Thrust 1 - Efficiency

Title

1B.2: Surface Effects on Motor Start-up Friction

Project Leader

Prof. Ashlie Martini (Purdue)

Statement of Project Goals

The specific objective of this project is to develop and experimentally validate a model for static friction to improve the start-up efficiency of hydraulic components.  The resulting modeling tool will be the first experimentally validated start-up friction model that incorporates the real surface profile characteristics and lubricant effects.  At its conclusion, a successful project will result in a fundamental understanding of the relationship between characteristics of a component's interfaces and the friction it must overcome at start-up.  The modeling tools and corresponding experimental test rig developed for the project will be used to evaluate existing and novel (e.g. textured) surfaces to improve start-up efficiency in fluid power machinery.

Figure 1: Flowchart and pictorial representation of the static friction model

Project's Role in Support of the Strategic Plan

In the context of the CCEFP strategic plan, this project will contribute to overcoming the transformational technical barrier of efficient components.  Many hydraulic motors exhibit extremely poor start-up efficiency, forcing OEM manufacturers to specify larger motors than necessary, which in turn makes the overall cost and weight of the machines greater.  This project will provide an understanding of the physical mechanisms underlying static friction which will lead to specific approaches for minimizing start-up friction.  This research is relevant not only in terms of the start-up efficiency of fluid power applications, but also in terms of its fundamental focus on understanding static friction from a tribological perspective.

Description and explanation of research approach

Description and explanation of research approach
One of the challenges for equipment designed to operate intermittently is static friction.  Static friction, the resistance to the onset of motion, results in large inefficiencies at start-up and often requires engineers to oversize machine components for the sole purpose of overcoming start-up conditions.  Unfortunately, static friction is a physical phenomenon that is not yet well understood, particularly when dealing with complex, lubricated interfaces.  Consistent parameter definitions and reproducible experimental methods are necessary to understand static friction and its dependence on material properties and operating conditions.  Characterization of static friction is a critical step towards enabling machine interfaces optimized for improved start-up efficiency.

Static friction is typically quantified by the static friction coefficient, the ratio of the force required to initiate movement to the normal load.  Researchers have used various experimental and model-based techniques to try to measure and understand the dependence of this parameter on material and operating conditions.

The following is a summary of experimental methods reported in the literature.  One of the first approaches was the inclined plane in which a flat test surface is placed on top of a tilting flat surface, and the tangent of the angle at which the top test surface starts sliding is identified as the static friction coefficient (1-5).  In another early approach, a rotational device was used to measure both static friction and the deformation of test specimens under very light loads (~mN) (6).  More recently, static friction has been measured using an instrument referred to as the centrifugal friction apparatus which measures the friction between a block and a rotating disk (7,8).  Static friction between flat surfaces at higher loads has been studied using instrumentation with hydraulic cylinders introduced as a clamping mechanism (9,10).  Lastly, static friction of point contacts under light loads (~mN) has been measured using modified pin-on-disk instruments (11-13).  The applicability of previously reported experimental methods is limited by (i) restrictions on contact geometry, (ii) the range of accessible normal loads, and (iii) the ability to accurately determine the onset of motion.  Perhaps the most significant limitation of nearly all previously reported methods for static friction measurement is their inability to accurately and consistently determine the onset of motion.  Test standards to measure the coefficient of friction do not specify how this point should be identified (14).  Therefore, most studies utilize visual inspection or other indirect methods to determine when displacement begins and therefore at what point the static friction coefficient should be calculated.

The following paragraph summarizes model-based approaches for predicting static friction.  Typical static friction models consider interfaces at the asperity level using statistically generated profiles to capture surface effects.  The static friction coefficient is obtained as the ratio of the normal force input to the maximum shear force the material can withstand before sliding.  The first models assumed elastic deformation of the asperities and used Hertz contact theory for predicting the contact area (15).  Later, a model accounting for plastic deformation was proposed (16).  Shortly after, a model including adhesive effects in metallic contact was proposed (17) and was later extended to incorporate lubricant adhesion (18).  Subsequent models focused on expanding the theories proposed in (16-23).  Unfortunately, nearly all of these models assume the surface profile can be represented by a statistical distribution of the asperity heights, which cannot capture the behavior of a real surface with scratches, grooves and deformations that are typical of surfaces in hydraulic components.  In addition, most of these studies are focused on extremely small scale devices and so are based on assumptions that are not necessarily applicable to macro-scale components.

Thus, there are significant limitations in to both the experiments and models reported previously to study static friction. Perhaps the most critical issue is that these limitations preclude direct comparison of model predictions with experimental measurements.

In this project we are developing both models and experiments that address the issues described above.  The experimental test rig will be capable of handling multiple contact geometries, high normal loads, and will allow precise determination of the onset of motion.  The model will be incorporate real surface profiles into the calculation and will be designed to capture the physics underlying static friction.  In addition, the models and experiments will be developed synergistically such that the model predictions can be validated by experiments, phenomena observed experimentally can be explained by reference to the simulations, and both can form the basis for reliable predictive models describing start-up friction.

(a)    (b)    (c)  

Figure 2: (a) Schematic and (b, c) photographs of the apparatus for measurement of the static friction coefficient.

References

(1)  Sameshima, J. and Tsubuku, Y. (1937), "Studies on the oiliness of liquids. IV. Measurements of the static frcition coefficients by the method of inclination," Tokyo Imperial University, Vol. 12.

(2)  Aida, R. (1950), "Measurment of coefficient of static friction of metals," The Research Institute for Scientific Measurements.

(3)  Rabinowicz, E. (1971), Tribology Transactions, Vol. 14, pp. 198-205.

(4)  Rabinowicz, E. (1992), Wear, Vol. 159, pp. 89-94.

(5) Garcia J, Martini A, and Lumkes J, Tribology and Lubrication Technology, March 2010, p 17.

(6)  Whitehead, J.R. (1949) Royal Science Academy, Vol. 201.

(7)  Dunkin, J.E. and Kim, D.E. (1996), Wear, Vol. 193, pp. 186-192.

(8)  Benabdallah, H.S. (2007), Tribology International, Vol. 40, pp. 64-73.

(9)  Gassenfeit, E.H. and Soom, A. (1988), ASME Journal of Tribology, Vol. 110, pp. 533-538.

(10)            Xie, W., De Meter, E.C. and Trethewey, M.W. (2000), International journal of machine tools & manufacture, Vol. 40, pp. 467-488.

(11)            Etsion, I. and Amit, M. (1993), ASME Journal of Tribology, Vol. 115, pp. 406-410.

(12)            Ovcharenko, A., Halperin, G. and Etsion, I. (2008), ASME Jounral of Tribology, Vol. 130, pp. 1-6.

(13)            Chang, W.R., Etsion, I. and Bogy, D.B. (1988), ASME Journal of Tribology, Vol. 110, pp. 50-56.

(14)            D4918-97 (2007), D4917-97 (2007), D12894-08 (2008), G164-99 (2009), ASTM International, West Coshohocken, PA.

(15)            Greenwood, J. A. and Williamson, J.B. P. 1442, Dec. 1966, Proceedings of the Royal Society of London, Vol. 295, pp. 300-319.

(16)            Chang, W. R., Etsion, I. and Bogy, D. B. April 1987, ASME Journal of Tribology, Vol. 109, pp. 257-263.

(17)            Chang, W. R., Etsion, I. and Bogy, D. B. Jan. 1988, ASME Journal of Tribology, Vol. 110, pp. 50-56.

(18)            Stanley, H. M., Etsion, I. and Bogy, D. B. 1990, ASME Journal of Tribology, Vol. 112, pp. 98-104.

(19)            Kogut, L and Etsion, I. 2002, ASME Journal of Applied Mechanics, Vol. 69, pp. 657-662.

(20)            Kogut, L and Etsion, I. 2003, Journal of Colloid and Interface Science, Vol. 261, pp. 372-378.

(21)            Kogut, L and Etsion, I. 2004, ASME Journal of Tribology, Vol. 126, pp. 34-40.

(22)            Polycarpou, A A and Etsion, Izhak: ASME, April 1998, ASME Tribology Transactions, Vol. 120.

(23)            Yu, Ning, Pergande, Shaun R and Polycarpou, Andreas A. 2004, ASME Journal of Tribology, Vol. 126, pp. 626-629.

(24)            Garcia, J.M., Martini, A., and Lumkes, J.H. (2010), Paper 1(ii), Proceedings of the 6th FPNI PhD Symposium, West Lafayette, IN.

(25)            Garcia J, Lumkes J, Heckaman B and Martini A, Tribology Transactions, In Press.

(26)            Michael P, Burgess K, Martini A, Garcia JM, Bair S and Devlin M, Proceedings of the STLE/ASME 2010 International Joint Tribology Conference, October 17-20, 2010, San Francisco, California.