KOSZAŁKA Grzegorz, NIEWCZAS Andrzej: Prediction of ic engine power system durability with the use of ring pack model. (Przewidywanie trwałości układu tłok - pierścienie - cylinder silnika spalinoweg o z wykorzystaniem modelu uszczelnienia TPC.)

PREDICTION OF IC ENGINE POWER SYSTEM

DURABILITY WITH THE USE OF

RING PACK MODEL

PRZEWIDYWANIE TRWAŁOŚCI UKŁADU

TŁOK-PIERŚCIENIE-CYLINDER SILNIKA

SPALINOWEGO Z WYKORZYSTANIEM

MODELU USZCZELNIENIA TPC

Lublin University of Technology, Dept. of Combustion Engines and Transport

Ul. Nadbystrzycka 36, 20-618 Lublin, Poland E-mails: g.koszalka@pollub.pl, a.niewczas@pollub.pl

Abstract: The paper presents the new method of engine life

prediction, which bases on the results of wear measurements taken on an investigated engine and simulations carried out with the use of the analytical model of ring pack of the engine. In the contrary to traditional methods the proposed method does not require the prior knowledge of the limit wear value, which is essential for the prediction, however its reliable establishing for new engines is difficult. In the presented method its value is determined with the use of the ring pack model. The example of durability calculation for an automotive diesel engine is also presented.

Keywords: IC engine, blowby, ring pack, durability prediction Streszczenie. W artykule przedstawiono nową metodę

prognozowania trwałości silnika, wykorzystującą wyniki pomiarów zużycia elementów układu tłok-pierścienie-cylinder oraz wyniki badań symulacyjnych modelu uszczelnienia TPC silnika. W przeciwieństwie do tradycyjnych metod, proponowana metoda nie wymaga wyprzedzającej znajomości zużycia granicznego, która jest kluczowa dla dokładności prognozy, a której wiarygodne określenie dla nowych konstrukcji jest trudne. W prezentowanej metodzie wartość zużycia granicznego wyznaczana jest z wykorzystaniem modelu uszczelnienia TPC. W artykule

przedstawiono przykład wykorzystania metody do prognozowania trwałości samochodowego silnika o zapłonie samoczynnym.

Słowa kluczowe: silnik spalinowy, przedmuchy spalin,

1. Introduction

Piston-rings-cylinder assembly (PRC) is a basic functional system of an internal combustion engine, and its most significant function is to guarantee tight, movable closing of the combustion chamber. Owing to different conditions of operation, especially thermal, components of the PRC assembly cannot be matched too tightly and some clearances between them exist. Therefore, described seal is not completely tight. The gas from the combustion chamber can leak to the crankcase through the gaps between the cylinder liner, piston and piston rings. In the opposite direction engine oil can enter the combustion chamber. As a result of components wear clearances increase what leads to the drop in the tightness. A good measures of the drop in the tightness of the PRC kit are increases in the blowby (gases leaking to the crankcase) and oil consumption. The increases in blowby and oil consumption are disadvantageous for the engine as they cause decrease in engine performance, increase in fuel consumption and toxic exhaust emissions, accelerated deterioration of engine oil quality and wear of various engine components as well as drop in the start-up capabilities of diesel engines (Andersson, et al., 2002). Repair of used PRC kit is very time- and cost-consuming and if it is done, it takes place during general overhaul. In many engines, especially smaller, such repair is unreasonable from the economic point of view. Thus, wear of the PRC assembly usually determines durability of the whole engine or, in certain reasonable cases, the necessity for general overhaul.

Methods of durability prediction of the PRC assembly allow shortening the time and lowering the cost of the research. Classical methods of durability prediction consist in evaluation of the wear course of chosen components of the PRC kit on the basis of shortened research and then extrapolation of the obtained wear course and estimation of the time when boundary wear value is reached (Figure 1). The conditions of obtaining reliable results are correct estimation of wear course and knowledge of boundary wear value. Usually the first condition is fulfilled if operating conditions of the engine during the research do not lead to the different wear, in quality terms, from that met in reality, and if wear course is determined on the basis of measurements made on the engine which has undergone its run-in period. Experiments show that wear course, especially in the case of cylinder liner, after completing the run-in process is linear. More difficulties occur when boundary wear needs to be determined, as there is no linear dependence between wear of the components and drop in the tightness of the PRC kit. Therefore, boundary values of the engine components wear are usually determined with statistical methods, using results of measurements made on the similar

objects withdrawn from the operation. However, utilizing such values of boundary wear for durability prediction can lead to significant errors resulting from various influence of wear on the sealing operation, even in similar designs of the PRC assembly. This is caused by complex mechanisms of the ring pack operation – even small modifications in the design can bring about significant changes in the efficiency of the seal. Moreover, boundary wear is usually determined in the described way on the engines which are at least one generation older than the ones for which the durability is forecasted. Another shortcoming of empirical models of boundary state (statistically determined values of wear for objects withdrawn from service) is that they do not allow for more detailed analysis of causes of the tightness drop – for instance how wear of particular components and increase of individual clearances affects the tightness.

Fig. 1 Principle of durability prediction

Taking above into consideration the authors decided to use analytical model of the ring pack for determination of the boundary wear. This model describes cause and effect relations between dimension of particular clearances and blowby intensity and allows for evaluation of tightness changes caused by wear. Thus it can be used for forecasting durability of engine. Analytical models have already been successfully used for analyzing of ring pack operation and improving the design (Keribar, et al., 1991; Tian, et al., 1998; Tian, 2005; Wolff, 2009).

2. Analytical model of piston ring cylinder kit

The analytical model of ring pack used in this research integrates model of the gas flow through the crevices of the RPC seal and model of rings motions in piston grooves. The model of gas flow treats the seal as a

labyrinth consisting of several stages linked together by throttling passages. The stages are created by inter-ring and behind ring volumes. The throttling passages are created by ring end-gaps and crevices between side surfaces of the rings and the grooves (Figure 2). The model takes into account the influence of thermal deformations and wear of the PRC components while determining instantaneous values (as a function of crank angle) of the stages volumes and passages areas. Cross-sectional areas of the crevices between side surfaces of the rings and the grooves result from instantaneous axial positions of the rings in the grooves. Axial positions of the rings in their grooves are calculated with consideration of the forces of gas pressure, ring friction against the cylinder and inertia. Pressures and temperatures of the gas in the stages are calculated using energy, continuity and state equations. The mass rates of the gas flow through crevices are calculated assuming that the flow is isentropic through the orifice with the consideration for sub-critical and sub-critical flows and taking into account discharge coefficients. Detailed description of the model has been previously presented (Koszałka, 2004a).

Calculations performed with the use of the numerical application of the model give, among other quantities, pressure courses in the particular stages of the labyrinth, displacements of the rings in their grooves and instantaneous gas flow intensities through the crevices of the ring pack – all as a function of crankshaft position (Figure 3). Blowby intensity is achieved by integration of instantaneous gas flowrates through the oil ring end-gap and crevice between oil ring and its groove shelf (m5-7 and m5-6 – see Figure 2).

Input data necessary to perform calculations include, among others: dimensions of engine components and course of pressure in the combustion chamber as a function of crankshaft angular position. The components dimensions can be determined using technical documentation of the engine or direct measurements. In the case of dimensions which directly influence the cross-sections of the crevices and volumes of the stages, values inserted into the computer program should take into account thermal deformations of the components. Thermal deformations can be calculated for the given engine operating conditions using Finite Elements Method. Obtained deformations are then added to the dimensions resulting from the documentation or obtained from the direct measurements of “cold” components. The most advisable is to use pressure course obtained from direct measurements of indicated pressure in the investigated engine.

Geometrical dimensions used in the calculations can include wear of particular components of the PRC kit, in the similar way as thermal deformations are considered (Koszałka, 2004a). This makes possible to estimate how wear value influences tightness of the PRC kit. Usability of the described model for the evaluation of wear influence on the blowby intensity was previously confirmed by the comparison of results obtained in numerical calculations with blowby measured on real engines (Koszałka and Niewczas, 2006; Koszałka, et al., 2008).

0 0 . 4 0 . 8 1 . 2 1 . 6 2 P re ss ur e [M P a] 0 0 . 0 2 0 . 0 4 0 . 0 6 0 . 0 8 R in g po si tio n [m m ] 0 1 8 0 3 6 0 5 4 0 7 2 0 C r a n k a n g l e - 0 . 4 0 0 . 4 0 . 8 1 . 2 1 . 6 2 F lo w r at e [g /s ] p1 p2 p3 p4 p5 m1 - 3 m3 - 5 m5 - 7 m5 - 6 xI xI I xI I I

Fig. 3 Pressure courses in inter and behind ring spaces, axial position of rings in piston grooves and flow rates of gas through ring end gaps and oil ring side

clearance versus crank angle determined with the use of the model

3. Method of durability prediction

Before starting the research on prediction of durability of the PRC kit, or its operation time until the general overhaul, using the proposed method, running-in of the engine is necessary. Engine operation time in this period should be sufficient to be certain that unstable tribological processes related to the running-in are over.

The main part of the research begins with the evaluation of the initial tightness and wear of the PRC kit. It is done by the measurement of blowby flowrate on operating engine and then measurements of the PRC components wear, after the engine disassembly.

Next, engine should be used during the period of time allowing for the evaluation of wear intensity of its components. The longer the time, the

better results can be obtained. The engine during this time can work either in a vehicle or on a test stand. Conditions of engine operation should not considerably differ from those typical for normal operation. After finishing this phase of research tightness and wear of the PRC kit need to be evaluated once more. Measurements of engine blowby and wear should be made in the same way as during the estimation of initial values.

Basing on the measurements of wear intensities of wear w of the engine components needs to be calculated (see Figure 1).

The second area of work connected with estimation of engine durability includes research of the ring pack model. In this area first of all, all input data necessary for calculations need to be gathered, including results of wear and blowby measurements previously taken. Next, numerical calculations of the blowby for the input data corresponding to the initial engine wear should be done and then – for the input data corresponding to the wear measured after the test. Comparison of the simulated increase in the blowby with actual enable an evaluation of correctness of the model identification. In the case of agreement of the numerical calculations with the measurements, simulations should be continued for the higher values of the PRC wear. Values of wear should be determined assuming that wear intensities of particular components are the same as the ones obtained on the basis of measurements (w). The aim of these calculations is to find such values of wear Wlim for which increase in the blowby reaches the previously assumed

permissible value of blowby. Permissible blowby value can be assumed considering its negative effects on the engine operation. Estimated value of wear Wlim is the boundary wear and corresponding time T is the searched

engine durability (see Figure 1). 4. Example of calculations

The object of the research was a commercially available inline 6 cylinder diesel engine with swept volume of 6.8 dm3 _{and rated power of 110 kW at}

2800 rpm. The engine was equipped with wet, cast iron cylinder liners with nominal inside diameter of 110 mm. Piston stroke was 120 mm. The piston ring pack of the engine comprised a keystone barrel-shaped face top ring, a taper face second ring and a twin-land spring-backed oil control ring. The face surfaces of the top and oil rings were chrome-plated. The aluminum pistons had a cast iron carrier for the top ring groove.

In order to avoid errors connected with deviation of the chosen copy of the engine from the average, the research was carried out on 5 engines. Engines were mounted in medium-sized motor trucks belonging to one company. All five vehicles were used in similar conditions, typical for the company, with

average monthly mileage of 10,000 km. About one third of the distance was covered by the vehicles in towns and cities and the rest on inter-city roads. The load of the vehicles never exceeded maximum manufacturer’s payload of 6,500 kg. The engines were fueled with standard diesel fuel and lubricated with the same type of oil CE/SF SAE 15W-40.

Measurements of initial wear and blowby were taken when vehicles achieved kilometrage of 50,000. Such kilometrage guaranteed that the engines were fully run-in. Blowby rate was measured during engine idling. Then the engine was partially dismantled and wear of cylinder kit elements wear measured. Diameters of cylinder liners were measured at two directions: parallel (A-A) and perpendicular (B-B) to the main engine axis at four depths: 20 mm (top dead center TDC of the top ring), 35 mm (TDC of the second compression ring), 50 mm and 95 mm (Figure 4) (Koszałka and Niewczas, 2007). Then engines were mantled and further used in the company in the same conditions as before measurements. When vehicles achieved kilometrage of 150,000 the blowby rate and wear were again checked in the same way, using the same instruments, as at 50,000 km.

Fig. 4 Planes of cylinder liner diameter measurements; A-A – direction parallel to the engine axis, B-B – direction perpendicular to the engine axis

Average values of PRC components wear were calculated on the basis of the results of measurements made on all five engines. As the ring pack model assumes that the piston with rings and cylinder are symmetrical axially it was not possible to take into account different wear values in parallel AA and perpendicular BB planes of the PRC components and so they were also averaged. Then the courses of wear were extrapolated assuming that they are linear. Mean diameters of cylinder liner at different depth obtained in measurements at vehicles kilometrages of 50,000 and 150,000 and predicted courses of wear are presented in Figure 6.

0 2 0 0 0 0 0 4 0 0 0 0 0 6 0 0 0 0 0 K i l o m e t r a g e [ k m ] 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 D ia m et er o f c yl in de r o ve r 1 10 m m [ m ] D i s t a n c e f r o m h e a d 2 0 m m 3 5 m m 5 0 m m 9 5 m m

Fig. 5 Mean diameters of cylinder liner at different depth obtained in measurements at vehicles kilometrages of 50,000 and 150,000 km

and predicted courses of wear

All necessary input data for calculations were established. Dimensions of components were determined from technical specification and measurements described above. Thermal deformations were calculated using FEM and they were added to “cold” values. The indicated pressure came from the measurements made on another copy of the same type of engine during test stand research.

Numerical simulations were done for dimensions of engine components corresponding to kilometrage of 50,000 and 150,000. Input data for these two series of calculations differed only in dimensions of elements resulting from wear. The wear of cylinder liner (Figure 6), face and side surfaces of the rings and flanks of the piston grooves were considered. Simulated blowby rate for kilometrage of 150,000 was 25% higher than for kilometrage of 50,000 and this increase was consistent with the actual.

Thus, simulations were carried out for higher values of wear. Values of wear for bigger kilometrages were determined from previously established linear courses of wear (example for cylinder liner in Figures 5 and 6). As investigated type of engine was not a new design the blowby rates in some copies of such engine withdrawn from service was measured. Those blowby were 2.5 times higher than for investigated 5 engines at kilometrage of 50,000. Thus, such increase in blowby was taken as permissible. In numerical simulations such increase in blowby was achieved for value of wear predicted for kilometrage of 590,000. So the predicted durability of the engine determined with proposed method was 590,000 km. The actual durability of investigated type of engines were between 500,000 and 700,000 km. Predicted profile of cylinder liner at 590,000 km, which is also predicted boundary wear, is presented in Figure 6.

0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 D i s t a n c e f r o m t h e h e a d [ m m ] 0 4 0 8 0 1 2 0 D ia m et e r o f c yl in de r o ve r 1 1 0 m m [ m ] 5 0 , 0 0 0 k m 1 5 0 , 0 0 0 k m 5 9 0 , 0 0 0 k m m e a s u r e d p r e d i c t e d

Fig. 7 Profiles of cylinder liner at different kilometrages

5. Conclusion

Presented method of durability prediction bases on empirically determined intensities of wear of piston kit components and on the results of simulations made with the use of the analytical model of the ping pack. The intensities of wear can be established using results of wear measurements taken on engines operating either on test stands or in vehicles. The advantage of the proposed method is no necessity of knowing of the boundary wear value. Correct assuming of its value is essential for accuracy of prediction in traditional methods. Unfortunately, reliable evaluation of the boundary wear value for newly designed engines is very difficult. Presented method does not require prior knowledge of the boundary wear

because its value is determined with the use of the analytical model of the ring pack. However it requires knowledge of permissible drop in tightness of PRC assembly. Permissible drop in tightness can be established taking into consideration acceptable decrease in engine performance and ecology. Prediction of durability made with the use of the presented method for automotive diesel engine was correct. That proved that the method is useful. Additional advantage of using the analytical model is possibility of direct utilization of results of calculations for improving the design of piston kit. Praca naukowa finansowana ze środków na naukę w latach 2010-2012 jako projekt badawczy Nr N N509 479538.

References

1. Andersson P., Tamminen J., Sandstrom C-E.: Piston ring tribology. A literature survey. VTT Tiedotteita - Research Notes 2178, Espoo (Finland), 2002.

2. Keribar R., Dursunkaya Z., Flemming M. F.: An Integrated Model of Ring Pack Performance. Journal of Engineering for Gas Turbines and Power, v. 113, p. 382-389, 1991.

3. Koszałka G.: Modelling the Blowby in Internal Combustion Engine, Part 1: A Mathematical Model. The Archive of Mechanical Engineering, v. LI, no 2, p. 245-257, 2004.

4. Koszałka G.: Modelling the Blowby in Internal Combustion Engine, Part 2: Primary Calculations and Verification of the Model. The Archive of Mechanical Engineering, v. LI, no 4, p. 595-607, 2004. 5. Koszałka G., Niewczas A.: The Influence of Compression Ring

Clearances on the Blowby in a Diesel Engine. SAE Paper 2006-01-3356, 2006.

6. Koszałka G., Niewczas A.: Wear profile of the cylinder liner in a motor truck diesel engine. Journal of KONES Powertrain and Transport, v. 14, no 4, p. 183-190, 2007.

7. Koszałka G., Niewczas A., Guzik M.: Predicted and Actual Effect of Cylinder Liner Wear on the Blowby in a Truck Diesel Engine. SAE Paper 2008-01-1717, 2008.

8. Tian T., Noordzij L. B., Wong V. W., Heywood J. B.: Modeling Piston-Ring Dynamics, Blowby and Piston-Ring-Twist Effects. Journal of Engineering for Gas Turbines and Power, v. 120, p. 843-854, 1998.

9. Tian T.: Dynamic behaviours of piston rings and their practical impact. Part 1: ring flutter and ring collapse and their effects on gas flow and oil transport. Journal of Engineering Tribology, v. 216, no 4, p. 209-227, 2005.

10. Wolff A.: Numerical analysis of piston ring pack operation. Combustion Engines – Silniki Spalinowe 2(137), p. 128-141, 2009.

Grzegorz Koszałka PhD – Assistant Professor (adiunkt) in Mechanical

Engineering Faculty, Lublin University of Technology. MSc., 1993, PhD. 2002, Lublin University of Technology. Research interests: Reliability and durability of IC engines, Analysis and modeling of piston ring pack.

Andrzej Niewczas Prof. – Professor in Mechanical Engineering

Faculty, Lublin University of Technology, Head of Dept. of Internal Combustion Engines and Transport. MSc., 1970, PhD. 1977, Warsaw University of Technology, DSc. (hab.), 1990, Poznan University of Technology. Research interests: Reliability and durability of IC engines and road vehicles, Stochastic models in wear and failures analysis.