ABSTRACTPartially Premixed Combustion (PPC) has demonstratedsubstantially higher efficiency compared to conventionaldiesel combustion (CDC) and gasoline engines (SI). Bycombining experiments and modeling the presented workinvestigates the underlying reasons for the improvedefficiency, and quantifies the loss terms. The results indicatethat it is possible to operate a HD-PPC engine with aproduction two-stage boost system over the EuropeanStationary Cycle while likely meeting Euro VI and US10emissions with a peak brake efficiency above 48%. Amajority of the ESC can be operated with brake efficiencyabove 44%.

The loss analysis reveals that low in-cylinder heat transferlosses are the most important reason for the high efficienciesof PPC. In-cylinder heat losses are basically halved in PPCcompared to CDC, as a consequence of substantially reducedcombustion temperature gradients, especially close to thecombustion chamber walls. Pumping losses are on the otherhand three times higher than for CDC due to the increasedmass flow rate over the valves from the charge dilution andthe high amounts of EGR. Friction losses remain uncertainwith respect to the direct injection of gasoline instead ofdiesel, but have been estimated to be slightly higher than forCDC in this work.

A sensitivity analysis demonstrates that further reductions ofin-cylinder heat transfer losses are possibly the most

beneficial for further increases in brake efficiency. Furtherimprovements can also be reached by reducing exhaust portand manifold heat transfer losses and optimized gas exchangeand boosting systems. A PPC engine with 57% grossindicated efficiency is likely to reach more than 50% brakeefficiency.

INTRODUCTIONThe requirements for the future combustion engines are bothcomplex and challenging. Obviously future engines will beexpected to perform the work with no less productivity andcomfort than current engines. Engine combustion andemissions aftertreatment systems need to be increasinglyefficient to reduce the specific emissions of unburnthydrocarbons (HC), carbon monoxide (CO), nitrous oxides(NOx) and particulate matter (PM). The engine totalefficiency needs to be maximized to reduce the carbondioxide (CO2) emissions and operating costs. Engine relatednoise, for instance combustion noise, needs to be undercontrol as well. Further these requirements needs to be metwith engines that can both be started and operated untroubledunder various climatic conditions throughout the designedlife of the engine, a designed life that is unlikely to be shorterthan current engines due to consumer and regulatorydemands. Certainly there is also an interest to reach all thesetargets with systems that rely to a minimum on costly andcomplex auxiliary sub-systems.

Loss Analysis of a HD-PPC Engine with Two-Stage Turbocharging Operating in the EuropeanStationary Cycle

2013-01-2700Published

10/14/2013

Martin Tuner and Bengt JohanssonLund University

Philip Keller and Michael BeckerBorgWarner Inc

Copyright © 2013 SAE International and Copyright © 2013 KSAE

doi:10.4271/2013-01-2700

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

In response, several low temperature combustion (LTC)concepts have been developed. Possibly the most famous andmost researched is homogenous charge compression ignition(HCCI), but currently the focus is on more practical solutionsthan on pure HCCI. To name some of the most successful;reactivity controlled compression ignition (RCCI) [1], highefficiency dilute gasoline engine (HEDGE) [2], gasolinedirect injected compression ignition (GDCI) [3] and partiallypremixed combustion (PPC) [4]. All these concepts originatesfrom single cylinder engine research and show impressivegross indicated efficiencies due to dilute low temperaturecombustion that reduce in-cylinder heat transfer losses andexhaust losses. Further they all depend on boosting which canprove to be challenging when going from externally boostedsingle cylinder experimental systems to self-sustained multi-cylinder engines with real boosting systems, especially if theimproved indicated efficiencies are to be materialized inimproved brake efficiencies [5].

Possibly the most impressive implementation into a practicalmulti-cylinder LTC engine is that on GDCI described byMark Sellnau at the SAE High Efficiency Symposium 2013.Since this presentation is not easily available the reader isdirected to Hoyer et al [3] that describes the systeminvestigations of that work.

Regarding PPC, the early research attempts to transfer thePPC concept into a production viable engine with brakeefficiencies higher than state of the art HD diesel engines hasnot been that successful. In [6] a Volvo TD13 commercialengine was modified for PPC operation with a low pressureEGR system but maintaining the standard VGT turbocharger.The turbocharger was too small for the increased mass flowand combined with problems with internal leakage of the fuelinjectors, leading to very high friction losses, brake efficiencynot higher than 42% could be reached. The Scania engineused for much of the single cylinder engine research on PPChas not demonstrated any particular problems while operatingon gasoline. Thus to investigate the system performance of aScania engine a detailed theoretical investigation wasperformed in [7]. That study indicated that the standard singlestage turbocharger was not sufficient for PPC operation, butif a boosting system of increased size and slightly increasedefficiency could be employed, close to 48% brake efficiencyshould be within reach. This work expands on that byimplementing an off-the-shelf production two-stage boostingsystem from Borg-Warner.

OBJECTIVE OF INVESTIGATIONThe objective of this work is two-fold. One objective is todevelop a realistic PPC engine system model that can be usedto investigate the potential of different PPC strategies andguide the implementation of for instance EGR and boostingstrategies for a real application in further experimentalresearch. The second objective is to quantify and analyze theloss sources in a heavy duty PPC engine and to comparethese with state-of-the-art heavy duty engines.

METHODThe investigation is carried out with 1-D tools (GT-Power)[8] of complete and functional Scania D13 PPC multi-cylinder system with an off-the-shelf Borg-Warner two-stageboosting system. The system model is applied for operationover the European 13 mode stationary cycle. Engineexperiments support the modeling as to provide accurate in-cylinder conditions. Due to the detailed level of the modelvarious losses can be investigated in detail in the enginesystem network.

ENGINE EXPERIMENTSThe experiments were performed by Manente et al. on aScania D13 engine modified for single cylinder operation [4].The four load cases in that work using 89RON gasoline wereselected for this investigation, due to the commercialavailability of that fuel. The experiments were all run at 1250rpm. The experimental conditions are presented in Table 1.

Table 1. Experimental operating conditions.

MODELING APPROACHDetails on the modeling approach, model development,calibration and performance was presented in a previouspublication [7].

A single-cylinder model was developed to replicate theexperiments for calibration of various parameters. The modeluses the experimental RoHR, fuel flow, lambda, inlet andexhaust temperatures and pressures. The compression ratio,heat transfer parameters, pipe and valve flow coefficientswhere adjusted until a satisfying agreement was reachedbetween the measured and simulated in-cylinder pressureover both the closed and open cycle, as well as for theaveraged inlet and exhaust temperatures and pressures. Incomparison to the previous work [7] the simplified treatmentof modeling the cylinder with only a single exhaust valve andport (common approach in 1-D) was replaced with a moredetailed description with both exhaust valves and ports,giving a much improved agreement on the exhausttemperatures (Table 2). Engine data and cylinder pressurecomparisons can be found in the appendix.

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

Table 2. Simulated operating conditions.

The multi-cylinder model was developed based on thecalibrated single-cylinder model together with a productionScania D13 Euro V model to include realistic heatexchangers and muffler/SCR. Finally a two-stage boostingsystem from Borg-Warner was included. The model wasoperated to cover the ESC 13 mode cycle. Since theexperiments only cover the lowest engine speed in the ESC(A-speed) combustion and in-cylinder data was assumed forB- and C-speed points, which is described in detail in [7]. Foreach of the operating points the relevant conditions for

combustion from the single-cylinder model calibration (A-speed) or assumed conditions (B-C-speeds) in terms oflambda, EGR amount, in-cylinder temperature and pressurewhere maintained by using PID regulators operating on theHP-turbine bypass valve and EGR valve.

Loss analysis is only performed on operating pointssupported by experimental data.

Multi-Cylinder PPC Engine ModelThe PPC system model can be seen in figure 1. The intakeuses the same air filter as the production engine.

The inter stage cooler (ISC) and charge air cooler (CAC) areof the same design as the standard Scania CAC, althoughwith 60% increased number of pipes as to maintain a similarpressure drop. The intake pipe passing boosting system, ISCand CAC is at 105 mm diameter, larger than for comparableCDC engines. Intake and exhaust ports, valves and valvetimings and intake and exhaust manifold are the same as for

Figure 1. The multi-cylinder PPC engine system model.

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

the production engine. The design of the intake and exhaustmanifolds does introduce slight variations of the chargebetween cylinders and the combustion. Of greaterconsequence are the larger variations in exhaust pressures dueto the different length of the individual exhaust runners.

Even though PPC demonstrates very low NOx emissionsduring steady state operation these levels could be difficult tomaintain during transients. The SCR could most likely bereduced in size compared to CDC. Further it is likely that anoxidizing catalyst may be required at lower loads. For thesetwo reasons the muffler/SCR from the GT-Power model ofthe production engine is retained, with similar regulated backpressure as a function of load and speed. The model does notcontain any particulate filter since it is judged that such willnot be needed for a well calibrated PPC engine.

The EGR system is contrary to the production engine a longroute low pressure EGR system with the junctions carefullysized as to minimize pressure losses. The internal pipes of theEGR-C are increased 20% in diameter while the EGR andexhaust system pipes are sized equally at 105 mm diameter.All pipes, coolers and cylinders use wall temperature solversto improve the prediction of heat transfer. The sinktemperatures of the EGR-C and ISC are set to achieve around70°C gas temperatures as to avoid condensation before thecompressors.

The CAC sink temperature is adjusted for the intake charge toreach the inlet manifold temperatures in the experiments(table 1). These temperatures are rather low and in practicalapplications these could likely only be reached with ambienttemperatures less than 10-15 °C depending on CACconfiguration. Since the inlet temperature will affect thecombustion, the experimental inlet temperatures have beenmaintained in this work. Future work will however aim atproducing experimental combustion data with varying inlettemperatures from the current 25-30 °C to 80 °C.

The three zone in-cylinder wall temperature solver uses adetailed description of the in-cylinder geometry, oil cooling,wall materials and thicknesses and sink temperatures. Portsand pipes are handled in a similar fashion. The Woschni-Huber heat transfer model is used to determine the in-cylinder heat transfer using the same multipliers as in thesingle cylinder calibrations to achieve similar pressure tracesduring expansion as for the experiments. The port heattransfer coefficients are slightly adjusted to reach the sameexhaust temperatures at the same location in the exhaustmanifold as for the experiments. The friction model is tunedto provide the same friction losses at each operating point asthe highest friction values from any of the compared CDCengines. The crank train is modeled to account for theflexibility of the connecting rod.

Boosting SystemThe dual demands of high power density and EGR flow ratesrequire a two-stage turbocharger system. The current first

iteration two-stage system chosen utilizes BorgWarner off-the-shelf production components. The high pressure stageturbocharger uses a 84T compressor with a wheel diameter of84 mm. The fixed geometry wastegated turbine for this stageis a 73QK with a diameter of 73 mm. The low pressure stagecompressor is a 4871N with a diameter of 123 mm. The lowpressure stage fixed geometry turbine is a 96M with adiameter of 96 mm. No efficiency or mass multipliers wereemployed in any of the simulations.

RESULTSBoosting System PerformanceThe performance of the low pressure stage and high pressurestage can be seen in Figure 2 and Figure 3 respectively. Dueto confidentiality, absolute values are normalized with darkred being unity and dark blue being nil. It can be seen that thematching of both the LP compressor and LP turbine isreasonable.

Figure 2. Normalized efficiency maps for the LPcompressor and turbine.

Figure 3 reveals that the HP turbine is reasonably wellmatched for the operating conditions while the HPcompressor is not utilized in the most efficient way,especially at higher engine speeds where the HP stage is moreor less bypassed.

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

Figure 3. Normalized efficiency maps for the HPcompressor and turbine.

ESC PerformanceThe two-stage boosting system is capable of providing thetarget inlet conditions, for instance lambda and EGR rates,required for operation over the complete ESC (Figures 4, 5,6, 7, 8, 9, Each figure shows the calculated points as whitedots over load and speed while gradients in between dots areinterpolated by GT-power.)

Figure 4. Lambda for the operating range covering ESCexcept idle.

The pumping losses in terms of PMEP are presented inFigure 6. Pumping losses are substantially higher for PPCthan for CDC. Especially at high loads and B-C-speedspumping losses are very high. The high pumping losses canpartly be explained by high mass flows due to the highamounts of EGR. There is likely potential for improvementswith a second iteration boosting system, updated valve-portinterfaces and other EGR calibrations.

Figure 5. EGR for the operating range covering ESCexcept idle.

Figure 6. PMEP for the operating range covering ESCexcept idle.

Figure 7 shows that brake efficiencies above 48% are reachedwhile an operating window covering the majority of A- andB-speeds reaches above 44% brake efficiency. According tothe simulations higher brake efficiencies are reached for amajority of the operating points than the typically reported43% peak brake efficiency of state-of-the-art diesel engines.The resulting BSFC figures can be seen in Figure 8.

To be certain that the catalysts ignite the exhaust temperatureshould be above 175°C [10]. Figure 9 indicates that thisrequirement is met for the complete operating regime (idlenot included).

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

Figure 7. Brake efficiency for the operating rangecovering ESC except idle.

Figure 8. BSFC for the operating range covering ESCexcept idle.

Figure 9. Exhaust temperatures before SCR for theoperating range covering ESC except idle.

Energy Balance and Loss AnalysisThis section presents energy losses for PPC at various loadsand in comparison with a CDC engine. The presented data forCDC consists of a compilation of data for a number ofdifferent state-of-the-art production diesel engines operatingat A75. The speeds and loads vary slightly in the ESC for thedifferent engines but are close to those of the PPC. Due toconfidentiality individual production engine data are not

revealed. For clarification: the presented heat transfer lossesare in-cylinder heat transfer losses, unlike often publishedheat transfer losses that actually includes exhaust port andexhaust manifold exhaust heat transfer losses, etc, dependingon where the exhaust temperature sensor was placed andexhaust mass flow determined [1, 9].

Figure 10 shows a comparison between PPC and CDC.Combustion losses are very small in both cases, due to verylimited crevice losses, while the in-cylinder heat transferlosses are basically halved in PPC. Exhaust losses are slightlylower while pumping losses are almost three times higher forPPC. The friction losses are slightly higher for PPC accordingto the simulations. The resulting brake efficiency issubstantially higher for PPC mainly due to the well timed anddiluted low temperature combustion that is less detrimental togas data, extends the expansion and lowers the in-cylinderheat transfer losses.

Figure 10. Comparison of loss terms in % of fuel energybetween a typical CDC and PPC operating at similar

load and speed.

Figure 11. Comparison of loss terms in % of fuel energyfor the PPC engine operating 40-100% load and1250

rpm.

A comparison was also made for the four load cases thatwhere supported experimentally (Figure 11). The resultsshow that combustion efficiency is reduced with load whilethe relative in-cylinder heat transfer losses increases. Bothrelative pumping and friction losses increase while relativeexhaust losses decrease with load. The brake efficiency is

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

fairly similar in the 40-100% load range with a peak of 48.3%at 60% load and minimum of 47.2% at 40% load.

The above presented exhaust and pumping losses can bebroken down in further detail. Figure 12 shows a sketch ofthe complete engine system with boxes defining specific subsystems or parts. The losses from each sub-system/part forthe A75 operating point are presented in Table 3.

Figure 12. Definition of location for the energy sourcespresented in Table 3.

1 defines the friction losses, 2 defines the cylinders and theirrelated heat transfer losses (HTL), 3 defines all the exhaustports as they merge from each exhaust valve pair into eachsingle outlet with their associated convective heat transferlosses. 4 is the complete exhaust manifold while 5 comprisesboth the high pressure and low pressure turbines. 6 is theEGR-C including the pipes into it as these also transferssubstantial amounts of heat. 7 is the exhaust system includingthe muffler and SCR unit. 8 contain both compressors while 9and 10 contains the ISC and CAC respectively.

One interesting result is that the combined heat transfer lossesfrom the exhaust ports (3) and exhaust manifold (4) exceedsthe in-cylinder heat transfer losses (2). The boosting system(5, 8) consumes 70 kW of the exhaust energy. The EGR-Ctransfers 42 kW. The energy that is lost with the tail pipeexhaust adds up to 51 kW based on the exhaust temperature,specific heats, mass flow and ambient temperature.

To complete the energy balance the energy transferred fromthe exhaust to the intake side needs to be considered. Thisconsists of energy from the boost system and from the EGRwhich is not cooled down to ambient temperature. The ISCand CAC then transfer heat with a rate of 50 and 34 kWrespectively to the atmosphere while the incoming charge isheated with 3 kW in the intake ports (11). There are a numberof other small loss sources, for instance intake manifold heattransfer, but these sources are comparably small and seem tobalance out each other since the balance of the major sourcescomes to 0.

Table 3. Losses and power in kW.

Sensitivity AnalysisVariation of In-Cylinder Heat TransferDue to the strong impact from the in-cylinder heat transferlosses on the system performance, a sensitivity analysis wasperformed by varying the heat transfer multiplier (HTM) forthe Woschni-Huber model for the A75 case. The calibrationin the single cylinder model had provided a suitable value of0.32 for the HTM. Two additional simulations whereperformed with the HTM set to 0.16 and 0.64. Already inFigure 10 it can be seen that in-cylinder heat transfer lossesare basically halved for PPC versus CDC. Although it isuncertain whether a further 50% reduction is possible, theauthors still wanted to investigate and demonstrate the impact[11].

Figure 13 shows the different cylinder pressures achievedcompared to the experimental pressure.

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

Figure 13. Cylinder pressure during expansion due todifferent cylinder heat transfer rates.

The results indicate a couple of interesting features. The firstis that the calibration is possibly overly conservative and thatin-cylinder heat transfer could be even lower than firstestimated. One needs, however, address the results withcertain cautioun since small differences in the phasing of theapplied RoHR as well as in the estimation of the connectingrod flexibility do have sufficient impact to alter thecalibration.

Table 4. Results due to variation of in-cylinder heattransfer

The second interesting feature is that the brake efficiencyincrease/decrease with a higher rate than the gross indicatedefficiency. With the decreased heat transfer losses there isslightly increased exhaust energy that reduces the pumpinglosses

Variation of Exhaust Port and Manifold HeatTransferThe loss analysis revealed that the combined heat losses fromthe exhaust ports and manifold exceeded that from thecylinders. This lost energy can be much better used in theboosting system. By applying a thermal barrier coating on theinside of the exhaust ports and exhaust manifold and alsoisolating the outside of the exhaust manifold (for instancewith double layer sheet metal as in passenger cars) the heattransfer losses can be reduced. How much, is not easy toestimate. A brief literature review do not provide clearindications so instead the HTM was reduced ad hoc in twosteps, 25% reduction and 50% reduction. As can be seen inTable 5 there is a small but clear impact from isolating theports and manifold. For a similar 50% decrease of around 13

kW heat losses, a reduction at the ports and manifold onlyprovides a 1/10 gain in brake efficiency compared to havingthe heat loss reduction within the cylinders. This can beexplained with that reduction of heat losses during theexpansion period is directly equivalent to increased workwhile heat transfer reduction on the ports and manifold isconverted to work indirectly through the boosting system.Still the reduction of heat losses through ports and manifoldcan be very important to achieve required boosting whenfurther reductions of exhaust losses have been achieved.

Table 5. Results due to variation of exhaust port andmanifold heat transfer

Variation of FrictionAlready the baseline case uses slightly higher friction settingsthan for CDC. The motivation for this is the high recordedfrictions while operating the pump injectors in the Volvo D13engine on gasoline [6]. The Scania common rail fuel injectionsystem has not limited the experimental work when usinggasoline, but the exact work required for pumping dieselversus gasoline has not been determined yet in our laboratory.Since the friction loss due to pumping gasoline is unknown atthis stage a variation with increased friction in two steps wasperformed. Table 6 shows that a change of friction multiplier(FM) gives a slightly smaller change in FMEP. Nevertheless,a FMEP increase of 15% yields less than 1% (0.4%-unit)reduction in brake efficiency. Whether this increase infriction is realistic or not remains to be seen experimentally.

Table 6. Results due to variation of friction

DISCUSSIONIn-Cylinder Heat TransferThe results indicate that the main benefit with PPC over CDCis the substantially reduced in-cylinder heat transfer lossesthat eventually lead to higher brake efficiency. Even thoughthe reduction is substantial it is not that surprising. The lastdecade of research into low temperature combustion engineshas on several occasions demonstrated a similar substantialreduction of in-cylinder heat transfer. For instance it has been

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

widely accepted that the much used Woschni heat transfercorrelation model overestimates heat transfer duringcombustion in HCCI engines. A common approach in thekinetics modeling community as well as in the postprocessing of HCCI experiments has been to omit theWoschni model C2 factor altogether thus giving an 80%reduction in heat transfer, giving a better match betweenmodeling and experiments. A more scientific approach hasbeen the work directed at developing new heat transfercorrelation models showing similar reductions [12, 13]. Workrelated to PCCI and RCCI do confirm the findings from theHCCI work. For instance in Splitter et al.[1] a HTM of 0.2 isused with Woschni-Huber for calibration of a 1D modelagainst experiments. There is no specific value of in-cylinderheat transfer losses presented, but the combined heat transferlosses from the cylinder exhaust port, manifold and exhausttank is estimated by the authors to 8.2% of the fuel energy.

The work by Fridriksson et al. provides an insight to whyheat transfer is so much lower in PPC compared to CDC [14].The temperature fields in a cutting plane through one fuelspray of half of the combustion chamber can be viewed forCDC (Figure 14) and PPC (Figure 15).

Figure 14. CFD calculated in-cylinder temperaturedistribution for a standard Scania D13 engine at 1250

rpm at 26 bar IMEP gross.

The CDC case reveals temperatures above 2000 K in areas inthe spray as well as very close to the surface along the pistonbowl and the squish area. In comparison the PPC case doesnot exceed 2000 K in any region and furthermore the bowland squish surfaces are exposed to 400-500 K lowertemperatures. The surface exposure to high gas temperatureshas also a shorter duration. The CFD simulations predicted5.4% of fuel energy as in-cylinder heat transfer losses for thePPC case with twice that figure for the CDC. The GIE of thisdiesel-PPC case was 53.8% for the experiment and 54.0% forthe CFD calculations thus providing similar data as for thegasoline-PPC A75 case investigated in this work.

Figure 15. CFD calculated in-cylinder temperaturedistribution for a Scania D13 PPC engine at 1250 rpm

21 bar IMEP gross.

SUMMARY/CONCLUSIONSEngine loss sources were investigated with a Scania D13 PPCengine system model based on single cylinder experiments, avalidated single cylinder model and a validated Scania D13,EuroV, production engine model and a first iteration, off-the-shelf BorgWarner two-stage boost system.

The results indicate that:

1. A realistic PPC engine system operated in steady state oncommercial gasoline is capable of reaching above 48% peakbrake efficiency and above 44% brake efficiency for most ofA and B-speeds while likely meeting EuroVI and US10emissions regulations.

2. The reduced (typically halved) in-cylinder heat transferloss is the major source of improvement for PPC compared toCDC.

3. The reduced in-cylinder heat transfer losses can beexplained by the lower combustion temperatures but also bythe partially premixed combustion that avoids hightemperature combustion close to the combustion chamberwalls. There is a lower overall temperature gradient.

4. Reduction of in-cylinder heat transfer losses may beattributed to increased exhaust losses, which can be used forreduced boost work providing a combined effect forincreased brake efficiency.

5. Heat transfer loss is higher for the combined exhaust portsand manifold than for the cylinders.

6. Reduction of heat transfer losses in the exhaust ports andmanifold has only 1/10 of the impact on brake efficiencycompared to reduced heat losses within the cylinders. Stillsuch reduction may be very useful where exhaust power forboosting is limited.

7. PPC has a factor three higher pumping losses than CDC.Potential improvements of the pumping through a seconditeration boosting system is planned to be investigated.

8. A potential 15% increase in friction loss, from for instanceincreased work for injecting gasoline, could reduce brake

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

efficiency 1%. Whether friction loss for PPC actually ishigher, or not, than for CDC has to be determined with futureexperiments.

9. Considering that 48.4% brake efficiency could be reachedwith 54.1% gross indicated efficiency, 50% brake efficiencycan probably be reached with 57% gross indicated efficiency.

Future work will aim at producing additional experimentalsingle-cylinder engine data and validating the models for alloperating points in the world harmonized stationary cycle(WHSC). With the updated models a second iterationboosting system can be investigated. Investigations with areal boosting system in multi-cylinder heavy duty engine canthen be performed.

REFERENCES1. Splitter, D., Wissink, M., DelVescovo, D., and Reitz, R.,“RCCI Engine Operation Towards 60% Thermal Efficiency,”SAE Technical Paper 2013-01-0279, 2013, doi:10.4271/2013-01-0279.

2. Joo, S., Alger, T., Chadwell, C., De Ojeda, W. et al., “AHigh Efficiency, Dilute Gasoline Engine for the Heavy- DutyMarket,” SAE Int. J. Engines 5(4):1768-1789, 2012, doi:10.4271/2012-01-1979.

3. Hoyer, K., Sellnau, M., Sinnamon, J., and Husted, H.,“Boost System Development for Gasoline Direct- InjectionCompression-Ignition (GDCI),” SAE Int. J. Engines 6(2):815-826, 2013, doi:10.4271/2013-01-0928.

4. Manente, V., Zander, C., Johansson, B., Tunestal, P. et al.,“An Advanced Internal Combustion Engine Concept for LowEmissions and High Efficiency from Idle to Max Load UsingGasoline Partially Premixed Combustion,” SAE TechnicalPaper 2010-01-2198, 2010, doi:10.4271/2010-01-2198.

5. Chadwell, C., Alger, T., Roberts, C., and Arnold, S.,“Boosting Simulation of High Efficiency AlternativeCombustion Mode Engines,” SAE Int. J. Engines 4(1):375-393, 2011, doi:10.4271/2011-01-0358. doi:10.4271/2011-01-0358.

6. Lewander, M., Johansson, B., Tunestål, P., Keeler, N. etal., “Evaluation of the Operating Range of Partially PremixedCombustion in a Multi Cylinder Heavy Duty Engine withExtensive EGR,” SAE Technical Paper 2009-01-1127, 2009,doi:10.4271/2009-01-1127.

7. Tuner, M., “Potential ESC Performance of a Multi-Cylinder Heavy Duty PPC Truck Engine: SystemSimulations based on Single Cylinder Experiments,” SAETechnical Paper 2013-01-0268, 2013, doi:10.4271/2013-01-0268.

8. http://www.gtisoft.com/

9. Shen, M., Tuner, M., and Johansson, B., “Close toStoichiometric Partially Premixed Combustion -The Benefitof Ethanol in Comparison to Conventional Fuels,” SAETechnical Paper 2013-01-0277, 2013, doi:10.4271/2013-01-0277.

10. Girard, J., Montreuil, C., Kim, J., Cavataio, G. et al.,“Technical Advantages of Vanadium SCR Systems for DieselNOx Control in Emerging Markets,” SAE Int. J. Fuels Lubr.1(1):488-494, 2008, doi:10.4271/2008-01-1029.11. Kosaka, H., Wakisaka, Y., Nomura, Y., Hotta, Y. et al.,“Concept of “Temperature Swing Heat Insulation” inCombustion Chamber Walls, and Appropriate Thermo-Physical Properties for Heat Insulation Coat,” SAE Int. J.Engines 6(1):142-149, 2013, doi:10.4271/2013-01-0274.12. Chang, J., Güralp, O., Filipi, Z., Assanis, D. et al., “NewHeat Transfer Correlation for an HCCI Engine Derived fromMeasurements of Instantaneous Surface Heat Flux,” SAETechnical Paper 2004-01-2996, 2004, doi:10.4271/2004-01-2996.13. Soyhan H.S., Yasar H., Walmsley H., Head B., KalghatgiG.T., Sorusbay C., Evaluation of heat transfer correlations forHCCI engine modeling, Applied Thermal Engineering,Volume 29, Issues 2-3, February 2009, Pages 541-549, ISSN1359-4311, 10.1016/j.applthermaleng.2008.03.014.14. Fridriksson, H., Sundén, B., Hajireza, S., and Tunér, M.,“CFD Investigation of Heat Transfer in a Diesel Engine withDiesel and PPC Combustion Modes,” SAE Technical Paper2011-01-1838, 2011, doi:10.4271/2011-01-1838.

CONTACT INFORMATIONmartin.tuner@energy.lth.se

ACKNOWLEDGMENTSSupport for this research was provided from the CompetenceCenter for Combustion Processes, KCFP.

Dr. John Gaynor at Scania is kindly acknowledged for hissupport.

ABBREVIATIONSAIT - Auto-ignition TemperatureATDC - After Top Dead CenterCAC - Charge Air CoolerCAD - Crank Angle DegreeEGR - Exhaust Gas RecyclingEGR-C - EGR-CoolerESC - European Stationary CycleGIE - Gross Indicated EfficiencyHCCI - Homogeneous Charge Compression IgnitionHEDGE - High Efficiency Dilute Gasoline EngineHTL - Heat Transfer LossesHTM - Heat Transfer MultiplierIMEPg - Gross Indicated Mean Effective PressureISC - Inter Stage CoolerIVC - Inlet Valve ClosureNVO - Negative Valve Overlap

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

PPC - Partially Premixed CombustionRCCI - Reactivity Controlled Compression IgnitionRoHR - Rate of Heat ReleaseRON - Research Octane NumberSCR - Selective Catalytic ReductionVGT - Variable Geometry Turbine

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

Table A1. Engine specifications.Compression ratio was adjusted to 17.0 in the calculations.

Figure A1. RoHR from the four single cylinder PPC engine experiments using 89RON gasoline.

Figure A2. Measured versus calculated cylinder pressure for A100.

APPENDIX

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM

Figure A3. Measured versus calculated cylinder pressure for A75.

Figure A4. Measured versus calculated cylinder pressure for A60.

Figure A5. Measured versus calculated cylinder pressure for A40.

The Engineering Meetings Board has approved this paper for publication. It hassuccessfully completed SAE's peer review process under the supervision of the sessionorganizer. This process requires a minimum of three (3) reviews by industry experts.

All rights reserved. No part of this publication may be reproduced, stored in aretrieval system, or transmitted, in any form or by any means, electronic, mechanical,photocopying, recording, or otherwise, without the prior written permission of SAE.

ISSN 0148-7191

Positions and opinions advanced in this paper are those of the author(s) and notnecessarily those of SAE. The author is solely responsible for the content of the paper.

SAE Customer Service:Tel: 877-606-7323 (inside USA and Canada)Tel: 724-776-4970 (outside USA)Fax: 724-776-0790Email: CustomerService@sae.orgSAE Web Address: http://www.sae.orgPrinted in USA

Downloaded from SAE International by Lund University, Friday, November 01, 2013 06:48:14 AM