Tribology ­ meeting the demands of internal combustion engine design

Citation

Wilson, B. (1998), "Tribology ­ meeting the demands of internal combustion engine design", Industrial Lubrication and Tribology, Vol. 50 No. 5. https://doi.org/10.1108/ilt.1998.01850eab.002

Tribology ­ meeting the demands of internal combustion engine design

Tribology ­ meeting the demands of internal combustion engine design

Keywords: Combustion, Institution of Mechanical Engineers, Tribology

The second seminar on engine tribology in 17 months organised by the Tribology Group of the Institution of Mechanical Engineers (I. Mech. E.)was held on 24 March 1998 in London. The meeting, co-sponsored by the Combustion Engines Group and the Automobile Division of the I. Mech. E., attracted around 40 delegates. The event was also supported by the European Community through the Leonardo da Vinci programme and the EUROMOTOR training network. Nine papers were presented for discussion on topics including:

• •

  lubricant modelling at Shell Research;

• •

  predictive techniques in bearing design (T&N Technology);

• •

  full toroidal traction drives (Torotrak Developments);

• •

  oil sampling from piston ring packs (De Montfort University);

• •

  two papers on materials for engine components, one from the Bolton Institute and the other from Neale Consulting Engineers

• •

  modelling of piston ring performance (University of Leeds);

• •

  rotary diesel fuel injection pump distribution rotor tribology (University of Leeds);

• •

  electronic information sources at the I. Mech. E.

Lubricants

After a brief introduction by Professor W. Rowe, the chairman for the morning session, Dr Ian Taylor described some of the lubrication models developed at the Shell Research and Technology Centre, Thornton, to enable prediction of the performance of internal combustion engine lubricants. They used the Non Equilibrium Molecular Dynamics equation to simulate the friction of Newtonian oils under elastohydrodynamic lubrication conditions of high shear stress and shear rates. For multigrade oils they used a White-Metzner model to explore viscoelastic effects in dynamically loaded plain journal bearings. Models for the transition from hydrodynamic to boundary lubrication have been developed and applied to engine systems. These theoretical models have been checked against the results of tests on petrol and diesel engines, including the Mercedes Benz M 111 E engine for valve train performance (see Table I). Using the results obtained, Shell has developed a 5W/20 engine oil that was relatively unaffected by rate of shear and which performs as well as a US fuel efficiency OW/40 oil. This product has been launched in Germany although its viscosity is lower than that required by existing specifications. The addition of polymers to oils has been found to give increased load capacity for the same viscosity.

The relatively small contribution of the valves to diesel engine friction is possibly the reason that friction modifier additives are not very effective in diesel engine oils. In MB M111 E20 petrol engine tests the power loss due to friction was 1.5kW with a typical 15W/40 oil.

Professor Malcolm Fox, De Montfort University, Leicester, followed up his presentation at the November 1996 seminar by giving some more results from his project to sample oil from the ring packs of a CAT 3406 B diesel engine. The work is sponsored by the US Department of Energy as part of an engine efficiency and emission reduction project. He found that increasing the engine speed from 750 to 1500rpm tended to reduce the oil and gas flows past the piston rings. The measured flow rates were from 2 to less than 0.05 grammes per hour in the form of an aerosol.

Bearings

Dr Hao Xu, T & N Technology, Rugby, outlined recent developments in the performance prediction techniques used in the design of automotive engine bearings, especially heavily loaded connecting rod bearings. Rapid predictive techniques such as the Mobility and Impulse methods have been improved to include some of the features of modern engine designs and have been integrated into computer packages. The flexibility of computational numerical methods and the power of modern computers have enabled the inclusion of more design features and more realistic boundary conditions. The introduction of elastohydrodynamic lubrication to engine bearing analysis has had the most significant impact on the understanding of engine bearing lubrication. Housing deflection under load has a large effect on the performance of connecting rod bearings in modern engines. Many problems that could not be explained by conventional predictive techniques can now be understood. The current challenge is how to interpret the numerical predictions and relate them to engine bearing problems.

Materials

Dr Keith Holmes, Faculty of Technology, Bolton Institute, gave an overview of the demands made on modern engines and how they affect the tribological design and materials selection (see Table II).

Michael Neale, Neale Consulting Engineers, enlarged on the topic concentrating on the characteristics required of materials and surfaces for crankshaft bearings, cylinder liners and cams and followers.

Piston rings

Dr Martin Priest, University of Leeds, described ways of bridging the gap between mathematical models and practical piston ring performance prediction. Areas of concern are friction loss, oil consumption and blow-by emissions. Wear life is not a problem. The shape of the ring face profile is critical, as has been provided in tests using a Caterpillar 1Y73 engine. A worn ring was found to operate with hydrodynamic lubrication over 93 per cent of the cycle compared with only 59 per cent with a new ring. During the other 7 or 41 per cent of the cycle mixed or boundary lubrication regimes exist. It is possible that there is elastohydrodynamic lubrication of the piston rings on the highly loaded expansion stroke after firing. There is need for a design method that includes the optimum ring profile and surface topographies of both ring and cylinder wall for the operating conditions. A basic problem is that these optima change with changes in conditions.

Diesel fuel injection pumps

Dr David Jilbert, Castrol International, described initial work in a new research area at Leeds University sponsored by Lucas Diesel Systems. He reminded delegates that Rudolf Diesel's 1893 diesel engine used an air blast fuel injection system. This was superseded in the 1920s by a cam operated "jerk pump", later developed by Robert Bosch into the "in-line" pump in 1927. This had a separate pumping element for each cylinder. The rotary or distributor pump developed in the 1950s offered a cheaper, lighter and more compact alternative to the in-line pump. Today "unit injectors" and the "common rail systems" are the fuel injection pump competitors of the rotary pump.

The fuel distributor rotor is a central element of the rotary fuel pump. It distributes precise quantities of high pressure diesel fuel sequentially to each fuel injector. Each matched pair of 19mm diameter rotors and sleeves has a diametral clearance of only 2mm, roundness within 0.4m and a surface roughness of 0.04m Ra. Diesel fuel at 200 to 300 bar pressure is the only lubricant. Speeds of rotation are up to 2,100rpm. Work to date on the test rig has looked at the effect of surface roughness, diametral clearance, speed and pressure on the rotor orbit measured using capacitance gauges. It is desirable for the rotor to remain as central in the bore as possible because leakage increases with the cube of the displacement. In the tests the optimum clearance in terms of film thickness was found to be 3mm. The mechanical and thermal stresses in fuel injection pumps keep increasing as high speeds, operating temperatures and pressures (up to 2,000 bar) are experienced, not to mention changes in fuel composition which reduce its boundary lubricating ability.

Infinitely variable transmission (IVT)

Rather out of place in a symposium on internal combustion engine design, the presentation on the Torotrak toroidal traction drive (TTD) was relevant because its commercial success depends on its acceptance as an automotive transmission. TTDs were offered by Austin and considered by General Motors in the 1930s. They are of interest today because they offer up to 20 per cent improvement in fuel consumption by allowing the engine to operate under optimum conditions at all times. The Torotrak full toroidal variator consists of two sets of three rollers sandwiched between discs with toroidal profiled faces. Torque is transmitted by shear of the fluid trapped between the rollers and discs. A special fluid with a high traction coefficient of 0.05 to 0.06 is needed as the lubricant. Overall efficiency of the toroidal drive is 91 to 92 per cent with the heat generated dissipated by low pressure traction fluid sprayed on to the running surfaces. A large capacity disc machine with two 65kW motor drives is to be used for further development work under realistic operating conditions. The target market is cars with an engine size greater than 2 litres, leaving the small (cheap) car market to the Van Doorne metal push belt continuously variable transmissions (CVT). Unlike the CVT, toroidal transmissions do not need a clutch or torque converter for starting, making for simplicity, reduced weight and lower production costs. Torotrak expect their transmissions to be fitted first in year 2003 model cars.

Electronic information sources in tribology

The I. Mech. E. library has online databases of information and can carry out searches on request. Fluidex produced by Elsevier is particularly relevant to tribology. On the Internet there is Tribology Information Service edited by Dr Dwyer-Jones, Sheffield University, and Tribology, produced by David Allsop, Cambridge University.

Tribology listserver, supported by the Swiss Federal Institute of Technology, Lausanne, provides a discussion forum for tribologists. The contact at the I. Mech. E. is Mrs Liz Gooday, Information Officer.

Comment

This seminar compared unfavourably with the packed and lively meeting held in Birmingham in November 1996. On that occasion most of the papers emphasised practical aspects and were of greater current interest, so stimulating debate that had to be cut short. Perhaps this was also because the venue was close to the centre of the UK motor industry.

A report of that meeting was published in the March/April 1997 issue of this journal.

Bill Wilson