Keeping a close eye on the life-blood of your lubricated equipment systems pays off in many ways, all of them crucial.
By Ken Bannister, MEch Eng (UK) CMRP, MLE, Contributing Editor
When a doctor wants to assess the condition of your health, he or she may order a blood test. Similarly, oil analysis, sometimes referred to as “wear-particle analysis,” is a mature condition-based maintenance approach used to determine the health of a machine and its lubricating oil. The process involves taking a small sample of oil from the equipment’s lubrication system, comparing it to a virgin stock sample through a series of laboratory tests, and examining the results to ascertain the “wellness” of machinery and oil.
Oil analysis reflects a highly effective and inexpensive means of deciding when to change lubricants based on condition; predicting incipient bearing failure so that appropriate action can be taken in a timely manner to avert failure; and diagnosing bearing failure should it occur. Yet, despite its availability and proven track record since the 1940s, oil analysis is still misunderstood and overlooked as a proactive strategy in many of today’s industrial plants. Relatively easy to set up, this type of program should be implemented in any facility that purchases, stores, dispenses, changes, uses, or recycles lubricants as part of its manufacturing or maintenance process.
Basic implementation steps
A successful oil-analysis program can pay for itself in a matter of weeks, given the fact that it delivers multiple benefits, including:
- oil change intervals (often extended) that are optimized to the machine’s ambient conditions and operational use requirements
- probable reduction in lubricant-inventory purchase costs and spent-lubricant disposal costs
- enhanced understanding of how bearings can fail (or are failing) in their operating environment so that such incidents can be controlled or eliminated
- increased asset reliability, availability, and production throughput.
(NOTE: The potential for program success is greater if a site already has a work-management approach in place, thereby assuring completion of corrective actions in a timely manner whenever oil-analysis reports recommend them.)
Similar to other successful change-management initiatives rolled out across the organization, an oil-analysis program will benefit from a piloted, phased implementation. Taking a stepped approach allows management and workforce alike to become accustomed to the new sampling and reporting processes and quickly iron out any problems prior to a full-scale launch.
Step 1: Appoint a program champion.
All programs require a “go to” decision-making person who advocates on the initiative’s behalf and is committed to making the implementation a success. The champion should be at a supervisor or manager level.
Step 2: Choose a suitable pilot area/machine.
Oil analysis begins with sampling the oil and can include lubricating and hydraulic fluids. Choosing a suitable program pilot will depend on the type of industry and business operation. Typical starting points to evaluate might include:
- critical product, process, line, or major piece of equipment, i.e., criticality determined by constraint and/or lack of back up, downtime costs, and product quality
- mechanical equipment with moving components that include lubricant reservoirs for re-circulating-oil-transmission systems that are mechanical and/or hydraulic in design.
Step 3: Conduct a lubricant audit.
A lubricant audit, required to identify what lubricants are currently employed in service at the plant, calls for the following:
- Check work-order system PM (preventive maintenance) job plans for lubricant specification(s).
- Check on or near the lubricant reservoir for lubricant identification labels or stickers.
- Check for matching MSDS (Material Safety Data Sheets).
If a discrepancy is found at this stage, outside assistance from a lubrication expert or supplier may be needed to determine if the correct lubricants are being specified for particular applications.
Step 4: Choose a laboratory.
Not all oil-analysis laboratories are created equal, making your choice of one an important step. Most oil-analysis reports are divided into four major sections that provide:
- sampling and virgin-oil specification data
- spectral-analysis testing results for wear elements identified as lubricant additives or contaminants
- additional physical test results for viscosity, water, glycol, fuel, soot, and acidity
- associated conclusions and recommendations.
Some laboratories specialize in engine-oil analyses that focus more on physical testing for water, glycol, fuel, and soot. Others specialize in industrial-sample analyses that focus more on wear-particle evaluations and some physical tests for viscosity, water, and acidity, and post-mortem testing for root-failure causes using ferrographic techniques. Some laboratories have technicians that specialize in both areas.
The key to any testing program is receiving results in a timely and consistent manner, especially where critical equipment is involved. When interviewing laboratories, be sure to rate their sample “turnaround” time and how they can assure testing consistency (usually through use of dedicated technicians to test your samples). Working with a laboratory should be viewed as a long-term relationship. The chosen facility will build and analyze your complete data history and make conclusions and recommendations based not only on your current sample versus its virgin sample counterpart, but also on an understanding of your plant ambient conditions and overall trending history of each sample.
Step 5: Set up a pilot sampling program.
A good laboratory will work with you to set up your sampling program, supply (in some way) sampling-point hardware, extraction pumps, and quality sample bottles, as well as train your staff to consistently collect “clean” oil samples.
The best oil samples contain maximum data density with minimum data disturbance—meaning the sample should best represent the oil’s condition and particulate levels as it flows through the system or as it sits in a reservoir. For example, if you extract a sample from the bottom of a reservoir in a non-pressurized gearbox lubrication system, the particulate fallout will be dense due to large wear particles and/or sludge accumulation and not correctly represent the remaining 80% to 90% of reservoir lubricant that actually lubricates the gears.
In a pressurized re-circulating lubrication system, samples are best taken as the machine is running and at operating temperature, from a live fluid zone where the lubricant is flowing freely. Whenever possible, the sample should be extracted from an elbow, thereby taking advantage of the data density caused by fluid turbulence. Sample points are best located downstream of the lubricated areas to catch any wear elements before they’re filtered out by inline pressure or gravity filters.
Virgin samples of all lubricants in the pilot program will need to be collected and sent to the laboratory for checking. They’ll be used as a benchmark for the laboratory to measure and understand what additive ingredients and lubricant condition represents a normal state. This type of benchmarking will lead to easier identification of additive depletion and wear elements in subsequent samples.
Outside assistance/training from a lubrication expert or oil-analysis laboratory is advisable when setting up the pilot sample points.
Step 6: Set up a work-management approach to sampling.
Lubricant sampling must be performed consistently, on a frequent basis—making it a suitable candidate for a maintenance/asset-management work-order system. Using the written sampling procedure as a job plan, the task can be set up effectively through PM scheduling software.
Extracting and sending a sample to the laboratory is only the first half of the oil-analysis process. Someone (usually the planner, if one exists) has to receive the results electronically by email, read the recommendations, and take any necessary corrective action and/or file the laboratory report electronically to history, usually as an attachment to the PM sampling work order. This will require development of a workflow procedure—and training all maintenance staff involved in the program on the procedure.
Step 8: Commence sampling and program roll-out.
An oil-analysis program will identify major contamination and wear problems with the first sample set. Sample trending can begin with the third set, wherein the site starts identifying/predicting any negative trend toward potential failure and schedule corrective action before failure occurs. Ideally, a pilot program should be allowed to run for approximately three months or longer to show basic results before tweaking it and rolling out to the next area within a plant.
Once a program is working and providing results, larger-sized enterprises may wish to consider investing in an in-house staffed laboratory that will deliver faster results turnaround. MT
Contributing editor Ken Bannister is a Certified Maintenance and Reliability Professional and certified Machinery Lubrication Engineer (Canada). He is the author of Lubrication for Industry (Industrial Press, South Norwalk, CT) and the Lubrication Section of the 28th Ed. of Machinery’s Handbook (Industrial Press). Contact him at firstname.lastname@example.org.