Developing An Effective Oil-Analysis Program

Like vibration monitoring and analysis, infrared thermography, ultrasonics, etc., oil analysis has proven itself time and again to be an invaluable technology in industry’s quest to wring longer and longer service life out of equipment. As a predictive tool, for example, it analyzes wear debris. As a proactive technology, it is on the lookout for oil condition changes and contamination.
Two previous articles published in Lubrication Fluid & Power (now LMT) and Lubrication Management & Technology discussed in detail the various predictive and proactive tests associated with oil analysis. These articles were:

• “Proactive Maintenance Through Condition Monitoring,” pgs. 28-31, Lubrication & Fluid Power (now LMT), November/December 2005
• “Improving Predictive Maintenance Through Wear Debris Analysis,” pgs. 10-15, LMT, November/December 2006

Before you can reap the benefits of timely and appropriate oil analysis, however, you first must correctly set up your program—with an emphasis on “correctly.” This is crucial to your program’s future success, and it all starts with selection of the right oil-analysis laboratory.

Any program, though, begins with the establishing of objectives: Where are you now? Where do you want to go? How do you get there? In the area of oil analysis, your objectives and strategy can be summed up as follows:

1. Improve asset reliability, which includes extended life and breakdown avoidance.
2. Identify and eliminate repetitive problems.
3. Reduce unscheduled maintenance.
4. Maximize lubricant life and performance, resulting in reduced lubrication and maintenance costs.
5. Achieve fault-free component life extension.
6. Utilize oil analysis to maximize proactive maintenance along with predictive maintenance technology.
7. Achieve the overall goal of condition-based maintenance.

Laboratory selection
The most important element in establishing a successful oil-analysis program is the selecting of a high-quality laboratory. Since you have so many from which to choose, your evaluation process should focus on selection of the operation that will help you best meet your objectives. The following criteria have been identified as the most important in the selection process:

Data quality…
Without quality data, you can’t effectively make decisions on improving your reliability program. Most laboratories utilize similar laboratory tests and equipment, but results can differ greatly on the same sample.

How do you know that your data accurately reflects your oil and equipment condition? One way is to use a laboratory that has achieved a high level of accreditation. ISO 9000 is a certification that deals with documentation of procedures and good recordkeeping—but it doesn’t address measuring the accuracy and precision of the data. The highest level of accreditation used for analysis laboratories is ISO 17025, where good recordkeeping and documentation are established, along with evaluation of laboratory test data for precision and accuracy. The definitions for data measurement are as follows:

• Accuracy is the closeness to the true value, as indicated by a National Institute of Standards and Testing (NIST) traceable standard. This can be likened to shooting arrows at a target and hitting a bulls eye—which would be representative of the standard. When looking at defined limits, accuracy is very important.
• Precision is when values are clustered closely together—but not necessarily accurate. Having precision, which equates to low scatter, is important when trending data. There are two ways to evaluate precision:
  • Repeatability is the closeness of the data run by the same operator on the same testing equipment.
  • Reproducibility is the closeness of the data on the same test run by two different laboratories.

A quality oil-analysis laboratory should be expected to provide accurate and precise results within the parameters of the testing procedure. Different tests have different levels of accuracy. If a laboratory has not been accredited, it may still provide accurate and precise data. Before making your selection—and if you are unsure about the quality of your data—visit the laboratory and ask to see the quality-assurance records on data accuracy and repeatability versus known standards.

Logistics…
A high-quality laboratory should make it as easy as possible to send in oil samples—and the turnaround time should be no more than 24 to 72 hours from the time that a sample is received by the facility. In fact, many labs are achieving 24- to 48-hour turnarounds on routine tests. As more sophisticated tests are required (such as analytical ferrography, for example), the turnaround time will increase.

Most laboratories will send sample bottles with preprinted labels along with shipping containers to facilitate your sample-collecting procedure. Quality labs do everything they can to simplify this process for you.

Remember, too, that laboratory proximity can be an advantage in more timely delivery of samples. This means that during emergency situations, samples can be delivered quickly to the lab for evaluation.

Technical service…
The ability of a laboratory to help set up a program is very important. Some labs provide field support in identifying and assisting in implementing optimum equipment sample points. Some will recommend the proper sampling points and help in the correct installation.

Training—so critical in maximizing an oil-analysis program—is supplied by many of the leading laboratories. During program inception, it is important for personnel to be trained as quickly as possible on procedures in sampling techniques, information transfer, data management and data interpretation.

Setting condemning limits is a key component in evaluating equipment and oil condition. Some labs use very sophisticated statistical techniques, while others base their recommendations on experience and OEM recommendations. So you can feel comfortable with their recommendations, question laboratories on the techniques they use to set their limits. The key to setting good limits is to provide the laboratory with very complete information on the equipment, including type , model number, sump capacity, filtration, etc.

Most laboratories provide recommendations on their reports when problems exist—but many of these recommendations are computer-generated. The ability to call the lab and speak to an equipment expert about problems is an important service. Some facilities have experts in different equipment areas that can help in data interpretation and recommendations for preventing future problems. When selecting a laboratory, the ability to access personal support on equipment problems should be heavily weighted in the selection process.

Some oil-analysis operations specialize in particular industries. For example, selecting a laboratory that specializes in your industry could provide a higher level of technical service. The ability of a laboratory to be flexible in providing non-routine tests—such as grease analysis—may be a consideration in the selection process.

Data retrieval and management also is a chief component in any oil-analysis program. Paper reports are history. Most major laboratories have easy-to-use Web-based data-retrieval and data-management programs. Some very sophisticated programs provide powerful tools in the overall data-evaluation process.
 

Cost…
The old adage that “you get what you pay for” is so true when it comes to selecting an oil-analysis laboratory. There are more than 100 of these operations in the U.S. Of these, 10 are considered major labs. Prices can be as low as $4 per sample—and as high as almost $100.

The following section on implementation discusses recommended programs by equipment types. For these programs, the average cost is $25 to $40 per sample. Analytical ferrography, however, the most sophisticated wear debris test used for justifying equipment shutdown, costs between $35 and $75. The costs for these special analyses, of course, would be added to the basic cost of a sample. Most labs offer a service where, for an additional, average nominal cost of around $5, the facility will run analytical ferrography as needed. This cost is based on running the test once every 10 samples.

The cost per sample becomes lower the more samples that are run. For large accounts, the discount can be significant. Another way to save money is to prepay for the samples when receiving the bottles—that can result in savings of up to 10%. Unused samples, which normally have a time limit, are not refunded. This fact should be taken under consideration whenever you are thinking about using this option.

NEVER select a laboratory based strictly on the cost of samples. Cost is the least important of the four major criteria discussed here. Instead, consider the overall value that the laboratory is capable of providing.

Program implementation

Develop an equipment list…

Once your objectives have been established and you’ve selected the right oil-analysis laboratory, it is time to implement the program—always the most difficult step. To assure a smooth implementation, it is important that one person from the end-user/client organization be named as the program champion and key contact with the lab. The following steps are recommended in program implementation:

1. Establish a list of equipment to be sampled. Don’t try to sample all of your equipment at once. Prioritize based on equipment criticality and add units as the program develops.
2. Complete an equipment list for the oil-analysis lab listing the following items:
   a. Equipment identification
   b. Component type
   c. Component make and model
   d. Lubricant brand and grade
   e. Oil reservoir capacity
   f. Filter type and rating
   g. Sampling frequency

Figure 1 is an abbreviated example of a master equipment list. (Make yours long enough to accommodate all equipment from which you will be collecting samples.) Once filled in (typically electronically), it should be submitted to your chosen oil-analysis lab—which will then supply preprinted labels for correct sampling. Additional details that need to go on the label include the time that the oil has been on the machine; how much oil has been added; and if the oil has been changed. By supplying complete information, a truly effective diagnosis can be made of the equipment condition.

Develop a sampling strategy…
This is one of the most important steps in the overall process. Bad data is worse than no data. (Please refer to a previous article in Lubrication Management & Technology published in the March/April 2007 issue, entitled “Proper Sampling Techniques: The Key To A Successful Oil Analysis Program.”)

Utilize the oil-analysis laboratory in the development of proper sampling locations and procedures and to provide training. Try to ensure that the same people sample the equipment. Consistency is very important. Sampling frequency is another important consideration. In order to identify potential problems early, and to develop meaningful trends, continuously operating industrial equipment should be sampled monthly.

Select the appropriate test packages…
Selecting the right tests for different equipment types is vital to the success of your oil-analysis program. The following are recommended tests:

• Industrial circulating systems (turbines, bearing reservoirs, paper machines, rotary compressors, pumps and motors)
  • Viscosity @ 40 C
  • Water (Karl Fischer)
  • Acid Number
  • Spectrometals (ICP or RDE)
  • Particle Count with ISO Code

Because of the importance of oil cleanliness, particle counts should be run on all circulating filtered systems.

• Hydraulic systems
  • Viscosity @ 40 C
  • Water (Karl Fischer)
  • Acid Number
  • Spectrometals (ICP or RDE)
  • Particle Count with ISO Code

Running particle counts on hydraulic systems is a requirement because of the importance of oil cleanliness in directional valve performance.

• Gearboxes
  • Viscosity @ 40 C
  • Water (Karl Fischer)
  • Acid Number
  • Spectrometals (ICP or RDE)
  • Particle Count with ISO Code (Filtered Gearboxes)
  • Direct Reading Ferrograph (DR) or Particle Quantifier (PQ) (Unfiltered Gearboxes)

Analytical ferrography should be run if one of the following conditions appears:

• Significant increase in wear metals detected by emission spectroscopy (ICP or RDE)
• Jump in DR Large or PQ
• Increase in particle count, especially in the larger size ranges

As mentioned earlier, the foregoing tests can range in price from $25 to $40-which does include analytical ferrography.

Allocate personnel…
The most successful programs maintain consistency by having a program champion who coordinates the program. He/she is the coordinator between the laboratory and the end-user/client operation(s). This person disseminates the oil-analysis information to the appropriate parties in the organization and also helps coordinate the sampling program. The key to maintaining consistency is to have a dedicated group doing the sampling.

Tracking and documenting cost/benefit…
Without proper documentation, many programs fail. Unless management is convinced of the program benefits, money will not be spent on maintaining the program.

Remember that what you measure, you manage. Good measurement techniques will assure program success. In one major manufacturing facility, all costs associated with oil analysis—including manpower and training—are measured and continuously compared to equipment breakdown savings associated with the program. Conservatively, this plant is realizing more than a 400% return on its investment. Moreover, despite a tough economic climate, this program hasn’t simply been maintained, it’s been enhanced.

Unfortunately, many successful oil-analysis programs in industry are eliminated because no compelling documentation is presented. At these plants, management typically—and incorrectly—has assumed that there were no major problems, so why incur the costs. In some of these cases, however, after eliminating the program, equipment breakdowns have occurred more frequently.

Conclusion
Oil analysis is a key component to any reliability condition-monitoring program. But it requires much thought and planning to implement a successful program—starting with the selection of the oil-analysis laboratory. The laboratory will be a valued partner in your overall oil-analysis process, so choose carefully. Once it has been selected, let your lab help in the implementation of your program through assistance in sampling, test-slate selection, data-evaluating techniques and training.

Consistency is an enormous factor in the success of your program. Pick an overall program coordinator and give him/her the tools to make the program pay off. Use the same people to do the oil sampling.

In a down economy, unless the program is justified on a cost/benefit basis, it will not survive. Make sure that you track and document the results and let everyone—especially management—know about the return on investment.

For more specific information on oil analysis, please refer to the three published articles on sampling, oil condition and equipment selection that were cited earlier in this article. LMT

Contributing Editor Ray Thibault is based in Cypress (Houston), TX. An STLE-Certified Lubrication Specialist and Oil Monitoring Analyst, he conducts extensive training in a number of industries. Telephone: (281) 257-1526; e-mail: rlthibault@msn.com