Applied Energy 205 (2017) 903–910

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

A computational analysis of the impact of bore-to-stroke ratio on emissions MARK

and efficiency of a HSDI engine
⁎
J. Benajesa, R. Novellaa, J.M. Pastora, , A. Hernández-Lópeza, T. Duvergerb
a
CMT – Motores Térmicos, Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain
b
Combustion and Fuel Supply Functions, PSA Groupe, 18 rue des Fauvelles BP16, 92256 La Garenne Colombes Cedex, France

H I G H L I G H T S

• AEfficiency
CFD study of the impact of bore-to-stroke on engine performance has been done.
• Higher NOxincreases with smaller B/S ratio due to lower heat losses.
• Lower B/S ratio
emissions for lower B/S ratio mainly due to faster mixing and combustion.
• decreases fuel rich pockets in squish region and then soot formation.

A R T I C L E I N F O A B S T R A C T

Keywords: Research on combustion systems for Internal Combustion Engines (ICE) is guided by the necessity of improving
Diesel engine engine efficiency while achieving the pollutant regulations. In this framework, this study identifies and describes
CFD model the effect of the bore-to-stroke ratio (B/S) on the combustion system performance and emissions by means of
Bore-to-stroke ratio computational fluid dynamics (CFD).
Engine efficiency
The study is applied to a 4-cylinder 4-stroke High Speed Direct Injection (HSDI) CI engine. It is divided in two
Emissions control
parts, the first part is focused on one operating point and presents a detailed description of the main effects of
different B/S ratios configurations, and the second part compares the results with different engine operating
conditions. For both parts the air management, injection settings and compression ratio were kept constant in
order to isolate the impact of the B/S ratio.
The results confirmed that the indicated thermal efficiency was increased for lower B/S ratio because of the
combustion chamber surface area decrease and faster combustion. Regarding the emissions, NOx and soot
presented a strong and opposed dependence on B/S ratio generated mostly due to enhanced air–fuel mixing for
lower B/S ratio. Finally, those trends were proven to be independent from the operating condition, giving the
study a more general value.

1. Introduction Historically, the optimization strategies in compression ignition (CI)

engines have been focused on improving air management [1], injection
Optimizing the Internal Combustion Engines (ICE) for transport settings [2], nozzle definition or piston bowl geometry [3–5]. This trend
applications is a major research topic due to the needs of fulfilling the continues due to the increasingly stringent emission standards and
current and future pollutant emission regulations. For that reason, the further CI engine investigation focuses on including dual fuel strategies
investigation in this field has been driven by the need of decreasing [6] or fuel blends coupled with injection settings optimization [7].
pollutants and overcome mechanical limitations. Nowadays, especially However, the basic geometric architecture, such as bore-to-stroke, tend
for Compression Ignition (CI) engines, keeping their fuel consumption to be kept unchanged or modified within a quite limited range for small
levels competitive compared to the Spark Ignition (SI) engines to con- and some medium sized diesel engines. This fact has generated a recent
trol CO2 emissions have gained importance and some aspects, such as interest since a fine tuning of B/S ratio has potential to improve heat
heat transfer, have become even more relevant in the current frame- transfer, mixing time or friction [8], resulting in an overall improve-
work of ICE research. ment in engine efficiency or power density. It is interesting, therefore,

⁎
Corresponding author.
E-mail address: jopasen@mot.upv.es (J.M. Pastor).

http://dx.doi.org/10.1016/j.apenergy.2017.08.023
Received 18 May 2017; Received in revised form 31 July 2017; Accepted 8 August 2017
0306-2619/ © 2017 Elsevier Ltd. All rights reserved.
J. Benajes et al. Applied Energy 205 (2017) 903–910

Nomenclature HRR heat release rate

HRL accumulated heat release law
aTDC After Top Dead Center ICE Internal Combustion Engines
BDC Bottom Dead Center IMEP indicated mean effective pressure
B/S bore-to-stroke IVC Intake Valve Closing (angle)
CA90 crank angle for 90% of fuel burnt mIVC in-cylinder mass at IVC
CI Compression Ignition PIVC pressure at IVC
CFD computational fluid dynamics TDC Top Dead Centre
DI Direct Injection TIVC temperature at IVC
EGR Exhaust Gas Recirculation YO2 in-cylinder oxygen concentration
EVO Exhaust Valve Opening (angle) YN2 in-cylinder nitrogen concentration
EXP experimental YRES in-cylinder residuals concentration

to re-examine those scaling relationships traditionally used to describe even use correlations to simulate its behavior. Vassallo et al. [17] car-
engine performance, and assess their impact on 4-cylinder 4-stroke ried out an investigation of the impact of B/S ratio also on a light-duty
High Speed Direct Injection (HSDI) CI engine. engine. The results were obtained by employing a 1D approach based
Experimental optimization is a well-known method due to the on Design of Experiments (DOE) methodology. They show that B/S
simplicity of adjusting air management, injection setting or fuel com- ratio close to 1, enables a good equilibrium between specific power
position aiming for an improved combustion process. Therefore, in the rating and low-end torque thanks to the volumetric efficiency profile
past years a lot of the research works focused on combustion system and the lower heat transfer by convection in the pipes. The higher
optimization, have been performed experimentally [9–11]. However, volumetric efficiency for this ‘square design’ and lower convection is
when engine hardware, such as the combustion chamber or the injector proven to compensate its unfavorable surface-to-volume ratio. In ad-
nozzle definition, is involved in the optimization tasks, the process is dition, the longer stroke length of low B/S ratio generates higher fric-
costly in terms of time and resources since it involves manufacturing tion losses leading to better thermal efficiencies for higher B/S ratios.
and assembling parts, together with weeks or even months of intensive The next step in the computational modeling approach, which is the use
testing. For that reason, limited experimental research has been re- of 3D CFD simulations, have also been performed [18]. As proven, even
ported in the literature discussing the optimum bore-to-stroke (B/S) operating with advanced combustion concepts they offer a more de-
ratio for engines equipped with state-of-the-art air management and tailed insight on the effect of B/S ratios and its interaction with the
injection systems, and also restricted to the current emissions con- combustion process, in-cylinder pollutants distributions and heat
straints. Kermani et al. [12] performed an experimental study of the transfer, allowing a better understanding of the complete problem.
influence of the B/S ratio on fuel consumption, performance and From the previous discussion, due to the recent interest on the po-
emissions on a light-duty Diesel engine. Experiments proved how tential benefits in terms of engine efficiency attainable by redefining
higher B/S ratios increase heat transfer losses, due to the higher com- the optimum B/S ratio and the high computational cost of integrating
bustion chamber area-to-volume ratio at TDC, resulting in a decrease of CFD modeling activities in the engine predesign stage, it is evident how
the indicated mean effective pressure (IMEP). Additionally, it was also the key effects of B/S ratio and how they interact are still not well
shown that decreasing B/S ratios increase the piston swept surface re- understood. In this framework, the research work reported in the pre-
sulting in an increase of friction losses. The study converges to an op- sent paper focuses on the identification and description of the impact of
timum B/S ratio range of 0.9–0.95, being 0.85 the smallest B/S ratio the B/S ratio on the trade-off between thermal efficiency and emissions
evaluated. Also from an experimental study, Lavoisier et al. [13] by means of CFD modeling, this is, taking into account the detailed
pointed out that higher B/S ratio will allow wider bowls that improve interaction between B/S ratio and the combustion process. The process
combustion process at high speed and load. This could be counteracted carried out in this paper has two main blocks, the first one focuses on
by the less favorable k-factor (ratio of bowl volume to total volume of one operating point and presents all the important effects of different B/
the combustion chamber at TDC), which limits the air utilization. In this S ratios configurations. On the second block, the previous results are
study was found an optimal B/S ratio close to 0.93, taking into account compared against other operating points of the same engine.
maximum rated power, fuel consumption and pollutants.
In terms of engine efficiency, the B/S ratio has an impact on both 2. Experimental tools
heat losses to the walls and friction losses [14], so a balance has to be
found between those two counteracting effects. Several studies have 2.1. Engine characteristics
been performed investigating the mechanical losses, Payri et al. [15]
showed that the mechanical losses represents around 4–10% of the total The experimental data required for the calibration and validation of
fuel energy injected. Subsequently, those authors [16] further devel- the CFD model was obtained from a 4-cylinder 4-stroke High Speed
oped their research, proving that the friction between the piston and Direct Injection (HSDI) CI engine, equipped with a common-rail injec-
the liner is responsible for 50% of the total mechanical losses, so that tion system. Table 1 contains the main engine characteristic and Table 2
the piston friction results in 2–5% of the total energy. contains the main fuel characteristics.
Recently, computational modeling is gaining reliability in pre- All the experimental results used in this paper, and additional
dicting combustion and pollutants by using properly calibrated and
validated models, not only because the predicted results have a high Table 1
accuracy but also because it offers an in depth analysis that cannot be Engine main characteristics.
performed experimentally with the current measurement techniques.
Then, computational modeling has been used to further study the effect Engine data
Bore × stroke [mm] 85 × 88
of B/S ratio in CI engines, mostly for predesign purposes. During this
Unitary Displacement [cm3] 499.36
stage of the engine design, 1D models are dominant since they offer a Connecting rod length [mm] 145
quick and robust prediction of the whole engine behavior. However, Geometric compression ratio [–] 16.7
these models tend to excessively simplify the combustion process or Nozzle hole number 7

904
J. Benajes et al. Applied Energy 205 (2017) 903–910

Table 2 The diesel spray was simulated by using the standard Droplet
Fuel main characteristics. Discrete Model [19]. Spray atomization and break-up were simulated
by means of the Huh-Gosman [26] and Reitz-Diwakar [27] models,
Engine data
Cetane number [–] 46.6 respectively. Diesel fuel physical properties were given by the DF1 fuel
LHV [MJ/kg] 42.124 surrogate [28].
Stoichiometric air/fuel ratio [–] 14.802 In these simulations, turbulent flow was modelled by means of the
Density [kg/m3] 843.3 RNG k-ε model [29], with wall-functions based on the model from
Kinematic viscosity [cSt] 2.46
Angelberger [30] in order to account for wall heat transfer. An implicit
scheme was used for time discretization, while divergence terms used
Table 3
the second order Monotone Advection and Reconstruction Scheme
In-cylinder thermodynamic conditions at IVC. (MARS) [19]. Velocity-pressure coupling was solved by means of a
Pressure-Implicit with Splitting of Operators (PISO) algorithm [31].
Case PIVC [bar] mIVC [g] TIVC [K] YO2 [%] YN2 [%] YRES [%] The CFD model was thoroughly validated by simulating the three
1750 rpm @ 3.89 1.28 422.6 22.9 75.3 1.8
operating conditions under investigation, described in Table 3. The
Full Load mesh used for the CFD cases was a hybrid mesh with about 160,000
2280 rpm @ 2.63 0.94 417.48 20.3 66.7 13 cells of 0.3 mm near the wall and 0.5 mm for the rest of the domain.
Part Load The simulations were run with 16 processors and each simulation took
3750 rpm @ 3.97 1.24 442.97 22.7 74.7 2.6
between 20 h and 36 h depending on the operating condition. The re-
Full Load
sults of the CFD model compared with the experimental data in terms of
performance and pollutants, are included in Figs. 1–3 and Table 4.
information required to perform the study of the B/S ratio, were pro- Those figures show a fair agreement in terms of performance (IMEP)
vided by the engine manufacturer. Additionally, a database of injection and combustion characteristics (HRR). In addition, the final soot levels
ratio profiles for different injection settings, using the engine’s injection were close to experimental data after calibrating the model constants on
system, was created. An in-house 0-D model code capable of reprodu- the 3750 rpm @ Full Load operating condition. An over-prediction of
cing any injection rate profile [1] was used to generate the needed NOx values is observed for the 1750 rpm @Full-load load condition,
injection rate profiles to perform the calculations. probably related with the fast reduction in HRR during the main in-
jection that was not able to be captured properly by the CFD model
compared to experimental data. Nevertheless, the quality of the CFD
3. Modeling tools model was considered as suitable for carrying out the optimization
activities.
The section below describes the experimental and theoretical tools
used to carry out the research. This brief description focuses only on
4. Methodology
their most relevant characteristics.
This section shows how the bore-to-stroke variations were done and
3.1. CFD model the assumptions that were taken for performing the CFD calculations.
An important aspect of the present study is to isolate as much as
The StarCD code version 4.22 [19] was used to perform the CFD possible the effect of the B/S ratio. In this case, when changing this
simulations of the engine combustion system. The axisymmetry of the parameter, the boundary conditions and initialization setup were all
combustion chamber allows to create a sector mesh comprising 157412 kept constant and only the geometry was modified, so injection settings,
cells at BDC with periodic boundary conditions after performing a grid volumetric compression ratio and displacement were kept constant at
convergence study. Each case was calculated as a closed combustion the reference value. The study was focused on four B/S ratio levels,
cycle, this is from the intake valves closing (IVC) to the exhaust valves ranging from 0.8 to 1.1, using the original engine value, B/S = 0.966 as
opening (EVO) (from 247 to 519° aTDC with the TDC at 360 deg). The the reference value.
initial thermodynamic conditions and wall temperatures are obtained As seen in Fig. 4, the shape of the piston bowl was kept unchanged
from the experimental cylinder pressure signal by means of the in-house for all the cases, in order to isolate B/S effects. In order to keep the
combustion diagnosis code CALMEC [20,21]. The reference values used engine compression ratio, a compensation volume, which accounts for
for the three operating conditions considered are shown in Table 3. valve recesses and other not resolved cylinder head geometric details,
The combustion model was the ECFM-CLEH [22,23]. Concerning was generated in the cylinder head following the same approach as
pollutants, NOx were calculated using a built-in NOx NORA model, found in the literature [32]. However, the squish gap between the cy-
which is based on the tabulation of equilibrium values of three NOx linder head and the flat top of the piston (from just before the end of
species and their relaxation times following a perturbation [24]. A bowl to the full piston bore) was kept constant for all B/S ratios, and
sectional method based on a description of sections containing soot only this compensation volume was adjusted to keep constant the
particles of equal volume was used for soot formation and oxidation compression ratio for all cases. As shown in Table 5, the connecting rod
[25]. was modified in order to keep as constant as possible the evolution of

200 0.04 Fig. 1. Experimental vs CFD results at 1750 rpm@ Full Load. In-cylinder
CFD pressure (a) and normalized heat release rate (b).
HRR [-/cad]

EXP
Pcyl [bar]

Inj. rate
100 0.02

0 0
-50 0 50 100 -20 0 20 40 60
Crank angle [deg a TDC] Crank angle [deg a TDC]

905
J. Benajes et al. Applied Energy 205 (2017) 903–910

100 0.08 Fig. 2. Experimental vs CFD results at 2280 rpm @ Part Load. In-cylinder
CFD pressure (a) and normalized heat release rate (b).

HRR [-/cad]
0.06 EXP
Pcyl [bar]

Inj. rate
50 0.04

0.02

0 0
-50 0 50 100 -20 0 20 40 60
Crank angle [deg a TDC] Crank angle [deg a TDC]

200 0.04 Fig. 3. Experimental vs CFD results at 3750 rpm@ Full Load. In-cylinder

HRR [-/cad] CFD pressure (a) and normalized heat release rate (b).
EXP
Pcyl [bar]

Inj. rate
100 0.02

0 0
-50 0 50 100 -20 0 20 40 60
Crank angle [deg a TDC] Crank angle [deg a TDC]
(a) (b)

Table 4 identify if volumetric efficiency is significantly affected by the man-

Experimental vs CFD pollutants emissions and IMEP results for the engine. datory reduction of inlet valves diameter, the average Mach number of
the air flowing across the inlet valves during the intake stroke was
Case IMEP [bar] SOOT [g/h] NOx [g/h]
calculated following the classical Taylor formulation, and proven to be
1750 rpm @ Full-load EXP 25.96 7.23 480.0 lower than 0.5 in the worst case (max rpm and min B/S ratio), what is
CFD 25.37 7.3 739.1 lower than the limit for Mach number equal to 0.5 where compressi-
2280 rpm @ Part-load EXP 8.7 0.54 77.2 bility effects arise [14]. Regarding the rest gas contents, it has been
CFD 9.02 1.04 63.6
3750 rpm @ Full-load EXP 23.32 17.92 991
estimated that represents 4–6% of total trapped mass for the operating
CFD 22.7 17.51 1076.5 conditions evaluated, and then the variations due to B/S ratio has been
neglected. In addition, as the air-loop system was not modeled, no
pumping work losses were taken into account. The effect of decreasing
B/S ratio over the pumping work and then over specific parameters is
expected to be negative and non-negligible due to the extended sonic
flow period during the blow-down through smaller exhaust valves, but
the detailed evaluation of the specific parameters is beyond the scope of
the present research work.
Additionally, some friction loses estimations were performed.
Following Miles and Andersson [8], friction losses scale inversely with
B/S. In this case the B/S ratio is modified from 0.8 to 1.1 which re-
presents a maximum variation of 28% of the total friction loses, this is,
0.5–1.25% of the total injected energy [33]. However, as the analysis of
specific parameters is out of the scope of the present investigation, the
impact of this variation was not discussed.

5. Results and discussion

Fig. 4. Geometries with different B/S values. A detailed study was carried out with the aim of investigating the
impact of the B/S ratio on engine performance and emissions. Firstly, a
detailed analysis for the high-speed/load point (3750 rpm @ Full-load)
Table 5
Geometric characteristics for each B/S ratio case. has been performed. After that the results for the other two operating
conditions are presented.
B/S ratio 0.8 0.88 0.966 1.1
Bore [mm] 79.8 82.4 85 88.8
5.1. B/S impact at high-speed/load
Stroke [mm] 99.8 93.6 88 80.7
Connecting Rod length [mm] 164.4 154.3 145 133
k-factor [–] 0.88 0.87 0.86 0.845 Focusing on the combustion process, Fig. 5 shows accumulated heat
release law (HRL) for all the B/S ratios considered and the differences
relative to the reference B/S. It can be seen how reducing B/S ratio
the volume during the combustion process. produces an increase in accumulated heat release up to around 80 cad
It has to be pointed out that the air-loop system has not been si- after the start of combustion, when more than of 90% of the total HRL
mulated, then the impact of B/S ratio on open-loop cycle was not di- has been achieved. However, this advantage is later recovered by larger
rectly assessed. However some aspects about the hypothesis of un- B/S ratios achieving nearly the same value of HRL at the end of the
changed initial conditions have been considered. As a method to combustion.

906
J. Benajes et al. Applied Energy 205 (2017) 903–910

Fig. 5. Normalized HRL (a) and HRL differences relative to the reference B/S
case (b) at high-speed/load operating point.

shown in Fig. 8(b).

Differences in heat transfer can be explained by the results pre-
sented in Fig. 9. It is shown that despite lower B/S ratios have higher
wall heat transfer (HT) per surface area, probably due to faster HRL, but
noticeably lower surface wall area. As expected this is the parameter
that affects the most the heat losses when modifying B/S ratio.
In addition to combustion and efficiency, pollutants are also af-
fected by different B/S ratios, as presented in Fig. 10. Lower B/S ratios
produce higher NOx and lower soot emissions. The relative variations
are higher for soot than for NOx, especially when B/S increases.
Fig. 6. Combustion duration (CA90-CA10) normalized at high-speed/load operating
Focusing on NOx emissions, it could be expected from previous
point. results that better mixing leads to faster combustion and higher in-cy-
linder temperatures. In particular, Fig. 11 shows that for lower B/S the
amount of mass with high temperature and lean conditions increases,
and consequently [14], the NOx formation.
Regarding soot emissions, the general trend plotted in Fig. 12(a)
shows that lower B/S ratios generate lower soot but also that there is a
big difference between the formation and oxidation stages. The main
difference in soot mass is generated during the formation stage, be-
tween 0 and 25 cad, the difference for 0.8 and 1.1 B/S ratios is almost
50% of the maximum soot value. However, during the oxidation stage
that difference is reduced to around 15%. This is, the difference in soot
formation cannot be compensated by the oxidation stage where higher
B/S ratios have a higher oxidation rate, so the main difference in the
Fig. 7. Percentage of fuel mass with equivalence ratio over stoichiometry at high-speed/ final soot emissions comes from the formation stage.
load operating point.
Soot formation rate is affected by mass with high temperatures and
rich equivalence ratio. Then, Fig. 13 shows the mass with equivalence
In terms of combustion duration, defined here as the difference ratio over two and high temperature within bowl and squish regions, to
between the crank angles corresponding the 90% and 10% of total HRL better locate the main difference in soot formation. It can be seen that
(named as CA90 and CA10 respectively), Fig. 6 clearly indicates how is there the main differences between B/S ratios takes place within squish
reduced when B/S is decreased. region, where the lower B/S ratio has less rich and high temperature
Faster combustion for lower B/S ratios is due to improved air/fuel mass throughout the combustion process. Also Fig. 13(b) shows that the
mixing, as indicated in Fig. 7, where the amount of fuel mass over difference in rich mass generates nearly a 50% increase of peak soot
stoichiometric conditions is shown (only extremes plotted to facilitate mass within the squish region, being the main source for the soot var-
understanding). These differences in air/fuel mixing cannot be attrib- iations for different B/S ratios.
uted to different piston bowl characteristics such as in [13], but to the
different squish dimensions and k-factor (see Table 5) which results in 5.2. B/S impact at different operating conditions
modified in-cylinder spray and air flow interaction.
Concerning heat losses, Fig. 8(a) shows the total heat transfer (HT) In order to give a wider view of the effect of B/S ratio, the results
where it can be seen how lower B/S ratios results in lower heat transfer. obtained for both full- and part-load operating points with different B/S
Thus lower B/S ratio have faster mixing and combustion and have ratios are presented in this sub-section. All the plots show the values
lower heat transfer values, leading to higher indicated efficiency, as normalized by the reference B/S ratio in order to facilitate the

Fig. 8. Heat transfer (a) and Gross indicated efficiency (b) normalized by the
reference B/S case at high-speed/load operating point.

907
J. Benajes et al. Applied Energy 205 (2017) 903–910

Fig. 9. Wall heat transfer per surface area (a) and total surface wall area (b) at
high-speed/load operating point.

Fig. 10. NOx (a) and soot (b) normalized by the reference B/S case at high-
speed/load operating point.

Fig. 11. In-cylinder mass with temperature over 1700 K and ϕ < 1 (a) and
NOx mass normalized by the max. value (b) at high-speed/load operating
point.

Fig. 12. Soot mass (normalized by the max. value) at high-speed/load operating point. Fig. 14. Normalized combustion length for all the operating conditions of the engine.

comparison. terms of combustion duration.

Fig. 14 shows how the combustion duration is shortened for lower The heat losses and GIE are plotted in Fig. 15 for all the operating
B/S ratios for full-load conditions both peak torque (low-speed) and points. The general trends are kept for all operating conditions: lower
power (high-speed). This effect has not been found for the part-load B/S ratios results in lower heat losses and increased high-pressure cycle
point, which has shorter injection duration and also lower global indicated efficiency. For part-load condition the improvement comes
equivalence ratio, and it is less sensitive to the B/S ratio variations in basically from decreased heat losses, while for full load cases is due to

Fig. 13. Mass-averaged with Ф > 1 (a) and soot mass (b) in squish and bowl
regions at high-speed/load operating point.

908
J. Benajes et al. Applied Energy 205 (2017) 903–910

Fig. 15. Heat transfer losses (a) and GIE (b) for all operating conditions nor-
malized by the value of the reference B/S ratio for each operating condition.

Fig. 16. NOx (a) and soot (b) for all operating conditions normalized by the
value of the reference B/S ratio for each operating condition.

bore-to-stroke setup ended with higher efficiencies mainly because of

having lower heat transfer losses and faster combustion. Despite the
heat transfer per unit area was higher for lower B/S due to better
mixing and then faster combustion, the surface area (larger for high B/S
values) turned out to be the main reason for the differences in heat
transfer.
The NOx emissions presented a strong dependence on the B/S ratio,
lower B/S ratios generated higher NOx emissions. This trend was gen-
erated mainly because of the better mixing for lower B/S ratio config-
urations that ended up having higher amount of low equivalence ratio
Fig. 17. In-cylinder max temperature for all operation conditions. mass with high temperatures compared to the higher B/S ratio cases.
The soot emissions showed a substantial dependence on the B/S
ratio as well, lower B/S ratio generated lower soot emissions. The main
both reduced heat losses and combustion duration. The results and
soot generation difference was located in the squish region where the
trends obtained in CFD calculations are in close agreement with test
soot mass formed in the 0.8 B/S ratio case was almost 50% of the soot
bench results from [12], where higher IMEP and then indicated effi-
mass formed in the 1.1 B/S ratio case. That variation is coherent with
ciencies were obtained for the B/S = 0.85 engine, which was the
the noticeable difference in rich mass with high temperature trapped in
smaller B/S tested.
the squish region generated by the distinction in mixing.
Fig. 16 compares the NOx and soot emissions for the different op-
The efficiency and emissions trends exposed for the high-speed/load
erating points. The same trends as those observed for the high-speed/
(peak power) case were proven to be general for other operating con-
load point can be also observes for the other operating conditions: NOx
ditions of the studied engine. It has to be highlighted that for part load
increases with lower B/S ratios and soot emissions are increased for
conditions the trends were kept but the absolute effect of the B/S ratio
higher B/S ratios.
on the NOx emissions was almost negligible due to the lower average
The impact of B/S in NOx emissions is lower for the part load case.
temperature in the combustion process. This results in a more favorable
That case is the only case with EGR meaning that the temperatures
NOx-soot trade-off when using lower B/S configurations.
during the combustion are lower than in any other case under high-load
conditions. As it was said, the NOx formation ratio grows exponentially
Acknowledgments
with temperature, but due to the lower in-cylinder temperatures of the
part load case, as seen in Fig. 17, the increase in temperature due to
Authors acknowledge that this work was possible thanks to the
different B/S ratios has a lower impact. This results in a more favorable
Ayuda para la Formación de Profesorado Universitario (FPU 13/02817)
NOx-soot trade-off for part-load conditions when using lower B/S
belonging to the Subprogramas de Formación y de Movilidad del
configurations, also found in [12].
Ministerio de Educación, Cultura y Deporte from Spain. The authors
would also like to recognize the funding and technical support from
6. Conclusions PSA Groupe.

A CFD study of the impact of different bore-to-stroke configurations References

on HSDI CI engine efficiency and pollutants has been performed. Air
management, injection settings and compression ratio have kept con- [1] Zamboni G, Moggia S, Capobianco M. Hybrid EGR and turbocharging systems
stant in order to isolate the effect of the B/S. The B/S ratio study has control for low NOX and fuel consumption in an automotive diesel engine. Appl
Energy 2016;165:839–48.
been performed for three different operating conditions of the same [2] Poorghasemi K, Saray RK, Ansari E, Irdmousa BK, Shahbakhti M, Naber JD. Effect of
engine (full-torque, full-power and one part-load). diesel injection strategies on natural gas/diesel RCCI combustion characteristics in a
The GIE proved to be noticeably affected by the B/S ratio. Lower light duty diesel engine. Appl Energy 2017;199:430–46.

909
J. Benajes et al. Applied Energy 205 (2017) 903–910

[3] Benajes J, Novella R, Pastor JM, Hernández-López A, Hasegawa M, Tsuji N, et al. SAE Technical Paper 2013-24-0013; 2013.
Optimization of the combustion system of a medium duty direct injection diesel [18] Lee C, Goel S, Babajimopoulos A. The effects of stroke-to-bore ratio on HCCI
engine by combining CFD modeling with experimental validation. Energy Convers combustion. SAE Technical Paper 2010-01-0842; 2010.
Manage 2016;110:212–29. [19] Dukowicz J. A particle fluid numerical model for liquid sprays. J Comput Phys
[4] Shi Y, Reitz RD. Assessment of optimization methodologies to study the effects of 1980;35:229–53.
bowl geometry, spray targeting and swirl ratio for a heavy-duty diesel engine op- [20] Lapuerta M, Armas O, Hernández JJ. Diagnosis of DI Diesel combustion from in-
erated at high-load. SAE Int J Engines 2008;1:537–57. cylinder pressure signal by estimation of mean thermodynamic properties of the
[5] Ge HW, Shi Y, Reitz RD, Wickman DD, Willems W. Optimization of a HSDI diesel gas. Appl Therm Eng 1999;19(5):513–29.
engine for passenger cars using a multi-objective genetic algorithm and multi-di- [21] Payri F, Molina S, Martín J, Armas O. Influence of measurement errors and esti-
mensional modeling. SAE Int J Engines 2009;2:691–713. mated parameters on combustion diagnosis. Appl Therm Eng 2006;26(2):226–36.
[6] Pedrozo VB, May I, Dalla NM, Cairns A, Zhao H. Experimental analysis of ethanol [22] Subramanian G, Vervisch L, Ravet F. New developments in turbulent combustion
dual-fuel combustion in a heavy-duty diesel engine: An optimization at low load. modeling for engine design: ECFM-CLEH combustion submodel. SAE Technical
Appl Energy 2016;165:166–82. Paper 2007-01-0154; 2007.
[7] Beatrice C, Napolitano P, Guido C. Injection parameter optimization by DoE of a [23] Abouri D, Zellat M, Duranti S, Ravet F. Advances in combustion modeling in STAR-
light-duty diesel engine fed by Bio-ethanol/RME/diesel blend. Appl Energy CD: validation of ECFM- CLEH model to engine analysis. In: 18th International
2014;113:373–84. mutidimentional engine user's meeting at the SAE congress; 2008.
[8] Miles PC, Andersson Ö. A review of design considerations for light-duty diesel [24] Vervisch PE, Colin O, Michel J-B, Darabiha N. NO Relaxation Approach (NORA) to
combustion systems. Int J Engine Res 2016;17(1):6–15. predict thermal NO in combustion chambers. Combust Flame
[9] Genzale CL, Reitz RD, Musculus MPB. Effects of piston bowl geometry on mixture 2011;158(8):1480–90.
development and late-injection low-temperature combustion in a heavy-duty diesel [25] Marchal C. Modélisation de la formation et de l'oxydation des suies dans un moteur
engine. SAE Int J Engines 2009;1(1):913–37. automobile, PhD Thesis. Université d'Orléans; 2008.
[10] Benajes J, Pastor JV, García A, Monsalve-Serrano J. An experimental investigation [26] Huh KY, Gosman AD. A phenomenological model of diesel spray atomization. In:
on the influence of piston bowl geometry on RCCI performance and emissions in a Proceedings of the international conference on multiphase flows; 1991.
heavy-duty engine. Energy Convers Manage 2015;103:1019–30. [27] Reitz RD, Diwakar R. Structure of high-pressure fuel sprays. SAE Technical Paper
[11] Choi S, Shin S, Lee J, Min K, Choi H. The effects of the combustion chamber geo- 870598; 1987.
metry and a double-row nozzle on the diesel engine emissions. Proc Inst Mech Eng, [28] Habchi C, Lafissas FA, Béard P, Broseta D. Formulation of a one-component fuel
Part D: J Automobile Eng 2015;229(5):590–8. lumping model to assess the effects of fuel thermodynamic properties on internal
[12] Kermani J, De Paola G, Knop V, Garsi C, Ruhland H, Willems W, Kaudewitz T, Mork combustion engine mixture preparation and combustion. SAE Technical Paper
A. An experimental investigation of the effect of bore-to-stroke ratio on a diesel 2004-01-1996; 2004.
engine. SAE Technical Paper 2013-24-0065; 2013. [29] Yakhot V, Orszag SA. Renormalization group analysis of turbulence. I. Basic theory.
[13] Fasolo B, Doisy A, Dupont A, Lavoisier F. Combustion system optimization of a new J Sci Comput 1986;1(1):3–51.
2 liter diesel engine for EURO IV. SAE Technical Paper 2005-01-0652; 2005. [30] Angelberger C, Poinsot T, Delhay B. Improving near-wall combustion and wall heat
[14] Heywood JB. Internal combustion engine fundamentals. New York: Mcgraw-Hill; transfer modeling in SI engine computations. SAE Technical Paper 972881; 1997.
1988. [31] Issa RI. Solution of the implicitly discretised fluid flow equations by operator-
[15] Payri F, Olmeda P, Martin J, Carreño R. A new tool to perform global energy bal- splitting. J Comput Phys 1986;62(1):40–65.
ances in DI diesel engines. SAE Int J Engines 2014;7(1):43–59. [32] Lucchini T, Della Torre A, D'errico G, Montenegro et al. Automatic mesh generation
[16] Payri F, Olmeda P, Martin J, Carreño R. Experimental analysis of the global energy for CFD simulations of direct-injection engines. SAE Technical Paper 2015-01-0376;
balance in a DI diesel engine. Appl Therm Eng 2015;89:545–57. 2015.
[17] Vassallo A, Gopalakrishnan V, Arrigoni S, Cavallo R, Turcato R, Racca A. Impact of [33] Benajes J, Martín J, García A, Villalta D, Warey A. In-cylinder soot radiation heat
bore-to-stroke ratio over light-duty di diesel engine performance, emissions and fuel transfer in direct-injection diesel engines. Energy Convers Manage
consumption: an analytical study using 1D-CFD coupled with DOE methodology. 2015;106:414–27.

910