CC BY-NC-ND
0Alexandria Engineering Journal (2017) xxx, xxx-xxx
HOSTED BY
Alexandria University Alexandria Engineering Journal
www.elsevier.com/locate/aej www.sciencedirect.com
ORIGINAL ARTICLE
Simulating the effects of turbocharging on the emission levels of a gasoline engine
Amir Reza Mahmoudia, Iman Khazaeeb'*, Mohsen Ghazikhanic
a Department of Renewable Energies and Environment, University of Tehran, Iran b Faculty of Mechanical and Energy Engineering, Shahid Beheshti University, A.C., Tehran, Iran c Department of Mechanical Engineering, Ferdowsi University of Mashhad, Iran
Received 13 July 2016; revised 28 January 2017; accepted 4 March 2017
KEYWORDS
Emission;
Gasoline SI engine; Turbocharging; GT-Power; 1-D simulation; Brake specific
Abstract The main objective of this work was to respond to the global concern for the rise of the emissions and the necessity of preventing them to form rather than dealing with their after-effects. Therefore, the production levels of four main emissions, namely NOx, CO2, CO and UHC in gasoline engine of Nissan Maxima 1994 is assessed via 1-D simulation with the GT-Power code. Then, a proper matching of turbine-compressor is carried out to propose a turbocharger for the engine, and the resultant emissions are compared to the naturally aspirated engine. It is found that the emission levels of NOx, CO, and CO2 are higher in terms of their concentration in the exhaust fume of the turbocharged engine, in comparison with the naturally aspirated engine. However, at the same time, the brake power and the brake specific emissions produced by the turbocharged engine are respectively higher and lower than those of the naturally aspirated engine. Therefore, it is concluded that, for a specific application, turbocharging provides the chance to achieve the performance of a potential naturally aspirated engine while producing lower emissions.
© 2017 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Spark ignition engines (SI, also known as gasoline engines) and compression ignition (CI, also known as diesel engines or direct injection (DI) engines) are the two major categories of internal combustion engines. SI engines have been widely used across Europe, while CI engines are wide spread globally. Their main application is for transportation. The major
* Corresponding author.
E-mail addresses: i_khazaei@sbu.ac.ir, Imankhazaee@yahoo.com (I. Khazaee).
Peer review under responsibility of Faculty of Engineering, Alexandria University.
emissions of the SI engines are NOx, CO, CO2, and unburned hydrocarbons (UHCs). NOx is a general term referring to nitrogen oxides (i.e. NO (nitrous oxide), NO2 (nitrogen dioxide)). These pollutants form in extremely high temperature in-cylinder conditions of SI engines in the presence of nitrogen and oxygen from the air and the fuel [1]. Nitrogen oxides form acid rain which harms human health, deteriorates structures, pollutes waters and marine ecosystems, and destructs forests [2,3]. Moreover, they can penetrate deeply into lung tissues and cause irritations, cause or worsen respiratory diseases, or deteriorate heart diseases [4]. Carbon dioxide (CO2) is the main product of the hydrocarbon combustion reactions. The anthropogenic carbon dioxide is known as the main reason for aggravation of the
http://dx.doi.org/10.1016/j.aej.2017.03.005
1110-0168 © 2017 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
greenhouse effect and global warming [5,6]. Carbon monoxide is a colorless, odorless, and tasteless gas that is slightly less dense than air. It is a product of partial oxidation of carbohydrates and forms when there is not enough oxygen available to combustion. In the presence of enough oxygen CO burns to produce carbon dioxide (CO2). CO has about 220 times greater potential to combine with hemoglobin in blood and thus lessens the oxygen delivery to the tissues and causes suffocation [7]. Moreover, CO contributes to formation of the tropospheric ozone [5]. Unburned hydrocarbons are actually the fuel molecules not burned during the combustion process in cylinder. Situated inside the crevices of the combustion chamber or precipitate on the walls of the chamber, they avoid the flame zone and thus remain unburned. These unburned fuel molecules then find their way to the exhaust fume and eventually the atmosphere [1]. The unburned hydrocarbons can cause respiratory problems, allergic issues or even immune deficiencies in children [8]. It is worth noting that there are significant evidence on the effects of generic air pollutants on fetal growth, preterm delivery, cardiac arrhythmia, and mortality [9-11].
One of the methods to improve the performance of internal combustion engines is to turbocharge them. Turbocharging was invented between 1909 and 1915 by Alfred Buechi, the Swiss engineer [12]. Movahed et al. researched the effects of concomitant injection of gasoline and CNG in 1.6 L tur-bocharged spark-ignition (SI) engine [13]. They revealed that BSFC, HC and CO emissions of gasoline mode at 4500 rpm and full load conditions are higher than 30% CNG mass fraction by about 24%, 80% and 75%, respectively. However, the cylinder peak pressure and raw NOx emissions of gasoline mode are lower by about 10% and 66%, respectively. Raw NOx emissions at 4500 rpm and full load conditions are 68% higher in the CNG mode, while BSFC, HC and CO emissions are 19%, 44% and 83% lower, consecutively. Silva et al. investigated the effects of turbocharging on a 2 L, light-duty gasoline engine. Their results reveal that turbocharging is an inexpensive and effective way for reducing fuel consumption and greenhouse gases production. 20% downsizing the engine, and boosting 20-50% the inlet pressure, they could lessen the fuel consumption about 6-14%. In addition, there is a decrease in pollutants: 2-4% for HC, 7-20% for CO, and 823% for NO [14]. Petitjean et al. researched fuel consumption and emissions in turbocharged SI gasoline engine experimentally [15]. They observed that the smaller turbocharged engine produces higher torques in the majority of the working speeds. They claim that CO2 emissions are significantly lower in tur-bocharged engines and the reason behind this is the improved fuel economy in turbocharged operations. That is, downsized turbocharged engines are capable of producing the same amount of power and torque as those of naturally aspirated engines, consuming less fuel and, therefore, releasing less carbon dioxide. Leduc et al. also suggested downsizing of gasoline engines as an efficient way to reduce CO2 emissions [16]. Their results revealed that use of small 0.8 L instead of genuine 1.6 L engine for powering the reference car, all other parameters being constant, leads to a reduction in vehicle fuel consumption. Olsson et al. claimed that using turbocharging on an HCCI engine is an effective way to reach high loads [17]. The ability of HCCI to provide high loads with very low NOx emissions is proven. The low exhaust temperature, which is characteristic of HCCI, combined with turbo charging
causes pump losses. To keep efficiency high, it is therefore important not to use higher boost than needed to reach desired load. Gharehghani et al. experimentally investigate the thermal balance and performance of a turbocharged gas spark ignition engine [18]. The first law of thermodynamics was used for control volume around the engine to compute the output power, transferred energy to the cooling fluid, exhaust gases and also unaccounted losses through convection and radiation heat transfer. They also investigated the effect of shifting from gasoline to CNG fuel on production of NOx, HC and CO as well as performance characteristics of a turbocharged SI engine. They could reach about 85% increase in power using a turbocharger on SI engine running on CNG. Performance and emission improving of turbocharged DI diesel engine is evaluated by Mohebbi et al. [19]. Their results suggest that, although EGR is effective to reduce NOx by lowering peak of cylinder pressure and temperature, there is a substantial trade-off in increased BSFC and PM emissions due to the reduction of oxygen concentration in the cylinder intake air. Karabektas revealed in his paper the effects of turbocharger on the performance of a diesel engine using diesel fuel and biodiesel, in terms of brake power, torque, brake specific consumption and thermal efficiency, as well as CO and NOx emissions [20]. The evaluation of experimental data showed that the brake thermal efficiency of biodiesel was slightly higher than that of diesel fuel in both naturally aspirated and turbocharged conditions, while biodiesel yielded slightly lower brake power and torque along with higher fuel consumption values. It was also observed that emissions of CO in the operations with biodiesel were lower than those in the operations with diesel fuel, whereas NOx emission in biodiesel operation was higher. Shahed and Bauer confirmed that 40% downsizing of a DI diesel engine, along with turbocharging, yields a 23% reduction in fuel consumption [21]. Less fuel consumption means less CO2 emissions. Also, Pakale and Patel confirmed that turbocharging substantially reduces nitrogen oxides and improves fuel efficiency and power density [22]. Vitek et al. dealt with the investigation of turbocharger optimization procedures using amended 1-D simulation tools. The proposed method uses scaled flow rate/efficiency maps for different sizes of a radial turbine together with a fictitious compressor map [23]. King incorporated an invented 1-D model into the GT-Power engine simulation code to predict on-engine performance of the turbocharger. He investigated the effects of unsteady operation of turbocharger on internal combustion engine's performance [24].
There seems to exist a gap in the previous literature about the effects of turbocharger on the emissions of gasoline engines, considering the high useful powers turbocharging will provide. The main objective of this work was to fill this gap while responding to the global concern for the rise of the emissions and the necessity of preventing them to form rather than dealing with their after-effects. Therefore, the production levels of four main emissions, namely nitrogen oxides (NOx), carbon dioxide (CO2), carbon monoxide (CO) and unburned hydrocarbons (UHC) in gasoline engine of Nissan Maxima 1994 are assessed via 1-D simulation with the GT-Power code. Then, a proper matching of turbine-compressor is carried out to propose a turbocharger for the engine, and the resultant emissions are compared to the naturally aspirated engine, after a 1-D simulation of the turbocharged engine is carried out, utilizing the same code.
2. Methodology
A brief description of the combustion, emission, heat transfer and friction losses calculation methods and equations available in the simulation software which are used in this work is provided in this section. Moreover, the engine and turbo specifications, atmospheric conditions and throttle designing considerations are described. These are followed by the validation of the model comparing the performance results with those of similar previous works in the literature. Further information on these subjects is available in [25].
In GT-POWER, the combustion rate is controlled by the burn rate. This burn rate input may be either imposed or predicted, depending on the combustion model selected. The available combustion models are non-predictive and predictive models.
A non-predictive combustion model simply imposes a burn rate as a function of crank angle. This prescribed burn rate will be followed regardless of the conditions in the cylinder, assuming that there is sufficient fuel available in the cylinder to support the burn rate. Therefore, the burn rate will not be affected by factors such as residual fraction or injection timing. The Wiebe function used by the software to impose the burn rate for SI engines is as follows:
Inputs: AA = Anchor Angle, D = Duration, E = Wiebe Exponent ("def" = 2.0), CE = Fraction of Fuel Burned (also known as ''Combustion Efficiency"), BM = Burned Fuel Percentage at Anchor Angle ("def' = 50%), BS = Burned Fuel Percentage at Duration Start ("def" = 10%), BE = Burned Fuel Percentage at Duration End ("def" = 90%).
Calculated Constants: Burned Midpoint Constant (BMC) = -ln(1-BM), Burned Start Constant (BSC) = -ln(1-BS), Burned End Constant (BEC) = -ln(1-BE).
Then the Wiebe constant (WC) and the Start of Combustion (SOC) parameters can be calculated with Eqs. (1) and (2).
BEC1/e+1 - BSC1/E+1
-(-E+1)
SOC = AA
(D)(BMC)
BEC^+0 _ BSC1/(E+!)
From which the combustion function at any crank angle is calculated with Eq. (3).
Combustion (h) = (CE) - e~
(WC)(9-SOC)(i
where Combustion (u) is the burned fraction ranging from 0.0 to 1.0, and u is the instantaneous crank angle.
The predictive model predicts the burn rate. In this model, input data about geometry of the cylinder, spark timing, air flow, and fuel properties are required. The mass flow of the fuel that enters the flame, and burn rate are calculated with these equations [25-28]:
d-M — PuAe(Sr + Sl)
dMb _ (Me - Mb) dt s
CkLt Re1'2
Puu'L l
in which, Me is the mass of unburned mixture entering the flame, t is time, pu is density of the unburned mixture, Ae is the flame front area, ST is the speed of turbulent flame assumed to be proportional to the turbulence intensity, SL is the speed of laminar flame, Mb is the mass of burned mixture, s is the time constant, k is Taylor scale length which is taken to be the representative dimension of the ignition sites within the flame front, Lt is turbulent length scale, Ret is the turbulent Reynolds number, and i is the absolute viscosity of the unburned gas. In theory, predictive combustion models are an appropriate choice for all simulations [25]. The laminar flame speed as a function of unburned gas conditions and equivalence ratio is computed from this correlation [28]:
S = sLr
(1 - aF-
where SLo is the laminar flame speed versus equivalence ratio at the reference conditions (To, Po), Tu is the instantaneous unburned gas temperature, P is the instantaneous cylinder pressure, and Fd is the diluent volume fraction. To is a reference temperature (298 K), Po is a reference pressure (1 atm), and a and b are constants and depend on equivalence ratio.
The turbulent flame speed is determined with the following equation [28]:
(1 1 + CkR}/L])
where u0 is turbulence intensity, Rf is the flame radius, and Cs and Ck are constants.
The flame front is assumed to be spherical in shape, and as it grows it gradually intersects combustion chamber surfaces. The sphere radius corresponding to the enflamed volume is determined by iteration. Further information on flame front determination method can be found in [28].
GT-Power uses two different strategies for predicting the in-cylinder species: equilibrium chemistry and chemical kinetics. The former assumes that the concentrations of species are as would occur if the current pressure and temperature were held constant for a long period of time. The latter takes into account the time required for the species to react and combine. This method is required for the calculation of NOx.
All GT-POWER combustion models include the capability to calculate NOx, CO, and CO2 concentrations. The concentration of unburned hydrocarbons in an SI engine, however, needs to be calculated only by a predictive model. The NOx emissions are calculated using the Extended Zeldovich mechanism, results of which are very sensitive to the trapped cylinder mass (i.e. engine airflow, EGR fraction, trapping ratio), fuel-to-air ratio, and combustion rate. NOx is also very sensitive to maximum cylinder temperature. Carbon dioxide and carbon monoxide are the primary products of hydrocarbon combustion and are always included in the results automatically. In
St — Csu
this work, the predictions were made using the equilibrium model. In order to raise the precision of the results, the method used in this work for predicting the UHC emissions is the more complex option available in which two plate quenching model is used along with a simple kinetic model after the flame is quenched. Some of the fuel and air mixture is pushed into the crevice volume and trapped there throughout compression. During the expansion process the trapped mixture starts to reenter the main cylinder volume. Any mixture that re-enters before the end of combustion will be burned according to the combustion model. Any mixture that re-enters after the end of combustion will be burned according to a kinetic model.
NOx includes all forms of oxides of nitrogen; however, in combustion engines it is dominated by nitric oxide NO followed by a small amount of nitrogen dioxide NO2. The main source for nitric oxide in internal combustion engines is the thermal NO which forms due to the dissociation of air nitrogen during combustion [29]. The reaction can be explained by the so-called Extended Zeldovich mechanism as follows [30]
N2 + O + N (11)
N + O2 $ NO + O (12)
-1-3 /—3
N + OH $ NO + H (13)
And the corresponding rate constants are
K+1 = 1.8 * 108 e"38370/Tm3 mol"1 s"1
= 3.8 * 107 e"425/T m3 mol"1 s"1 K+2 = 1.8 * 104 e"4680/T m3 mol"1 s"1 K_2 = 3.8 * 103 e"20820/T m3 mol"1 s"1 K+3 = 7.1 * 107 e"450/T m3 mol"1 s"1 K"3 = 1.7 * 108 e"24560/T m3 mol"1 s"1
The in-cylinder heat transfer is calculated by a model which imposes cylinder wall temperatures Tw to be 575 K, 575 K, and 375 K for cylinder head, piston, and cylinder wall, respectively [25]. The model starts with instantaneous heat flux written in terms of the heat transfer coefficient
q(t) = hg(t)[Tg(t) - Tw(t)]
where hg is heat transfer coefficient and varies spatially, Tg varies spatially depending on the location of the burned zone, and Tw varies spatially. Heat transfer coefficient hg is assumed to be correlated with the strength of the flow motions and is derived from Colburn analogy
hg = 1 CfpUeffCpP-2'3 (15)
Heat transfer coefficient is mainly driven by the product of pUeff. The effective velocity is defined as
Ueff =(U + uy + 2k) 1/2
where Ux and Uy are the two flow components parallel to the plane in question outside the boundary layer and k is the turbulence kinetic energy. According to the boundary layer theory, Cf is defined as
Cf = a(pUeffd /1)"1/4 (17)
where the constant a depends on the surface geometry and is assumed to be 0.0565. Values for P, and 1 are gas properties
and can be calculated knowing the gas temperature and composition. The model calculates velocities and turbulence near each of the resolved in-cylinder surfaces, and then finds the heat transfer coefficient through Colburn analogy. The flow model divides the combustion chambers' volume into regions, the temperatures of which are calculated based on the inputs provided to the software (aforementioned). Radial and axial velocities, swirl, turbulence intensity, and turbulence length scale need to be calculated in each subzone. Further details on the model and its submodels can be found in [28].
GT-POWER uses the Chen-Flynn model to calculate engine friction using the attributes as follows:
FMEP = C +(PF * Pm
(MPSF * Speedmp)
(MPSSF * Speed;;
where FMEP is Friction Mean Effective Pressure, Pmax is Maximum Cylinder pressure, Speedmp is Mean Piston Speed, C is Constant part of FMEP, PF is Peak Cylinder Pressure Factor, MPSF is Mean Piston Speed Factor, and MPSSF is Mean Piston Speed Squared Factor. This is an empirically derived model that states the total engine friction is a function of peak cylinder pressure, mean piston speed, and mean piston speed squared. FMEP is then used to calculate the output power and torque of the engine.
General specifications of the engine, and the fuel specifications are presented in Tables 1 and 2, respectively.
Air enters the intake system from the atmospheric conditions of the working environment that are introduced in Table 3.
Throttle is modeled with an orifice of variable coefficient of discharge. These coefficients are manipulated as recommended in the manuals of the software itself. Table 4 bears these coefficients and how they are dependent on the throttle angel.
Air-fuel ratios for different engine working conditions are defined between 12.6 for the most fuel-rich to 19.3 for the most fuel-lean combustions. For Indolene, this puts the fuel- air equivalent ratio / in the typical range of 0.75-1.15. The fuel-air ratio elevates with an increase in engine speed and/or load. Ignition delay is input in the range of 3.5-16 degrees of crank angle and is inversely related to the engine speed and load.
In order to select the proper turbocharger for the application, horsepower target, engine displacement, maximum rpm, and ambient condition are the parameters that need to be known. Some parameters including engine volumetric efficiency, intake manifold temperature, and BSFC of the
Table 1 Engine's specifications.
Brand Nissan maxima
Production date 1998
Engine type Gasoline V6
Bore 93 mm
Stroke 73.6 mm
Displacement 3.0 L
Compression ratio 10
Crevice volume 5 cc
Clearance 0.86 mm
Valve quantity in each cylinder 2*2
Max. brake torque 205 N m @ 4000 rpm
Max. brake power 190 hp @ 6400 rpm
Table 2 Fuel specifications.
Fuel name Indolene (Neat gasoline)
Molecular formula Ci8H25NO
Molecular weight 271.39 g/mol
Boiling point 398.9 C
Flash point 195 C
Density 0.98 g/cm3
Table 3 Atmospheric conditions.
Parameter Measure
Absolute pressure 100 kPa
Temperature 25 °C
Dry air pressure 99 kpa
turbocharged engine need to be assumed wisely. The assumed values for these parameters are presented in Table 5. Then, guided by the manufacturer, a procedure is followed in order to find the proper turbo model. Technical specifications of the selected turbo model are presented in Table 6 [31]. This model will not face surge or choke problems under the particular working conditions of this work.
Validation of the simulation results is based on the practical data reported and used in [12] as practical data from industry. A graph is drawn in Fig. 1 to make a comparison between the results of the simulation and those of the experimental researches. This graph depicts the full-load brake power curve
Table 5 Assumed parameters used in turbo selection.
Engine volumetric efficiency 95%
Intake manifold temperature 94 °C
BSFC of the turbocharger engine 0.5
of the engine in various speeds. As is evident, the minimum and the maximum deviation percentages from the practical data are 5.7% and 20.2%, respectively. In order to further facilitate the investigation of the deviation, Table 7 reports the error percentages of the simulation in 12 discrepant engine speeds.
Also, brake torque derived from the simulation is put in comparison with the results from a standard 3.0 L V6 engine, which is identical to Nissan Maxima 1998 engine, as reported by Lecointe and Monnier [32]. Fig. 2 depicts both curves in one graph for comparing purposes, and Table 8 facilitates this comparison by revealing the data points. As can be clearly seen, the minimum and the maximum deviation percentages from the standard engine are 3.3% and 14.8%, respectively.
3. Results and discussion
Figs. 3 and 4 reveal that brake power and brake torque in the turbocharged engine increase substantially to about double those in the naturally aspirated engine. Maximum brake power in the turbocharged condition is 371 HP at 6000 rpm, while maximum brake power in the naturally aspirated engine is 193 HP at 6000 rpm. Maximum brake torque in the turbocharged engine is 543 N m at 4000 rpm, while that of the naturally aspirated engine is 275 N m at 4000 rpm. In addition it is interpreted from these figures that greater ranges of brake power and torque are accessible with the turbocharged engine.
Fig. 5 shows two quasi-quadratic curves of NOx concentrations in 4500 rpm engine speed, side by side. The curve to the left belongs to the naturally aspirated engine, while the one to the right belongs to the turbocharged engine.
NOx stands for two different nitrogen oxides, NO and NO2, together. The formation of NOx emissions is substantially dependent on in-cylinder temperature during combustion and exhaust strokes, and also, to a lesser extent, on the concentration of oxygen inside the cylinder. In higher temperatures and oxygen concentrations, more NOx will be emitted. The climax of NOx concentration in exhaust gases is reached at just under the stoichiometric fuel-to-air ratio, since a compromise between maximum temperature and sufficient oxygen is established there [1]. Production of NO is of far higher level than that of NO2; therefore, the trend of NOx concentrations is accurately traceable, considering the trend of NO concentrations. As is shown in Fig. 5, peaks in concentrations of the NOx emissions are observable; thus, the fuel-air equivalence ratios for both naturally aspirated and turbocharged engines can be perceived to be close to the stoichiometric ratio, where NOx emissions typically peak [1].
The fact that these two curves are standing very far apart each other suggests that the turbocharged engine produces significantly higher concentrations of NOx emission at incomparably higher brake torques (roughly 2000 ppm and 300 N m
Table 4 Coefficient of discharge in different throttle angles.
Throttle Coefficient of Throttle Coefficient of
angle (deg) discharge angle (deg) discharge
0 0.01 50 0.375
5 0.016 60 0.507
10 0.04 70 0.63
20 0.106 80 0.714
30 0.177 85 0.8
40 0.263 90 0.99
Table 6 Technical specifications of GT3076R.
GTR3076R Compressor Turbine
Turbo PN CHRA PN Inducer Wheel Dia. Exducer Wheel Dia. Trim A/R Wheel Dia. Trim
700382-12 700177-7 57 mm 76.2 mm 56 0.6 60 mm 84
2501---——--1-i-—---
ID 1000 2000 3000 4000 5000 ~ 6000 7000
Engine speed (rpm)
Figure 1 Practical and simulation power results.
Table 7 Practical and simulation data points for power. Engine speed (rpm) Brake power (hp)
Practical data Simulation results % error
1000 33.4 27.7 17.2
1500 53.8 44.0 18.2
2000 75.5 62.6 17.1
2500 100.9 81.6 19.1
3000 123.8 103.3 16.6
3600 145.2 135.6 6.6
4000 163.5 154.2 5.7
4500 180.5 166.0 8.0
5000 197.5 181.9 7.9
5500 214.4 191.6 10.6
6000 226.6 193.7 14.5
6400 230.4 183.8 20.2
increments are observed at the peak points). Therefore, this type of comparison with concentrations does not seem very wise and dependable to an engineer, for it does not maintain a balanced and normalized perspective on the two prominent factors of power and emissions. In this day and age, reaching very high powers is sometimes essential and this will compromise the higher emission levels.
Consequently, a normalized figure, called brake specific emission level, is also utilized for reporting the comparison between the emission levels. The unit for brake specific emission is grams per kilowatt-hour (g/kw-h) and is derived from dividing the mass flow of the pollutant in g/h by the power in kW in the corresponding working condition.
Fig. 6, facilitates comparison between the naturally aspirated, the bar group on the left, and the turbocharged engine, on the right, in terms of g/kw-h. Different colors represent discrepant throttle angles, or loads, as introduced in the legend. As is evident, the brake specific NOx emissions of the naturally aspirated and the turbocharged engine are of the same order. This suggests that power and NOx emissions rise with roughly the same ratios when shifting to a turbocharged situation. Considering the graph more meticulously, we can infer that the emission levels in the turbocharged engine are even a trifle lower at higher loads.
Considering one of the bar groups, take the right one for example, for full load down to half load, the brake specific figures show a significant growth because of the rise in NOx concentrations. This rise in NOx concentrations is due to the falling fuel-air ratios toward stoichiometric ratio and rising temperature inside the cylinder. Also, power dwindles and encourages the changes, since it makes the denominator of the fraction used in calculating the brake specific emissions. From this point on, the excessive dilution of fuel-air mixture triggers a dramatic fall in temperature and, subsequently, NOx concentrations. This nosedive in the numerator is so potent that overwhelms the decrease in power which is the denominator, and the outcome is an obvious decrease in brake specific NOx emissions.
Fig. 7 depicts the carbon dioxide emission concentrations for the naturally aspirated and the turbocharged engine, at 4500 rpm and different loads. Generally, when plotted against fuel-air equivalent ratio, CO2 trend is a quasi-quadratic curve with its peak in moderate fuel-air equivalence ratios [1]. It can be perceived that in 4500 rpm, the corresponding fuel-air ratios are distributed around the extremum of the curve so that maximums of emission levels are observed at roughly 190 and 520 N m of brake torque, respectively in the naturally aspirated and the turbocharged engines. Moreover, it is declared by this figure, that carbon dioxide concentrations in the tur-bocharged engine are significantly higher, either are the torques. This discrepancy is about 1.5 x 104 PPM in peak points that occur in the 75% load condition.
Fig. 8 includes the brake-specific CO2 emissions of both engines at 4500 rpm. The naturally aspirated engine stands considerably higher than the turbocharged engine in brake-specific carbon dioxide emission levels. Generally speaking, this difference is an increase of about 150 g/(kw-h), that is roughly 17 percent of the naturally aspirated engine's CO2 emissions in the half-load condition. In addition this bar graph reveals that brake-specific CO2 emissions are not highly responsive to the changes in percentage of load.
According to the graph in Fig. 9, we can assert that carbon monoxide concentration levels are not relatively higher in tur-bocharged engine, when putting in perspective the results from the turbocharged and the naturally aspirated engine. The utmost difference is about 2000 ppm in the full load condition that is a change of roughly 20 percent. This suggests that
£ 200
-«-3 01 engine [28]
.....i..„ -"-Simulation results
__M \\
/ NTs.
! rj j : : :............: .........J
• .......:.......t..................
. i . ................. , . . . . 1
1000 2000 3000 4000 5000 Engine speed (rpm)
Figure 2 Practical and simulation torque results.
Table 8 Practical and simulation data points for torque. Engine speed (rpm) Brake torque (N m)
Data from [32] Simulation
% error
results
1000 227.9 197.0 13.5
1500 243.4 209.0 14.1
2000 253 222.7 12.0
2500 272.7 232.4 14.8
3000 282.2 245.1 13.2
3600 285.9 268.2 6.2
4000 284.7 274.5 3.6
4500 282.3 262.7 7.0
5000 271 259.1 4.4
5500 256.7 248.1 3.3
6000 240.7 229.9 4.5
6400 224 204.5 8.7
turbocharging the Nissan Maxima engine, as a model for the spark-ignition engine, will not substantially influence the concentrations of carbon monoxide in the exhaust gas.
Carbon monoxide (CO) emissions from internal combustion engines are controlled primarily by the fuel air equivalence ratio. For fuel-rich mixtures, CO concentrations in the exhaust fume increase steadily with increasing equivalence ratio, as the
amount of excess fuel increases and combustion faces a more severe lack of oxygen [1]. For fuel-lean mixtures, CO concentrations in the exhaust vary little with equivalence ratio. Since Spark ignition engines often operate close to stoichiometric at part load and fuel-rich at full load, CO emissions are significant and must be curbed.
As is expected based on what is perceived from the graph of CO concentrations, the brake specific amounts of carbon monoxide dwindle in the turbocharged condition (Fig. 10). This is caused when the concentration rises only by minute amounts, while the power dwindles more influentially, as applies to this case.
Overall, considering the boost in power and torque in the turbocharged engine, we can assert that this engine produces less greenhouse gases per unit power, and, therefore, is more efficient in this aspect.
Four possible HC emission formation mechanisms for spark-ignition engines (where the fuel-air mixture is essentially premixed) have been proposed [1]: (1) flame quenching at the combustion chamber walls, leaving a layer of unburned fuel-air mixture adjacent to the wall; (2) the filling of crevice volumes with unburned mixture which, since flame quenches at the crevice entrance, escapes the primary combustion process; (3) absorption of fuel vapor into oil layers on the cylinder wall during intake and compression, followed by desorption of fuel vapor into the cylinder during expansion and exhaust; and (4) incomplete combustion in a fraction of the engine's operating
350 300
100 so
i —1Ttnbcxjhargßd -^Naturally aspirated i .......... ^---?
^^ Max 371 ©6000
|Max193@6000|
3000 -1000
Engine Speed (rpm)
Figure 3 Brake power in turbocharged and naturally aspirated engine.
Engine Speed (rpm)
Figure 4 Brake torque in turbocharged and naturally aspirated engine.
о 5500
1 □ Turbo
250 350
Break Torque (N.m)
Figure 5 NOx concentration dependency on brake torque, in naturally aspirated (left) and turbocharged (right) engine, at 4500 rpm.
Figure 6 Brake specific NOx in naturally aspirated (left) and turbocharged (right) engine at 4500 rpm. Different loads are drawn in different colors.
cycles (either partial burning or complete misfire), occurring when combustion quality is poor.
When plotting unburned hydrocarbon concentration against equivalence fuel-air ratio, it firstly experiences a fall until the fuel-air ratios well under that of stoichiometric, and then recovers rapidly. This behavior is justifiable with taking into account the misfiring phenomenon that occurs in excessively lean mixtures where flame propagation faces difficulties and engine's performance becomes erratic. Moreover, later in richer fuel-air mixtures, the denser presence of fuel molecules
brings up lower temperatures, and both contribute to higher concentrations of unburned hydrocarbons in the exhaust fume.
The concentrations of unburned hydrocarbons in the exhaust of the naturally aspirated and the turbocharged engines at 4500 rpm are observable in Fig. 11. In very fuel-lean mixtures, the concentration of UHC is relatively higher due to misfiring. That is, combustion cannot propagate properly from one molecule to another, since they are too far apart each other. Also in fuel-rich mixtures UHC concentration rises again due to the reduction in temperature. Another major
Brake Torque i
Figure 7 CO2 concentration dependency on brake torque, in naturally aspirated (left) and turbocharged (right) engine, at 4500 rpm.
Figure 8 Brake specific CO2 in naturally aspirated (left) and turbocharged (right) engine at 4500 rpm. Different loads are drawn in different colors.
Brake Torque |
Figure 9 CO concentrations dependency on brake torque, in naturally aspirated (left) and turbocharged (right) engine, at 4500 rpm.
point in this figure is that the concentration of unburnt hydrocarbons is exceptionally lower in the exhaust fume that outflows the tail pipe of the turbocharged engine. This is apparently due to more pressure and temperature caused by the intense combustion in the turbocharged engine, that impels the chemical equilibrium coefficient to further move toward oxidization of these molecules later in the power stroke and during the exhaust stroke. Enough oxygen should exist adja-
cent to the fuel molecules to oxidize them during the aforementioned procedure.
Brake specific unburned hydrocarbon (HC) emission, follows the trend of concentrations and is minimum in half load (Fig. 12). In Fig. 12, starting from the full load condition and following the graph of either engines down to the half load condition, the HC concentration is nose-diving in an extent that fully takes over the decrease in power, resulting in the
Figure 10 Brake specific CO in naturally aspirated (left) and turbocharged (right) engine at 4500 rpm. Different loads are drawn in different colors.
Figure 11 UHC concentration dependency on brake torque, in naturally aspirated (left) and turbocharged (right) engine, at 4500 rpm.
Figure 12 Brake specific UHC in naturally aspirated (left) and turbocharged (right) engine at 4500 rpm. Different loads are drawn in different colors.
decrease in brake specific emissions. From the half load condition on, the page turns over and the HC concentration, as confirmed by Fig. 11, is on a growth that causes the numerator of the brake specific to become greater and the brake specific graph (Fig. 12) depicts a rise in 25 percent load.
In addition, having been scrutinized, the graph in Fig. 12 reveals that the turbocharged engine leaves less unburned hydrocarbons for production of unit power. This is because of less HC concentrations due to higher temperatures in the
turbocharged condition that as elaborated in the previous section burns the hydrocarbon leftovers and prevents them to enter exhaust. Also, the power levels in the turbocharged engine are considerably higher.
4. Conclusion
In this work, 1-D simulation code of GT-Power software has been utilized to investigate the effects of turbocharging on
emissions as well as performance of V6 gasoline engine of Nissan Maxima. In order to accomplish this goal, a model of the engine was built and validated with experimental data. The primary simulation declares that this model acceptably represents the engine and deviates insignificantly from that. The most important conclusions of this research are as follows.
For the naturally aspirated engine:
10. The performance levels of a naturally aspirated engine are achievable more environmentally benign, with the same engine turbocharged. This is possible with utilizing a smaller turbocharged engine which is capable of delivering the same power range as the naturally aspirated engine. This will lead to significantly lower brake specific emissions.
1. In the naturally aspirated engine, the maximum torque and power are 275 Nm and 193 HP at 4000 and 6000 rpm, respectively. These figures are close to those reported by the manufacturer (205 Nm, 190HP) (Figs. 3 and 4).
For the turbocharged engine:
2. In the turbocharged engine, the maximum brake torque, 543 N m, is reached at 4000 rpm, whereas the maximum horsepower, 371 HP, is produced at 6000 rpm (Figs. 3 and 4).
3. Turbocharging provides near-to-peak powers in a wider range of engine speeds, in comparison with the same engine, naturally aspirated (Fig. 3).
4. NOx emissions show different trends in different speed zones. In 4000 rpm, there is a peak in the curve at about 490 N m of brake torque (Fig. 5).
5. CO2 emissions show different trends in different speed zones. In 4500 rpm, there is a peak in the curve at about 520 N m of brake torque (Fig. 7).
6. In constant speed, CO concentrations in exhaust gas increase with the increase in torque. This escalating trend is relatively slow at the beginning but becomes steeper later on (Fig. 9).
7. HC emissions increase in fuel-rich mixtures (higher loads) due to the falling temperatures and the dense presence of fuel (Fig. 11).
For comparison between turbocharged and naturally aspirated:
8. The peak torque and power are boosted roughly 97% and 95%, respectively. Their increase in other points of the engine speed range is lower, especially in low-end speeds. This is due to the policy in choosing the turbocharger which has been focused on boosting the maximum power. This policy leads to larger tur-bochargers which operate well in high-end speeds (Figs. 3 and 4).
9. Brake specific emissions of the turbocharged engine are all lower than those of the naturally aspirated engine, correspondingly; however, emission concentrations in the turbocharged engine are higher, except for HC concentrations. This suggests that the increase in power will compensate for the increase in emissions. Note that the reason behind the emission concentrations being higher in the turbocharged engine is that more fuel is consumed (leading to more carbon oxides) and higher in-cylinder temperatures occur (leading to higher NOx levels). However, this higher temperature eliminates the unburned hydrocarbons after the flame quench, and leads to lower UHC emissions in the turbocharged engine.
Acknowledgment
We would like to thank Mr. Hossein Behzadan and Mr. Hojjat
Shabgard for their support, without whose help this work
would never have been possible.
References
[1] J.B. Heywood, Internal Combustion Engines Fundamentals, McGraw-Hill, New York, 1988.
[2] G.E. Likens, C.T. Driscoll, D.C. Buso, Long-term effects of acid rain: response and recovery of a forest ecosystem, Science 272 (5259) (1996) 244-246.
[3] S. Eyssautier-Chuine, B. Marin, C. Thomachot-Schneider, G. Fronteau, A. Schneider, S. Gibeaux, P. Vazquez, Simulation of acid rain weathering effect on natural and artificial carbonate stones, Environ. Earth Sci. 75 (9) (2016).
[4] U.S. Environmental Protection Agency, NOx: How nitrogen oxides affect the way we live and breathe, 1998.
[5] K.P. Shine, Radiative forcing of climate change, Space Sci. Rev. 94 (2000) 363-373.
[6] A. Berger, C. Tricot, The greenhouse effect, Surv. Geophys. 13 (1992) 523-549.
[7] D.G. Penney, Carbon Monoxide Toxicity, CRC Press, 2000.
[8] M.J. Mendell, Indoor residential chemical emissions as risk factors for respiratory and allergic effects in children: a review, Indoor Air 17 (4) (2007) 259-277.
[9] M. Maisonet, A. Correa, D. Misra, J.J.K. Jaakkola, A review of the literature on the effects of ambient air pollution on fetal growth, Environ. Res. 95 (1) (2004) 106-115.
[10] P.S. Shah, T. Balkhair, Air pollution and birth outcomes: a systematic review, Environ. Int. 37 (2) (2011) 498-516.
[11]J.P. Langrish, S.J. Watts, A.J. Hunter, A.S.V. Shah, J.A. Bosson, J. Unosson, S. Barath, M. Lundback, F.R. Cassee, K. Donaldson, T. Sandstrom, A. Blomberg, D.E. Newby, N.L. Mills, Controlled exposures to air pollutants and risk of cardiac arrhythmia, Environ. Health Perspect. 122 (7) (2014) 747-753.
[12] M. Falahaty, Simulation of a Turbocharged Nissan Gasoline Engine: the Study of Compression Ratio and Variable Valve Timing with GT-Power Software, Ferdowsi University of Mashhad, Iran, 2014.
[13] M.M. Movahed, H.B. Tabrizi, M. Mirsalim, Experimental investigation of the concomitant injection of gasoline and CNG in a turbocharged spark ignition engine, Energy Conver. Manage. 80 (2014) 126-136.
[14] C. Silva, M. Ross, T. Farias, Analysis and simulation of 'low-cost' strategies to reduce fuel consumption and emissions in conventional gasoline light-duty vehicles, Energy Conver. Manage. 50 (2) (2009) 215-222.
[15] D. Petitjean, L. Bernardini, C. Middlemass, S.M. Shahed, Advanced Gasoline Engine Turbocharging Technology for Fuel Economy Improvements. Paper No. 2004-01-0988, in SAE World Congress, 2004.
[16] P. Leduc, B. Dubar, A. Ranini, G. Monnier, Downsizing of gasoline engine: an efficient way to reduce CO2 emissions, Oil Gas Sci. Technol. 58 (2003) 115-127.
[17] J. Olsson, P. Tunestal, G. Haraldsson, B. Johansson, A Turbo Charged Dual Fuel HCCI Engine. Paper No. 2001-01-1896, in: International Spring Fuels & Lubricants Meeting & Exhibition, 2001.
[18] A. Gharehghani, M. Koochak, M. Mirsalim, T. Yusaf, Experimental investigation of thermal balance of a turbocharged SI engine operating on natural gas, Appl. Therm. Eng. 60 (2013) 200-207.
[19] A. Mohebbi, S. Jafarmadar, J. Pashae, Performance evaluation and emissions improving of turbocharged DI diesel engine with exhaust gas recirculation (EGR), Int. J. Autom. Eng. 2 (2) (2012) 95-106.
[20] M. Karabektas, The effects of turbocharger on the performance and exhaust emissions of a diesel engine fuelled with biodiesel, Renew. Energy 34 (4) (2009) 989-993.
[21] S.M. Shahed, K. Bauer, Parametric studies of the impact of turbocharging on gasoline engine downsizing, SAE Int. J. Eng. 2 (1) (2009) 1347-1358.
[22] P.N. Pakale, S.U. Patel, Performance Analysis of IC Engine using Supercharger and Turbocharger-a Review, 2015, pp. 17-22.
[23] O. Vitek, J. Macek, M. Polasek, New Approach to Turbocharger Optimization using 1-D simulation tools, SAE Technical Paper, 2006.
[24] A.J. King, A Turbocharger Unsteady Performance Model for the GT-Power Internal Combustion Engine Simulation, Purdue University, 2002.
[25] GT, GT-Power Engine Performance Application Manual, Gamma Technologies, 2009.
[26] S. Hires, R. Tabaczynski, J. Novak, The Prediction of Ignition Delay and Combustion Intervals for a Homogeneous Charge, spark ignition engine, SAE International, 1978.
[27] N.C. Blizard, J.C. Keck, Experimental and Theoretical Investigation of Turbulent Burning Model for Internal Combustion Engines, SAE Technical Paper, 1974.
[28] T. Morel, C.I. Rackmil, R. Keribar, M.J. Jennings, Model for Heat Transfer and Combustion in Spark Ignited Engines and Its Comparison with Experiments, in: International Congress and Exposition, 1988.
[29] S. Liviu-constantin, M. Daniela-elena, Simplified mechanism used to estimate the NOx emission of Diesel engine, in: 2nd International Conference on Manufactureing Engineering, Quality and Production Systems, no. 104, pp. 61-64.
[30] R.C. Flagan, J.H. Seinfeld, Pollutant formation and control in combustion, in: Fundamentals of Air Pollution Engineering, Prentice-Hall, Englewood Cliffs, New Jersey, 1988.
[31] Garrett, Garrett Catalog, vol. 5. Garrett by Honeywell, 2015.
[32] B. Lecointe, G. Monnier, Downsizing a Gasoline Engine Using Turbocharging with Direct Injection. Paper No. 2003-01-0542, in SAE World Congress, 2003.
