THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47th St., New York, N.Y

THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47th St., New York, N.Y. 10017 98-071 -358 The Society shall not be responsible for statements or opinions advanced in papers or discussion at meetings of the Society or of its Divisions or Sectorzs, or printed in its publications_ Discussion is printed only if the paper is published in an ASME Journal. Authorization to photocopy for internal or personal use is granted to libraries and other users registered with the Copyright Clearance Center (CCC) provided $3/article or $4/page is paid to CCC, 222 Rosewood Dr., Danvers, MA 01923. Requests for special permission or bulk reproduction should be addressed to the ASME Technical Publishing Department. Copyright C 1998 by ASME All Rights Reserved Printed in U.S.A. DETERMINATION OF THE REALISTIC TURBOCHARGER EFFICIENCY WITH PULSATING GAS-FLOW COMPARED ON A 4-CYLINDER ENGINE Ferdinand Trenc Matej der Francisek Bizjan Department of Mechanical Engineering University of Ljubljana Ljubljana Slovenia ABSTRACT Small single or twin entry radial turbines are mostly used to drive compressors of the turbocharged internal combustion engines. There are two general possibilities to feed the turbine of a four-stroke, 4-cylinder turbocharged Diesel engine: 1) by preserving most of the available exhaust kinetic energy, or 2) by mixing exhaust pulses from all cylinders in one common manifold. In the first case, better utilization of the dynamic pulse energy increases efficiency of the turbine; highly unsteady mass flow of the exhaust gasses, on the other hand, and thus periods of partial exhaust flow admission at the turbine inlet simultaneously reduces this gain in the turbine efficiency. More steady mass-flow of the exhaust gasses is created in the case of the exhaust system 2), however some kinetic energy is lost during the mixing phase in the common exhaust manifold. Calculation of the overall turbocharger and turbine efficiency is normally based on average values of the measured pressures and temperatures. As the result apparent efficiencies are obtained; the more the flow is pulsating, the bigger is the difference between the real and the apparent efficiency. The ratio between these two efficiencies is known as the energy pulsation factor 3. It depends generally on the "pulse intensity"- pressure deviation from its mean value, shape of the pressure pulse, number of the individual pulses feeding separate gas turbine inlets, turbocharger, and can be successfully used to determine real efficiency of a turbocharger and to define some working parameters of the engine. A field of 0 factors for different engine running conditions and for the 4-cylinder engine with 2-cylinder group pulse system (rarely applied), and the commonly applied exhaust system with 4-cylinder group and moderate pressure fluctuation is presented in the paper. Influence of the dynamic exhaust temperatures on the 0 value is discussed as well. Ale Hribernik Department of Mechanical Engineering University of Maribor Maribor Slovenia NOMENCLATURE T1 air temperature at the compressor inlet T3 instantaneous exhaust gas temperature at the turbine inlet T3average measured exhaust gas temperature at the turbine inlet ai relative length of particular pulse energy segments at the turbine inlet based on crank angle increment cp specific heat at constant pressure h5t total available enthalpy of the exhaust pulses at the turbine inlet Nhsr instantaneous enthalpy value of the exhaust pulses at the turbine inlet m mass flow pi air pressure at the compressor inlet P2 air pressure at the compressor outlet p3 measured dynamic exhaust gas pressure at the turbine inlet p3 measured average exhaust gas pressure at the turbine inlet P4 measured average exhaust gas pressure at the turbine outlet rc specific heat ratio c p/c,. overall turbocharger efficiency 77TC p correction factor of the exhaust pulse energy air air - compressor side exhaust gasses - turbine side sumation factor of the subsequent pulse energy segments max maximum value minimum value $ based on static pressure value Presented at the International Gas Turbine & Aeroengine Congress & Exhibition Stockholm, Sweden June 2-June 5, 1998

INTRODUCTION Two different exhaust systems are generally applied to serve a gas turbine of a four-stroke, 4-cylinder turbocharged Diesel engine. The first has a more sophisticated design with two pairs of selected cylinders, directed by the firing order and having particular exhaust periods shifted by 360 0 CA, filling two short exhaust manifold branches and two turbine entries. The second system is commonly used by the 4-cylinder engines because of its simpler design: four engine cylinders feed one relatively larger capacity exhaust manifold and one single turbine entry. In the first case, a highly nonsteady pulsating flow is transmitted to the turbine without losing too much of its kinetic energy, to the turbine. In addition two separated manifold branches with two particular turbine inlets prevent mutual interference between different cylinders especially during the fmal sequence of the exhaust period. Due to the pressure pulsation from the two cylinders in one manifold branch, this system is also called "two-pulse" system. On the other hand, mixing process of a 4- cylinder gas flow within one common exhaust manifold results in a more stable flow with less intensive pressure fluctuations and the system is therefore called constant pressure system. Both exhaust systems together with principal pressures and temperature locations (according to the equation (1)) are schematically presented in Fig. l; the 2-pulse system with a twin entry turbine on the left, and the constant pressure common manifold system with the single entry turbine on the right side. Turbocharger and turbine efficiency can be determined from the measured data. However the nature of the equipment normally used for measurements in an I.C.Engine laboratory and highly unstable pressure and temperature situation in the exhaust system lead to "apparent" efficiencies that are mostly far from realistic values. Unfortunately there is a non-linear relation between the magnitude and the shape of the pressure pulses and the energy pulsation factor 0, determined by Zinner (1980), also controls the power balance of the compressor and the turbine. Different types of pulsating flows with 3, 2, or 4 cylinders feeding one manifold, together with particular designs of the engine components, exhaust system, and running conditions give different values of the factor (3. There is a shortage of the available 0i factor data in the existing literature, especially for the 2-pulse and for the constant pressure system of 4-cylinder engines. Zinner (1980) described mechanisms and proposed methods to determine quantitatively factor (3 for steady and pulsating flow of the exhaust gasses. Most of the presented (i values correspond to the exhaust system of a 6-cylinder engine, where three selected engine cylinders feed one common exhaust manifold branch. Limited information on the 2-pulse exhaust system and its influence on the turbocharger and engine performance was given by Watson et al (1984). Optimistic results obtained during research work on a prototype 4-cylinder, air-cooled, 7 litre displacement, turbocharged and aftercooled Diesel engine applying 2-pulse exhaust system were obtained and reported by Hribernik et al (1995, and 1997), and by Trenc et al (1997). The authors concentrated their work also on finding out the influence of the applied exhaust system on the turbocharger operation and on quantitative determination of 0 factor. This in turn made calculation of the real turbocharger efficiency very simple and FIG.1 2-PULSE (LEFT) AND CONSTANT PRESSURE (RIGHT) EXHAUST SYSTEM made possible better comparison of the overall turbocharger and turbine efficiency data for different exhaust systems and the same baseline exhaust driven turbocharger. More realistic engine performance data can be predicted during the optimisation procedure of an engine, taking into account different exhaust systems. Fig.2 represents results of measurements performed on the above mentioned prototype engine for the 2-pulse and the constant pressure exhaust system. Relative average pressure difference (between the air boost pressure p2 and the back-pressure in the exhaust manifold p3) is often used as the measure of the average engine scavenging capability. Efficiency of the turbocharger is implicitly contained in the exhaust gas temperature T3 (parameter in Fig. 2), according to the turbo power balance - equation (1). Too large exhaust temperature differences in Fig. 2 between the observed two systems are the direct consequence of the uncorrected apparent turbocharger efficiency. Detailed analysis of the scavenging process and other engine parameters showed that the results from Fig.2 should be either taken into account only as a guideline, or should be corrected by determining energy pulse correction factor for both exhaust system. As the consequence, it is therefore resonable to calculate real turbocharger efficiency and also real mean pressure differences from the compressor and turbine power balance, equation (1). DETERMINATION OF THE ENERGY PULSATION FACTOR Overall real turbocharger efficiency can be calculated from the measured data and by applying equation (1). * may. T s cp. s TITC 6 muh T3 CP,ezh. z.,-1 P2 " ' [(j l] P1 I [* x^ ]] -1 (1) Energy pulsation factor 0 must be determined for different turbochargers, engine running conditions, and for different exhaust systems, reports Zinner (1980). If 0 is unknown only apparent turbocharger efficiencies based on average measured pressure and temperature values can be determined. In this case

c.i 0 40 - Single entry manifold (4 cyt/man.) Two branch manifold (2 cyl./branch) i 0.30 Cp T3, 1 P41 l 2 \ x^,+l I * s P4 Ida P4 I : P31 s ai 1J li tl 3V 0.20 OO K- -1l " r sc st s npa '^' Pa '^ * pj+ *a 0.10 \ ^K P3) p 3 r1 fair -1/./ 73^ r.00 `\K \ \ K^ (3) -0.10 CO X 20 x.40 1.60 1 B0 x (10 9 7 (P2/P l ass FIG.2 RELATIVE ENGINE SCAVENGING CAPABILITY FOR TWO EXHAUST SYSTEMS comparisons between different engines (although exactly the same turbocharger group is applied) are not possible. Real turbocharger efficiency is therefore influenced by the design of the exhaust system (factor 13), compressor and turbine pressure ratios, mass flows and inlet temperatures. In fact dynamic pressures and temperatures during the engine working cycle are replaced by average (measured) p and T values. Factor 13 is the function of the applied exhaust system and depends generally on the dynamic pressure and temperature history in the exhaust manifold. According to Zinner (1980) 3 is defmed as the ratio of the sum of real (dynamic) specific exhaust energy and the average available specific exhaust energy (calculated and based on the mean measured pressures, temperatures, and mass flows), and is defmed by the equation 2.,6n (2) h,., s Specific means that adequate mass portion of the exhaust gasses should be used together with the appropriate dynamic portion of the pulse energy. Equation (2) can be further expanded into its final form - equation (3). It should be noted that mass and energy flows are expressed by average and instantaneous values of pressures and temperatures at the turbine inlet: p3, p3.;, T3, and by the turbine outlet pressure p 4. whereas a. represents relative length (duration) of a particular section during the entire engine working cycle (7200 CA). If isothermal processes were assumed in the exhaust manifold, equation (3) would change to a simpler form and there would be no need for sophisticated measurements of dynamic gas temperatures. An error is certainly introduced by this simplification and will be discussed later. DISCUSSION OF THE RESULTS In addition to the above described 4-cylinder turbocharged Diesel engine an 82 mm diameter compressor, driven by a radial turbine was used for the experiments. Appropriate size of single and twin entry turbine housings were alternatively selected to assure approximately the same compressor boost pressure for the two exhaust systems, and for better comparison of the engine results, reported Trenc et al (1997). As mentioned before, pressure time history at the turbine inlet is required for different engine working conditions and different exhaust systems to determine 0 values. A typical (measured) pressure distribution for constant (common exhaust manifold), 2- pulse exhaust system (one manifold branch), at the engine peak torque conditions is presented in Fig.3. As the consequence of better preservation of the available exhaust pulses, higher pressure peaks, displaced by 360 CA were detected in particular exhaust manifold branches of the 2- pulse system. Lower pressure peaks of the same exhaust system (caused by the pressure wave reflections and the interference of partial flows in the twin-flow turbine housing) do not significantly affect the final sequence of the exhaust stroke in the adjoining cylinders. Uniform pressure fluctuation is, on the other hand, typic.l for the common-manifold, constant pressure system. Relatively high pressure peaks close to the TDC position (valve overlap period) can substantially affect scavenging process in neighbour cylinders, reported Trenc et al (1997). Over 40 dynamic pressure diagrams were analysed together with other engine and turbocharger data for different engine speeds and loads for each of the described exhaust systems. Computation of the 03 factor was performed according to the equation (3); relative lengths a. of particular segments in the computation process were limited to 1 0 CA. In the first calculation, step static to static pressure ratios were taken into account in the exhaust manifold and isothermal situation, based on measured mass-averaged T3 values was assumed. According to Zinner (1980) this assumption introduces an error of a few percent in (i value. 0 values were computed for both exhaust systems and are presented in the engine performance diagram, Fig. 4 a and b. Largest 0 values correspond to the lowest engine speeds and the highest loads, where relative exhaust pressure amplitudes reach their peak values. From both diagrams it can also be concluded that 0 approaches 1.0 for the constant pressure system at the engine rated power. Values of apparent and real turbocharger 3

1.8 1.7 1.6 1.5 1.4-1.3 1.2 1.1 _:" - lu 1II I11i ^ U ri u r/^ 1 r^ IILI I1 IiiI U 2 1,9 1,8 1,7 1,6 0.9 0 90 180 270 360 450 540 630 720 (p ['CA] FIG. 3 PRESSURE DISTRIBUTION IN THE EXHAUST MANIFOLD FOR CONSTANT AND 2-PULSE SYSTEM AT PEAK TORQUE; MEASURED efficiencies coincide there. Maximum absolute p values of the 2-pulse system are almost 30 % higher in comparison to those of the constant pressure system. 1,5 beta 1,4 1,3 1,2 1,1 1 0 1 2 3 4 (P 3ma^C P7m ir)i'(p3m -P4) FIG. 5 COMPARISON OF R VALUES FOR DIFFERENT EXHAUST SYSTEMS AND DIFFERENT AUTHORS 0 1.50 a) 2-pulse exhaust system s 1. 0 b) constant pressure system FIG.4 (3 VALUES FOR TWO EXHAUST SYSTEMS AT DIFFERENT ENGINE RUNNING CONDITIONS 1 In the second step of the calculations, isothermal situation in the exhaust manifolds was replaced by the calculated dynamic exhaust temperatures ( T3,; ). Isentropic change was assumed (large mass-flows in short and narrow pipes) in this case. Results of comparisons agree with the above mentioned statements of Zinner (1980): "isentropic" 0 values are by 4.9% higher at the peak torque engine conditions and for the 2-pulse system in comparison to the "isothermal" case. This difference was reduced to 2.2% by the constant pressure system and the same engine operating conditions. At the engine rated conditions the 2-pulse system and isentropic temperature change led to 2.0% higher 0 values. This change reduced to only 0.5% when the constant pressure system and engine rated power was concerned. In the third step, total-to-static turbine pressure ratio (p3t/p4s) was applied for the calculations of P. Quality of the obtained results can be tested by calculating real turbine (and turbocharger) efficiencies from the apparent (measured) ones and comparing them to the laboratory data determined by the manufacturers of the turbocharger (for the same compressor, turbine wheel, and turbine housing). In our case up to (max.) 4% efficiency point difference was noticed when comparing suitable engine test data with the same turbocharger units. Comparison of 0 values for different engine runing conditions, different exhaust systems, and different authors is presented in fig. 5. Relative size of the pressure pulses is used as the independent variable (horizontal axis). Higher values represent highly unstable, pulsating exhaust flow at relatively low pressure ratios typical for the "pure" pulse system, lower engine speeds and higher engine loads. The lower left side of the diagram refers to the rated operating engine and turbocharger conditions, where high pressure ratios and lower 4

relative pressure amplitudes are typical. Zinner (1980) published (3 values for the 2 and 3-pulse system (Fig.5, curve 0 A) that are 14 % lower in comparison to the present published data. The difference is probably caused by the assumption of the isothermal flow and probably by the lower density of the calculated pressure segments. It is also surprising that values for the 2 and 3-pulse system, and even constant pressure data presented here, lie on the same above mentioned curve. Constant pressure (4 cylinders connected to one common exhaust manifold) data 0 should generally not exceed suitable pulsating flow data. Although pressure amplitudes of the constant pressure system are damped they are still present and ^i values still exceed 1.0 except at the rated engine conditions. Correction of the apparent turbine and turbocharger efficiencies can therefore also be applied for the most commonly used constant pressure exhaust systems of the turbocharged 4-cylinder engines. CONCLUSION Two different exhaust systems can be successfully applied to a turbocharged, 4-cylinder engine. Due to the highly unstable mass and energy flow, the 2-pulse system requires a 40% larger size turbine housing than one that could cover the requirements of the constant pressure system to give the same air boost pressure. Apparent turbocharger efficiency, based on the mass averaged measured pressures and temperatures, differs by 50% from the real values. The more the flow is pulsating, the bigger is this difference. Apparent turbocharger efficiencies for the two described systems can differ even by 25% for the same equipment and the same running conditions. Correction is therefore necessary for any comparisons of the engine performance, especially during its optimisation procedure. Energy pulsation factor 0i that takes into account the type of the exhaust system, its shape and relative size of pressure amplitudes, can be calculated and real efficiency data can be obtained. REFERENCES Hribernik,A., 1995: "Primerjava nic in eno-dimenzionalnih metod za simulacijo procesov v tlacno poinjenem dizelskem motorju", Conference L4T"95, Proceedings, No.952204, Radenci, 1995, Slovenia, pp.165-173 Hribernik,A., 1997: "Comparison of Performances of a 4 - Cylinder Supercharged Diesel Engine with Single or Twin Entry Turbine", Conference L4T"97, Proceedings, No.971104, Otocec, 1997, Slovenia, pp.29-36 Watson,N., Janota, M.S.,1984: " Turbocharging the Internal Combustion Engine; Macmillan Publishers Ltd.. London, ISBN 0 333 24290 4 Zinner, K, 1980: "Aufladung von Verbrennungsmotoren", Springer-Verlag, Berlin, ISBN 3-540-10088-1 Trenc,F. et al,1997: "Influence of the Exhaust System on Performance of a 4-Cylinder Supercharged Engine", ASME Fall I.C.E. Division Conference, ICE-Vol.29-1, Paper No.97-ICE- 44, Madison, pp.95-100 5