Elsevier Editorial System(tm) for Energy Conversion and Management Manuscript Draft

Transcription

1 Elsevier Editorial System(tm) for Energy Conversion and Management Manuscript Draft Manuscript Number: ECM-D R1 Title: Experimental Investigations of the Effect of Pilot Injection on Performance, Emissions and Combustion Characteristics of Karanja Biodiesel Fuelled CRDI Engine Article Type: Original research paper Section/Category: 6. Fuels, Combustion, and Chemical Processes Keywords: Multiple injections; biodiesel; fuel injection pressure; emissions; combustion characteristics. Corresponding Author: Prof. Avinash Kumar Agarwal, PhD Corresponding Author's Institution: IIT Kanpur First Author: Atul Dhar Order of Authors: Atul Dhar; Avinash Kumar Agarwal, PhD Abstract: Pilot and post injections are being used in modern diesel engines for improving engine performance in addition to meeting stringent emission norms. Biodiesel produced from different feedstocks is gaining global recognition as partial replacement for mineral diesel in compression ignition (CI) engines. In this study, 10%, 20% and 50% Karanja biodiesel blends were used for investigation of pilot injections, injection pressures and injection timings on biodiesel blends. Experiments were carried out in a single cylinder CRDI research engine in multiple injection mode at 500 and 1000 bar FIP under varying start of pilot injection (SOPI) and start of main injection (SOMI) timings. Brake specific fuel consumption (BSFC) increased with increasing Karanja biodiesel concentration in test fuels however brake thermal efficiency (BTE) of biodiesel blends was slightly higher than mineral diesel. Lower biodiesel blends showed lower brake specific carbon monoxide (BSCO) and brake specific hydrocarbon (BSHC) emissions than mineral diesel. Brake specific nitrogen oxides (BSNOx) emissions from KOME20 and KOME10 were higher than mineral diesel. Combustion duration of was also higher than mineral diesel.

2 Cover letter IINDIIAN IINSTIITUTE OF TECHNOLOGY KANPUR DEPARTMENT OF MECHANICAL ENGINEERING KANPUR , INDIA Dr. Avinash Kumar Agarwal, Tel: (O), (R) Poonam and Prabhu Goyal Endowed Chair Professor Fax: akag@iitk.ac.in Editor, Energy Conversion and Management November 8th, 2014 Dear Sir, I am submitting a revised manuscript manuscript entitled "Experimental Investigations of the Effect of Pilot Injection on Performance, Emissions and Combustion Characteristics of Karanja Biodiesel Fuelled CRDI Engine" by Atul Dhar, and Avinash Kumar Agarwal for possible inclusion in "Energy Conversion and Management". We have incorporated all changes suggested by the reviewers and a rebuttal to the reviewer's comments is also attached for your ready reference. Submission of this article implies that the work described has not been published previously (except in the form of an abstract or as part of a published lecture or academic thesis), that it is not under consideration for publication elsewhere, that its publication is approved by all authors and that, if accepted, it will not be published elsewhere in the same form, in English or in any other language, without the written consent of the Publisher. Looking forward for a positive decision soon. Best regards Dr A K Agarwal

3

4 *Checklist for New Submissions Energy Conversion and Management Submission Checklist Please save a copy of this form to your computer, complete and upload as Checklist for New Submissions. Your manuscript will be considered incomplete unless all the below requirements have been met. Please initial each step to confirm. (1) I Avinash kumar Agarwal confirm that the work described has not been published previously (except in the form of an abstract or as part of a published lecture or academic thesis), that it is not under consideration for publication elsewhere, that its publication is approved by all authors and that, if accepted, it will not be published elsewhere in the same form, in English or in any other language, without the written consent of the Publisher. Authors found to be deliberately contravening the submission guidelines on originality and exclusivity shall not be considered for future publication in this journal. (2) The source document is editable, i.e. Word, WordPerfect or Latex AKA {enter initials} (3) The source document is double-spaced AKA {enter initials} (4) The source document has been prepared prepared in 12 or 10 point font size, preferably 12 points AKA {enter initials} (5) The source document is in one column per page AKA {enter initials} (6) The figures and tables have been supplied: either integrated with the text file or as separate files. AKA {enter initials}

5 *Highlights (for review) Experimental Investigations of the Effect of Pilot Injection on Performance, Emissions and Combustion Characteristics of Karanja Biodiesel Fuelled CRDI Engine Atul Dhar, Avinash Kumar Agarwal* Engine Research Laboratory, Department of Mechanical Engineering Indian Institute of Technology Kanpur, Kanpur , India *Corresponding Author s akag@iitk.ac.in Research Highlights Effect of multiple injections on CRDI engine performance, emission & Combustion. Effect of Multiple injections, injection pressures &injection timings on Biodiesel. Lower biodiesel blends showed lower BSCO, BSHC but higher BSNOx emissions. Maximum cylinder pressure at higher FIP was higher at same SOPI and SOMI. Combustion duration of was higher than mineral diesel

6 *Revised Manuscript with no changes marked Click here to view linked References Experimental Investigations of the Effect of Pilot Injection on Performance, Emissions and Combustion Characteristics of Karanja Biodiesel Fuelled CRDI Engine Atul Dhar, Avinash Kumar Agarwal* Engine Research Laboratory, Department of Mechanical Engineering Indian Institute of Technology Kanpur, Kanpur , India *Corresponding Author s akag@iitk.ac.in Abstract Pilot and post injections are being used in modern diesel engines for improving engine performance in addition to meeting stringent emission norms. Biodiesel produced from different feedstocks is gaining global recognition as partial replacement for mineral diesel in compression ignition (CI) engines. In this study, 10%, 20% and 50% Karanja biodiesel blends were used for investigation of pilot injections, injection pressures and injection timings on biodiesel blends. Experiments were carried out in a single cylinder CRDI research engine in multiple injection mode at 500 and 1000 bar FIP under varying start of pilot injection (SOPI) and start of main injection (SOMI) timings. Brake specific fuel consumption (BSFC) increased with increasing Karanja biodiesel concentration in test fuels however brake thermal efficiency (BTE) of biodiesel blends was slightly higher than mineral diesel. Lower biodiesel blends showed lower brake specific carbon monoxide (BSCO) and brake specific hydrocarbon (BSHC) emissions than mineral diesel. Brake specific nitrogen oxides (BSNOx) emissions from KOME20 and KOME10 were higher than mineral diesel. Combustion duration of was also higher than mineral diesel. Keywords: Multiple injections; biodiesel; fuel injection pressure; emissions; combustion characteristics. 1

7 Introduction Rising global energy demand and dwindling fossil fuel resources have catalyzed global interest in finding sustainable alternative fuels for transport sector. Compression ignition (CI) engines are most widely used prime movers in transport sector worldwide. Large number of investigations carried out on biodiesel obtained from different feedstocks has indicated that biodiesel is a promising renewable alternative fuel candidate for partial replacement of mineral diesel [1-3]. Tziourtzioumis et al. studied transient behavior of engine with CRDI system using B70. They observed that lower heating value of biodiesel limits engine s ability to attain the highest-load points on the engine operating map [4]. For achieving same power output, authors proposed to extend maximum fuel delivery per stroke by roughly 10% [4]. Kousoulidou et al. also compared HC emissions from palm oil biodiesel and Rapeseed methyl ester (RME) vis-a-vis mineral diesel in an engine equipped with CRDI system [5]. For palm biodiesel and RME, HC emissions increased by 40% and 15% respectively compared to mineral diesel [5]. Kousoulidou et al. also reported +6 to -4% variations in NOx emission for 10% Palm biodiesel fuelled engine compared to baseline mineral diesel [5]. Ye et al. observed that for all loads, an increase in FIP in a CRDI engine significantly increased NOx emissions from both ultra-low sulfur diesel (ULSD) as well as B40 of Soybean methyl ester (SME). With the same injection strategy as baseline diesel, biodiesel usage increased NOx emissions [6]. Yoon et al. observed significantly lower filter smoke number (FSN) for biodiesel-ethanol blend (90-10) in comparison to mineral diesel in all cases of double injection strategy [7]. Park et al. reported that fuel injection in a multiple-injection mode with short injection interval between the two pulses led to a reduction in soot, HC, and CO emissions, while NOx emissions increased from a soybean biodiesel fuelled CRDI engine [8]. Ye et al. studied the combustion characteristics of ULSD and B40 in a CRDI engine. Heat release rate (HRR) analysis showed that at all load conditions, retarding the SOI timing towards the ATDC side enhanced premixed combustion. Authors observed that at the same SOI, higher FIP led to higher HRR and a slightly earlier start of combustion (SOC) due to better air-fuel mixing [6]. At lower loads, biodiesel fueling advanced the SOC and decreased the premixed heat release. At moderate to high loads, however, biodiesel fueling did not make a significant impact on the HRR [6]. Yamane et al. found that difference in the bulk modulus of biodiesel and diesel are higher at lower temperature and higher pressure [9]. In a direct injection compression ignition (DICI) engine, differences in the injection characteristics of biodiesel and diesel become more 2

8 significant at higher engine speeds and loads [9]. Boudy et al. investigated the effect of fuel properties on injection quantity and duration of a CRDI system [10]. They observed that fuel density is the main property, which influences fuel quantity injected. Pressurewave velocity affects the injection quantity in the second injection pulse during a multiple injection strategy. They observed that quantity of injected fuel quantity decreases with increasing bulk modulus and increases with increasing fuel density [10]. Soybean biodiesel showed faster ignition, lower premixed heat release, and lower peak cylinder pressure in a CRDI engine compared to diesel due to its higher cetane number and lower calorific value [11-12]. It is generally accepted that fuels containing upto 20% (v/v) biodiesel do not require modifications in engine hardware [3, 13-15]. With significant improvement in fuel injection and control technologies, mechanical fuel injection systems are being replaced by electronic fuel injection systems. Optimization of fuel injection parameters is an important and powerful tool available for control of NOx emissions from such elecrnic fuel injected compression ignition engines e.g. common rail direct injection (CRDI) systems [16]. Electronic fuel injection systems allow use of relatively higher injection pressures and offer flexible and precise control of fuel injection timings [14-15]. For large-scale efficient utilization of biodiesel in existing CI engines, its effect on performance, emissions and combustion characteristics of modern CRDI engines in a wide range of control parameters needs to be experimentally investigated and optimized. Depending on agricultural climatic conditions, different countries use specific locally available feedstocks for biodiesel production. In the background of shortage of edible oils and with a view of discouraging use of fertile agricultural land for biofuel production, biofuel policy of India focused on biodiesel production from non-edible oils such as Jatropha and Karanja, which can grow on marginal lands with very small water requirement [17]. Karanja tree naturally grows in almost entire Indian sub-continent hence its utilization for large-scale biodiesel production is expected to ensure stability and sustainability of fuel supply because it is well adapted to local climatic conditions [18]. For ensuring efficient utilization of Karanja biodiesel blends in modern transportation engines equipped with CRDI system, effect of multiple injections, injection timings and injection pressures for various biodiesel blends has been experimentally investigated in this study. 3

9 Experimental Setup Schematic of the experimental setup used for evaluating performance, emissions and combustion characteristics of Karanja biodiesel blends vis-à-vis baseline mineral diesel is given in figure 1. Figure 1: Schematic of single cylinder research engine experimental setup Effect of FIP in multiple injection mode, start of pilot injection (SOPI), and start of main injection (SOMI) timings on engine performance, emissions and combustion characteristics of all test fuels was investigated in a single cylinder research engine (AVL; 5402), which was coupled to an AC dynamometer (AVL; TLBDURASTA.0101). Technical specifications of the test engine with the CRDI fuel injection system are given in Table 1. This research engine is flexible for user defined settings of FIP, injection timings and injected fuel quantity. Fuel injection controller is capable of supplying upto four injection pulses in split injection mode at 1400 bar FIP (max.). Experiments were performed at 500 and 1000 bar FIP with varying SOPI and SOMI timings. During the experiments, temperature of test fuels was maintained at 20 C by the fuel conditioning unit (AVL, 753CH). Temperature of the coolant was maintained at 80 C by coolant conditioning condition unit (Yantrashilpa, YS4027). Temperature and pressure of lubricating oil was maintained at 90 C and 3.5 bar respectively by the oil condition unit (Yantrashilpa, YS4312). Intake air flow rate was measured by the rotary gas flow meter (Elster Instromrt, RVG G160), which was installed upstream of a charge air vessel, used for dampening the fluctuations in the air flow. Fuel flow rate was measured by gravimetric fuel flow meter (AVL, Fuel Balance 733S.18). For combustion analysis, a water cooled piezoelectric pressure transducer (AVL, QC34C) was installed in the engine cylinder head. Rotation of crank shaft was recorded by an optical encoder (AVL, 365CC/ 365X). Cylinder pressure-crank angle data history was acquired and analysed by high speed data acquisition system (AVL, Indismart 611). Table 1: Technical specifications of the test engine Performance, emissions and combustion characteristics of 10, 20 and 50% Karanja biodiesel blends (KOME10, KOME20, KOME20) at constant engine speed (1500 rpm) were compared with baseline mineral diesel. Important physical properties of all test fuels are given in table 2. Table 2: Physical properties of test fuels Fuel energy input to the engine was kept constant for all engine operating conditions, which was corresponding to an air-fuel ratio (AFR) of 23 for mineral diesel. For all engine operating conditions, 10% fuel quantity injected/ cycle was injected in the pilot 4

10 injection and remaining 90% fuel quantity was injected during the main injection. Engine operating point corresponding to 5 bar brake mean effective pressure (BMEP) at 1500 rpm was chosen for detailed investigations of engine performance, emissions and combustion characterization in this study. 3. Results and Discussion Effect of fuel injection pressure, SOPI and SOMI timings on engine performance, emissions and combustion characteristics of Karanja biodiesel blends with mineral diesel (, KOME20, and KOME10) vis-à-vis baseline mineral diesel were investigated in a single cylinder research engine at 1500 rpm engine speed at a fixed input fuel energy at all investigated operating conditions. For controlling the fuel energy input in every engine cycle precisely, mass of fuel injected/cycle was controlled by adjusting the duration of injection pulse with the help of an open ECU, which was used for controlling the fuel injection. By experimental measurement of calorific value of different fuel blends used in the investigation using a bomb calorimeter, quantity of injected fuel/ cycle was calculated. 3.1 Performance characteristics Figure 2 shows the variation of BSFC in multiple injection mode with SOMI timings at 500 and 1000 bar FIP for -21 CA and -18 CA SOPI timings. BSFC was calculated from the measured fuel consumption rate (kg/h) and power output (P). Power output was calculated by measuring engine rpm and torque. BSFC (kg/kwh) = Fuel consumption rate (kg/h)/ P (1) BTE (%) = (360 P)/ (Calorific value of fuel (MJ/kg) Fuel consumption rate (kg/h)) (2) At 500 bar FIP for all test fuels, BSFC was lowest at -15 CA and -12 CA SOMI timings at -21 CA and -18 CA SOPI timings respectively (figure 2(a)-(b)). Figure 2: Variations in BSFC with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing Among all combinations of injection timings investigated at 500 bar FIP with multiple injections, BSFC was lowest at -12 CA SOMI timings at -18 CA SOPI timing for all test fuels. At -21 CA and -18 CA SOPI timings for all test fuels at 1000 bar FIP, BSFC was lowest at -9 CA and -6 CA SOMI timings respectively (figure 2(c)-(d)). BSFC was observed to increase with increasing biodiesel fraction in the test blends due to reduction in the calorific value for all SOI timings. In multiple injection mode at 1000 bar FIP, BSFC was lowest at -21 CA SOPI and -9 CA SOMI timing for all test fuels 5

11 (figure 2 (a)). In multiple injection mode, fuel injected during pilot injection helps in reducing combustion noise, which makes it possible to advance the fuel injection for main injection vis-a-vis single injection mode, while maintaining noise level under control. Earlier injection of fuel helps in achieving lower BSFC in multiple injection mode vis-à-vis single injection mode. Figure 3: Variations in BTE with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and - 21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing Figure 3 shows the variation of BTE in multiple injection mode at 500 and 1000 bar FIP at different SOPI and SOMI timings. At -21 CA and -18 CA SOPI timings for all test fuels, BTE reduced with retarded main injection timings (figure 3(a)-(d)). At 1000 bar and 500 bar FIPs, maximum BTE was obtained at -21 CA (-9 CA SOMI timing) and - 18 CA (-12 CA SOMI timing) SOPI timings respectively (figure 3(b) and 3(c)). This shows that BTE is more sensitive to SOMI timings. SOPI timings are used to control the thermodynamic condition of the cylinder charge during main injection, which controls the rate of heat release (ROHR) during main injection and combustion. Higher FIP leads to finer atomization of test fuels into smaller droplets. This subsequently improves vaporization therefore rate of pressure rise becomes higher for advanced SOI timings. Multiple injections were effective in extending this range of SOI timing by keeping rate of pressure rise within acceptable limits, which resulted in achieving higher BTE at 1000 bar and 500 bar FIPs in multiple injection mode. Despite higher BSFC, thermal efficiency of all biodiesel blends was observed to be higher than mineral diesel. This indicates that oxygen content of biodiesel helps in improving the combustion of fuel in the engine cylinder. It was also observed that thermal efficiency of KOME20 and KOME10 were higher than mineral diesel as well as. This can be explained by deterioration of atomization and air-fuel mixing characteristics of test fuels with higher concentration of biodiesel due to their higher viscosity, density and relatively inferior volatility characteristics of biodiesel compared to mineral diesel. 3.2 Emission characteristics Figure 4 shows the variation of brake specific carbon monoxide (BSCO) emissions in multiple injection mode at 500 and 1000 bar FIP. In split injection mode at 500 and 1000 bar FIP, BSCO emissions increased with retarded SOMI timings for all tested SOPI timings (figure 4). Lowest BSCO emissions observed at 500 bar FIP in split injection mode were higher than lowest BSCO emission observed at 1000 bar FIP in 6

12 split injection mode for all test fuels, which indicates improvement in air-fuel mixing with increasing FIP. Lowest BSCO emissions for all test fuels were observed at -18 CA SOPI and -6 CA SOMI timings at 1000 bar FIP. In split injection at 500 bar injection pressure, BSCO emissions for biodiesel blends were lower than mineral diesel. At 500 bar FIP, effect of higher concentration of biodiesel in test fuel was significant, which adversely affected the air fuel mixing, resulting in increased BSCO emissions from in comparison to lower biodiesel blends. Lowest BSCO emissions were observed for KOME20 and KOME10 at all fuel injection timings. Ong et al. also reported lower CO emissions for lower concentration biodiesel blends but CO emissions increased for higher concentration biodiesel blends [19]. Mofijur et al. reported that 10 and 20% biodiesel blends reduced CO emission by 10.60% and 22.93% compared with mineral diesel, respectively [20]. Higher concentration of Karanja biodiesel in the test fuel tends to increase the liquid length of spray and larger droplets, leading to inferior air-fuel mixing, which increases BSCO emissions [21]. Spray characteristics of lower biodiesel blends may be almost similar to mineral diesel however their oxygen content improves combustion, which results in lower BSCO emissions for KOME20 and KOME10 vis-a-vis mineral diesel. Figure 4: Variations in BSCO emissions with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing Figure 5: Variations in BSHC emissions with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing Figure 5 shows the variation of brake specific hydrocarbon (BSHC) emissions with SOMI timing at different SOPI timings for 500 and 1000 bar FIP. At 500 bar FIP for all SOPI timings, BSHC emissions increased with retarded as well as too much advanced SOMI timings. For all tested SOPI timings at 1000 bar FIP, BSHC emissions increased with retarded SOMI timings. KOME10 and KOME20 showed lower BSHC emissions in comparison to mineral diesel for all SOI timings. BSHC emissions for were higher than BSHC emission for mineral diesel. Silitonga et al. [22] reported decreasing HC emissions with increasing concentration of biodiesel in the fuel. Mofijur et al. reported that 10 and 20% biodiesel blends reduced HC emission by 9.21% and 23.68% in comparison to mineral diesel, respectively [20]. BSHC emissions in multiple injection mode were significantly lower than BSHC emissions for single injection mode. Lowest BSHC emissions at -21 CA SOPI timing were higher than the lowest BSHC emissions 7

13 at -18 CA SOPI timing for all test fuels. This indicates that too advanced injection timing of small fuel quantity injected during the pilot injection may result in overmixing, forming fuel lean zones. These fuel lean zones are responsible for increased hydrocarbon emissions at advanced SOPI timings. Figure 6: Variations in BSNOx emissions with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing Figure 6 shows the variation of brake specific nitrogen oxides (BSNOx) emissions with SOMI timings at different SOPI timings for 500 and 1000 bar FIP. For all SOPI timings, BSNOx emissions decreased with retarded SOMI timings. Retarded injection timings reduce the maximum in-cylinder temperature and pressure during combustion, which results in lower BSNOx emissions. In split injection mode also, BSNOx emissions of KOME20 and KOME10 were higher than mineral diesel. Ong et al. [19], Silitonga et al. [22] and Fattah et al. [23] reported increased NOx emissions with increasing biodiesel concentration in the fuel. At different SOPI timings, BSNOx emissions at a fixed SOMI timing were almost similar for all test fuels. BSNOx emissions are more sensitive to variations in SOMI timings. BSNOx emissions were higher for 1000 bar FIP in comparison to 500 bar FIP because improved fuel-air mixing at higher FIP increases the peak in-cylinder temperature, resulting in higher NOx formation. 3.3 Combustion characteristics Effects of FIP, injection strategy and SOI timings on the combustion characteristics of, KOME20 and KOME10 vis-à-vis mineral diesel were analyzed by measuring in-cylinder pressure w.r.t. crank angle in a single cylinder research engine equipped with CRDI system. Measured cylinder pressure data for 200 consecutive cycles was averaged in order to eliminate the effect of cyclic variations of combustion parameters and then averaged data was analyzed to calculate HRR, mass burn fractions (MBF) and combustion duration. Figure 7: Cylinder pressure and HRR for different SOPI and SOMI timings at 500 bar FIP Figure 7 shows in-cylinder pressure and HRR for test fuels at -6 and -3 CA SOMI timings and at -21 and -18 CA SOPI timings at 500 bar FIP. Two distinct heat release peaks are seen for all test fuels, which correspond to two fuel injection pulses. Figure 8: Cylinder pressure and HRR for different SOPI and SOMI timings at 1000 bar FIP 8

14 Figure 8 shows HRR and in-cylinder pressure variation at -6 and -3 CA SOMI timings at 1000 bar FIP. At retarded injection timings, two premixed heat release peaks were observed for all test fuels corresponding to the two fuel injection pulses, similar to 500 bar FIP. At 500 bar FIP, peak of premixed heat release for the first injection pulse was higher in comparison to 1000 bar at the same injection timing due to longer ignition delay at lower FIP, which increases the fuel mass burnt in the premixed combustion phase. Peak of premixed heat release for the main injection pulse was higher at 1000 bar FIP. This indicates that earlier combustion completion for fuel injected in pilot pulse reduces its effectiveness in reducing the ignition delay for the main injection. Accumulation of larger fuel quantity during the premixed combustion phase at 1000 bar FIP increases the peak heat release. In multiple injection mode at 500 and 1000 bar FIP, maximum in-cylinder pressure decreased and crank angle position of maximum pressure retarded with retarding SOMI timings at all SOPI timings (figure 9) for all test fuels. Position of maximum cylinder pressure was relatively retarded for and this retard was higher for retarded SOMI timings. Figure 9: Variation of maximum cylinder pressure and its position with SOMI timing at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing Increasing separation between pilot and main injection pulses resulted in cooling of cylinder charge before the main injection. Comparatively lower in-cylinder temperature at the time of main injection results in relatively inferior mixing of lower volatility fuel (biodiesel), which results in retarded maximum pressure position. Maximum in-cylinder pressure at 1000 bar FIP in comparison to 500 bar FIP was higher at same SOPI and SOMI timings. At the same SOPI and SOMI timings, position of maximum in-cylinder pressure was almost same at 1000 and 500 bar FIPs. Combustion of larger fuel fraction at maximum cylinder pressure condition due to higher HRR (at higher FIP) results in higher maximum cylinder pressure at 1000 bar FIP, in comparison to 500 bar FIP. Start of combustion (SOC) is characterized by crank angle position of 10% mass burn fraction (MBF). End of combustion (EOC) is characterized by crank angle position of 90% MBF. Figure 10: Variation of 10% and 90% MBF position with SOMI timing for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing 9

15 In multiple injection mode at 500 and 1000 bar injection pressure, SOC retarded with retarded SOMI timing for all test fuels at all SOPI timings (figure 10). At a fixed SOMI timing, retarded SOPI timing also resulted in retarded SOC. At 1000 bar FIP, SOC timing for all test fuels was almost identical at all injection timings. At -21 CA SOPI timing, 90% MBF position of all test fuels were almost same (10(c)). At -18 CA SOPI timings, EOC for was delayed in comparison to mineral diesel and lower biodiesel blends, which had comparable EOC positions (figure 10(d)). Higher fuel requirement for biodiesel blends (due to lower calorific value) and their relatively inferior mixing characteristics may be the cause for delayed EOC for biodiesel blends. 90% MBF position was delayed at 500 bar FIP in comparison to 1000 bar FIP. At 500 bar FIP, difference in 90% MBF position of higher biodiesel blends (KOME20 and ) and mineral diesel were higher in comparison to 1000 bar FIP. This indicates that increasing FIP is more effective in improving the HRR of lower biodiesel blends and mineral diesel in comparison to higher biodiesel blends. Combustion duration has been calculated as the difference between 90% MBF and 10% MBF positions in terms of crank angle degrees. Figure 11: Variation of combustion duration with SOMI timing for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing In multiple injection mode at 500 and 1000 bar FIP, combustion duration increased for retarded SOMI timing for all test fuels at -21 and -18 CA SOPI timings (figure 11). Retarded SOMI timings delayed both start and end of combustion (figure 10) but delay in SOC timing was shorter than delay in EOC timing because it was mainly affected by SOPI timing (which remained fixed). Shorter delay in SOC timing resulted in increased combustion duration with retarding SOMI timing. For all injection timings, combustion duration of and KOME20 were higher than mineral diesel due to relatively inferior mixing characteristics and requirement of larger fuel quantity in comparison to mineral diesel with increasing concentration of biodiesel in the test fuel. Combustion duration at 500 bar FIP was longer than the combustion duration at 1000 bar FIP. Increased injection pressure resulted in atomization of fuel into finer droplets, which improved air-fuel mixing. Improved air-fuel mixing resulted in shorter combustion duration at higher FIP. At 500 bar FIP, retarded SOPI timing (from -21 CA to -18 CA) resulted in reduction in combustion duration. Probably at advanced SOPI timings, combustion of fuel injected during pilot pulse was completed earlier therefore it was not instrumental in improving the combustion characteristics of main injection. At retarded 10

16 SOPI timing, effect of heat released during pilot injection pulse increased the pressure and temperature of the combustion chamber during main injection, which improved the combustion characteristics of test fuels. 4. Conclusions Effects of start of pilot injection, start of main injection and fuel injection pressure on engine performance, emissions and combustion characteristics of Karanja biodiesel blends vis-a-vis mineral diesel were investigated at 1500 rpm in a single cylinder CRDI engine. Important findings are summarized below: BSFC of test fuels increased with increasing concentration of Karnaja biodiesel in the test fuel. Brake thermal efficiency of Karanja biodiesel blends was slightly higher than mineral diesel. At 1000 bar and 500 bar FIPs, maximum BTE was obtained at SOPI timings of -21 CA (-9 CA SOMI timing) and -18 CA (-12 CA SOMI timing) respectively. Lower Karanja biodiesel blends showed lower brake specific CO and HC emissions in comparison to mineral diesel but BSHC emissions of were higher than mineral diesel at some operating conditions. Brake specific NOx emissions KOME20 and KOME10 were higher than mineral diesel. At different SOPI timings, BSNOx emissions at a fixed SOMI timing were almost similar for all test fuels. BSNOx emissions were higher for 1000 bar FIP in comparison to 500 bar FIP. Maximum in-cylinder pressure at 1000 bar FIP in comparison to 500 bar FIP was higher at same SOPI and SOMI timings. Combustion duration of was higher than mineral diesel due to relatively inferior mixing characteristics and requirement of larger fuel quantity. Overall this experimental investigation suggested that utilization of upto 10% or 20% Karanja biodiesel blends in CRDI engines with pilot injection can be useful in improving engine efficiency while reducing emissions. 11

17 Acknowledgements Support provided by Council of Scientific and Industrial Research (CSIR), Government of India under Senior Research Associate (Pool Scientist) scheme to Mr. Atul Dhar for conducting this research under the supervision of Prof. Avinash Kumar Agarwal at Engine Research Laboratory, IIT Kanpur is gratefully acknowledged. References 1. Lapuerta M, Armas O, Fernàndez JR. Effect of biodiesel fuels on diesel engine emissions. Progress in Energy and Combustion Science 2008; 34: Agarwal AK. Biofuels (alcohols and biodiesel) applications as fuels for internal combustion engines. Progress in Energy and Combustion Science 2007; 33: Agarwal AK, Dhar A. Comparative performance, emission and combustion characteristics of rice-bran oil and its biodiesel in a transportation diesel engine. Journal of Engineering for Gas Turbines and Power, Transactions of ASME 2010; 132: Tziourtzioumis D, Stamatelos A. Effects of a 70% biodiesel blend on the fuel injection system operation during steady-state and transient performance of a common rail diesel engine. Energy Conversion and Management 2012; 60: Kousoulidou M, Fontaras G, Ntziachristos L, Samaras Z. Biodiesel blend effects on common-rail diesel combustion and emissions. Fuel 2010; 89(11): Ye P, Boehman AL. An investigation of the impact of injection strategy and biodiesel on engine NOx and particulate matter emissions with a common-rail turbocharged DI diesel engine. Fuel 2012; 97: Yoon SH, Hwang JW, Lee CS. Effect of injection strategy on the combustion and exhaust emission characteristics of biodiesel-ethanol blend in a DI diesel engine. Journal of Engineering for Gas Turbines and Power 2010; 132: Park SH, Yoon SH, Lee CS. Effects of multiple-injection strategies on overall spray behavior, combustion, and emissions reduction characteristics of biodiesel fuel. Applied Energy 2011; 88: Yamane K, Ueta A, Shimamoto Y. Influence of physical and chemical properties of biodiesel fuels on injection, combustion and exhaust emission characteristics in a direct injection compression ignition engine. International Journal of Engine Research 2001; 2:

18 Boudy F, Seers P. Impact of physical properties of biodiesel on the injection process in a common-rail direct injection system. Energy Conversion and Management 2009; 50 (12): Yoon SH, Suh HK, Lee CS. Effect of spray and EGR rate on the combustion and emission characteristics of biodiesel fuel in a compression ignition engine. Energy & Fuels 2009; 23(3): Kim MY, Yoon SH, Hwang JW, Lee CS. Characteristics of particulate emissions of compression ignition engine fueled with biodiesel derived from soybean. Journal of Engineering for Gas Turbines and Power 2008; 130: Dhar A, Kevin R, Agarwal AK. Production of biodiesel from high-ffaneem oil and its performance, emission and combustion characterization in a single cylinder DICI engine. Fuel Processing Technology 2012; 97: Caresana F. Impact of biodiesel bulk modulus on injection pressure and injection timing. The effect of residual pressure. Fuel 2011, 90 (2): Dhar A, Agarwal AK. Effect of Multiple Injections on Particulate Size-Number Distributions in a Common Rail Direct Injection Engine Fueled with Karanja Biodiesel Blends. SAE Paper ; Palash SM, Masjuki HH, Kalam MA, Masum BM, Sanjid A, Abedin MJ. State of the art of NOx mitigation technologies and their effect on theperformance and emission characteristics of biodiesel-fueledcompression Ignition engines.energy Conversion and Management 2013; 76: Ministry of New & Renewable Energy, Government of India; National Policy on Biofuels, accessed on Agarwal AK, Dhar A. Experimental investigations of performance, emission and combustion characteristics of Karanja oil blends fuelled DICI engine. Renewable Energy 2013; 52: Ong HC, Masjuki HH, Mahlia TMI, Silitonga AS, Chong WT, Leong KY. Optimization of biodiesel production and engine performance from high free fatty acid Calophylluminophyllum oil in CI diesel engine.energy Conversion and Management 2014; 81: Mofijur M, Masjuki HH, Kalam MA, Atabani AE, Arbab MI, Cheng SF, Gouk SW. Properties and use of Moringa oleifera biodiesel and diesel fuel blendsin a multi-cylinder diesel engine. Energy Conversion and Management 2014; 82: Som S, Longman DE, Ramírez AI, Aggarwal SK. A comparison of injector flow and spray characteristics of biodiesel with petrodiesel. Fuel 2010; 89:

19 Silitonga AS, Masjuki HH, Mahlia TMI, Ong HC, Chong WT. Experimental study on performance and exhaust emissions of a dieselengine fuelled with Ceibapentandra biodiesel blends. Energy Conversion and Management 2013; 76: Fattah IMR, Masjuki HH, Kalam MA., Wakil MA, Ashraful AM, Shahir SA. Experimental investigation of performance and regulated emissionsof a diesel engine with Calophylluminophyllum biodiesel blends accompanied by oxidation inhibitors. Energy Conversion and Management2014; 83:

20 Figure(s) Figure 1: Schematic of single cylinder research engine experimental setup

21 BSFC (kg/kwh) BSFC (kg/kwh) BSFC (kg/kwh) BSFC (kg/kwh) SOPI: -21 º CA, FIP: 500 bar KOME KOME Start of injection (deg) SOPI: -18ºCA, FIP: 500 bar KOME KOME Start of injection (deg) 0 (a) (b) SOPI: -21ºCA, FIP: 1000 bar KOME20 KOME Start of injection (deg) SOPI: -18ºCA, FIP: 1000 bar KOME20 KOME Start of injection (deg) (c) (d) Figure 2: Variations in BSFC with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

22 BTE (%) BTE (%) BTE (%) BTE (%) 28 SOPI: -21 º CA, FIP: 500 bar 28 SOPI: -18 º CA, FIP: 500 bar KOME20 KOME Start of injection (deg) KOME20 24 KOME Start of injection (deg) 0 (a) (b) SOPI: -18ºCA, FIP: 1000 bar SOPI: -21ºCA, FIP: 1000 bar KOME20 KOME Start of injection (deg) KOME20 KOME Start of injection (deg) 1.5 (c) (d) Figure 3: Variations in BTE with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and - 21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

23 BSCO (g/kwh) BSCO (g/kwh) BSCO (g/kwh) BSCO (g/kwh) SOPI: -21 º CA, FIP: 500 bar SOPI: -18 º CA, FIP: 500 bar KOME20 0 KOME Start of injection (deg) KOME20 0 KOME Start of injection (deg) 0 (a) (b) SOPI: -21ºCA, FIP: 1000 bar SOPI: -18ºCA, FIP: 1000 bar KOME20 KOME KOME20 KOME Start of injection (deg) Start of injection (deg) (c) (d) Figure 4: Variations in BSCO emissions with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

24 BSHC (g/kwh) BSHC (g/kwh) BSHC (g/kwh) BSHC (g/kwh) 0.3 SOPI: -21 º CA, FIP: 500 bar 0.3 SOPI: -18 º CA, FIP: 500 bar KOME KOME Start of injection (deg) KOME KOME Start of injection (deg) 0 (a) (b) 0.3 SOPI: -21ºCA, FIP: 1000 bar 0.3 SOPI: -18ºCA, FIP: 1000 bar KOME20 KOME KOME20 KOME Start of injection (deg) Start of injection (deg) (c) (d) Figure 5: Variations in BSHC emissions with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

25 BSNOx (g/kwh) BSNOx (g/kwh) BSNOx (g/kwh) BSNOx (g/kwh) SOPI: -21 º CA, FIP: 500 bar KOME20 KOME Start of injection (deg) SOPI: -18 º CA, FIP: 500 bar KOME20 KOME Start of injection (deg) 0 (a) (b) SOPI: -21ºCA, FIP: 1000 bar KOME20 KOME SOPI: -18ºCA, FIP: 1000 bar KOME20 KOME Start of injection (deg) Start of injection (deg) (c) (d) Figure 6: Variations in BSNOx emissions with varying SOMI timings for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

26 Figure 7: Cylinder pressure and HRR for different SOPI and SOMI timings at 500 bar FIP

27 Figure 8: Cylinder pressure and HRR for different SOPI and SOMI timings at 1000 bar FIP

28 Maximum Pressure (bar) Max. Pressure Position (deg) Maximum Pressure (bar) Max. Pressure Position (deg) Maximum Pressure (bar) Max. Pressure Position (deg) Maximum Pressure (bar) Max. Pressure Position (deg) SOPI: -21 CA, FIP: 500 bar SOPI: -18 CA, FIP: 500 bar KOME20 50 KOME Start of main injection (deg) KOME20 50 KOME Start of main injection (deg) (a) (b) SOPI: -21 CA, FIP: 1000 bar KOME20 KOME Start of main injection (deg) SOPI: -18 CA, FIP: 1000 bar 4 KOME20 2 KOME Start of main injection (deg) (c) (d) Figure 9: Variation of maximum cylinder pressure and its position with SOMI timing at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

29 CA 10 (deg) CA 90 (deg) CA 10 (deg) CA 90 (deg) CA 10 (deg) CA 90 (deg) CA 10 (deg) CA 90 (deg) 10 6 KOME20 KOME SOPI = -18 CA, FIP: 500 bar SOPI: -21 CA, FIP: 500 bar Start of main injection (deg) KOME20 KOME Start of main injection (deg) (a) (b) 10 6 SOPI: -21 CA, FIP: 1000 bar SOPI: -18 CA, FIP: 1000 bar KOME20 KOME Start of main injection (deg) KOME20 KOME Start of main injection (deg) 1.5 (c) (d) Figure 10: Variation of 10% and 90% MBF position with SOMI timing for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

30 Combustion duration (deg) Combustion duration (deg) Combustion duration (deg) Combustion duration (deg) SOPI = -18 CA, FIP: 500 bar KOME20 KOME SOPI: -21 CA, FIP: 500 bar Start of main injection (deg) KOME20 KOME Start of main injection (deg) 0 (a) (b) 30 SOPI: -21 CA, FIP: 1000 bar KOME20 KOME Start of main injection (deg) SOPI: -18 CA, FIP: 1000 bar KOME20 KOME Start of main injection (deg) (c) (d) Figure 11: Variation of combustion duration with SOMI timing for test fuels at (a) 500 bar FIP and -21 CA SOPI timing (b) 500 bar FIP and -18 CA SOPI timing (c) 1000 bar FIP and -21 CA SOPI timing (d) 1000 bar FIP and -18 CA SOPI timing

31 *Detailed Response to Reviewer Comments Response to Reviewer s Comments Experimental Investigations of the Effect of Pilot Injection on Performance, Emissions and Combustion Characteristics of Karanja Biodiesel Fuelled CRDI Engine Atul Dhar, Avinash Kumar Agarwal* a Engine Research Laboratory Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur , India *Corresponding Author s akag@iitk.ac.in Reviewer #1: The level of research work carried out is ok. The technical language is also ok, except for a few errors that have been pointed out inthe text. Kindly address these errors. The work could be improved if the authors can compare their results with those obtained by other authors, specially with regards to emissions from the biodiesel blends. Note: Please see the attached marked copy. Response: Thanks for recommending the paper for publication. Typographical mistakes indicated in the marked pdf file have been corrected. Results for emissions are compared with other biodiesel blends, as suggested by the reviewer. This part of the modified discussion is reproduced below for your ready reference. Ong et al. also reported lower CO emissions for lower concentration biodiesel blends but CO emissions increased for higher concentration biodiesel blends [19]. Mofijur et al. reported that 10 and 20% biodiesel blends reduced CO emission by 10.60% and 22.93% compared with mineral diesel, respectively [20]. Silitonga et al. [22] reported decreasing HC emissions with increasing concentration of biodiesel in the fuel. Mofijur et al. reported that 10 and 20% biodiesel blends reduced HC emission by 9.21% and 23.68% in comparison to mineral diesel, respectively [20]. Ong et al. [19], Silitonga et al. [22] and Fattah et al. [23] reported increased NOx emissions with increasing biodiesel concentration in the fuel. 1

32 Reviewer #2: Aim of this paper is to investigate the effect of pilot + main injection strategy on conversion, emissions and combustion characteristics of a CRDI engine fed with blends of mineral diesel and Karanja Biodiesel. Results are compared with those obtained feeding the engine with only mineral diesel. The subject is interesting and up-to-date, and the work is generally well organized and exposed. However, there are several adjustments required in order it to be eligible for publication: Response: Thanks for the words of encouragement. Details of improvements based on your suggestions are given below, point by point: 1) The expression "Multiple Injections" throughout the paper is confusing; in fact, multiple injections usually refers to more than two injections,while, in case only pilot + main injections are performed, usually it is referred as pilot injection; plus, to my understanding, the paper does not tackle the problem of performance, given that no data are provided concerning the maximum power; in this case, conversion efficiency (in terms of BSFC and BTE) is more appropriate; my suggestion is to change the title of the paper as "Experimental Investigation of the Effect of Pilot Injection on Conversion efficiency, emissions and combustion characteristics of Karanja Biodiesel fuelled CRDI engine. Response: We agree with your suggestion of using the word pilot injection rather than multiple injections. Term performance is used for description of maximum power, BTE and BSFC. Maximum power estimation was not in the scope of this study because the fuel quantity was fixed for all experimental conditions and fuels. Still we would like to use term performance. Hence, title of the manuscript has been changed to Experimental Investigations of the Effect of Pilot Injections on Performance, Emissions and Combustion Characteristics of Karanja Biodiesel Fuelled CRDI Engine. Thanks for the suggestion. 2) - row 149: the Authors state that the experimental characterization was done "at a fixed input fuel energy"; in this case, they shall specify how the guaranteed this; plus, the Authors shall detail how they estimated BSFC and BTE; Response: For controlling the fuel energy input in every engine cycle precisely, mass of fuel injected/cycle was controlled by adjusting the duration of injection pulse with the help of an open ECU, which was used for controlling the fuel injection. By experimental measurement of calorific value of different fuel blends used in the investigation using a bomb calorimeter, quantity of injected fuel/ cycle was calculated. (line ). 2