Testing for the Evaluation of the Oils
Lubricants are essential in maintaining the condition of fixed or mobile assets. Any degradation in the lubricant’s properties impairs its ability to protect the equipments’ dynamic components. By having the ability to monitor oil condition in real time and on equipment, the user can change the oil at the optimum schedule and when unexpected contaminants are detected.
Contaminated oil can degrade and corrode dynamic gearbox components. Thus, oil changing timing becomes critical as, left unchecked, degraded oil can damage critical parts the longer it remains in the equipment. Contamination of oil can also cause seals to expand and develop fire hazards.
Currently, to ascertain an asset’s comprehensive operating oil condition, manufacturing personnel must send a sample to an oil analysis laboratory. Depending on the distance of the laboratory from the manufacturing plant, the process can take several days. Conversely, an on-line oil monitoring system will produce results instantaneously, not in days. In a scenario when contaminants are found in the oil by the lab, maintenance personnel at the plant must flush the system, change the oil, take a sample, and send it to the lab for analysis. If the lab determines the oil’s condition is still unacceptable, the maintenance personnel are required to repeat the above course of action until the oil meets specifications. Such an operation can potentially consume weeks, while having the oil analysis on the asset itself potentially reduces the process to no more than the time it takes to flush and change the oil for this iterative process.
Measuring the viscosity of oil is a rapid method of determining oil condition and is considered an important parameter in assessing asset readiness. Conventional mechanical and electro-mechanical viscometers designed primarily for laboratory measurements are difficult to integrate into the control and monitoring environment. As a consequence, many companies rely on decisions based on intermittent “snapshot” data acquired from periodic sampling, where conventional instrumentation can be affected by temperature, shear rate, and other variables.
Acoustic wave (AW) sensors offer a number of advantages over conventional mechanical and electromechanical viscometers as they are small solid-state devices that can be completely immersed in the oil, providing an instantaneous viscosity data stream for embedded OEM or end user spot–check applications. The sensors are unaffected by shock or vibration or by flow conditions, so they can be used in harsh operating conditions with a temperature range of -25°C to 125°C with a high degree of accuracy. At the same time, sensor measurements are not affected by particulates in the oil.
The viscosity sensor — which can compliment IR spectroscopy and other bulk property sensors — can provide instantaneous on-line viscosity and temperature data and offers universal plug-n-play connectivity for integration to control platforms. The sensors have been tested in actual commercial specified oils in order for a correlation function to be established between the ASTM methods acquired dataset and the sensor generated viscosity values; these correlation functions can be stored on any handheld for automatic conversion. The viscosity sensor is currently installed in commercial market applications, in rigorous environments where ROI benefits have been realized, and are now being evaluated for mobile and fixed assets where oil condition monitoring is of paramount importance.
The function of oil is to enable a mechanism to transfer heat from moving components, protect the various parts of the asset from external contaminants and provide a hydrodynamic layer for smooth operation. Given that the degradation of viscosity is an important indicator of these characteristics, it is important to not rely on snapshot data alone. For the proper operation of machinery, the viscosity of the oil has to be kept to the specifications determined by the manufacturer and plant personnel relevant to the operating conditions of the equipment. Knowledge of viscosity in real time provides a significant benefit to measuring the aging of oil, the ingress of contaminants during commercial operations, and to prevent incipient mechanical failure due to loss of oil lubrication properties.
This article describes a next-generation sensor that is packaged as an in-line, real-time threaded bolt solution that is targeted at embedded integration to fixed and mobile equipment. Also presented will be data from a customer that has tested the sensor on three specific oils for application into industrial gearboxes. The data has been correlated to lab measurements by this and other customers, thereby resulting in viscosity data that bridges the traditional dataset to the new methods of measurement.
Vectron, a designer and and manufacturer of acoustic wave (AW) based sensor products, has developed a unique method to offer a viscosity sensor with a wide dynamic range (air to several thousand cP) in a single sensor (Figure 1).
The Vectron ViSmart™ is a commercially available, robust, reliable, and cost-effective surface acoustic wave solid-state viscometer for integration into in-line, real-time monitoring and process control systems for scalable applications via a threaded interface design (Figure 2).
The sensor has no moving parts other than the atomic scale vibration of the surface, and due to the high frequency of the vibration — several million vibrations per second — it is independent of flow conditions of the liquid and immune to vibration effects of the environment. High temperature electronics are utilized that allow a very wide operating temperature range for the sensor.
The importance of these acoustic sensors lies in the distinctly different measurement principle. Whereas one class of mechanical devices measures kinematic (flow) viscosity and the other class measures intrinsic (friction) viscosity, the acoustic wave (AW) sensors measure acoustic impedance, (ωρη)1/2, where ω is the radian frequency (2πF), ρ is the density, and η is the intrinsic viscosity.
The viscosity measurement is made by placing the quartz crystal wave resonator in contact with liquid. The liquid’s viscosity determines the thickness of the fluid hydro-dynamically coupled to the surface of the sensor. The sensor surface is in uniform motion at frequency, ω=2πF, with amplitude, U. The frequency is known by design, and amplitude is determined by the power level of the electrical signal applied to the sensor. As the shear wave penetrates into the adjacent fluid to a depth, d, determined by the frequency, viscosity, and density of the liquid as d=(2η/ωρ)1/2, as depicted in Figure 3.
Acoustic viscosity is calculated using power loss from the quartz resonator into the fluid. The unit of measure is acoustic viscosity (AV) and is equal to ρη, (g/cm3 • cP) (density times dynamic viscosity).
The acoustic wave resonator supports a standing wave through its thickness. The wave pattern interacts with electrodes on the lower surface (hermetically sealed from the liquid) and interacts with the fluid on the upper surface. The bulk of the liquid is unaffected by the acoustic signal, and a thin layer — on the order of microns or micro inches — is moved by the vibrating surface. Also present is a proprietary hard coat surface that is scratch proof and abrasion resistant, which allows the sensor to be operable in extreme environments and enables the AW sensor to be a suitable candidate for oil condition based monitoring applications in mobile and fixed asset markets.
Testing for the Evaluation of the Oils
Significant testing as been accomplished by a commercial customer in order to ascertain the performance for the solid-state viscometer. The tested oils are Mobil SHC XMP 320, Kluber GH 6-220, and Aral Degol BG 68, and they represent a wide variety of viscosity values and lubrication characteristics. The viscosity values for the oils are measured at ASTM approved rheometer equipment (at 512 1/s shear rate) and with the solid-state sensor over a temperature range from 20 to 100°C. Based on this data, functions are generated to interpolate the viscosity for intermediate temperatures.
The goals of the customer testing are twofold: first, to observe the performance of the sensor over a temperature and validate that the sensor can differentiate between the oil types and be able to track the change of viscosity over temperature; and second, to verify if the sensor can provide the same data as from a lab instrument via the implementation of a correlation function.
The Vectron low shear solid-state viscosity sensor measures the acoustic viscosity (AV), which is the product of dynamic viscosity and mass density. Dynamic or kinematic viscosity is more commonly used in industry. The goal by customers is to establish correlation between acoustic viscosity and kinematic viscosity.
Correlation with All Oil Samples
The change of viscosity as a function of temperature for the three oils is presented in Figures 4-6. It important to keep in mind that Mobil SHC XMP 320 is a synthetic gear oil, Kluber GH 6-220 is a synthetic gear oil for spur, worm, and planetary gears, and Aral Degol BG 68 is a zinc-free, closed loop transmissions (CLP) type gear oil.
It is seen that the sensor can track the viscosity change as a function of temperature and can also differentiate between the sensors. This performance characteristic clearly demonstrates that solid-state viscometers can operate in these environments and be an important tool for the industry in its efforts to bring oil condition monitoring online for real-time decision making.
In order to create the correlation function, a simple methodology can be employed, as shown in Figure 7. The key to the methodology is:
• Acquiring the lab instrument data for the viscosity (and density) for the oil either via the manufacturers’ specification sheet, or by independent testing (needs to be done only once).
• Converting the native measurements acoustic viscosity for the solid-state sensor to kinematic viscosity with the relationship noted above.
• Plotting the lab and solid-state sensor kinematic viscosity value and observing the accuracy of fit.
The results for the three oils are demonstrated in Figures 8-10. As seen, the R-squared value for the oils indicates an accuracy level of 97 percent or better. Employing the correlation function in a database that can be embedded into any host control platform, manufacturing personnel can track the performance of the oil as a function of its viscosity value and take preventative or corrective action as deemed necessary.
It is also important to note that correlation functions do not need to be created for scenarios where a given relative shift from an established baseline value is deemed acceptable for purposes of conducting oil condition monitoring activities. If Vectron does not have any information about oil type, commercial testing has shown that with global correlation functions that are specific to the families of oils, functions with correlation R-squared values of 0.9107 and better can be created.
The conclusions that can be drawn from this data and the customer testing are:
• There is correlation between the lab method and Vectron viscosity sensor for each oil (or group of oil), and that a library of “fresh oil” correlations is practical.
• The Vectron ViSmart™ viscosity sensors operate at repeatable shear rates that are relevant to the assets being lubricated under normal operating conditions.
• The ViSmart™ sensors offer acceptable correlation to lab measurements at these shear rates.
• The correlation created from virgin oil provides the reference points from which changes in viscosity due to aging and contamination of the oil can be easily determined.
• Deviations of the oil from a predefined interpolation function at any temperature is a significant means of screening oil quality and is more accurate than “compensating” to 40°C.
• Evaluating deviations in “acoustic viscosity” is of comparable value to using kinematic viscosity. That is, independent and accurate knowledge of density is only important for correlation between on-site sensor testing and lab data.
The viscosity sensor can indeed provide a measurement of the resistance of the lubricant to flow. Changes in viscosity level indicate contamination by having the incorrect oil, fuel, or oxidation by-products. In fixed or mobile diesel assets, fuel contamination reduces the oil’s viscosity and flashpoint temperature and significantly reduces its load carrying ability. A high fuel dilution over a short period of time, or a moderate fuel dilution over an extended period of time, can severely damage oil wetted components (bearings, gears, pistons, etc.). In addition, fuel dilution promotes other failure mechanisms, including:
• Increased wear of oil wetted components
• Lubrication breakdown and component seizure
• Increased oil oxidation, sludge, and deposits
Keeping in mind that solid-state viscometers leverage standard semiconductor manufacturing processes, the results are commercially available products that are robust and reliable, high quality to yield repeatable products, and scalable for embedded applications where cost, functionality, and space limitations are paramount considerations. With these benefits, solid-state viscometers are an additional tool the gear industry can utilize in addressing end-customer needs. Such efforts yield the addition of value-added features that extend the life cycle of the components and equipment and increase operational efficiency for the manufacturing industry at large.