Applied Energy 212 (2018) 1–12
Contents lists available at ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Injection strategies for reducing smoke and improving the performance of a butanol-diesel common rail dual fuel engine
Jaykumar Yadav , A. Ramesh Internal Combustion Engines Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai, India
H I G H L I G H T S study on multiple injection of diesel in a butanol-diesel dual fuel engine. • Detailed plus post injection enhances energy eﬃciency of butanol dual fuel engines. • Main reduction of smoke, NO and fuel consumption were achieved. • Simultaneous injection quantity of 1.5 mg and oﬀset of 9.7° found to be optimal. • Post • Auto-ignition of butanol before diesel injection limits butanol energy share.
A R T I C L E I N F O
A B S T R A C T
Keywords: Butanol Dual fuel engines Post injection for soot reduction Injection strategies for dual fuel engines Engine emissions Alternative fuels for engines
In dual fuel engines auto-ignition of the inducted butanol creates a high temperature environment prior to the injection of diesel. This results in enhanced smoke emissions. This work was aimed at controlling the smoke level in a butanol diesel common rail turbocharged dual fuel engine through multiple fuel injections. Experiments were performed on a three cylinder turbocharged common rail diesel engine at a speed of 1800 rpm and BMEPs corresponding to 75% and 100% of full load (BMEP of 11.8 bar). Port fuel injectors along with dedicated circuitry were employed to control the quantity and timing of butanol introduction into the intake air. An open engine controller was used to vary the rail pressure, injection timing and number of pulses of the diesel that was directly injected into the combustion chamber. The injection timing of diesel was always set for best eﬃciency. First the eﬀect of Main plus Post Injection (MPI) of diesel at a ﬁxed butanol to diesel energy share (BDES) of 30% was evaluated at diﬀerent post injection quantities and main to post oﬀsets. Subsequently the inﬂuence of BDES was studied at a ﬁxed post injection quantity and oﬀset from the main injection. Finally Pilot plus Main Injection (PMI) of diesel, Main plus Post Injection (MPI) of diesel and Main plus Two Post Injections (MPTPI) of diesel were compared in the dual fuel mode. MPI resulted in improved brake thermal eﬃciency (BTE) and drastically reduced the smoke level because of enhanced mixing by the momentum of the post injected fuel. NO and CO2 were also reduced. Using high BDES values along with optimised post injection quantities and main to post oﬀsets reduced the smoke level. PMI of diesel resulted in lower BTE and higher smoke, while the only advantage was reduced NO levels. MPI was better than MPTPI with respect to all the parameters. On the whole, in a dual fuel engine that uses butanol and diesel the main plus post strategy is eﬀective in improving energy eﬃciency, reducing smoke and also in increasing the amount of butanol that can be utilized.
1. Introduction The increase in the demand for motive power, depletion of fossil
fuels and the need to reduce emissions including CO2 are motivating researchers to look for alternative engine fuels. On the other hand current electronically controlled fuel injection systems oﬀer the
Abbreviations: BDES, butanol to diesel energy share; BMEP, brake mean eﬀective pressure; BSFC, brake speciﬁc fuel consumption; BTE, brake thermal eﬃciency; CA50,, combustion phasing angle; CO, carbon monoxide; CO2, carbon dioxide; COV of IMEP, coeﬃcient of variation of indicated mean eﬀective pressure; deg. CA, degree crank angle; ECU, engine control unit; EGTs, exhaust gas temperatures; EVO, exhaust valve opening; FID, ﬂame ionization detector; FSN, ﬁlter smoke number; FPGA, ﬁeld-programmable gate array; HC, hydrocarbon; HRR, heat release rate; IMT, intake manifold temperature; MBT, maximum brake torque; MPI, main plus post injection; MPTPI, main plus two posts injection; NO, nitric oxide; MRPR, maximum rate of pressure rise; PMI, pilot plus main injection; PM, particulate matter; P1, Post-1; P2, Post-2; SPI, single pulse injection; TDC, top dead center ⁎ Corresponding author at: MS Scholar, Internal Combustion Engine Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai 600 036, Tamil Nadu, India. E-mail address: [email protected]
(J. Yadav). https://doi.org/10.1016/j.apenergy.2017.12.027 Received 1 October 2017; Received in revised form 24 November 2017; Accepted 2 December 2017 0306-2619/ © 2017 Elsevier Ltd. All rights reserved.
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Similar observations were made by other researchers [24,25]. The common rail fuel injection system for diesel with electronic controls has brought in great ﬂexibility and precision in control of fuel injection parameters. It has enabled injection in multiple pulses within the same cycle [26,27]. Multiple injection strategies in diesel engines generally consist of an early pilot injection of a small quantity followed by main injection and ﬁnally by post injection of a small quantity. In automotive engines it is common to use more than three injection pulses. The pilot injection helps in reducing combustion noise; however it can result in an increase in particulate matter (PM) because it reduces the ignition delay of the diesel that follows during the main injection . Post injection that comes after the main injection enhances mixing due to its kinetic energy and increases the in-cylinder temperature which reduces soot by improved oxidation [29,30,31]. Reduced particulate emission without much eﬀect on fuel consumption and HC emission has been observed in a diesel engine with post injection . Another study has also indicated the eﬀectiveness of post injection in lowering soot, with no eﬀect on NOx emission . Careful control of the timing of post injection was needed for reducing smoke . Post injection too close (oﬀset below 8°CA deg. CA) to the main injection resulted in an increase in smoke due to elevated temperatures during combustion . This could also elevate the NOx levels . However, in another work it was observed that too retarded post injection (above 13°CA) also did not help in lowering the smoke level. Further, with three injections (pilot-main-post) NOx emission was slightly decreased as compared to single injection . Smoke was found to decrease when a ﬁxed quantity of post injected diesel was used in a butanol dual fuel engine . Butanol helped in reducing particulate matter and NOx emissions with no signiﬁcant effect on brake thermal eﬃciency in an automotive dual fuel engine. However, it had a negative impact on CO and THC emissions . Oxygen in butanol was found to be the main reason for reduction in soot. Prolonged ignition delay with butanol induction also had a positive eﬀect . Increase in the ethanol fraction reduced smoke and NOx emissions in a dual fuel engine. However, HC and CO levels were found to increase. Ethanol is more eﬀective in reducing smoke and NOx emissions as compared to gasoline as it is an oxygenated fuel .When a butanol diesel blend (B20) was used in a compression ignition engine reduction in particle number and diameter were observed. The presence of the hydroxyl group in the butanol molecule led to a lower rate of PM formation and enhanced rate of PM oxidation as compared to neat diesel operation. Further, post injection enhanced the oxidation of soot and lowered the PM levels . It has also been reported that increase in the butanol content in the blend with diesel reduces soot emissions due to the presence of oxygen in the fuel . In the case of butanol diesel blends, the lower ﬂame temperatures reached due to the higher latent heat of vaporization and lower energy content of butanol will hinder NOx formation. On the other hand butanol increases the amount of oxygen and also decreases the cetane number, both of which can enhance NO formation. Experiments indicated that NOx emissions with n-butanol/diesel blends reduced with increase in the butanol content as compared to reference diesel . On the whole previous studies have indicated that at high outputs, butanol diesel dual fuel operation results in problems of enhanced smoke and high maximum rate of pressure rise. This is due to the autoignition of butanol prior to the injection of diesel. On the other hand injecting diesel in multiple pulses with in a combustion cycle has been eﬀectively used to control smoke, NOx and rate of combustion in modern diesel engines. These injection strategies can also be employed in dual fuel engines in order to mitigate some of the drawbacks explained earlier. Since this aspect has not been explored in detail so far, experimental investigations which can assess the potential of multiple injection strategies for diesel in enhancing the energy eﬃciency and utilization of butanol in dual fuel engines have been conducted in this work.
possibility of utilizing alternative fuels like biogas, biodiesel, butanol, ethanol, methanol and hydrogen eﬀectively in internal combustion engines. Biofuels will reduce greenhouse gas emissions [1,2]. Alcohols can be easily stored and distributed. They have oxygen in their molecule that will aid combustion. On account of their high octane numbers they are excellent fuels for SI engines. However, because they have low cetane numbers, they have mostly been used in the dual fuel mode in CI engines [3–6]. In the dual fuel mode a high octane number primary fuel like alcohol is inducted along with air and then compressed like in a CI engine. This compressed charge does not generally auto-ignite. Hence, it is ignited by injecting a small quantity of a high cetane number fuel (secondary fuel) like diesel. In this method the energy share of the inducted fuel can be varied. The premixed inducted charge helps in reducing smoke emissions. However, the amounts the primary and secondary fuels and their properties signiﬁcantly inﬂuence the performance of dual fuel engines [7,8]. Neat alcohols have also been used directly in CI engines with ignition assistance in the HCCI mode [9,10]. Though ethanol and methanol have been widely investigated the use of butanol is gaining importance. Butanol is a viable renewable fuel as it can be produced by the fermentation of agricultural feed stock that are normally used for producing ethanol [11–14]. It has several properties that are closer to both gasoline and diesel as compared to methanol and ethanol. Properties of ethanol, butanol and diesel are compared in Table 1 [15,16]. Butanol has a higher caloriﬁc value and density compared to the other alcohols. Its low vapour pressure leads to diﬃculties in starting in SI engines . It is less corrosive and less prone to water contamination than ethanol and so can be easily transported and distributed using existing fuel supply infrastructure. Its high octane number makes it a good SI engine fuel  while its cetane number which is much higher than other alcohols has enabled it to be used even as the sole fuel in compression ignition (CI) engines in some cases [19,20]. Since the cetane number of butanol is only 25, its use as the sole fuel is limited to low loads in CI engines. Thus butanol is generally used in the dual fuel mode or in the blended form along with diesel in CI engines. On the other hand it can also auto-ignite under certain conditions when premixed and compressed like in dual fuel engines . This leads to problems related to combustion control and high smoke emissions. In dual fuel engines smoke levels fall when the quantity of diesel is reduced i.e. when the quantity of the inducted primary fuel is increased . However, at high BMEPs (Brake mean eﬀective pressures), when butanol was inducted along with air and diesel was used for ignition, high smoke levels were observed particularly when the butanol to diesel energy share (BDES) was high. This is because the inducted butanol auto-ignited during compression and thus the diesel that was injected entered into the high temperature environment that was created. This reduced the ignition delay of the injected diesel and left less time for its mixing with air and hence elevated the smoke level . Table 1 Properties of fuels [15,16]. Property Molecular formula Lower heating value Cetane number Oxygen content Latent heating at 25 °C Stoichiometric A/F Motor octane number Auto-ignition temperature Density at 20 °C Boiling point Viscosity at 40 °C Flash point at closed cup Vapour pressure at 38 °C
MJ/kg % mass kJ/kg mass basis °C kg/m3 °C mm2/s °C kPa
C12–C25 42.5 40–55 – 270 14.3 20–30 ∼316 810–890 180–360 1.9–4.1 65–88 ∼1
C4H9OH 33.1 25 21.6 582 11.21 96 343 810 117 2.63 35 11.9
C2H5OH 26.8 8 34.8 904 9.02 108 423 789 78.5 1.08 8 15.9
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
experiments. In Table 2 detailed speciﬁcations of the engine are given. Fig. 1 shows a schematic view of the experimental setup. The engine was coupled to an eddy current dynamometer with closed loop control of speed. An in-house developed Field-Programmable Gate Array (FPGA) based open engine controller along with commercially available driver modules was used to vary the diesel fuel rail pressure, injection timing and number of diesel injection pulses. The diesel fuel injection pulse diagram shown in Fig. 2 indicates the nomenclature followed with respect to the diﬀerent injection strategies studied in this work. A specially developed port injection system was incorporated in the intake manifold after the compressor of the turbocharger for injecting butanol into the air stream. This system employed another in-house developed electronic controller for varying the quantity of the injected butanol. Both the controllers used crank and TDC signals from the angle encoder mounted on the crank shaft. Cam shaft position signals were also fed to the open controller for control purposes. Since this work focuses on reducing the problems arising because of the auto-ignition of butanol, experiments were limited to medium to high loads. Hence, all experiments were conducted at brake mean effective pressures (BMEP) of 11.8 bar (full load) and 8.8 bar (75% load) at varying butanol to diesel energy share (BDES) values. The speed of the engine was maintained at 1800 rpm while the coolant temperature was regulated at 60 ± 1 °C. The uncertainties of the measured parameters estimated based on the standard procedure are provided in Table 3 . BDES was calculated as given below.
Table 2 Speciﬁcations of the engine. Type of engine
4 Stroke water cooled turbocharged three cylinder engine with common rail direct injection system
Bore × Stroke Compression ratio (geometric) Displacement volume Peak torque Rated power EGR% Coolant temp
80 mm × 98 mm 17.1:1 1478 cc 150 N-m @1600–2400 rpm 40.5 [email protected]
rpm Nil 60 ± 1 °C
1.1. Details of the work In this work various diesel multiple injection strategies were applied for assessing their potential to reduce smoke and improve energy eﬃciency of a butanol diesel dual fuel common rail automotive engine. Initially the eﬀect of post injection was studied by varying its quantity and oﬀset from the main injection at diﬀerent BDESs (butanol diesel energy shares) and at a ﬁxed BMEP. Subsequent studies were done with two other injection strategies of diesel (Pilot plus Main and Main plus two Post injections) while butanol was injected into the intake ports. All these results have also been compared with the previously published results of the authors on the same engine in the butanol diesel dual fuel mode when diesel was injected as a single pulse instead of multiple pulses .
Butanol to diesel energy share(BDES) =
EB EB + ED
Where EB is the energy input from butanol and ED is the energy input from diesel. The measurement of air ﬂow into engine was done with a turbine ﬂow meter. The ﬂow rates of diesel and butanol were measured on the
2. Experimental setup and experiments A three-cylinder, 1.5 L turbocharged common rail automotive diesel engine run at a constant speed of 1800 rpm was used for the
Fig. 1. Schematic diagram of experimental setup.
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Fig. 2. The diesel fuel injection pulse diagram.
In the second phase of experiments the eﬀect of BDES was studied at the best main to post oﬀset and post injection quantity of diesel selected from the experiments in Phase-1. Thus these experiments were done at 100% load with a post quantity of 1.5 mg/cycle and main to post oﬀsets of 9.7 and 13°CA both of which were found to be suitable in Phase-1. Again in all the cases both the pulses were moved with respect to crank angle and positioned for highest torque using the open engine controller. For studying the eﬀect of post injection at medium load experiments were also conducted at BMEP of 8.8 bar (75% load) with the best main to post oﬀset value of 9.7°CA and post injection quantity of 1.5 mg/cycle. In the third phase, three diﬀerent injection strategies namely Pilot plus Main Injection (PMI) of diesel, Main plus Post Injection (MPI) of diesel and Main Plus Two Post Injections (MPTPI) of diesel along with port fuel injection of butanol were tried. The oﬀset of the pilot and post injections from the main were kept at the best values obtained from the experiments in Phases 1 and 2. Comparison with diﬀerent injection strategies were done at two BDES values namely 15% and 25% while the rail pressure was maintained at 700 bar. The main injection quantity was adjusted for attaining the BMEP of 11.8 bar. The injection pulses were moved with respect to crank position till the highest torque was attained. The Pilot, Post and Post-1 (P1) & Post-2 (P2) quantities were 2.5 mg/cycle, 1.5 mg/cycle and 1.5 mg/cycle & 1 mg/cycle respectively. A few results of SPI from earlier reported work by the authors  have been used for comparison here as mentioned earlier.
Table 3 Uncertainty in measurement. Measured parameter
Torque (N m) Speed (rpm) Brake thermal eﬃciency (%) Smoke (FSN) NO (ppm) HC (ppm)
± 0.5% ± 0.1% ± 1.3% ± 5.2% ± 2.1% ± 4.7%
mass basis. Temperatures of intake air, coolant and exhaust gas were measured using K type thermocouples. HC, CO and NO emissions were measured using FID, NDIR and Chemiluminescence based analysers respectively. Smoke intensity was measured using an AVL ﬁlter paper based measuring system. A ﬂush mounted piezo electric pressure transducer was used along with an optical crank angle encoder for cylinder pressure measurements. In-cylinder pressure data was acquired on the crank angle basis using a data acquisition system along with in-house developed software. Cylinder pressure data from 100 consecutive engine cycles was ensemble averaged. This was then post processed using in-house developed software for referencing and calculation of combustion parameters like ignition delay, peak pressure and heat release rate. Heat release rate was computed based on the method indicated in literature . The Hohenberg’s heat transfer correlation was used for estimation of heat transfer . All instruments were periodically calibrated as per the instructions of the manufacturers. In the ﬁrst phase of experiments, diesel was directly injected as two pulses per cycle (Main plus Post Injection - MPI) and butanol was injected into the intake port at full load (BMEP of 11.8 bar at 1800 rpm) under a butanol to diesel energy share (BDES) of 30%. This is around the maximum BDES (28.5%) that could be used in the conventional dual fuel mode (single pulse injection of diesel) as restricted by high combustion rates which will be described later. The eﬀects of quantity of the post injected diesel (1, 1.5 and 2.5 mg/cycle – amounting to 4%, 6% and 10% of the total amount of diesel respectively) and its oﬀset in crank angle degrees from the main injection (6.5–32.5°CA) were also studied. The rail pressure of diesel was kept constant at 700 bar. This was found to be the most suitable based on the thermal eﬃciency and NO emission considerations in the dual fuel mode when diesel was injected as a single pulse (Single Pulse Injection – SPI). In these experiments while the oﬀset between the main and post injections was ﬁxed both the pulses were simultaneously moved with respect to crank position till the highest torque was achieved. It is also to be noted that some of the already published  experimental results (only brake thermal eﬃciency, NO, Smoke and HC emission at diﬀerent BDES values at 75% and 100% loads) of the same authors on the same engine in the SPI mode alone have been used for comparison with the other modes evaluated here.
3. Results and discussion The results of experiments conducted in all the three phases are presented and discussed in this section. Comparisons have also been made with the results of Single Pulse Injection (SPI) of diesel taken from the earlier work of the authors which was also conducted on the same engine . 3.1. Phase-1: Eﬀect of quantity of post injection and oﬀset between main and post injection The brake thermal eﬃciency (BTE) shown in Fig. 3 was not signiﬁcantly aﬀected by the main to post injection oﬀset or the quantity of fuel in the post injection phase. There was a small increase in the BTE with Main plus Post Injection (MPI) as compared to SPI of diesel because of better combustion phasing. The BTE for the pure diesel case is also indicated on the same ﬁgure (shown as a black point on the primary Y axis at zero oﬀset – Fig. 3). Fig. 4 indicates the Heat Release Rates (HRR) with diﬀerent main to post oﬀset values and also with SPI. The main and the post pulses (injector driving signals) for each oﬀset are also indicated on the same ﬁgure. We see that in all the cases combustion started before the main diesel pulse i.e. before the diesel was injected. This indicates auto-ignition of the butanol that was injected into the ports before the diesel that was directly injected into the 4
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Table 4 In-cylinder conditions at EVO. Main to post oﬀset
Single pulse injection of diesel
In-cylinder Pressure at EVO (bar) Estimated temperature at EVO (K)
output of the turbine of the turbocharger and as a result the boost pressure and intake air temperature will go up. In addition the retained exhaust gas also will be at a higher temperature in such cases. These will favour auto-ignition of butanol and will raise the temperature after combustion which has a cascading eﬀect. Table 4 shows the measured in-cylinder pressure and estimated averaged in-cylinder temperature at the time of exhaust valve opening (EVO) (pressure and temperature at the inlet of the turbine). In MPI strategy of diesel as the oﬀset between the main and post was increased auto-ignition of butanol was also advanced (Fig. 4 – shown by the black ellipse). This was because pressure and temperature at the inlet of the turbine also increased as explained earlier (Table. 4). It is also noted that with the MPI strategy of diesel pressure and temperature at the time of EVO were always lower as compared to SPI of diesel. Because of this auto-ignition of butanol with the MPI strategy of diesel was delayed. It is also seen in Figs. 5 and 6 that advancing the injection timing of diesel in both MPI and SPI modes reduced the in-cylinder gas temperatures in the expansion stroke and thus retarded the auto-ignition of butanol. It however advanced the second heat release portion which was contributed by the combustion of diesel and the remaining butanol. This reduced the exhaust gas temperature and in-turn the intake temperature also was reduced as explained earlier. This retarded the ﬁrst peak in the HRR which is due to auto-ignition of butanol. Fig. 7 indicates that with MPI of diesel NO emission decreased due to reduced and delayed peak HRR which lowered the maximum temperature in the cycle (Fig. 4). As the main to post oﬀset was increased, the NO emission level decreased till an oﬀset of 13°CA and after that there was no signiﬁcant change. It is also reported in literature that increasing the oﬀset of the post injection from the main injection in case of neat diesel engines could reduce the cooling eﬀect the post injection produces after the main combustion and thus may lead to increase in
Fig. 3. Variation in BTE and IMT with post injection quantity and main to post oﬀset.
Fig. 4. HRR and estimated in-cylinder temperature with diﬀerent main to post oﬀsets (Post quantity = 2.5 mg/cycle).
combustion chamber. In the case of MPI we see that this auto-ignition of butanol was delayed (which could lead to improved BTE) due to the reduced boost temperature that resulted in lower exhaust gas temperatures (EGTs). The estimated in-cylinder gas temperatures shown in Fig. 4 also indicate that lower values were attained in the case of MPI as compared to SPI (reduction of 70 K) particularly during the expansion stroke. The boost pressure was also lower with MPI at compared to the SPI strategy. Hence, due to the coupled eﬀects of lower EGT, boost pressure and temperature, auto-ignition of butanol was delayed. With SPI the auto-ignition was quite early and this aﬀected the BTE adversely. The advanced combustion with SPI  was found to be due to the higher boost temperature (intake manifold temperature- IMT) which is also indicated in Fig. 3 (shown as a point on the Y axis at zero oﬀset). The advanced combustion of butanol by auto-ignition in the SPI mode also resulted in higher in-cylinder gas temperatures. Thus there was an inﬂuence of exhaust gas temperature on auto-ignition of butanol and vice versa. Higher exhaust gas temperatures will increase the
Fig. 5. HRR and estimated in-cylinder temperature at diﬀerent SOIs of diesel in the MPI mode.
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Fig. 6. HRR and estimated in-cylinder temperature at diﬀerent SOIs of diesel in the SPI mode.
Fig. 8. Variation in HC and CO emission with post injection quantity and main to post oﬀsets.
low smoke. Small main to post oﬀsets (post injected immediately after the main) did not reduce smoke as main and post combustion phases almost merged with each other. This lowered the eﬀect of enhanced mixing by post injection. This was because under these conditions the inﬂuence of the small momentum brought in by the post becomes insigniﬁcant as compared to what is introduced by the main injection. We also see that when the oﬀset was higher than about 13°CA the smoke level went up. This was because of the lower temperatures that were created by the late injection of the post which was inﬂuenced by the expansion of the gases in the cylinder. It has also been reported in literature that there is a possibility of the post injected fuel interfering with that part of the fuel injected during the main injection which rebounds from the combustion chamber . Thus the beneﬁts of the post injection can be realised only if the main to post oﬀset is such that the post does not enhance the fuel available in the soot formation region created by the main injection but helps elevate the temperature for aiding oxidation of the soot particles. The smoke emission for the pure diesel case is also indicated on the same ﬁgure (shown as a black point on the secondary Y axis at 40 oﬀset – Fig. 7). The HC and CO emissions (shown in Fig. 8) indicate that increased post injection quantities lead to elevated HC and CO levels. HC and CO emissions for the pure diesel case are also shown on same ﬁgure as a black point on Y axis. Fig. 9 indicates that the CO2 level with post injection was always below the level seen in the SPI mode  because of improved thermal eﬃciency. The CO2 emission for the pure diesel case is also indicated on the same ﬁgure (shown as a black point on the primary Y axis at zero oﬀset – Fig.9). The coeﬃcient of variation of indicated mean eﬀective pressure (COV of IMEP) calculated over 100 successive cycles shown in Fig. 9 indicates higher values than the SPI mode only when the post was too close to the main injection. Under this condition, the inﬂuence of the post injected fuel on torque generation was signiﬁcant because it was close to TDC. Thus any variation in the post quantity between cycles which is likely when the injected quantity is near the lower limit of the injector is inﬂuential when the main to post oﬀset is low. However, the COV of IMEP was always below 3.5% which indicated stable combustion. Under the best oﬀset the COV of IMEP was comparable to the SPI mode of operation. The COV of IMEP with the stock engine ECU was also around this value (Fig. 9).
Fig. 7. Variation in NO and SMOKE emission with post injection quantity and main to post oﬀsets.
the NO levels . In Fig. 7 in some cases it is seen that increasing the oﬀset increased the NO level slightly. The quantity of the post injected fuel had only a small inﬂuence on NO emission. However, with SPI  (shown as a point on the Y axis at zero oﬀset) the NO level was high since the auto-ignition of butanol was advanced which elevated the incylinder temperature (Fig. 4). NO emission for the pure diesel case is also indicated on the same ﬁgure (shown as a black point on the primary Y axis at zero oﬀset – Fig. 7). We see in Fig. 7 that there was a drastic drop in smoke emission with post injection as the oﬀset increased till 13°CA as compared to the SPI mode. This was because post injection which occurred after the diesel injected in the main pulse started burning, enhanced mixing within the cylinder due to its momentum which aided oxidation of soot . As the post injected fuel burns it will elevate the temperature in the cylinder. This also has aided the oxidation of soot . The quantity of fuel associated with post injection has to be optimised for 6
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Fig. 11. Variation in NO emission with butanol to diesel energy share at diﬀerent main to post oﬀsets at 75% and 100% loads.
Fig. 9. Variation in CO2 emission and COV of IMEP with post injection quantity and main to post oﬀsets.
which is one of the reasons for the increase in BTE. Fig. 11 indicates that the NO emission level fell signiﬁcantly with the introduction of butanol because of the charge cooling eﬀect arising out of the vaporization of butanol whose latent heat of vaporization is signiﬁcantly higher than conventional fuels. The NO level with post injection was lower than SPI  of diesel for both 75% and 100% loads which was due to lower peak temperatures as compared to SPI of diesel. In addition, combustion was also completed in a shorter duration with MPI . Fig. 12 indicates that with post injection of diesel a signiﬁcant reduction in smoke emission due to its eﬀects on in cylinder mixing and temperature as discussed earlier was obtained. For the main to post oﬀset of 9.7°CA smoke was lower as compared to the oﬀset of 13°CA. This is because main to post oﬀset inﬂuences the temperature and mixing as discussed earlier. Similar results were seen for diﬀerent post injection durations as discussed in Phase-1 (Fig. 5). With SPI  of
3.2. Phase-2: Eﬀect of post injection of diesel at 100% and 75% loads under diﬀerent BDES Experiments were done in this phase with a post injection quantity of 1.5 mg/cycle and main to post oﬀsets of 9.7°CA and 13°CA which were found to be the most suitable based on experiments in Phase-1 at 100% load while considering BTE, NO and smoke emissions. Eﬀect of post injection at 75% load was also studied in this phase at diﬀerent BDES values while the post injection parameters were kept at the same values (i.e. Post injection quantity 1.5 mg/cycle and a main to post oﬀset of 9.7°CA). Fig. 10 compares the BTEs obtained with MPI and SPI of diesel at 75% and 100% loads. The BTE increased with post injection as compared to SPI  at all values of BDES. Under the set of injection parameters used, the heat release occurred with better combustion phasing and at a faster rate. With MPI of diesel the crank angle at 50% HRR (CA 50) was advanced as compared to SPI of diesel (by 3°CA)
Fig. 10. Variation in BTE with butanol to diesel energy share at diﬀerent main to post oﬀsets at 75% and 100% loads.
Fig. 12. Variation in smoke emission with butanol to diesel energy share at diﬀerent main to post oﬀsets at 75% and 100% loads.
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Fig. 14. Variation in CO emission with butanol to diesel energy share at diﬀerent main to post oﬀsets at 75% and 100% loads.
Fig. 13. Variation in HC emission with butanol to diesel energy share at diﬀerent main to post oﬀsets at 75% and 100% loads.
diesel, smoke levels decreased initially with increase in BDES due to the reduction in the injected diesel which contributes towards it. However, beyond a BDES of 20% the smoke level increased with SPI of diesel as the butanol air mixture became richer and started to auto ignite which created a high temperature environment when the diesel was injected. This is discussed further in the later paragraphs using HRR diagrams. Smoke emission was found to decrease with post injection at 75% load also (Fig. 12). With increase in BDES, HC levels (Fig. 13) increased due to quenching and crevice eﬀects mainly related to the injected butanol. With post injection, HC levels were higher because of lower average gas temperatures in the expansion stroke which aﬀected post oxidation. Similar trends were found at 75% load (Fig. 13). For higher main to post oﬀsets, in-cylinder gas temperatures in the expansion stroke were higher as compared to smaller main to post oﬀsets (Fig. 18). However HC levels with 75% load were higher as compared to 100% load. This is because of the lower temperatures attained at 75% load. CO emission shown in Fig. 14 was found to be decrease with increase in post injection quantity. Similarly CO2 emission (Fig. 15) was also found to decrease because of improved BTE. The COV of IMEP was always below 1.5% at all BDES values. Fig. 16 indicates that the maximum rate of pressure rise (MRPR) was low when low values of BDES were used due to reduced rate of combustion and increased ignition delay of diesel. However, with increasing BDES values when butanol started to auto-ignite the MRPR rapidly increased. The values of MRPR were lower with MPI as it created conditions that were less conducive for auto-ignition of butanol as discussed earlier. With SPI of diesel MRPR was always higher as compared to MPI (because of advanced and rapid heat release due to autoignition with SPI, Fig. 4). This is one of the reasons for higher NO and smoke emissions in the SPI mode. We see that with increase in the BDES the butanol air mixture tended to auto-ignite with signiﬁcant high heat release rate values which led to knock. Thus the maximum BDES was limited to 32% at 100% load and 46% at 75% load. With MPI of diesel there was an increase of 4% in the BDES that could be used as compared to SPI  of diesel at both 75% and 100% loads. This is because lower MRPR values were observed with MPI of diesel. However, BDES was limited by knocking (MRPR) in both MPI and SPI of diesel. Fig. 17 shows the HRR with MPI (BDES = 28.3%) and SPI (BDES = 28.5%) of diesel cases at 100% load. There were two peaks in the HRR curves. With post injection the ﬁrst peak of HRR which is due
Fig. 15. Variation in CO2 emission with butanol to diesel energy share at diﬀerent main to post oﬀsets at 75% and 100% loads.
to auto-ignition was lower and a little retarded. This is because boost pressure and temperature were lower (reason for auto-ignition retarded) and hence lesser amount of butanol was auto-igniting as compared to SPI of diesel. The second peak which was dominated by the burning of diesel was advanced and higher as compared to SPI of diesel. This is because more butanol was burning in this phase which favourably inﬂuenced BTE. Post injection also accelerated the ﬁnal stage of combustion (Fig. 4) . Further, with post injection, between the ﬁrst and the second peaks the value of HRR did not drop oﬀ signiﬁcantly as compared to SPI. This is also one of the reasons for the higher BTE with post injection. With SPI the ﬁrst peak of HRR was advanced due to the reasons explained earlier. Fig. 18 shows that the estimated average in cylinder temperature reached higher values and stayed so for longer durations with SPI which led to increase NO levels. When the BDES was reduced as seen in Fig. 19 similar trends were observed but the ﬁrst peak was not so dominant. 8
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Fig. 16. Variation in MRPR emission with butanol to diesel energy share at diﬀerent main to post oﬀsets at 100% load.
Fig. 18. Estimated In-cylinder temperature traces for main to post oﬀset and SPI of diesel at 100% load.
Fig. 17. HRR and In-cylinder pressure traces at diﬀerent main to post oﬀsets at a BDES of 28.3% and 100% load.
Fig. 19. HRR and In-cylinder pressure traces at diﬀerent main to post oﬀsets at a BDES of 22% and 100% load.
Fig. 20 shows the HRR with MPI (BDES = 37.4%) and SPI (BDES = 36.5%) of diesel cases at 75% load. It may be noted that the actual start of injection of diesel will be slightly delayed as compared to the injection pulse (electrical signal) shown in Fig. 20. In both SPI and MPI cases auto-ignition of butanol occurs. Auto-ignition in the case of MPI is delayed as compared to SPI as a result of lower charge temperatures due to the reasons explained earlier. This trend was also observed at 100% load (Fig. 17). The initial peak in the HRR with MPI is higher than with SPI. This is because the start of combustion (autoignition) is more retarded as compared to the start of injection of diesel in the case of MPI. Hence, some of the injected diesel could also autoignite along with butanol leading to higher heat release rates.
diesel and the results are shown in Figs. 21 and 22. In this case there was an increase in smoke levels and reduction in BTE. This was because the pilot diesel resulted in very early auto-ignition of butanol and advanced the combustion process as seen in Figs. 23 and 24 particularly at high BDES values. Similar trends were also observed in the results presented in Phase-1. Smoke values were also higher as compared to SPI of diesel. Experiments with even pilot, main and post combination along with manifold injection of butanol were conducted. It was found that this mode was not beneﬁcial in reducing smoke. Thus it is not discussed further. Main plus post injection (MPI) of diesel was the most suitable. Hence, experiments were also done with Main plus two post injections (MPTPI) of diesel along with port fuel injection of butanol at 100% load. The oﬀset between main and post-1 (P1) was maintained at a value of 9.7°CA which was found to be the best from earlier results (Phase-2). The quantity of P1 was also kept at 1.5 mg/cycle which was obtained from the experimental results from Phase-1. The oﬀset of Post-
3.3. Phase-3: Comparison of diﬀerent injection strategies Experiments were also done with Pilot plus Main Injection (PMI) of 9
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
Fig. 20. HRR and In-cylinder pressure traces with diﬀerent injection strategies at a BDES of 37% and 75% load.
Fig. 22. BTE, MRPR, NO and smoke emission for diﬀerent injection strategies of diesel at a BDES of 25% and 100% load.
Fig. 21. BTE, MRPR, NO and smoke emission for diﬀerent injection strategies of diesel at a BDES of 15% and 100% load.
Fig. 23. HRR and In-cylinder pressure traces for diﬀerent injection strategies of diesel at a BDES of 15% and 100% load.
2 (P2) from P1 and the quantity of P2 were varied and the best values (6.5°CA and 1 mg/cycle) were obtained. The quantity of fuel injected in the main diesel pulse and its timing were varied till the torque was maximum (MBT-Timing) while the main to P1 and P1 to P2 oﬀsets and quantities of P1 and P2 were kept constant at the values indicated earlier. A comparison between all the methods is seen in Figs. 21 and 22 at two diﬀerent BDES values namely 15% and 25%. From Figs. 23 and 24 we see that SPI resulted in a clear ﬁrst peak in the heat release rate (HRR) even before injection of diesel and the second peak was inﬂuenced by diesel. In the case of PMI there were three peaks in the HRR (Fig. 23). The second peak (auto-ignition of butanol) was inﬂuenced by the pilot and BDES (richness of the inducted mixture). At higher BDES values the second and the ﬁrst peaks merged together in the case of PMI (Fig. 24). When the auto-ignition of butanol was signiﬁcant the main pulse of diesel had to be retarded in order to get the highest BTE with PMI. This eﬀect was also seen in the HRR
curves. Between MPI and MPTPI there was little diﬀerence in heat release rates. From Figs. 21 and 22 we see that at full load the highest BTE and lowest smoke levels were reached with MPI. As compared to SPI the BTE was higher by 1.6% and the smoke was lower by 0.73 FSN. The maximum rate of pressure rise and NO level were also lower. With MPTPI the smoke levels were higher than MPI this is because the second post (P2) was injected in expansion stroke and that was contributing to smoke while the other parameters were similar. The only advantage with PMI was low NO levels. MPI is better than Main Plus Two Post Injections with respect to all the parameters evaluated. On the whole the Main plus Post Injection (MPI) strategy seems to be the best for dual fuel operation with butanol and diesel at medium to high loads as it can result in signiﬁcant reductions in smoke, improved energy eﬃciency and no adverse eﬀect on NO. However, HC levels were elevated as compared to SPI.
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
In summary, Main plus Post (MPI) is the best amongst the diﬀerent strategies evaluated in this work. It resulted in the highest thermal efﬁciency and lowest smoke and maximum rate of pressure rise levels. Post injection quantity of 1.5 mg/cycle and oﬀset of 9.7°CA with the main injection were found to be generally suitable. A change in the injection strategy is needed when the engine shifts between neat diesel and butanol diesel dual fuel modes of operation. References  Nagy Karoly, Körmendi Krisztina. Use of renewable energy sources in light of the “New Energy Strategy for Europe 2011–2020”. Appl Energy 2012;96:393–9.  Koponen Kati, Tsupari Eemeli, Soimakallio Sampo, Thun Rabbe, Antikainen Riina. GHG emission performance of various liquid transportation biofuels in Finland in accordance with the EU sustainability criteria. Appl Energy 2013;102:440–8.  Karim GA, Amoozegar N. Determination of the performance of a dual fuel diesel engine with the addition of various liquid fuels to the intake charge (No. 830265). SAE Technical Paper 1983.  Abu-Qudais M, Haddad O, Qudaisat M. The eﬀect of alcohol fumigation on diesel engine performance and emissions. Energy Convers Manage 2000;41(4):389–99.  Fang W, Huang B, Kittelson DB, Northrop WF. Dual-fuel diesel engine combustion with hydrogen, gasoline, and ethanol as fumigants: eﬀect of diesel injection timing. J Eng Gas Turbines Power 2014;136(8):081502.  Kumar BR, Saravanan S. Use of higher alcohol biofuels in diesel engines: a review. Renew Sustain Energy Rev 2016;60:84–115.  Zhu Y, Chen Z, Liu J. Emission, eﬃciency, and inﬂuence in a diesel n-butanol dualinjection engine. Energy Convers Manage 2014;87:385–91.  Michikawauchi R, Tanno S, Ito Y, Kanda M. Combustion improvement of diesel engine by alcohol addition-investigation of port injection method and blended fuel method. SAE Int J Fuels Lubr 2011-01-0336;4(2011):48–57.  Maurya RK, Agarwal AK. Combustion and emission characterization of n-butanol fueled HCCI engine. J Energy Res Technol 2015;137(1):011101.  Maurya RK, Agarwal AK. Experimental study of combustion and emission characteristics of ethanol fuelled port injected homogeneous charge compression ignition (HCCI) combustion engine. Appl Energy 2011;88(4):1169–80.  Green EM. Fermentative production of butanol—the industrial perspective. Curr Opin Biotechnol 2011;22(3):337–43.  Weber C, Farwick A, Benisch F, Brat D, Dietz H, Subtil T, Boles E. Trends and challenges in the microbial production of lignocellulosicbioalcohol fuels. Appl Microbiol Biotechnol 2010;87(4):1303–15.  Ezeji TC, Qureshi N, Blaschek HP. Bioproduction of butanol from biomass: from genes to bioreactors. Curr Opin Biotechnol 2007;18(3):220–7.  Kumar Manish, Gayen Kalyan. Developments in biobutanol production: new insights. Appl Energy 2011;88(6):1999–2012.  Jin C, Yao M, Liu H, Chia-fon FL, Ji J. Progress in the production and application of n-butanol as a biofuel. Renew Sustain Energy Rev 2011;15(8):4080–106.  Bechtold RL. Alternative fuels guidebook: properties, storage, dispensing, and vehicle facility modiﬁcations. Warrendale, USA: Society of Automotive Engineers; 1997. p. 51–5. Chap. 2.  Szwaja S, Naber JD. Combustion of n-butanol in a spark-ignition IC engine. Fuel 2010;89(7):1573–82.  Cooney C, Wallner T, McConnell S, Gillen JC, Abell C, Miers SA, Naber JD. Eﬀects of blending gasoline with ethanol and butanol on engine eﬃciency and emissions using a direct-injection, spark-ignition engine. In: ASME 2009 internal combustion engine division spring technical conference (pp. 157–165). American Society of Mechanical Engineers. 2009 January.  Yanai T, Han X, Reader GT, Zheng M, Tjong J. Preliminary investigation of direct injection neat n-butanol in a diesel engine. J Energy Res Technol 2015;137(1):012205.  Jeftić M, Yang Z, Reader GT, Zheng M. Fuel eﬃciency analysis and peak pressure rise rate improvement for neat n-butanol injection strategies. In: Proceedings of the institution of mechanical engineers, Part D: J Automob Eng. 2016. p. 0954407016632141.  Han X, Zheng M, Wang J. Fuel suitability for low temperature combustion in compression ignition engines. Fuel 2013;109:336–49.  Papagiannakis RG, Hountalas DT. Combustion and exhaust emission characteristics of a dual fuel compression ignition engine operated with pilot diesel fuel and natural gas. Energy Convers Manage 2004;45(18):2971–87.  Yadav, Jaykumar, Asvathanarayanan Ramesh. Comparison of single and multiple injection strategies in a butanol diesel dual fuel engine. In: ASME 2017 Power Conference Joint With ICOPE-17 collocated with the ASME 2017 11th International Conference on Energy Sustainability, the ASME 2017 15th International Conference on Fuel Cell Science, Engineering and Technology, and the ASME 2017 Nuclear Forum, pp. V001T04A018–V001T04A018. American Society of Mechanical Engineers, 2017.  Gao T, Reader G, Tjong J, Zheng M. Energy eﬃciency comparison between butanol and ethanol combustion with diesel ignition (No. 2015-01-0859). SAE Technical Paper; 2015.  Gao T, Yu S, Zhu H, Li T, Tjong J, Reader G, Zheng M. A Feasibility study of using di butanol as an ignition source for dual-fuel combustion (No. 2017-01-0770). SAE Technical Paper; 2017.  Mahr B. Future and potential of diesel injection systems. In: thermo-and ﬂuid dynamic processes in diesel engines 2 2004 (pp. 3–17). Berlin, Heidelberg; Springer.
Fig. 24. HRR and In-cylinder pressure traces for diﬀerent injection strategies of diesel at a BDES of 25% and 100% load.
4. Conclusions Based on the experiments conducted at 75% and 100% loads in the butanol diesel dual fuel mode with diﬀerent injection strategies for diesel the following conclusions are derived.
• Main plus Post Injection (MPI) of diesel resulted in a small increase •
• • •
in brake thermal eﬃciency (BTE) as compared to the conventional Single Pulse Injection (SPI) of diesel both at 75% and 100% loads. At high BDES values, butanol always auto-ignited before the injection of diesel and two peaks were observed in the heat release rate (HRR). Retarding the start of injection of diesel advanced the autoignition of butanol (that is ﬁrst peak in the HRR) and delayed the second peak of HRR which is the combined burning of diesel and remaining butanol. This is because of the increase in the exhaust gas temperature and resulting increase in boost pressure and temperature. The maximum rate of pressure rise also went up signiﬁcantly with increase in BDES. By altering the injection strategy adopted for diesel auto-ignition of butanol that occurs at medium to high loads (at high BDES) cannot be prevented. However, smoke emission can be reduced and BTE can be improved. With MPI of diesel there was a drastic drop in smoke emission with post injection as the oﬀset was increased till 13°CA because of enhanced mixing within the cylinder which helps oxidation of soot. Too retarded post injection was not beneﬁcial as it resulted in lower gas temperatures. Higher BDES values along with optimised post injection quantity and main to post oﬀset were better for reduction of smoke. NO emission decreased with post injection and was inﬂuenced signiﬁcantly by the main to post oﬀset. NO emission decreased with increase in BDES. Values lower than diesel could be reached with MPI of diesel. The CO2 level with post injection was also below the levels with SPI because of improved thermal eﬃciency. HC emission increased in the dual fuel mode due to quenching and crevice eﬀects. MPI resulted in higher HC and CO than SPI because of lower temperatures during expansion. Pilot plus Main Injection (PMI) of diesel resulted in increased smoke levels (higher than SPI) and reduction in BTE because of very early auto-ignition of butanol which advanced the combustion process.
Applied Energy 212 (2018) 1–12
J. Yadav, A. Ramesh
 Liu Y, Reitz RD. Optimizing HSDI diesel combustion and emissions using multiple injection strategies (No. 2005-01-0212). SAE Technical Paper; 2005.  Park C, Kook S, Bae C. Eﬀects of multiple injections in a HSDI diesel engine equipped with common rail injection system (No. 2004-01-0127). SAE Technical Paper; 2004.  Hotta Y, Inayoshi M, Nakakita K, Fujiwara K, Sakata I. Achieving lower exhaust emissions and better performance in an HSDI diesel engine with multiple injection (No. 2005-01-0928). SAE Technical Paper; 2005.  Vanegas A, Won H, Felsch C, Gauding M, Peters N. Experimental investigation of the eﬀect of multiple injections on pollutant formation in a common-rail DI diesel engine (No. 2008-01-1191). SAE Technical Paper; 2008.  Mingfa Y, Hu W, Zunqing Z, Yan Y. Experimental study of multiple injections and coupling eﬀects of multi-injection and EGR in a HD diesel engine (No. 2009-012807). SAE Technical Paper; 2009.  Chen SK, Simultaneous reduction of NOx and particulate emissions by using multiple injections in a small diesel engine (No. 2000-01-3084). SAE Technical Paper; 2000.  Benajes J, Molina S, García JM. Inﬂuence of pre-and post-injection on the performance and pollutant emissions in a HD diesel engine (No. 2001-01-0526). SAE Technical Paper; 2001.  Desantes JM, Arrègle J, López JJ, García A, A comprehensive study of diesel combustion and emissions with post-injection (No. 2007-01-0915). SAE Technical Paper; 2007.  Martin J, Sun C, Boehman A, O'Connor J. Experimental study of post injection scheduling for soot reduction in a light-duty turbodiesel engine (No. 2016-010726). SAE Technical Paper; 2016.  Jeftić M, Zheng M. A study of the eﬀect of post injection on combustion and emissions with premixing enhanced fueling strategies. Appl Energy 2015;157:861–70.  López AF, Cadrazco M, Agudelo AF, Corredor LA, Vélez JA, Agudelo JR. Impact of n-butanol and hydrous ethanol fumigation on the performance and pollutant emissions of an automotive diesel engine. Fuel 2015;153:483–91.  Liu H, Li S, Zheng Z, Xu J, Yao M. Eﬀects of n-butanol, 2-butanol, and methyl
    
octynoate addition to diesel fuel on combustion and emissions over a wide range of exhaust gas recirculation (EGR) rates. Appl Energy 2013;112:246–56. Wu HW, Wang RH, Ou DJ, Chen YC, Chen TY. Reduction of smoke and nitrogen oxides of a partial HCCI engine using premixed gasoline and ethanol with air. Appl Energy 2011;88(11):3882–90. Fayad MA, Fernández-Rodríguez D, Herrerosa JM, Martos FJ, Lapuerta M. Manipulating modern diesel engine particulate emission characteristics through butanol fuel blending and fuel injection strategies for eﬃcient diesel oxidation catalysts. Appl Energy 2017;190:490–500. Rakopoulos DC, Rakopoulos CD, Giakoumis EG, Dimaratos AM, Kyritsis DC. Eﬀects of butanol–diesel fuel blends on the performance and emissions of a high-speed DI diesel engine. Energy Convers Manage 2010;51(10):1989–97. Doğan Oğuzhan. The inﬂuence of n-butanol/diesel fuel blends utilization on a small diesel engine performance and emissions. Fuel 2011;90(7):2467–72. Holman JP, Anderson Ann M. Experimental Methods for Engineers. Exp Therm Fluid Sci 1994;9(2). 250-250. Hayes TK, Savage LD, Sorenson SC. Cylinder pressure data acquisition and heat release analysis on a personal computer (No. 860029). SAE Technical Paper; 1986. Hohenberg GF. Advanced approaches for heat transfer calculations (No. 790825). SAE Technical Paper; 1979. Poorghasemi K, Ommi F, Yaghmaei H, Namaki A. An investigation on eﬀect of high pressure post injection on soot and NO emissions in a DI diesel engine. J Mech Sci Technol 2012;26(1):269–81. Bobba M, Musculus M, Neel W. Eﬀect of post injections on in-cylinder and exhaust soot for low-temperature combustion in a heavy-duty diesel engine. SAE Int J Engines 2010-01-0612;3(2010):496–516. O'Connor J, Musculus M. Post injections for soot reduction in diesel engines: a review of current understanding. SAE Int J Engines, 6(2013-01-0917); 2013: pp. 400–21. Horibe N, Komizo T, Sumimoto T, Wang H, Ishiyama T. Smoke Reduction Eﬀects by Post Injection for Various Injection Parameters and Combustion Chamber Shapes in a Diesel Engine (No. 2014-01-2634). SAE Technical Paper; 2014.