This paper was presented by Patrick Forgeron at the Maintenance and Reliability Technology Symposium (MARTS 2004) in Chicago, Illinois on May 25, 2004."
Background:
In order to quickly and accurately determine the "cause and effect" of oil and wear related problems, proper tests and procedures based on the equipment sampled and its application must be utilized. This paper details the appropriate test slates for various types of industrial equipment and applications as well as the additional factors necessary to manage a successful and cost effective oil analysis program.
Oil analysis has long been accepted as a valid predictive maintenance technique. The facts show that, when properly utilized, using a qualified laboratory with proper testing, and there is a full commitment of the user, oil analysis requires the lowest cost of implementation and the provides the highest rate of Return on Investment by any of the predictive maintenance disciplines.
Proper Utilization:
Oil analysis is most effective when used as a regular component of an ongoing Predictive or Reliability Centered Maintenance program. Users need to sample the right equipment at the right time using the right sampling method.
The decision process for the determination of the equipment to be placed on the program should be made based upon the criticality of each piece of equipment. Sample frequencies should be based on the usage or utilization of the equipment. Sampling location and technique should be determined by the ability to capture the most representative sample - - upstream or prior to filtration and away from "dead flow" areas.
Criticality of the equipment covers many items. Consideration should be given to the level of importance the piece of equipment has to the overall production of the plant (or line). The cost of repairs and the cost of replacement also factor in the decision-making process. These costs must include parts, labor and lost production due to the downtime. Of least importance is the sump capacity of the equipment. In many instances plant personnel forgo sampling of small-sump capacity units, that if or when they fail can bring an entire production process to a halt. It is not the cost of the lubricant or size of the sump that should drive the sampling but the cost of productivity, repairs and replacement.
Sampling frequency is typically based on hours of usage for the lubricant in use. Another key component in the setting of proper sample frequencies is factoring the rate of change in wear or lube degradation, that can occur over short periods of time.
An example is high speed units that develop abnormal wear patterns which may further accelerate the wear pattern by virtue of the speed of the rotating components, resulting in "wear creating wear". Or, equipment that has a tendency to experience lube contamination from outside sources may experience very quick changes in viscosity due to the type of contamination.
In these and similar situations sampling frequencies should be shortened in order to increase the chance of early detection of the problems.
Proper sampling location and technique is a critical element of a successful program. The lubricant is the medium holding the information about the condition of the equipment. Without a proper representative sample of the lube in use the diagnostics of the test results will be restricted. Wherever possible samples must be pulled from the circulating oil upstream of the filter housing. Samples should be taken with the oil at or near operating temperature and no later than fifteen minutes after a shutdown.
Whenever possible sample valves should be installed and used. This helps guarantee a proper location, with the sample being taken in the same manner each time, regardless of who is taking the sample. In order of preference, samples should be taken via:
Two last items in the utilization of a program are proper paperwork and prompt delivery of the samples to the laboratory.
The adage "the job's not over until the paperwork is complete" has never been so true as with oil analysis. Unless the laboratory has the proper and needed information regarding each piece of equipment enrolled in the program and on each individual sample, the ability to accurately evaluate the test results to determine equipment and lube condition is restricted.
Information required on the equipment includes:
Note: information on the equipment needs only to be provided once unless something changes.
Information on each sample includes:
When provided, this information allows diagnosticians to properly assess the equipment and lubricant condition based on utilizing the variable information that is specific to each sample. For example, a test result of 125 ppm of copper on a sample from a Cincinnati worm drive gearbox with 1000 hours of running time on the oil might be considered quite normal while the same unit with the same 125 ppm of copper reported on 200 hours of running time on oil may represent a significant wear related situation. In this case, the run time on oil would be the critical factor, representing the time over which the wear products accumulated before the sample was taken.
Qualified Laboratories:
There are three sources of laboratory services available to industry: in-house facilities, major suppliers, and independent organizations.
In-house facilities, which are manned and operated by company personnel, are most prevalent when the sampling facility has unique requirements for testing, are in an extremely remote area or have difficulty in shipping samples to an outside lab due to environmental or legal restrictions.
Major suppliers such as manufacturers and oil companies often provide an oil analysis as a requirement for warranty or a value added service. Testing through a major supplier can be on a fee basis or included with the purchase of product.
Independent organizations such as Analysts, Inc. are laboratories that provide oil analysis as a core business proposition.
Users have literally hundreds of choices of laboratories to choose from for their oil analysis requirements. The source of laboratory, in-house, major supplier or independent, is not as important as the qualification of the laboratory. Qualified laboratories are determined by two primary components: ability to provide accurate test results and ability to provide accurate and meaningful diagnostics.
The process begins with the need for accurate test results. The data must conform to established industry (ASTM) parameters of repeatability and reproducibility and must be based on the appropriate methodology for the equipment and lubricant being tested. Quality certifications such as ISO 9001:2000, ISO 17025 or 10CFR50, Appendix B are certainly differentiating factors which help identify quality laboratories. Beyond the certifications, a lab should have the knowledge to recommend and perform the appropriate tests on samples based upon the equipment type, application and lubricant in service. It is incumbent upon the lab to demonstrate this knowledge to the user, but it also incumbent upon the user to demand this knowledge and insure that it's utilized.
Meaningful diagnostics are the end point of the lab services. Once the quality data is produced the test results must be interpreted and translated from numbers and values into specific statements of conditions with recommended actions to correct the identified problems, and their causes.
It is the laboratory's responsibility to demonstrate the level of diagnostic ability and experience in the user's industry. It is the user's responsibility to insure they understand the level of expertise the lab possesses and that it is acceptable to the expectation and requirement of the user.
Individual certifications such as STLE's CLS or an academic accreditation such as a degree in engineering, physics or chemistry for a diagnostician (data analyst) will help differentiate the level of ability of a laboratory to provide meaningful diagnostics and recommendations of sample data.
Proper Testing:
Loosely defined oil analysis is simply the "analyzing of oil". This "analyzing" can be comprised of literally thousands of different tests to determine hundreds of different properties of the sample. However, logic dictates that commercial use for an RCM program defines oil analysis as a slate or series of tests performed on used lubricants to determine component wear of the equipment and, contamination and certain physical properties of the lube in use.
Generally, oil samples are provided based upon a pre-packaged slate of tests. It is important that the user understand the tests and the methodologies used to perform the tests when selecting the various slates. Lack of test knowledge or strictly price driven decisions can easily lead to misapplication of the testing to equipment and lackluster results in the overall program.
Tests can be categorized into three areas; contamination, physical properties and wear. The tests routinely utilized in used oil analysis include:
The key to a successful program is "appropriate" tests for the specific equipment and lubricant in service. Both of these items have a direct bearing on the validity of the analysis and diagnostics.
An example of this is when a moisture content reported in ppm is required for a piece of equipment where the lubricant in use is heavily additized as part of its normal additive package. The typical test to determine the moisture content is Water Karl Fischer. However, there is more then one technique in running a Karl Fischer. The most common technique used by laboratories is where the sample is injected directly into a titration vessel. Unfortunately, in these instances heavily additized, also known as "well fortified" lubricants cause interference with the methodology and false high readings are reported. The proper methodology is to use an oven method Karl Fischer where the sample is heated under controlled conditions. The moisture in the sample is vaporized and then carried by an inert gas to the vessel chamber for titration. The additives do not enter into the chamber and a correct moisture level is reported.
The following table lists recommended test slates for generic equipment types. Adjustments to these recommendations may be made based upon the specific type of equipment manufacturer, application and lubricant in use. Finalization of testing should be made after the user and laboratory discuss and define the program goals and objectives.
A successful oil analysis program will provide information to the user of that will identify wear rates of oil wetted components, levels of contamination present in the system and certain physical properties of the lubricant. Wear is generally attributed to a "cause and effect" scenario. Contamination and or changes in the properties of the lubricant are the "causes" that create accelerated wear, the "effect".
By properly testing for the appropriate contaminants, measuring changes in physical properties and monitoring wear metals a qualified laboratory can identify problem areas well in advance of outward signs, and should provide diagnostic alerts to recommend corrective actions that will correct not only the indicated problem(s) but more importantly the causes of the problem(s).
Commitment of the User:
Information is only as good as what we do with it.
A successful oil analysis program that significantly contributes to the reduction of maintenance costs requires a committed program by all levels of the user organization. Management must buy in to the program and budget the needed resources for the inclusion of all critical equipment, with testing at the proper frequencies and manpower to be able to react to the identified problems and alerts.
Reliability personnel must demand that their laboratory assist them in the identification of critical equipment, establishment of proper frequencies, provide immediate (24 hour maximum) response on sample turnaround and if applicable, integrate the data delivery into an existing CMMS program or provide electronic delivery of the data so that it can be easily managed.
Training, with the assistance of laboratory personnel, should be provided to the individuals charged with taking the samples and completing the necessary paperwork. A full understanding of what is involved in a program and why helps insure accurate and complete information is constantly exchanged between the user facility and the laboratory. For example, samples that are taken in the proper manner, with full information, that sit in a mail room for several day before shipping to the laboratory defeat the purpose of gaining information as soon as possible. Generally, an abnormal situation can be considered a critical situation waiting to happen. It's just a matter of time. Should a sample be delayed in getting to the laboratory it is quite possible that a costly problem might have been avoided. Unless the sampler is aware of this possibility the urgency in shipping a sample might not be there.
When problems are identified by the oil analysis corrective action needs to be put in place. Minor and scheduled maintenance activity is always less expensive then major or unscheduled activity. The productivity or earning capacity of a piece of equipment is always greater then the cost of repair. Downtime is a "dirty word" with operation personnel. Unscheduled downtime is an unacceptable event in a RCM program. Both can be controlled and managed with a successful oil analysis program.
The following is a fully documented case history of a committed organization's utilization of their oil analysis testing that encompasses all the components of a successful program: properly utilization, qualified laboratory, proper testing and a full commitment to the program by all levels of the organization.
Click here for a case history
