition can reveal the beginning of changes in friction, wear and protection against seizure or even first observations reported from operations. In this case, shorter tribological test times are sufficient, as stated in the standardized test methods. 

The challenge in the work described here was to develop methodologies to differentiate the tribological responses of used oils. A second challenge was to have reproducible values. Another important parameter in the later stage of the method’s application as a tool for controlling the quality of used oil is the test duration. 

Experimental Scenario

For tribometric evaluation, we selected a globally standardized ball-on-disk configuration in an SRV tribometer. For the upper specimen, we chose a 10 millimeter diameter ball made of 100Cr6 (SAE 52100) steel, sliding against a standard 7.9 mm high SRV disk also made of 100Cr6 steel with a roughness of 0.05 to 0.065 micrometers as the bottom specimen. It was necessary to clean specimens before and after tests in petroleum ether in an ultrasonic bath for five minutes. A tribological SRV test requires only a low oil sample volume of up to 0.3 milliliters.

We considered three gear oils used in wind turbines with high viscosity of ISO VG 320 from different plants for this study. The samples had different service times, but the key data from the oil analysis compared with fresh oil data are shown in Table 1.

The SRV standard friction and wear test parameters with a test time of two hours and high precision were assessed and adapted in this study. (See Table 2 for a summary of the elaborated test conditions.) 

The test durations of two hours, as used in the standard methods, are considered to be too long for the purposes of condition monitoring of used oils in an analytic lab or on site in the plant. For this reason, triplicated tests were run for all four gear oil samples with shorter times of 20 minutes to evaluate the data precision.

The Results

Figure 1 shows an example of the evolution of coefficient of friction over 20 minutes for fresh gear oil. Figure 2 displays the calculated mean values of CoF with their standard deviation. The CoF values in Figure 2 are somewhat comparable for all four gear oils independently of the service time. 

After running the test on the SRV, the wear volumes of ball and disk were determined according to ASTM D7755. (See Figure 3 for an example of wear morphologies on the ball scar and wear track on the disk for fresh oil and the oil from Plant 3.) It can be clearly seen that the lubricant’s wear protection ability has decreased significantly.

After 1,016 hours, the wear volumes of Plant 1 oil samples were similar to the wear performance of fresh oil. (See Figure 4.) Water content was similar for all four samples. (See Table 1.) Interesting results are derived from oil samples Plant 2 and 3. Plant 3 with the highest wear volumes has the lowest P content after 98,816 hours, which indicates an additive depletion and some actions concerning the maintenance should be taken. However, the Plant 2 oil sample with the same phosphorus content as fresh oil (see Table 1), shows a very high level of wear volumes, which may be originated by particles, an increase in total acid number or other factors. This example clearly shows that tribological quantities offer important additional information about the condition of operating gear oils and accurate data gathering cannot be limited only to chemical analysis.   

Figure 4 shows the wear volumes for all four samples with their standard deviations

Hydraulic Oils 

Two hydraulic oil samples – one fresh oil sample and one sample from the plant after 40,000 hours of service – were provided by partners for this study. 

The test parameters for investigating the tribological performance of used hydraulic oils using SRV test equipment were selected based on the results from a screening test campaign. (See Table 3.) Figure 5 compares the evolution of the CoF for fresh oil with oil from machinery after 40,000 hours of service.

There were slightly different evolutions in the CoF values at the beginning, which were negligible at the end of the test. Concerning the values of the wear volume at the ball and the disk (as shown in Figure 5), the used oil shows higher wear volume for the ball and similar wear volumes on the disk. This indicates a beginning drop regarding tribological properties, which cannot be seen from the chemical and viscometric analysis, even the iron content remained unchanged. 

Conventional oil characterization did not show major oil degradation compared with the fresh oil. (Also see Table 3.) Phosphorus and sulfur content, mainly related to extreme pressure and antiwear additives, was slightly reduced over time.

Tribometric test results of fresh and used oil show that this hydraulic oil had retained its quality level, even after 40,000 hours. As wear values increase, even with a high concentration of additive elements, another investigation is recommended into tribological properties with the next oil analysis.

Off the Cliff

Cliff testing aims to identify the induction time, or off-set point (the so-called “cliff“), after which wear and friction increased or failure occur in engine or gear tests. Such post-test explanations for friction and wear increases, as well as failures that occur during engine tests, can be derived from SRV testing of oil samples taken or collected at different engine test times and by correlating these with their friction, wear and extreme pressure data in respect to depleting curves for specific additives or other oil properties. A root-cause analysis was done by plotting SRV data versus any relevant chemical or viscometric properties. The collected oil samples must be fully SRV tested and chemically analyzed. 

As an auxiliary method, cliff testing supports the interpretation of engine tests, as it enhances the values of these expensive procedures. Sequence IIIG-engine-tested oil samples taken in discrete intervals (every 20 hours) were tribologically tested at 150 degrees Celsius using the ASTM D6425 standard test and the evolution of tribological and functional properties versus engine running time was monitored. 

The oil samples were also analyzed (additive depletion, oxidation, viscometrics and dispersancy/detergency). As temperature has a significant effect on a lubricant’s tribological performance, the temperature for the SRV test must correlate with the tested oil or component temperature in the respective engine test. Figure 7 shows the evolutions of coefficients of friction and the morphologies of wear tracks (disk) at test end for oil samples from a sequence IIIG engine test. The friction increased significantly between 40 hours and 60 hours.

Figure 8 plots different chemical and viscometric parameters versus wear coefficients in SRV tests according to D6425. The SRV test data indicate that after 60 hours of sequence IIG testing, the friction, wear, viscosity, iron content and total acid number are the highest. Even after the engine test reached 100 hours, it was obvious that the origin of failure was initiated before 60 hours and with phosphorus content below 600 parts per million.

In Conclusion

Chemical and viscometric analysis of oil samples and quantification of the oxidation, nitration and sulfation of a lubricant allow operators to predict the trend of the remaining oil life and can also reveal improper operating conditions. This is useful for indicating the premature failure of parts (piston seals, for example) or the wrong lubricant for the application. But these analyses do not indicate the end of life of lubricants by functional properties such as friction, wear and adhesive failure. 

By combining the traditional quantities determined in condition monitoring and oil analysis with tribological quantities, root-cause relationships can be established in order to safely identify an upcoming oil drain interval and determine the lowest effective additive concentration or additive concentration at which failure was initiated or excessive wear started in an engine test.

Monitoring the wear volumes of balls and disk are functional indicators of the suitability for purpose of lubricants during service. This additional information is interesting for assessing machine reliability as well as for optimizing lubricant
service life.

Ameneh Schneider is the Europe and Asia sales manager for Optimol Instruments Prueftechnik GmbH. She has 25 years of experience in engineering R&D, more than 20 publications and one patent, as well as a PhD in technical chemistry from the Technical University of Vienna.

Mathias Woydt is managing director of Matrilub and has more than 34 years of experience in the R&D of feasibility studies on disruptive technologies, root cause failure analysis, slip-rolling alloys, fine ceramics, ceramic composites, coatings, wear, lubricants and tribo-testing.