Tribology and Lubrication Technology May 2012 : Page 36
coVeR stoRy Jean Van Rensselar /Contributing Editor While uniform standards don’t exist, you’ll do well by establishing a statistical range for 10 key metrics and looking for deviations. Used oil KEY CONCEPTS • Widespread disagree-ment exists on used-oil condemning limits. • A uniform set of con-demning limits does not exist and would be nearly impossible to create. • The best procedure is to set limits using reliable statistics and watch for out-of-limit results. 36 “Principles of Metallurgy”—all-new education course (co-sponsored w/ASM International)
Jean Van Rensselar
While uniform standards don’t exist, you’ll do well by establishing a statistical range for 10 key metrics and looking for deviations.
Used oil condemning limits
Do you feel lucky? It’s more than a classic line from a Clint Eastwood movie, it’s what fleet managers—those responsible for determining when fluids need to be changed—regularly ask themselves. Why? Because uniform condemning limits are hard to come by.
Uniform oil analysis test procedures, on the other hand, are established and reviewed by such agencies as the International Organization for Standardization (ISO), the American Society of Testing Materials (ASTM) and the Society of Automotive Engineers (SAE).
Having uniform oil analysis test procedures without uniform condemning limits—is like having an accurate blood cholesterol test but no knowledge of how high is too high. You could spend years perseverating over every pat of butter or you could throw caution to the wind and end up succumbing to a single scoop of ice cream.
Establishing uniform condemning limits, though, is not really possible for a number of reasons including the variability of engine-use conditions. This leaves one or the other user in the position of having to make an educated guess as to when to change fluids. If they change fluids too late, they risk engine damage; too early and they waste money.
“Any report showing a severe condition will indicate that the problem needs to be addressed immediately and there is little margin of safety until the situation is addressed,” says Ted Savage, business manager-central for ALS Tribology in Kansas City, Kan. “Some action or investigation at some level should be instigated in a timely manner. The action may be as simple as asking yourself, ‘Can I live with this amount of fuel dilution if it does not get worse?’ As the rate of change of wear, contaminants and oil degradation increases, the safety margin decreases. The margin of safety may be very limited when rates of change for excessive wear and contaminants and oil degradation increase during an abnormal condition.”
The scope of this article is limited to oils for internal combustion engines large enough to warrant used oil analysis.
TESTS AND CONDEMNING LIMITS
All oil analysis programs are designed to be performed as a trend analysis, requiring a series of regularly scheduled samples. “We derive condemning limits from several sources,” says Savage. “ALS will work with lubricant manufacturers for condemning limits related to contaminants and lubricant properties as well as OEMs for condemning limits related to equipment design wear, contaminants and lubricant properties.”
In addition to the condition of the oil itself, other considerations include:
• The hours and/or mileage on the oil
• The OEM-specified oil drain interval
• The engine’s total run-time
• The expected lifespan of the machine.
Tracy Wolf, gas engine customer service for Caterpillar, Inc. in Lafayette, Ind., explains, “As the engine ages, items such as piston rings, liners, valve guides and valve stem wear. This leads to more rapid contamination of the oil and, hence, can lead to a shorter drain interval. Deposit build-up and internal component wear is especially critical in biogas fuel applications.”
Ideally, the oil analysis records of each piece of equipment includes baseline data and results of samples taken at regular intervals. Experts in most fields say that establishing a trend requires at least three samples.
“The goal of using trending analysis is to detect problems before they become major issues and plan accordingly,” says Savage. “The consequences of exceeding published limits are related to assessing the overall situation. Does the problem represent a potential catastrophic situation if it is not addressed immediately? Does the severity of the situation allow planned shutdown and repairs that will be most economical and least disruptive to business? Equipment reliability will not be compromised if addressed in a timely fashion. Sometimes a single test parameter needs to be addressed immediately if it represents a root cause that will lead to greater problems in the near future.
“Sometimes,” Savage adds, “a single test parameter is not going to justify immediate action if all other test results appear normal. Awareness of the condition is all that may be required at this time. Usually looking at multiple parameters in a test report gives the best assessment of equipment condition. There are times when no single test parameter has reached a condemning limit, but the overall trend for multiple parameters is showing an increased rate of detrimental change, indicating it would be prudent to take action before problems accelerate to a serious condition.”
But these tests are less significant without sensible condemning limits in place—set by OEMs, independent oil analysis labs and, in large operations, end-users.
COMMON OIL PROPERTY EVALUATION TESTS
Oil analysis programs address such areas as abnormal wear rates for internal engine parts, the state of oil additive packages, the level of contamination and the physical condition of the oil.
Disagreements exist regarding which of the multiple oil condition parameters are most critical to monitor and which thresholds determine the oil change interval. One of the most exhaustive data-gathering exercises on the variations among condemning limits set by different OEMs was conducted by the U.S. Army in 1999. Referred to as the Marbach-Frame Report, its purpose was to identify the best methods of field testing used oil. But the report was so thorough that it yielded a wealth of data relevant to condemning limits. This report is still cited today.1
The information revealed that the recommended oil change interval varied significantly among OEMs.2 Oil viscosity, however, was agreed upon as the most critical parameter related to both oil degradation and engine condition and Total Base Number (base number) was considered to be a significant parameter for determining the extent of oil degradation.3
Maintenance personnel identify set properties to evaluate the condition of the oil and equipment. These properties include the following:
Viscosity. This measures the oil’s resistance to flow. The ideal viscosity is extremely important to performance and equipment life. In addition, the Viscosity Index (VI) can indicate many abnormalities.
Water content. Water gets into the crankcase through gasket leaks, water pump seals and condensation that occurs during cool down. Moisture can promote oxidation and corrosion.
Coolant/glycol. Glycol breaks down at high engine temperatures, which leads to water evaporation. The glycol then forms sludge which reduces filterability. Large quantities corrode bearings and block oil flow.
Fuel Dilution. Fuel dilution is caused by incomplete combustion brought about by extended periods of idling or by using a heavy fuel. It may also be caused by a malfunctioning or improperly sized fuel injector or from worn liners or piston rings that end up dumping too much fuel into the system.
Acid number. This measures the total amount of acidic products in the oil, indicating the level of the oil’s oxidation.
Base number. This measures alkalinity. As oil ages, alkaline additives neutralize the combustion acids, and alkalinity decreases. Operating with low base number oil can lead to wear and deposit formation. The alkalinity can only be restored by adding or changing the oil.
Insolubles. In addition to oil-insoluble material, this also measures some oil-soluble resinous material that originates from oil and/or additive degradation.
Soot. Soot is usually caused by delayed injection timing and is an indication of poor air-to- fuel ratio. Along with increased viscosity, excessive soot causes abnormal valve and injection- train wear. It can increase exhaust emissions and clog filters.
Elemental analysis. This tracks the additive and wear elements, wear metals, and contaminants in oil. The concentration of wear metals can indicate abnormal wear and also indicate an incorrect oil is being used. In addition to monitoring equipment condition, it can detect coolant leakage.
Particle count. Particle count is recommended for use in conjunction with wear metal analysis because it detects wear and deterioration of synthetic friction materials on discs. It also detects the larger worn particles.
According to Bill Briggs, director-field service of GE’s Waukesha gas engines division in Waukesha, Wis. (and in line with Savage’s comments) within the parameters that are set for different types of oil, condemnation limits are generally well-established and conservative to protect the gas engine. However, condemnation limits of the individual constituents of the oil are reached at different times in different gas engines, depending on the combustion system of the engine, the characteristics of the fuel and the environment in which the gas engine is operating.
In practice, however, rarely, if ever, are all of the constituents of used oil tested to determine their condition. Once oil passes published condemnation limits, it is difficult to determine the point at which damage to the engine will occur because of the degradation of any one or more of the constituents in the oil. Therefore, once any oil constituent reaches its condemnation limit, it’s time for the oil to be changed.
Most experts agree that the No. 1 indicator of a fluid’s quality is viscosity. This goes decades back when maintenance workers used to “eyeball” the condition of the fluid. Another early method for testing viscosity was to place a drop of oil on a paper towel to see how far it spread. Even though today’s viscosity methods are far more sophisticated, expert eyeballs are still a good indicator.
An increase in viscosity is due partly to evaporation of lighter base oil fractions and also the loss of volatile oil degradation products, high-temperature oxidation, water, coolant contamination and soot accumulation. Excessive oil viscosity can hamper oil flow to critical engine components, especially during low temperature startups. Continued use of overly viscous engine oil can cause severe problems, leading to seizure and failure in the extreme.
A decrease in oil viscosity usually indicates fuel dilution but in multigrade oil can be caused by shearing of the VI improver. Using an oil with a viscosity that is too thin can result in high oil consumption, stress on critical parts and component failure. (Editor’s Note: For more on VI improvers, see the September 2011 TLT issue, available digitally at www.stle.org.)
OEMs, independent labs and large fleet operators use these tests to define condemning limits.
“Generally ALS does not override engine manufacturer limits, which would be the standard we go by,” says Savage. “ALS house limits are only incorporated when there is no available information from the OEM. We work directly with some equipment manufacturers to establish new or updated oil limits for their product.”
“It’s not unusual for ALS to work in conjunction with equipment manufacturers and partner to develop OEM limits,” Savage adds. “These normally consist of using data collected for the OEM through one of our testing programs and then applying the data to equipment manufacturing designs and applications. This is useful when new equipment designs are created and put on the market.”
SETTING LIMITS USING STATISTICS
Independent analysis labs and OEMs use several approaches to determine condemning limits. The most common, and most readily adaptable to customize user limits for some parameters (like wear metals), is to use a simple statistical model to analyze the data range of similar components with similar drain intervals. This model assumes that elemental concentrations from similar equipment fall into a normal distribution pattern. Calculating the average is as basic as summing all the values and dividing by the number of values. However, it’s also important to know the range of acceptable values and determine the standard deviation.
Once the normal distribution value range and standard deviation are established, any value falling outside the standard deviation or a multiple of the standard deviation is cause for investigation.
The value of this approach to setting wear metal limits is that it fairly accurately represents the expected wear metals concentrations in used oil samples, as long as the data set is large enough and appropriate. Today, most oil analysis software and even standard spreadsheet software can easily calculate the average and standard deviation for a selected sample population.
A critical key to establishing a condemning limit based on statistics is accurately defining the data set. That means identifying samples by factors that include:
• Equipment type
• Component type
• Oil type
• Functional environment.
As long as there is sufficient data available, the statistical approach for setting condemning limits applies to fleet equipment as well as stationary plant equipment.
“At GE, oil condemnation limits are reached at different times in different gas engines, depending on the combustion system of the gas engine, the characteristics of the fuel and the environment in which the gas engine is operating,” Briggs explains. “GE’s Waukesha gas engines division bases oil condemnation limits on a combination of long experience with the operation of the engines and working with large oil manufacturers who conduct extensive laboratory tests.”
Often engine oil manufacturers will work with their customers to set up a used oil test program. The customers will send samples of their used oils to the manufacturer who will run tests on the oil to determine its condition. Theoretically, this should allow fleet managers to extend the oil drain interval out as far as they can. But one of the big questions is whether OEMs are setting limits too high just to be on the safe side.
Savage says, “If no specific limits are provided by lubricant manufacturers or equipment manufacturers, ALS calculates limits using its database to statistically derive limits based on data population related to specific equipment design and lubricant application. It is not unusual to also develop limits by working with specific clients or end-users to meet their needs. We will work with clients to set optimal limits, which will ensure optimal equipment performance for their specific need or application.
THE MARBACH-FRAME STUDY
This groundbreaking work was undertaken by H.W. Marbach, Jr., and E.A. Frame at the U.S. Army TARDEC (Tank Automotive Research, Development and Engineering Center) Fuels and Lubricants Research Facility (SwRI) Southwest Research Institute in San Antonio, Texas.
Marbach and Frame began by compiling a list of the major component manufacturers of engines, transmissions, hydraulic systems, generators, and manual transmissions/final drives typical of U.S. Army ground equipment.
Next, they contacted these manufacturers to obtain their oil analysis recommendations, and change criteria for oil degradation and equipment condition (essentially condemning limits). Using this information, along with TFLRF (The Fuels and Lubricants Research Facility) personnel experience, they compiled a list of key parameters of used oil properties. They identified properties required for oil monitoring and/or component monitoring. Then they developed critical oil analysis requirements for compression ignition engines, turbine engines, power shift transmissions and hydraulic systems.
Once they received the manufacturers’ oil-change criteria/ guidelines, Marbach and Frame compiled tables for oilchange properties, along with the properties’ ranges for engines, transmissions, hydraulic systems, drive axles and manual transmissions.
The authors discovered that three of the major engine manufacturers had slightly different points of view about the importance of certain parameters in their engine recommendations. One author recommended wear metal analysis for detecting oil contamination. Another considered the following primary oil characteristics to be warning signals: viscosity, flash point and elements silicon, boron, sodium and potassium. A third manufacturer recommended using primarily viscosity and base number to establish engine oil condition.
The two researchers determined that most hydraulic system manufacturers required viscosity, water content and particle count, while the neutralization number and flash point are generally recommended for fire-resistant hydraulic fluid. All except two drive-axle and manual transmission OEMs recommended hard time intervals for lubricant changes.
Basically, Marbach and Frame found that there was significant variation in oil drain recommendations among OEMs.
Marbach and Frame then identified critical lubricant analyses properties based on equipment/component manufacturer oil analysis requirements. Next they categorized critical oil properties and contaminants into three levels based on cost and risk. Since the ultimate goal of the study was to create a portable (in the field) method for testing oil, the cost factor was the estimated cost of the required analysis equipment. The risk factor was defined as a subjective estimate of the risk of not detecting a problem.
level 1. This included the minimum used oil properties (viscosity, water/glycol, acid number, wear metals) that must be determined to provide reasonable equipment protection. This least costly set of properties would involve the highest level of risk that a problem would be missed.
level 2. This intermediate level included all Level 1 property analyses plus soot content, insolubles, base number and wear particle analysis.
level 3. This ultimate analysis included all Level 1 and
Level 2 analyses plus Fourier-Transform Infrared (FTIR) and spectrographic techniques. Level 3 has the highest estimated cost and lowest estimated risk. Level 3 provides the most detailed oil analysis, but test equipment costs rise commensurately.
While most oil analysis end-users don’t own their own analysis equipment, the cost of the equipment affects the cost of the test, meaning equipment cost in this study can be roughly proportional to test cost. The three test levels that Marbach and Frame developed are a good starting point for identifying test depth and frequency for various types of equipment.
DETERMINING DRAIN INTERVALS
The Marbach-Frame Study is extremely valuable for fleet managers and maintenance personnel in every industry because of the nature of the equipment under study. Because they were considering military equipment, they were fully aware that not only was the equipment at risk, but, far more important, lives. A diesel engine breaking down on the highway is one thing, a tank breaking down in the middle of a firefight is quite another.
That Marbach and Frame were asked to look for ways to not only speed and optimize reliability of oil analysis but also reduce the cost of oil analysis for critical battleground equipment is a telling indication of how costly the problem of drain interval ambiguity is. In fact, their goal was to reduce the number of used oil samples sent to remote testing facilities by approximately 80% through the use of onsite equipment and/or reducing the number of samples.
There is a movement in the trucking industry to extend drain oil intervals as far as possible—partly through adoption of API CJ-4 specifications. As mentioned above, the drain interval depends on a number of factors that qualify the condition of the oil but don’t necessarily show up in oil analysis results. One of those factors is the operating environment. For example, a diesel truck that is constantly hauling heavy loads up steep dusty roads is going to have a drain interval quite different from the same make and model that goes on long hauls over highways.
Wolf says that nearly all customers are very concerned about extending oil drain intervals for the following three reasons:
• Cost of equipment downtime to change oil and filters
• Cost of the oil and filters
• Cost of disposing of the used oil and filters.
One suggestion for determining (and possibly extending) drain intervals using a statistical method is to remove as many variables as possible. For example, remove the environmental variable by dedicating specific equipment to specific operations.
Susanne Chvatal, engine engineering and oil management of GE’s gas engines division, says, “A specific oil quality results in a typical oil drain interval. If a customer needs extended drain intervals, we can recommend products of a quality that provide a longer drain interval. Complaints about insufficient oil quality usually do not occur if customers follow our oil management guidelines, which include using an approved oil, analyzing oil samples in a defined interval and program and maintaining the oil below condemning limits.”
CREATING A BALANCE
When it comes to setting condemning limits and determining drain intervals, there are several interested parties:
• The OEM
• The independent test lab
• The fluid manufacturer
• The EPA and environmentalists
• The end-user.
Their interests aren’t all mutually exclusive. For example, it’s not in anyone’s best interest for equipment to fail or to pollute the environment with waste oil. Ultimately, it’s up to the end-user to determine which condemning limits make the most sense and whether it’s feasible or advisable to create a set of independent, organization-specific limits. Only the end-user can determine the organization’s margin of safety, its take on the risk versus cost continuum.
According to GE, some of the potential consequences of exceeding established condemning limits include:
1. When oil deteriorates, it can build up behind the piston rings. That causes the ring to push out, which wears out the ring and the liner against which it moves. As a result, there is no place for the oil to sit on the liner, which, in turn, leads to scoring of the liner and piston.
2. Deteriorated oil can leave varnish-like deposits on pistons that damage the rings and the liners.
3. These varnish-like deposits can form on valve stems, which can cause the valves to stick. A sticking valve can be hit by the piston and break, which could cause a major failure of the engine.
4. When the acid content of the engine oil exceeds condemnation limits, it can cause acid attack or bearing corrosion, which can lead to bearing failure.
5. Wear metal particles or environmental debris in the oil can damage bearings in both the engine and the turbocharger.
“The end result in exceeding a published limit can affect equipment reliability, work interruption, equipment life and resell value, repair costs, even safety and increase the economic impact of maintenance and repair costs,” says Savage. “A test result outside of an acceptable range should be investigated and not ignored so the reason for the out-of-limit test result is understood. It may be the published limit is too stringent. There may be a unique situation that forces a test result outside of the normal range due to the environment or local application.”
Concludes Savage: “With some parameters for test oil and equipment, the question to ask is, ‘Can I live with this situation and for how long?’”
Read the full article at http://onlinedigitalpublishing.com/article/Cover+Story/1024209/106542/article.html.
Previous Page Next PagePublication List