Life Cycle Tribology al. (Editors) (Editors) D. Dowson et al. 2005 Elsevier B.V. All All rights reserved reserved © 2005
Lubricants & Energy Efficiency: Life-Cycle Analysis R.I. Taylor", R.T. Dixona, F.D. Wayne" & S. Gunselb a b
Shell Global Solutions (UK), Cheshire Innovation Park, PO Box 1, Chester, CHI 3SH, UK Shell Global Solutions US Inc,Westhollow Technology Center, PO Box 4327, Houston, TX 77210-4327, USA
The use of optimised lubricants can lead to significant energy savings (i.e. reduced fuel or electricity costs) and also reduced CO2 emissions. Over the last 20-30 years, the drive to improve vehicle fuel consumption has led to the widespread availability of fuel economy lubricants (e.g. friction modified SAE 5W-30 grades compared to SAE 15W-40 grade lubricants). Over realistic driving cycles, vehicle fuel consumption can in some circumstances decrease by between 2-5% when fuel economy lubricants are used. For cold-start, short-trip driving, the fuel economy improvement may be even higher still. Energy efficient lubricants are less widely used in industrial applications, but the potential savings may be even greater. For instance, in hydraulic systems, friction is dominated by pipe losses and is therefore directly proportional to lubricant dynamic viscosity (whereas in an engine friction is proportional to the square root of viscosity). Hence changing the ISO viscosity grade, or moving to a higher Viscosity Index (VI) lubricant could lead to significant energy savings under cold-start conditions. Similarly, in gear lubricants, it is known that in moving from mineral base-stocks to more highly processed base-stocks, the decrease in the pressure-viscosity coefficient of the lubricant results in both decreased friction and decreased temperature rises. In all these examples, the energy efficient lubricants are generally more expensive than conventional lubricants, and a life-cycle cost analysis, which takes account of operating costs as well as the initial purchase cost of the lubricant, is required to bring out the true benefits of these products. When the cost of CO2 emissions becomes more apparent in future years, the life-cycle cost analysis will be weighted even more in favour of energy efficient lubricants.
1. INTRODUCTION Optimised lubricants can lead to significant energy savings. Jost1 reported that friction and wear accounted for roughly 10% of Gross National Product (GNP) for many developed countries, and that the use of good tribological principles, and the use of correct lubricants could lead to savings of the order of 1-2% of GNP. Clearly these are significant cost savings, and certainly worth striving for. Since the oil price shock of the 1970's, and more recently with the desire to reduce carbon dioxide emissions, there have been major drives to improve motor vehicle fuel economy. Many engineering changes have been made that have resulted in vehicles with improved fuel economy. The lubricant used in the vehicle can also influence the vehicle's fuel consumption, and there has in consequence been a move to friction modified lower viscosity lubricants. Industry standard engine tests (such as
the ASTM Sequence VI-B fuel economy test, and the CEC M i l l fuel economy test) are in place to check the fuel economy performance of a lubricant, relative to a reference oil. Friction modified SAE 5W-30 lubricants are now relatively commonplace, whereas a decade or so ago, the dominant lubricant viscosity grade was an SAE 15W-40. This paper discusses in detail the fuel economy benefits that may be achieved by using lower viscosity lubricants that contain friction modifiers. Energy efficient lubricants will also be beneficial for industrial lubricant applications. For hydraulic systems, friction is dominated by pipe losses, and so the friction losses are directly proportional to lubricant viscosity (whereas for an engine the friction is proportional to the square root of the viscosity). Therefore "cold start" friction losses are likely to be more significant in hydraulic systems than they are in internal combustion engines, and a
566 move to higher viscosity index hydraulic fluids (or to lower ISO viscosity grades) will lead to large energy savings. We also consider gear lubricants. Moving from gear oils formulated with mineral base-stocks to those formulated with low traction synthetic basestocks is often found to lead to both lower friction losses and lower lubricant temperatures. (Examples of low traction synthetic base-stocks include polyalphaolefins (PAO), severely hydroprocessed mineral oils (Group III), isomerised slack waxes such as XHVI™, and certain types of polyalkylene glycols (PAG) and esters). The lower temperature rises are due to the fact that gears operate in the elasto-hydrodynamic (EHD) regime, and the pressure-viscosity coefficient of the lubricant (a, in GPa"1) is important in determining friction losses and mineral oils have higher values of a than do synthetic base-stocks. Lubricants which give energy efficiency benefits are often more expensive than conventional lubricants. Therefore, a life-cycle cost analysis needs to be undertaken to justify their use. In other words, not only must the purchase price of the lubricant be considered, but so also must the impact on the operating costs. It is often the case that fuel (or electricity) cost savings outweigh the increased purchase cost. In future years, when carbon dioxide emissions will be more closely monitored, and carbon dioxide trading allowed, these costs would also need to be included in a life-cycle cost analysis. 2. FUEL ECONOMY ENGINE LUBRICANTS Figure 1 shows that vehicle fuel consumption has been improving over the last 20-30 years. Naturally, many engineering improvements have been made to achieve this improvement. The use of aluminium (rather than cast iron) engine blocks, the use of 4 valves per cylinder (rather than 2), lighter vehicles, smaller components, improved combustion technology (e.g. gasoline direct injection systems) have all played a role. It is also well known that lubricants can affect the vehicle fuel consumption2"5. In general, it is a good assumption that engine friction varies as the square root of the lubricant viscosity. Furuhama6 has reported that the friction
losses in the piston assembly vary as -\JJJCO , where T) is the lubricant dynamic viscosity (mPa.s) (calculated at a temperature representative of the piston assembly) and (0 is the angular speed (rad/s) of the engine. For journal bearings, under lightly loaded conditions, the Petroff equation7 suggests that the friction power loss would vary linearly with lubricant viscosity, as the equation below shows: C7
...(1) where F is the friction power loss (Watts), r| is the lubricant dynamic viscosity (mPa.s) appropriate to the bearing, a> is the engine's angular speed (rad/s), L is the bearing width (m), R is the bearing radius (m) and c is the bearing radial clearance (m). For a heavily loaded bearing, however, Taylor4 has shown that the friction power loss would vary as r|0 75:
...(2) For the valve train system, under fully warmed up conditions, the friction power loss is expected to increase as the viscosity decreases, since it operates in the mixed/boundary lubrication regime. Since the piston assembly friction is thought to dominate engine friction, for most of the engine operation, assuming that engine friction varies as the square root of the lubricant viscosity is not a bad approximation. Shayler et al8 have reported engine friction measurements, made on a motored 1.8 litre diesel engine. They report that the friction mean effective pressure (FMEP) of the engine varies with viscosity raised to the power n, where n lies between 0.190.35. This does not seem to agree with the value of 0.5 that we would expect from the previous discussion, but it is possible that there are viscosity independent parameters included in the FMEP measurement (accessories, pumps, etc) that dilute the viscosity effect.
Fuel Consumption (litres/100 km) 201 USA 18 16 14 12 Japan 10 Norway
6 4 2 0 1970
2% aai GolfdieselTDi 4% aai
Figure 1: Fuel consumption (litres/100 km) over the last 30 years ("aai" stands for "average annual increase") (This data has been compiled using Shell internal data)
If we assume that engine friction does scale with the square root of lubricant viscosity, then it would suggest that using a lower viscosity lubricant would lead to friction savings, which would result in lower fuel consumption. Table 1 shows typical kinematic viscosity values for different lubricant viscosity grades. Figure 2 then shows how the fuel consumption in an industry standard M i l l fuel economy test varies with lubricant viscosity. Figure 3 shows essentially the same results, but broken down by stage. It is clear that larger benefits are seen at the lower temperature stages of the test. Similar sensitivities to viscosity are seen in the US industry standard fuel economy engine test. It should be added that these tests are not just sensitive to viscosity. The addition of friction modifier additives is essential to pass these tests - such additives work by reducing friction in the mixed and boundary lubricated contacts in the engine.
SAE grade 10W-60 20W-50 OW-40 5W-40 1 OW-40 15W-40 OW-30 5W-30 10W-30 OW-20 5W-20 OW-10
Vk40 (cSt) 144 165.93 80 97.3 99.40 103.2 55.69 66.16 69.04 40.96 46.28 30.09
Vk100(cSt) 22 18.11 14.3 15.2 14.21 13.8 10.48 10.75 10.5 7.88 8.16 6.09
Table 1: Typical values of Vk40, Vk100 for different lubricant viscosity grades
5 4" i
feFM A With FM
HTHS Viscosity (mPa.s)
Figure 2: Sensitivity of M i l l Effective Fuel Economy Increase (EFEI) (%) (compared to RL-191 reference oil) to lubricant High Temperature High Shear (HTHS) viscosity
• Oil A • Oil B
3. ENERGY FLUIDS
\ 33 C
In 20 C
at the benefits of moving to lower viscosity engine and transmission lubricants (and axle greases) on the operating costs of a typical European truck. Finally, we note that the benefits will not just be lower fuel bills. A typical passenger car, in Europe, emits over 3 tonnes of carbon dioxide per year. (This calculation assumes an annual distance covered of 16,000 km, and a carbon dioxide emission rate of 190 g/km). If we assume that there are 25 million cars in the UK, the annual CO2 emissions (for the UK alone, just from passenger cars) are of the order of 75 million tonnes. If fuel consumption (and hence also carbon dioxide emissions) can be reduced by 5% by moving to fuel efficient lubricants, this would lead to an annual reduction in CO2 emissions of around 3.75 million tonnes. (We should also comment that the EU target for a fleet average CO2 emission level of 140 g/km will also result in significant reductions in CO2 emissions - and fuel economy lubricants are expected to play a part in helping OEMs to achieve these targets).
Figure 3: Breakdown of fuel economy benefit for the different portions of the CEC M i l l fuel economy test (the benefits are relative to an SAE-15W/40 reference oil, RL-191)
Although there is no industry standard test in place for heavy duty diesel engine fuel economy, similar considerations hold - i.e. we would expect lower engine friction with lower viscosity lubricants. For these vehicles, friction modifiers are less important since somewhat paradoxically heavy duty diesel engines operate in a more hydrodynamic lubrication regime than do passenger car engines3. In Section 5, on life-cycle cost analysis, we shall look
Hydraulic systems are widely used throughout the world. Earthmoving equipment, diggers, etc. are examples of mobile hydraulic systems, whereas plastic injection moulding machines are an example of an energy intensive process operating 24 hours per day in a factory environment. In both these applications the energy used by these systems depends on the hydraulic fluid used. In contrast to engines, where the friction losses depend roughly on the square root of the fluid viscosity, in hydraulic systems the friction is usually dominated by pipe losses, which vary linearly with viscosity. Therefore the potential for cold start energy savings in hydraulic systems is significantly greater than that for engines. In addition to energy losses, pump performance is also critical in hydraulic systems. If the hydraulic fluid is too viscous, then pump mechanical efficiency is too low. On the other hand if the lubricant has too low a viscosity, leakage within the
569 569 pump can occur, and the pump volumetric efficiency may become too low. One way to overcome these problems is to use a hydraulic fluid with a higher viscosity index (VI). Such a fluid has a flatter viscosity-temperature response. The idea behind the use of such a fluid is as follows illustrated in Figure 4. Low VI
Temperature Figure 4: Schematic viscosity temperature relationships (viscosity on a logarithmic scale) for a low VI and a high VI fluid. At low temperatures, the high VI fluid will have a relatively lower viscosity, leading to improved energy efficiency, whilst at higher temperatures, the high VI fluid will lead to relatively higher viscosities, leading to reduced pump leakage losses Therefore, the use of a higher VI hydraulic fluid, possibly combined with a change of ISO grade, can give significant energy benefits under cold-start conditions, and can also give volumetric efficiency benefits under high temperature conditions, compared to a low VI fluid. These effects do influence performance. In a hydraulically operated digger, it can take a few hours for the system to reach operating temperature, during which time much fuel is wasted overcoming friction in the hydraulic system. On the other hand, during the day, as the digger operates, the system can heat up, resulting in increased leakage losses and reduced volumetric efficiency of the pump. The use of a high VI hydraulic fluid could help to overcome both these issues. To conclude, some rough indication of the amount of energy savings that may be achieved by changing ISO viscosity grade is given. For a typical stationary hydraulic machine, operating at a
temperature of around 50°C, in-house Shell data indicates that changing from an ISO 68 hydraulic fluid to an ISO 46 hydraulic fluid would lead to electricity savings of up to 20%. At the 2004 Annual STLE meeting, data was presented13 which suggested that moving from an ISO 68 mineral oil to an ISO 32 synthetic oil resulted in an energy saving of 13% (this is an example using oil mist lubrication - the temperature of the oil was not given). Finally, we remark that changing the viscosity grade of the hydraulic fluid may impact on the friction and wear within the hydraulic pump. Naturally any decrease in lubricant viscosity must be assessed to ensure that the durability of the pump is not affected. 4. ENERGY EFFICIENT GEAR OILS In contrast to engine oils and hydraulic oils, gear and transmission oils operate for the majority of their time in the elasto-hydrodynamic (EHD) lubrication regime. It is known that gear friction losses are correlated with temperature rises of the gearbox lubricant14. It is also known that moving from a mineral based oil to a low traction synthetic based oil (where the low traction synthetic base-stock could be a polyalphaolefin (PAO), a severely hydroprocessed mineral oil (Group III), or an isomerised slack wax such as XHVI™, and certain types of polyalkylene glycols (PAG) and esters) results in lower gearbox oil temperature rises, and lower friction losses. The reason for this phenomenon is due to the detailed variations in the high pressure rheology for these synthetic base-stocks. Table 2 shows typical values of a for different base-stocks, from Infineum15. Base stock PAO 4 PAO 6 Group III, 4 cSt Group HI, 5 cSt Group I, 100 N Group I, 150N
a (GPa"1) at 60°C 10.6 11.9 11.9 13.4 15.5 16.7
a (GPa1) at 80°C 9.9 10.5 10.9 11.7 13.6 15.2
a (GPa"1) at 100°C 9.2 9.1 9.8 10.0 11.7 13.7
Table 2: a (GPa1) at different temperatures for different basestocks, according to Infineum15
570 570 Gunsel et al16 have demonstrated that in an EHD contact three separate lubricant properties must be considered for understanding and optimising oil film thickness and friction. These are: ambient viscosity, pressure-viscosity coefficient (a) and limiting shear stress. The first two properties determine the oil film thickness profile in the contact whilst the third property determines friction in the contact. It was found16 that, broadly speaking, there was a correlation between EHD friction coefficient and a. Clearly, there needs to be a check to ensure that moving to a lubricant with a lower value of a does not adversely affect the durability of the gears. 5. LIFE-CYCLE COST ANALYSIS Energy efficient lubricants are often more expensive than conventional lubricants. However, as explained in the previous sections of this paper, they should reduce operating costs since they will result in lower energy consumption. Therefore, rather than just concentrating on the initial purchase price of the lubricant, the customer should consider the life-cycle cost of using the product. This lifecycle cost would take account of: (1) the initial purchase price of the product, (2) the effect on operating costs, over the life time of the product, (3) any changes in costs due to different service intervals (eg the oil drain interval may be extended), (4) disposal costs. In future years, when CO2 emissions may be more closely monitored, and CO2 trading is permitted, any costs associated with these emissions should be included in the cost analysis. Figure 5 shows a typical life cycle cost analysis for a heavy duty truck. In the example shown, fuel efficient engine oil, gear oil, and axle grease are assumed to result in a 3% improvement in fuel consumption. The cost savings shown in Figure 5 are mainly the result of the lower fuel bill that arise from using more fuel efficient lubricants, and the example is based on typical European fuel costs (where fuel tax is relatively high). Clearly the life cycle cost analysis may arrive at a different result in a country with low fuel tax (such as the US). The same sort of cost analysis can be carried out for hydraulic machinery. In this example, the energy cost would typically be calculated from electricity
consumption, and so a move to energy efficient hydraulic fluids would be most likely in countries with high electricity prices (such as Japan). 6. FUTURE OPTIONS The reader may wonder why it is not possible to produce a lubricant which has a viscosity that is constant with temperature. For instance, water is such a fluid over the temperature range 0-100°C. However, water does not have any pressure induced viscosity thickening (i.e. a « 0) and a non-zero, positive, value of a is essential for good EHD lubrication. Also water is no good if the temperature of the system is less than 0°C or greater than 100°C. Is it possible to produce a hydrocarbon lubricant which has a flat viscosity-temperature relationship (and a non-zero value of a) ? A recent patent application suggests one way in which this may be possible17. The essence of the idea is that there are two oil reservoirs, one containing a lubricant which has been chosen to be optimum for "warm" operating conditions, and the other reservoir contains a diluent (essentially a lower viscosity base oil). As the temperature of the system decreases, more and more of the diluent is added to the lubricant, in an attempt to keep the viscosity constant as the temperature is lowered. It has been found that this approach sometimes allows two important properties to be controlled simultaneously, such as the viscosity and the friction (or traction) coefficient. Control of two properties requires careful selection of pairs of liquids. Clearly, this approach requires some method by which the diluent oil may be separated from the lubricant as the temperature rises and the patent explains how this may be done17. It is expected that if this approach works, the initial applications would be expected to be in the area of traction fluids for infinitely variable transmissions, and possibly in hydraulic applications. However if the same approach can also be used for engine oils, then the potential fuel consumption savings could be substantial, particularly since a large number of car journeys cover less than 5 miles, when the engine is not fully warmed up2.
571 571 MAINTENANCE COST SAVING CALCULATOR TRUCK DETAILS
Truck Make/Model Engine Make/Model Gearbox Make/Model Axle Make/Model
Cummins M11 Eaton Dana-Spicer
30 20 20
MAINTENANCE & OPERATING COSTS Annual Mileage, km 100,000 Avg fuel consumpt, km/1001 325 0.8 Avg Fuel Cost, $/litre 20 Labour cost, $/hr Off-road time for Oil Change, hrs 4 Loss of earnings, $/hr 30
Grade Recommended ODI, km Cost, S/l Brand Engine Standard Engine Oil 15W40 48,000 1.50 SAE90 80,000 1.60 Gearbox Standard Grease A 1.70 Axle Standard Grease B SAE90 80,000 Wheel Bearings Standard Grease C n/a n/a n/a Current Maintenance Interval, km 40,000 Lubricant Recommendation
Engine Super Engine Oil* 10W40 Gearbox Super Grease A* 80W 80W90 Axle Super Grease B* n/a Wheel Bearings Super Grease C* Proposed Maintenance Interval, km 'fuel economy grades Typical Fuel Saving (if applicable) ••I 3.0% I POTENTIAL Lubricant Cost, $ Labour Cost, $ Additional Up-Time Benefits, $ Fuel Saving, $
Maintenance Interval, km Cost, $/l 72,000 2.20 2.00 80,000 80,000 2.00 n/a
ANNUAL SAVINGS 12.50 100.00 150.00 738.46
Figure 5: Life cycle cost analysis for different lubricants used in a heavy duty diesel truck
7. CONCLUSIONS Lubricants have a role to play in improving the energy efficiency of both automotive and industrial machinery. The technology to do this is well understood, although care is needed, both by the machine designer, and by the lubricant formulator, to ensure that any reduction in lubricant viscosity does not result in decreased durability. One of the main reasons why such lubricants are not used more widely is that often lubricants are selected on the basis of their price alone, without much regard for the impact of the lubricant on operating costs. When a more sophisticated life-cycle cost analysis is performed, it is often found that energy efficient lubricants are more cost effective than conventional lubricants. Such a life-cycle cost
analysis will further favour energy efficient lubricants in future years when the true costs of CO2 emissions become more transparent, through CO2 limits on industry, and CO2 permit trading becomes more commonplace.
8. ACKNOWLEDGEMENTS The authors would like to thank Shell Global Solutions for permission to publish this paper, and would also like to thank their colleagues for useful discussions and support in writing this paper.
572 572 9. REFERENCES 1. The Jost Report, "Lubrication (Tribology), Education & Research", UK Dept. of Education & Science, HMSO, 1966 2. R.I. Taylor & R.C. Coy, "Improved Fuel Efficiency by Lubricant Design: A Review", Proc. Instn. Mech. Engrs., Vol 214, Part J, pp 1-15,2000 3. R.I. Taylor, "Heavy Duty Diesel Engine Fuel Economy: Lubricant Sensitivities", SAE 2000-01-2056 4. R.I. Taylor, "Lubrication, Tribology & Motorsport", SAE 2002-01-3355 5. R.I. Taylor, "The Development of Fuel Economy Lubricants", 8th Annual Fuels & Lubes Asia Conference & Exhibition, Singapore, 2002 6. S. Furuhama & S. Sasaki, "Effect of Oil Properties on Piston Frictional Forces", JSAE Review, pp 68-76, November 1984 7. G.W. Stachowiak & A.W. Batchelor, "Engineering Tribology", Elsevier Tribology Series, 24, 1993 8. P.J. Shayley, J.A. Burrows, C.R. Tindle & M. Murphy, "Engine Friction Characteristics Under Cold Start Conditions", Paper No. 2001-ICE-432, ICE-Vol. 37-3, 2001 Fall Technical Conference, ASME 2001 9. T.J. Sheahan & W.S. Romig, "Lubricant Related Fuel Savings in Short Trip, Cold Weather Service", SAE 750676 10. K. Fraelund, J. Schramm, B. Noordzij, T. Tian & V. Wong, "An Investigation of the Cylinder Wall Oil Film Development During Warm-Up of an Si-Engine Using Laser Induced Fluorescence", SAE 971699 11. D. Placek, S.N. Herzog, CD. Neveu, "Reducing Energy Consumption with Multigrade Hydraulic Fluids", Proceedings of the 9th Annual Fuels & Lubes Asia Conference & Exhibition, Singapore, 2003 12. L. Henriksson, CD. Neveu, S.N. Herzog, D. Placek, "Improving Pump Efficiency & System Performance by Selecting the Optimum Fluid Viscosity Grade", Scandinavian International Conference on Fluid Power, SICFP '03, May 7-9, 2003
13. H.P. Bloch & Per Arnold Elgqvist O., "TTD-IA: Strategies to Reduce Both Energy Costs and Maintenance Expenditures", STLE Annual Meeting, Toronto, May 2004 14. C. Wincierz, M. Muller, K. Hedrich & C. Neveu, "Influence of Viscosity Index Improvers on the Tribological Properties of Modern Transmission Fluids", 2nd World Tribology Congress, September 2001, Vienna 15. "Facing the Challenges of the Future in Crankcase Lubricants", article in Infineum Insight, Issue No. 10, June 2001 16. S. Gunsel, S. Korcek, M. Smeeth & H.A. Spikes, "The Elastohydrodynamic Friction and Film Forming Properties of Lubricant Base Oils", Tribology Transactions, Vol 42, No. 3, pp 559-569, 1999 17. WO 03/087650 Al, "Method of Controlling Lubricant Properties by Means of Diluting the Same", Inventor: F.D. Wayne