Authors Brian Thompson, Analysts, Inc. Greg Livingstone, Clarus Technologies Background Varnish is a widespread problem in many hydraulic and sensitive lubrication applications. Routine oil analysis cannot indicate whether the lubricant has a potential to form varnish. The purpose of this paper is to introduce Quantitative Spectrophotometric Analysis (QSA®) as an effective tool for measuring the varnishing potential of lubricating fluids. This new predictive tool will enhance an existing oil analysis program by alerting the user to dangerous fluid and system conditions that would otherwise go undetected. The basic principle behind the method is that certain types of oil degradation contaminants have distinct and measurable properties. By separating these contaminants from the oil and performing QSA® on the residue, one can assess the likelihood of harmful sludge and varnish buildup within the system. The authors present two case studies in which routine oil analysis failed to indicate any signs of varnish and demonstrate the effectiveness of Quantitative Spectrophotometric Analysis as a predictive tool. Introduction Varnish is composed of lubricant degradation by-products and can cause a host of performance problems. Sensitive hydraulic and lube systems can come to an abrupt halt if varnish goes undetected. Unfortunately, most traditional oil analysis methods are of little value in predicting varnish. Varnish can occur in lubricants that appear healthy and have no alarming laboratory reports. QSA® is the industry's first predictive tool that can identify a lubricant's potential to produce varnish. This test will augment routine oil analysis and improve existing predictive maintenance and reliability programs. The Impact of Varnish in a Lubrication System There are numerous types of insoluble contaminants found in lubricating systems. Insoluble contaminants are those materials that will not dissolve in the oil. The two most general classifications of insoluble contaminants in are hard contaminants, comprised of dirt, debris and wear particles, and soft contaminants, containing the various oil degradation by-products. Varnish originates from the latter of these insolubles - the soft contaminants. Varnish is a thin, insoluble film that, over time, deposits throughout the internal surfaces of a lubrication or hydraulic system. Varnish deposits are composed of degradation by-products that result from the oxidation and polymerization of lubricant molecules. The degradation process accelerates as the lubricant undergoes exposure to air, water, and high temperatures. Varnish comes in a wide range of colors and consistencies ranging from black tar-like lacquer to opaque petroleum jelly-like deposits. The presence of varnish in hydraulic and lubrication systems causes many serious problems, including but not limited to the following: Reduced clearance zones affecting lubrication regimes. Often, this means a transition from hydrodynamic lubrication to boundary lubrication, which increases wear rates of pumps, bearings and gears. Increased friction in components demanding higher energy requirements and causing valves to stick or seize. Higher run temperatures. Varnish acts as an insulator, lowering the effect of heat exchangers and lessening the ability of the lubricant to cool. Restriction or impedance of oil flow. Varnish can cause valves, strainers and filters to clog. Increased wear rates. Varnish captures hard contaminants creating a sandpaper surface that will accelerate wear. Increased maintenance resources required to clean varnished systems. Some images of the consequences of system varnish are shown below: Figure 1: Varnish found on a turbine-bearing pad. Figure 2: Plugged filter due to varnish formation Figure 3: The Sandpaper Effect of Varnish Figure 4: Bearing Surface covered with Varnish Figure 5: Varnish on Gears. Figure 6: Varnish on valves The financial impact of having system varnish depends upon the piece of equipment, but in many applications, is quite significant. A sticking valve in a gas turbine plant may cause the unit to go offline, resulting in potentially hundreds of thousands of dollars in lost power. Deposits in an injection-molding machine may result in scrapped finished parts, costing an operation thousands of dollars in lost time and material. Due to the potentially high costs associated with varnish, it is important for maintenance and reliability personnel to have a predictive tool to measure the lubricant's varnish potential. Determining the varnish potential of a fluid enables the user to investigate the root cause and implement corrective action before a catastrophic failure occurs. Oil analysis has become a valuable predictive tool, used in a wide range of lubricant applications. There are many industry examples of the tremendous savings realized by incorporating oil analysis into maintenance and reliability programs. Unfortunately, even the best oil analysis programs fail to provide valuable information on the varnish potential within the lubricating system. The authors have witnessed dozens of specific applications where a business has incurred high costs due to system varnish, even though their oil analysis data indicated no cause for concern. There has not been a comprehensive laboratory method to measure and monitor the countless by-products formed during oil degradation. Most significantly, there has been no effective test method for measuring those by-products responsible for the formation of resin and varnish. Conventional laboratory testing methods such as Particle Counting (ISO PC), Acid Number (AN), Fourier Transform Infrared (FTIR), Ultracentrifuge (UC) and Rotating Pressure Vessel Oxidation Test (RPVOT) have proven to be of limited value for predicting the varnish potential of a lubricant. Particle counting and the associated ISO Cleanliness Class rating measure a system's contamination level of particulate matter greater than 2 microns in size. This test method is of limited use for the soft insoluble contaminants that contribute to varnish are often less than one micron in size. Acid Number, which measures the acidic constituents in the fluid, is only useful for measuring certain types of degradation by-products. Acid Number has inherently low sensitivity and can be influenced by certain beneficial additives. FTIR, which measures a compounds absorbance of energy in the IR spectrum, is subject to numerous types of interference, making results very difficult to obtain and interpret. The Ultra Centrifuge test, which centrifugally separates the insoluble material from the oil, has very low sensitivity and the process of taking measurements is operator subjective. RPVOT, which measures the fluids oxidation resistance, is not only expensive, but fails to provide a direct correlation to the insoluble content in the fluid. It is evident that these conventional laboratory tests are limited in their usefulness as predictive tools for system varnish. In response to this concern, Analysts, Inc. has developed a testing methodology to fill in this important information gap in modern predictive maintenance programs. Quantitative Spectrophotometric Analysis is a new testing procedure that allows a lubricant user an accurate means to assess lubricant and system varnish potential by directly monitoring insoluble content. The Oil Degradation Process To determine the root cause of varnish formation and the potential for a lubricant to produce varnish, one must first understand some of the common mechanisms of lubricant degradation. Most conventional lubricants are refined from petroleum consisting of hydrocarbon molecules. The alteration of the hydrocarbon molecule is defined degradation. As the chemical composition of a lubricant changes from degradation, reactive by-products are formed which are insoluble and unstable in the oil. These insoluble compounds are precursors to sludge and varnish. Typically these reactions can be achieved through reacting molecules incorporating alcohol, amine or carboxylic acid (or other carboxyl derivative) functional groups. In these reactions, heat acts as a catalyst. Lubricant degradation is most often a direct result of oxidation or thermal decomposition of the hydrocarbon molecules. Oxidation is the relatively slow process of replacing the hydrogen atoms with oxygen atoms along the oil's carbon backbone. This often involves a chemical reaction between the hydrocarbon and ambient or entrained air trapped within the lubricant. Thermal degradation is the chain or step polymerization of the hydrocarbon molecule. Thermal degradation occurs much more quickly than oxidation and involves much higher initialization temperatures. Thermal degradation may occur from adiabatic compression, localized hot spots, chemical reactions involving various foreign contaminants, or spark discharges from mechanical filters. It is not possible to monitor lubricant degradation at the molecular level, but it is possible to observe the products formed. As hydrocarbon molecules undergo oxidation or thermal degradation, they produce free radicals, which eventually form polar compounds. Free radicals are uncharged atomic or molecular species with unpaired electrons or an otherwise open shell configuration. These unpaired electrons are highly reactive, so free radicals are likely to take part in chemical reactions. Free radicals play an important role in combustion, polymerization, and many other chemical processes. A polar molecule is a molecule in which the centers of positive and negative charge distribution do not converge, making it electrically imbalanced. Polar compounds are highly soluble in other polar compounds, and virtually insoluble in nonpolar compounds. Oil is non-polar making it a hostile environment for the newly formed polar contaminants. The degradation by-products are repelled from the oil and will agglomerate with other contaminants and eventually settle out on metal surfaces. The primary collection areas for varnish within a lubricant system are the cooler heat exchangers, metallic surfaces in hot regions of the system or in tight clearance zones as in valves. Quantitative Spectrophotometric Analysis Quantitative Spectrophotometric Analysis is a technique of purposely isolating and measuring the specific degradation byproducts responsible for sludge and varnish formation. Spectrophotometric determinations assess the relative amount of compounds that absorb specific visible light wave regions. The process begins by treating the lubricant sample with a specific chemical mixture designed to isolate insoluble by-product material. Next, a separation process collects the varnish forming insoluble oil degradation by-products. The process concludes with a spectral analysis on the isolated by-product. The evaluation technique used can directly correlate the insoluble level to the varnish potential of the fluid. A 1 to 100 severity rating scale indicates the propensity for the lubricant to form sludge and varnish. Quantitative Spectrophotometry Used as a Predictive Tool Following are two cases in different industries where routine oil analysis failed to predict a varnish problem to illustrate the benefits of Quantitative Spectrophotometry. Case Study 1: Combustion Turbine Power Plant A large power plant in the southeast U.S. complained of severe varnish buildup on valves and filters. The plant had been experiencing unit trips due to sticking electro-hydraulic servo control systems. Each unit trip cost the plant thousands of dollars. The combustion turbine had a common 6,000 gallon lubricant sump used for four purposes: lubrication and cooling of the bearings and other parts of the lube system, seal oil, a fluid to energize the valves and actuators in the hydraulic circuit, and as a lift oil providing hydrostatic lubrication to the bearings during start-up. The lubricant in question was premium R&O Turbine oil with an ISO viscosity grade of 32. The color of the used oil appeared normal and the test results were similar to other used turbine oil samples. Figure 7: Picture of Used Turbine Oil from a Varnished Combustion Turbine The following analytical results showed no signs of a problem. Figure 8: Routine Analysis of Used Turbine Oil QSA® revealed that the turbine oil does have a high varnish potential, as shown below: Figure 9: Quantitative Spectrophotometric Analysis of Turbine Oil Sample Case Study 2- Hydraulic System used for Injection Molding The client, who operates large injection molding machines in the Central US, was complaining of sticking control valves and had experienced several unscheduled shutdowns. The plant manufactures HDPE bottles for use in many applications in a continuous operation. The system capacity was approximately 300 gallons and the fluid in use was less than two years old. The client operated several other identical injection-molding machines, of the same age and service time, without incident. The lubricant used in these systems is a premium AW 46 hydraulic fluid. The customer had recently contracted with an outside vendor to side-stream filter the oil to a particular ISO cleanliness level. The vendor spent almost an entire week onsite without success of meeting the target. The color of the oil appeared normal and the analytical results were similar to other used hydraulic oil samples. The cause of the elevated particle count in the 2-5 micron range could not be determined using existing data. Figure 10: Picture of Used Hydraulic Oil from a Varnished Hydraulic System Figure 11: Routine Analysis of Used Hydraulic Oil QSA® revealed that the hydraulic fluid does have a high varnish potential, as shown below: Figure 12: Quantitative Spectrophotometric Analysis of Hydraulic Oil Sample Summary Reliability Engineers and maintenance personnel depend upon the data that they receive from oil analysis to make informed preventative maintenance decisions. Both of the case studies illustrated in this paper have "world class" maintenance and reliability programs and have successfully been using oil analysis as a predictive tool for years. As was illustrated, routine oil analysis did not show any signs that there might be varnish in their systems. Quantitative Spectrophotometric Analysis revealed that the lubricants in both case studies did indeed have a high propensity to produce deposits, demonstrating the predictive capabilities of the test. Both operations now use QSA® to measure the varnish potential on all of their critical systems, expecting to take preventative measures on systems with a high varnish potential during planned maintenance intervals. Conclusion Varnish formation is a costly problem across a wide range of industries. Routine oil analysis does not typically indicate that varnish a problem exists, leaving wide gaps in a company's predictive maintenance and reliability programs. Historically, you would have to perform a visual inspection of equipment during an outage to find varnish. The authors presented two case studies in which routine oil analysis failed to identify a varnish problem, even though the customers had incurred costs due to varnish. Analysts, Inc. developed a testing procedure called Quantitative Spectrophotometric Analysis and demonstrated the methods usefulness as a supplement to conventional testing methodologies. Quantitative Spectrophotometric Analysis is the only predictive tool in industry that provides the lubricant user a means to evaluate and monitor the varnishing potential of their lubricants. Quantitative Spectrophotometry has proven itself as a valuable predictive tool to improve traditional oil analysis testing programs.