In the classical approach to elastohydrodynamic lubrication (EHL) a single parameter, the pressure-viscosity coefficient, quantifies the isothermal pressure dependence of the viscosity for use in ...prediction of film thickness. Many definitions are in current use. Progress toward a successful definition of this property has been hampered by the refusal of those working in classical EHL to acknowledge the existence of accurate measurements of the piezoviscous effect that have existed for nearly a century. The Hamrock and Dowson pressure-viscosity coefficient at high temperature requires knowledge of the piezoviscous response at pressures which exceed the inlet pressure and may exceed the Hertz pressure. The definition of pressure-viscosity coefficient and the assumed equation of state must limit the use of the classical formulas, including Hamrock and Dowson, to liquids with high Newtonian limit and to low temperature. Given that this problem has existed for at least fifty years without resolution, it is reasonable to conclude that there is no definition of pressure-viscosity coefficient that will quantify the piezoviscous response for an analytical calculation of EHL film thickness at temperatures above ambient.
The field of elastohydrodynamic lubrication (EHL) now has two very different approaches to the problem. For the first approach, two assumptions regarding the pressure, temperature and shear ...dependence of viscosity have been essential to the way that classical EHL developed over the last forty years.
1. The liquid in the inlet zone responds in Newtonian fashion.
2. The shear stress versus shear rate relationship of the liquid has the same form as the average shear stress versus average shear rate obtained from a traction curve.
The new, quantitative, approach employs viscosities measured in instruments which do not rely upon these assumptions. There has been a rapid succession of advances in understanding of film forming and friction under the new approach.
•Two assumptions regarding rheology have been essential to classical EHL.•Classical EHL has avoided using viscosity obtained from viscometers.•Primary measurements of viscosity would have allowed testing of these assumptions.•Real pressure-viscosity response does not support the assumptions.
With the uncertainty regarding the global energy future, the ability to lubricate concentrated contacts with sufficiently thick liquid films while minimizing friction is of extreme importance. The ...assumptions of classical elastohydrodynamic lubrication have remained unchanged since the early days. It has not been possible to test many of these assumptions without the measurement of the viscosity at elastohydrodynamic lubrication (EHL) pressures, and viscometer measurements have been ignored. One of these assumptions has been the equivalence of a traction curve to a rheological flow curve for the lubricant. This notion should have been discarded forty years ago, simply because it required the pressure–viscosity behavior to be unlike the behavior observed in viscometers. At the heart of the problem is the fact that the pressure within the EHL contact is not homogeneous and the liquid properties are highly dependent on pressure, making the contact a very poor rheology laboratory. These past failures must be avoided in the future.
The field of EHL (elastohydrodynamic lubrication) may be the only one in science in which a model for shear-dependent viscosity would be evaluated by means other than viscometer measurements ...
Numerical models for thermal elastohydrodynamic lubricated contacts of steel surfaces have always employed a value for the thermal conductivity of steel corresponding to its soft annealed alloy ...state. However, steel in elastohydrodynamic lubricated contacts is usually hardened. It has been known for more than a century now (not within the Tribology community though) that the thermal conductivity of steel could be reduced by a factor of more than two when it is hardened. Only recently did the Tribology community realize this “mistake”, of which the impact on friction predictions is investigated in this work. The mistake is found to lead to significant overestimations of friction in the thermo-viscous regime.
•TEHL simulations of steel contacts have always used the thermal conductivity of soft steel.•Steel in EHL contacts of test rigs or real-life machine elements is usually hardened.•Hardened steel has a much lower thermal conductivity than soft annealed steel.•Using the wrong value of soft steel thermal conductivity leads to an overestimation of friction.•Friction prediction errors are linked to lubricant film and bounding solids' temperature variations.
In the classical study of elastohydrodynamic lubrication (EHL) which does not employ real, measurable viscosity in analysis, the possibility of a glass transition has not been considered in many ...years. Indeed, the two rheological assumptions of classical EHL, the Newtonian inlet and the equivalence of a traction curve to a flow curve, would not have persisted so long had the pressure dependence of the viscosity been accurately stated. With the recent appearance of viscosity obtained from viscometers in EHL analysis, the possibility of a glass transition in the contact should be reexamined, especially for the fragile traction fluids. This article employs published data for a synthetic cycloaliphatic hydrocarbon to estimate the glass transition viscosity so that, when using real viscosities in EHL simulations, the state of the liquid may be assessed. Far into the glassy state the liquid should be treated as an elastic solid with a yield stress.
The reliability and efficiency of refrigeration compressors is dependent on the lubricating effectiveness of the refrigeration oil, which is necessarily a solution of oil and refrigerant. In this ...article, the development of a high-pressure viscometer that can operate at elastohydrodynamic lubrication (EHL) inlet pressures for oil/refrigerant solutions is explained in detail. New viscosity measurements are presented for two compressor oils to 1.2 GPa. Preliminary results are given for a polyol ester with two concentrations of R134a refrigerant to 0.35 GPa. These are the first measurements of oil/refrigerant solutions to such high pressure. Temperature-pressure-viscosity correlations are applied to all materials for use in modeling. Correlations are provided for the temperature and pressure dependencies of the viscosity of the oils, refrigerant, and mixtures. The definition of pressure-viscosity coefficient is a pressing problem for elastohydrodynamics.
Ninety years of high-pressure measurements with many different types of viscometers have shown that faster-than-exponential (super-Arrhenius) pressure dependence of viscosity is universal for ...glass-forming liquids and, therefore, all typical liquid lubricants. Dielectric spectroscopy at elevated pressure also yields super-Arrhenius response in the dependence on pressure of the primary relaxation time. In contrast, classical elastohydrodynamic lubrication (EHL) has gone to great lengths to ignore this phenomenon, including fictional accounts of the results of viscometry. As a result of this, classical EHL is unable to quantitatively account for one of the most important properties affecting friction at low sliding velocity, the low-shear viscosity. Differences in friction between similar liquids at low sliding velocity can be explained by their different inflection pressures. Some observed liquid response to shear stress at high pressure can be explained with the measured super-Arrhenius pressure dependence. It should be clear that, had classical EHL employed realistic pressure dependence of viscosity from its beginning, the field would have been in a better position today to solve engineering problems which involve the differences among molecular structures.
Reference liquids can be useful in the new quantitative approach to elastohydrodynamic lubrication calculations only if the properties of these liquids are accurately modeled. The pressures in the ...contact sometimes exceed the pressures of the available viscometer measurements, requiring extrapolation. The previtreous piezoviscous response is always faster-than-exponential with an apparent divergence pressure. Therefore, any model used for extrapolation must possess these characteristics and up to now published work in EHL, where extrapolation of the viscosity of squalane is performed, has not. To date, the only model that has been shown to accurately extrapolate to high pressure is the Hybrid model. This expression was applied to six reference liquids for use in quantitative investigations in EHL. Viscosity measurements for trioctyl trimellitate were extended to 1.1 GPa. The results obtained with high pressure viscometers are preferable over molecular dynamics results if they are different. Molecular dynamics calculations are not currently capable of reliable results for viscosity at very high pressures.
The recent rapid progress in the understanding and prediction of film thickness in elastohydrodynamic lubrication (EHL) has been the result of a shift in method from the classical treatment of ...viscosity as an adjustable parameter to measuring viscosity in viscometers. See for example, Hooke et al. (2021) 1. For researchers outside of EHL the latter method would appear to be the obvious best approach. However, measuring shear-dependent viscosity in lubricating oils is fraught with great difficulty because there are other phenomena which appear to be constitutive behavior. For example, viscous heating appears in viscometers as the decrease in viscosity with increasing shear rate and in many viscometers this was accurately described by the sinh-law model. The lack of understanding of thermal feedback in viscometers resulted in much wasted effort in EHL friction.
PDF
