Combustion Characteristics of a Single-Cylinder Engine Equipped with Gasoline and Ethanol Dual-Fuel Systems

Transcription

1 SAE TECHNICAL PAPER SERIES Combustion Characteristics of a Single-Cylinder Engine Equipped with Gasoline and Ethanol Dual-Fuel Systems Guoming Zhu, Tom Stuecken, Harold Schock and Xiaojian Yang Michigan State University David L. S. Hung Visteon Corporation Andrew Fedewa Mid Michigan Research, LLC 28 SAE International Powertrains, Fuels and Lubricants Congress Shanghai, China June 23-25, 28 4 Commonwealth Drive, Warrendale, PA 596- U.S.A. Tel: (724) Fax: (724) Web:

2 By mandate of the Engineering Meetings Board, this paper has been approved for SAE publication upon completion of a peer review process by a minimum of three (3) industry experts under the supervision of the session organizer. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE. For permission and licensing requests contact: SAE Permissions 4 Commonwealth Drive Warrendale, PA 596--USA permissions@sae.org Tel: Fax: For multiple print copies contact: SAE Customer Service Tel: (inside USA and Canada) Tel: (outside USA) Fax: CustomerService@sae.org ISSN Copyright 28 SAE International Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in SAE Transactions. Persons wishing to submit papers to be considered for presentation or publication by SAE should send the manuscript or a 3 word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE. Printed in USA

3 Combustion Characteristics of a Single-Cylinder Engine Equipped with Gasoline and Ethanol Dual-Fuel Systems Copyright 28 SAE International Guoming Zhu, Tom Stuecken, Harold Schock and Xiaojian Yang Michigan State University David L. S. Hung Visteon Corporation Andrew Fedewa Mid Michigan Research, LLC ABSTRACT The requirement of reduced emissions and improved fuel economy led the introduction of direct-injection (DI) spark-ignited (SI) engines. Dual-fuel injection system (direct-injection and port-fuel-injection (PFI)) was also used to improve engine performance at high load and speed. Ethanol is one of the several alternative transportation fuels considered for replacing fossil fuels such as gasoline and diesel. Ethanol offers high octane quality but with lower energy density than fossil fuels. This paper presents the combustion characteristics of a single cylinder dual-fuel injection SI engine with the following fueling cases: a) gasoline for PFI and DI, b) PFI gasoline and DI ethanol, and c) PFI ethanol and DI gasoline. For this study, the DI fueling portion varied from to percentage of the total fueling over different engine operational conditions while the engine air-to-fuel ratio remained at a constant level. It was shown in all cases that the IMEP (indicated mean effective pressure) decreases by as much as % as DI fueling percentage increases, except in case b) where the IMEP increases by 2% at light load. The combustion burn duration increases significantly at light load as DI fueling percentage increases, but only moderately at WOT (wide open throttle). In addition, the percentage of the ethanol in the total fueling plays a dominant role in affecting the combustion characteristics at light load; but at heavy load (WOT), the DI fueling percentage becomes an important parameter, regardless of the percentage of ethanol content in the fuel. INTRODUCTION Increasing concerns about global climate change and ever-increasing demands on fossil fuel capacity call for reduced emissions and improved fuel economy. Vehicles equipped with direct-injection (DI) fuel system have been introduced to markets globally. In order to improve DI engine full load performance at high speed, Toyota introduced an engine with a stoichiometric direct injection system with two fuel injectors for each cylinder, see []. One is a DI injector generating a dual-fanshaped spray with wide dispersion, while the other is a port injector. The dual-fuel system introduces one additional degree of freedom for engine optimization to reduce emissions with improved fuel economy. Ethanol has been used widely as a fuel additive or an alternative fuel due to its high-octane and clean combustion. Early research in [2] provides the physical and chemical fuel properties of ethanol that affect spark and compression ignited engine performance. Recently, renewed research in ethanol is mainly due to the concerns of global warming and transportation energy shortage ([3], [4], and [5]). High concentrate ethanol effect on spark ignited (SI) engine cold startability can be found in [6]; the discussions of ethanol application to DI and turbocharged engines can be found in [7] and [8]; and finally, reference [9] presents the combustion and emission characteristics of a PFI ethanol HCCI (homogenous charge compression ignition) engine. Using gasoline PFI and ethanol DI dual-fuel system to increase gasoline engine efficiency substantially is described in []. The main idea is to use a highly boosted small turbocharged engine to match the performance of a much larger engine. Direct injection of ethanol is used to suppress engine knock due to its substantial air charge cooling resulting from its high heat of vaporization. This paper investigates the combustion characteristics of a single cylinder engine equipped with dual-fuel system when different combinations of fuels (gasoline and ethanol) are used for PFI and DI fuel systems. For a given air-to-fuel ratio (AFR), we studied the engine Indicated Mean Effective Pressure (IMEP), Mass Fraction Burned (MFB), and burn duration by varying the

4 DI fueling percentage from zero to one hundred percent while either maintaining a fixed spark timing or conducting a spark timing sweep around MBT (minimal advance for the best torque) timing. In the following sections, the engine test setup and the combustion characteristics of the dual fuel injection in the engine are presented. TEST SETUP The test data shown in this paper was obtained using a three-valve.675l single-cylinder engine equipped with a conventional Port-Fuel-Injection (PFI) and Visteon s Low Pressure Direct Injection (LPDI) fuel systems [, 2], see Figure. The rail pressures of the PFI and LPDI fuel systems were operated at 3.5 and 2 bar, respectively. The end of injection timings of the PFI and DI fuel systems are at 36 and 3 DBTDC (Degree Before Top Dead Center), respectively. The fuel injection system incorporates two injectors on the single cylinder head. The PFI injector, mounted on the intake port, was a production injector with a 2-hole director plate and it formed a dual plume spray pattern. The nominal fuel injection pressure was set to 35 kpa and the corresponding static flow rate was about 3 g/s. On the other hand, the DI injector was side-mounted onto the cylinder head at an angle of 35 from the horizontal axis. It was a multi-hole production-intent injector configured with a nine-hole orifice plate. The injector was pressurized to about 2 MPa, giving a static flow rate of approximately 2. g/s. The internal nozzle geometry and geometrical parameters were designed to offer different spray characteristics. The DI injector employed in this set of engine tests delivered a wide spray with a 6 spray angle and a 5 bent angle. Using n-heptane as the standard test fuel per SAE J275 [4] recommended practice, the Sauter Mean Diameters (SMD) measured across the spray at 5 mm below the injector tip were between 2 and 42 microns. These radial scan point-wise SMD measurements were then converted into a single, line-of-sight SMD value by weighing and normalizing the measurement at each location with its corresponding flux density. The calculated SMD value was 33.4 microns at 2 MPa. Details of other spray characterizations performed on the DI fuel injectors and the LPDI fuel injection systems can be found in [5]. The cylinder head was instrumented with a laboratory grade pressure sensor, and a UEGO (universal exhaust gas oxygen) sensor was installed for air-to-fuel ratio measurement and control. The single-cylinder engine was controlled by a prototype engine controller for spark timing and dual-fuel injections; and the engine throttle and speed were regulated by the engine dynamometer controller. The single-cylinder engine has a single camshaft for both intake and exhaust valve timing regulation. The camshaft timing was manually advanced to optimize engine combustion. There was no Exhaust- 2 Gas-Recirculation (EGR) was used for all tests conducted. In-cylinder pressure and air-to-fuel ratio signals were collected using a dynamometer data sampling system with one crank degree resolution. For each test point, 3 cycles of test data were collected with one crank degree resolution. In the rest of this paper, the pressure signals used are averaged over 3 cycles, or otherwise specified. The engine IMEP, MFB, and burn duration are calculated based upon the 3 cycle averaged pressure signals. PFI Injector LPDI Injector Figure : Test Setup The gasoline test fuel used for all consequent dynamometer tests is indolene, and the ethanol used for these tests are laboratory grade E, which contains 5 percent gasoline and percent ethanol by volume. CASE A: GASOLINE PFI AND DI In this section, we discuss the combustion characteristics of the dual-fuel single-cylinder engine when gasoline was used for both fuel injection systems. For each test point, engine throttle and speed were held at constant. Engine test started at percent PFI fueling with a constant AFR, and then, the PFI fueling was reduced to a desired level (for example, 7% of original fueling quantity) while the corresponding DI fueling was increased to maintain the same AFR. The test continued until percent DI fueling was reached. Five PFI fueling percentages were selected and they are, 7, 5, 3, and percent by mass. For spark timing sweep at each given PFI fueling percentage, MBT timing was selected by adjusting spark timing such that the 5% MFB location remains at around 8 to degrees after TDC (top dead center). The test matrix for this case is shown in Table, where λ (inverse of equivalence ratio) is defined as the engine air-to-fuel ratio divided by the stoichiometric air-to-fuel ratio for the fuel mixture used in the test. For the engine operated at 5 RPM with 3.3 bar IMEP load, the spark

5 timing (ST) was selected at 34, 37, 4, 43, and 46 DBTDC, where 4 DBTDC was the MBT timing with percent PFI fueling. Again, for each test point, the PFI fueling percentage was selected at, 7, 5, 3, and percent. Table : Test matrix - PFI and DI gasoline RPM IMEP λ ST (DBTDC) bar WOT WOT bar. Sweep Figure 2 shows the engine IMEP as a function of DI fueling percentage when the engine was operated at 5 RPM with 3.3 bar IMEP. The solid line with circles is the normalized IMEP associated with the engine operated at its MBT spark timing, and the dotted line with diamonds is the engine IMEP with a fixed spark timing (4 DBTDC). It can be observed that in both cases that the engine IMEP reduces as the DI fueling percentage increases, and the IMEP reduces even more when the spark timing remains unchanged. This is mainly due to the DI charge cooling that reduces the thermal efficiency of the combustion process RPM 3.3bar IMEP Dual Fuel - All Gasoline MBT timing Fixed timing Percentage of DI Fueling (%) Figure 2: IMEP vs. DI fueling percentage (case A) Figure 3 shows the similar information to that of Figure 2 for the engine operated at 25 and 3 RPM with WOT (wide open throttle), where the solid line with circles is the normalized IMEP for engine operated at 25 RPM with wide open throttle, and dotted line with diamonds is at 3 RPM with WOT. It is clear that the IMEP reduces as DI fueling percentage increases at both speeds, but as engine speed increases, the IMEP reduction becomes moderate. 5 8 WOT Dual Fuel - All Gasoline 25 RPM 3 RPM Percentage of DI Fueling (%) Figure 3: WOT IMEP vs. DI fueling percentage (case A) Figure 4 presents the averaged engine in-cylinder pressure signals on top plot and MFB signals calculated from averaged in-cylinder pressure signals on bottom plot when the engine was operated at 5 RPM with 3.3 bar IMEP. The spark timing was adjusted to its MBT timing at each given DI fueling percentage. Note that these operational conditions are associated with the solid line with circles in Figure 2. In Figure 4 the solid lines are for the pressure and MFB signals with percent DI fueling; the dashed lines are for 5 percent DI fueling; the dotted lines are for 7 percent DI fueling; and the dash-dotted lines are for percent DI fueling. The MFB signals were calculated by normalizing the net pressure signals presented in equations () and (2). As described in reference [3]. The net pressure change Δ P(i) between two crank angles is V ( i) ΔP( i) = P( i + ) P( i) V ( i + ) and the net pressure at each crank angle is P NET.3 V ( i), () V Ig ( i) = P ( i ) + ΔP( i), (2) NET where P is the in-cylinder pressure, V is the chamber volume, and V Ig is the chamber volume at the ignition point. It can be observed in Figure 4 that as the DI fueling percentage increases, the combustion process slows down and the corresponding peak cylinder pressure reduces. This confirms the fact that the engine IMEP reduces as DI fueling percentage increases; see Figure 2. 3

6 Pressure (Pa) Normalized MFB 2.5 x 6 5 RPM Dual Fuel - All Gasoline % DI 5% DI 7% DI % DI Figure 4: Pressure and MFB at 5 RPM (case A) Figure 5 shows the calculated burn duration from to percent MFB for the engine operated at 5 RPM with 3.3 bar IMEP, while 25 and 3 RPM were at WOT. The solid line with circles is for 5 RPM with 3.3 bar IMEP, dashed line with diamonds is for engine operated at 25 RPM with WOT, and the dash-dotted line with stars is for 3 RPM with WOT. As mentioned before, the spark timing at 5 RPM with 3.3 bar IMEP was adjusted to the engine MBT timing, while at 25 and 3 RPM the spark timing was fixed at 2 DBTDC that is very close to its MBT timing. As discussed before, the to percent burn duration increases in all three cases as the DI fueling percentage increases. As a summary of this section, we concluded that for case A (gasoline for both PFI and DI fuel systems), both engine IMEP and peak cylinder pressure decrease as the DI fueling percentage increases. In addition, the IMEP reduction becomes severe at WOT compared to part load conditions due to the heavier charge cooling effect at WOT than at light load. Similarly, to percent burn duration increases as the DI fueling percentage increases bar IMEP and WOT for Gasoline PFI and DI CASE B: GASOLINE PFI AND ETHANOL DI In this section, we discuss the combustion characteristics of the dual-fuel single-cylinder engine when gasoline was used for PFI fuel system and E was used for DI fuel system. Similar to all gasoline case described in the previous section, for each test point, engine throttle and speed were held at constant. Engine tests started at percentage PFI gasoline fueling with a given AFR, and then, the PFI fueling was reduced to a desired level while the corresponding DI E fueling percentage was increased to maintain the same AFR. The tests continued until percent DI E fueling was reached. Similar to case A, five PFI fueling percentages were selected at, 7, 5, 3, and percent. Spark timing sweep was also conducted at 5 RPM, where for a given PFI fueling percentage, MBT timing was selected by adjusting spark timing such that the 5% MFB location remained at about 8 to degrees after TDC. The test matrix for this case is shown in Table 2. For the operational condition at 5 RPM with 3.3 bar IMEP load, the swept spark timing was selected at 34, 37, 4, 43, and 46 DBTDC, where 4 DBTDC is the MBT timing for percent PFI fueling. At 5 RPM with WOT, the spark timing was selected at 4, 7, 2, 23, and 27 DBTDC, where MBT timing is at 2 DBTDC for percent PFI fueling. Again, for each test point, PFI fueling percentage was selected at, 7, 5, 3, and percent. Table 2: Test matrix - PFI gasoline and DI E RPM IMEP λ ST (DBTDC) bar bar. 3 5 WOT WOT WOT bar. Sweep 5 WOT.925 Sweep to % Burn Duration (deg) RPM 3.3bar IMEP 25 RPM WOT 3 RPM WOT Percentage of DI Fueling (%) Figure 5: Burn duration (case A) 4 Tests were also conducted with fixed spark timing at 5 RPM with WOT, 3.3 and 5.5 bar IMEP and at 25 and 3 RPM with WOT (see Table 2 for spark timing), where for each test point, the gasoline PFI fueling was selected to be at, 7, 5, 3, and percent. Figure 6 shows the engine IMEP as a function of DI E fueling percentage when the engine was operated at 5 RPM with 3.3 bar IMEP and WOT. The solid line with circles is the normalized IMEP associated with the engine operated at its MBT spark timing, and the dotted line with diamonds is the engine IMEP at MBT timing with WOT. It is interesting to see that for the engine operated with 3.3 bar IMEP, the actual engine IMEP increases as the DI E fueling increases; while when

7 the engine is operated at WOT, its IMEP decreases as the DI E percentage increases. It can be seen that the IMEP with 5% DI percentage is slightly lower than % and the IMEPs with 7% and % DI percentage are higher than % when the engine operates with 3.3bar IMEP. When the engine operates at WOT, the IMEP with 7% DI percentage is higher than the ones with 5% and %. 5 5RPM w/ 3.3bar IMEP and WOT (PFI-Gasoline & DI-E) Normalized Fuel Energy Injected (%) 5 5 5RPM w/ 3.3bar IMEP and WOT (PFI-Gasoline & DI-E) 3.3 bar IMEP WOT bar IMEP WOT Percentage of DI Fueling (%) Figure 6: IMEP vs. DI fueling percentage (case B) To investigate why at different load conditions the engine IMEP varies at opposite directions as the DI ethanol fueling percentage increases, the total fueling energy injected was calculated based upon the fuel injection quantity of both DI and PFI fuel injections. Specifically, the DI fueling percentage was adjusted according to the change of the PFI fueling dynamic flow (injected quantity per pulse) on a mass basis. Then, the injection duration for each injector was determined from the injector calibration curves. Therefore, the total energy injected ( TE Inj ) is the sum of the energy injected of each injector based on the Lower Heating Value (LHV) of the specific fuel and the injected quantity. TE + Inj = LHVgasolinemgasoline LHVEmE, (3) where mgasoline and m E are the injected masses of gasoline and ethanol (E), respectively. Figure 7 shows the calculated injected energies, normalized to the fuel energy with percent PFI fueling, in terms of the DI fuel injection percentage. One can see from this figure that in both cases, the total energy injected increases as the DI fuel injection percentage increases. But at light load (3.3 bar IMEP) the increment is more significant (9%) than that at WOT (2.6%). Percentage of DI Fueling (%) Figure 7: Fuel energy injected vs. DI fueling percentage Consider two main factors that affect the engine combustion efficiency: DI charge cooling and injected fuel energy. It becomes clear that at light load condition (3.3 bar IMEP), the increment of the injected energy (9%) due to increasing DI fueling percentage dominates the combustion process; while at the heavy load condition (WOT), the charge cooling effect becomes a key factor since the fueling energy increment is moderate at 2.6%. This indicates that for a dual-fuel (gasoline PFI and ethanol DI) system engine to improve the combustion efficiency at high DI injection percentage, it may be important to adjust the engine compression ratio for the best combustion efficiency. Figure 8 shows the similar information compared to Figure 6 for fixed spark timing tests at 5, 25 and 3 RPM with different load conditions. The dotted line with circles and solid line with diamonds are the normalized IMEP signals at 5 RPM with 3.3 and 5.5 bar IMEP, respectively; the solid line with star is the IMEP at 5 RPM with WOT; and the dash-dotted line with cross and the dashed line with pluses are the IMEP signals at 25 and 3 RPM with WOT. It can be observed that at 5 RPM with light load (3.3 bar IMEP) the normalized IMEP reduces and then increases as DI fueling percentage increases. As the load increases at 5 RPM, the IMEP reduces while the DI E fueling percentage increases. This effect becomes less severe as the engine speed increases. However, at 3 RPM, this effect reverses. Instead of decreasing IMEP, the IMEP remains relatively unchanged as the DI E percentage increases. This is mainly due to the fact that at high engine speed the high combustion chamber temperature makes DI charge cooling effect less dominated. 5

8 5 8 Fixed Timing Tests (PFI-Gasoline & DI-E) 5RPM 3.3bar IMEP 5RPM 5.5bar IMEP 5RPM WOT 25RPM WOT 3RPM WOT Percentage of DI Fueling (%) Similarly, Figure shows the averaged engine incylinder pressure signals on top graph and MFB signals calculated from averaged in-cylinder pressure signals on bottom graph when the engine was operated at 5 RPM with WOT. The spark timing was adjusted to its MBT timing at each given DI fueling percentage. These pressure and MFB signals are associated with the dotted line IMEP shown in Figure 6. Similarly, only, 5 and percent DI E fueling signals are plotted to make the figure less clouded. The dotted lines are for the pressure and MFB signals with percent DI E fueling; the dashed lines are for 5 percent; and the solid lines are for percent. Figure 8: IMEP vs. DI E fueling % (Fixed ST case B) Figure 9 presents the averaged engine in-cylinder pressure signals on top graph and MFB signals calculated from averaged in-cylinder pressure signals on bottom graph when the engine was operated at 5 RPM with 3.3 bar IMEP. The spark timing was adjusted to its MBT timing at each given DI fueling percentage. Note that these pressure and MFB signals are associated with the solid line in Figure 6. Only, 5 and percent DI E fueling test points are presented in the plot for clarity. The dotted lines are for the pressure and MFB signals with percent DI E fueling; the dashed lines are for 5 percent; and the solid lines are for percent. Pressure (Pa) MFB x 6 5RPM 3.3bar IMEP (PFI-Gasoline & DI-E) % DI 5% DI % DI Figure 9: Pressure and MFB vs. DI % (3.3bar, case B) Note that both dotted and solid lines (pressure and MFB signals) are almost overlaid each other with solid line slightly higher than the dotted line. This indicates that the in-cylinder pressure signal with percent DI E fueling is slightly higher than that of percent DI fueling; and the MFB signal for percent DI E fueling is faster than that of no DI fueling. Recall that Figure 6 indicates that the highest IMEP was reached at percent DI E fueling; and the lowest IMEP was at 5% percent DI fueling. 6 MFB Pressure (Pa) 6 x 6 5RPM WOT (PFI-Gasoline & DI-E) % DI 5% DI % DI Figure : Pressure and MFB vs. DI % (WOT, case B) Even though the MFB curves are very close to each other for all 3 conditions, their shapes are slightly different. For the in-cylinder pressure signals with, 5 and percent DI E fueling, the peak cylinder pressure reduces as the DI E fueling percentage increases. This confirms the fact shown in Figure 6 that the IMEP decreases as the DI E fueling percentage increases. Figure shows the calculated burn duration from to percent MFB for the engine operated at 5 RPM with 3.3 bar IMEP and WOT. The solid line with circles is at 5 RPM with 3.3 bar IMEP, and dashed line with diamonds is at 5 RPM with WOT. As discussed before, the spark timing was adjusted to the engine MBT timing. These test points are associated with the IMEP curves shown in Figure 6. It can be observed that the to percent burn duration increases at both load conditions as the DI E fueling percentage increases. This is also true for the cases of fixed timing shown in Table 2, and therefore, the plot is not presented.

9 Burn Duration (deg) RPM 3.3bar/WOT (PFI-Gasoline & DI-E) 3.3bar IMEP WOT Percentage of DI E Fueling (%) Figure : Burn duration vs. DI E fueling % (case B) The spark timing used for each test point was CT, CT±3, and CT±6. At 5 RPM with WOT, the spark timing sweep was fixed at 4, 7, 2, 23, and 27 DATDC, where engine MBT timing is around 2 DBTDC. Again, for each test point, PFI fueling percentage was selected at, 7, 5, 3, and percent. Table 3: Test matrix PFI E and DI gasoline RPM IMEP λ ST (DBTDC) bar bar. 3 5 WOT WOT WOT bar. Sweep 5 WOT.925 Sweep We can conclude that for the case of gasoline PFI and E DI at WOT, the engine IMEP and peak cylinder pressure decrease as the DI E fueling percentage increases. But at light load, the engine IMEP and peak cylinder pressure decrease first and then increases as the DI E fueling percentage increases. This is mainly due to the fact that at light load, the combustion process is dominated by the increase in total fuel energy injected, while at WOT, the charge cooling effect is more important. In general, the to percent burn duration increases as the DI E fueling percentage increases. CASE C: ETHANOL PFI AND GASOLINE DI In this section, we discuss the combustion characteristics of the dual-fuel single-cylinder engine when ethanol (E) was used for the PFI fuel system and gasoline was used for the DI fuel system. Similar to the two previous cases, for each test point, engine throttle and speed were held at constant; and all tests started at percentage PFI E fueling with a given AFR. Then, PFI fueling was reduced to a desired level while the corresponding DI gasoline fueling was increased to maintain a constant AFR. The test continued until percent DI gasoline fueling was reached. Five PFI fueling percentages were selected and they were, 7, 5, 3, and percent. Spark timing sweep was also conducted at two test points (5 RPM with 3.3 bar IMEP and WOT), where for a given PFI fueling percentage, MBT timing was selected by adjusting spark timing such that the 5% MFB location remains at around 8 to degrees after TDC. The test matrix for this test case is shown in Table 3. Due to the significant MBT timing variations for different DI gasoline fueling percentages at 5 RPM with 3.3 bar IMEP load, the spark timing sweep window was quite different. For each fuel ratio test point, nominal spark timing, called center timing (CT), was selected. For this test, CT was select at 35, 42, 44, and 47 DBTDC corresponding to, 5, 3, and percent PFI fueling. 7 Tests were also conducted with a fixed spark timing at 5 RPM with WOT, 3.3 and 5.5 bar IMEP and at 25 and 3 RPM with WOT, where for each test point, the ethanol PFI fueling was selected to be at, 7, 5, 3, and percent. Figure 2 shows the normalized engine IMEP as a function of DI gasoline fueling percentage when the engine was operated at 5 RPM with 3.3 bar IMEP and WOT. The solid curve with circles is the normalized IMEP associated with the engine operated at its MBT spark timing with 3.3 bar IMEP, and the dotted line with diamonds is the engine percent IMEP with WOT. It can be observed that when the engine was operated at WOT, the engine IMEP reduces by 3 percent at percent DI gasoline compared to that of percent PFI E; while at 3.3 bar IMEP, the engine IMEP decreases as much as percent at percent DI fueling compared to that of percent PFI fueling RPM 3.3bar IMEP & WOT (PFI-E & DI-Gasoline) 3.3bar IMEP WOT Percentage of DI Gasoline Fueling Figure 2: IMEP vs. DI fueling percentage (case C) Figure 3 shows the similar information to that of Figure 2 for the fixed spark timing tests at 5, 25 and

10 3 RPM with different loads. The solid line with circles and dashed line with diamonds are the normalized percent IMEP at 5 RPM with 3.3 and 5.5 bar IMEP, respectively; the dotted line with star is the IMEP at 5 RPM with WOT; and the dash-dotted line with pluses and the solid line with squares are the IMEP signals at 25 and 3 RPM with WOT. It can be observed that in general the normalized percentage IMEP reduces as DI gasoline fueling percentage increases. 5 8 Fixed Timing (PFI-E & DI-Gasoline) 5RPM 3.3bar IMEP 5RPM 5.5bar IMEP 5RPM WOT 25RPM WOT 3RPM WOT Percentage of DI Gasoline Fueling Figure 3: IMEP vs. DI fueling % (fixed timing, case C) For the case of PFI E fueling and DI gasoline fueling, Figure 4 presents the averaged engine in-cylinder pressure signals on top graph and MFB signals calculated from averaged in-cylinder pressure signals on bottom graph when the engine was operated at 5 RPM with 3.3 bar IMEP load. The spark timing was adjusted to its MBT timing at each given DI fueling percentage. Note that this corresponds to the solid line IMEP signal shown in Figure 2. Only these signals associated with, 5 and percent DI gasoline fueling are plotted. The dotted lines are for the pressure and MFB signals with percent DI gasoline fueling; the dashed lines are for 5 percent; and the solid lines are for percent. Pressure (Pa) MFB x 6 5RPM 3.3bar IMEP, PFI: E, DI: Gasoline % DI 5% DI % DI Figure 4: Pressure and MFB signals (3.3bar, case C) It is clear that as the DI percentage of gasoline fueling increases, the combustion slows down and peak cylinder pressure reduces. Figure 5 shows the averaged engine in-cylinder pressure signals on top graph and MFB signals calculated from averaged in-cylinder pressure signals on bottom graph when the engine was operated at 5 RPM with WOT. The engine was operated at its MBT spark timing at each given DI fueling percentage. Note that these operational conditions are associated with the dashed line IMEP shown in Figure 2. For clarity, only these signals associated with, 5 and percent DI E fueling are plotted. The dotted lines are for the pressure and MFB signals with percent DI gasoline fueling; the dashed lines are for 5 percent; and the solid lines are for. MFB Pressure (Pa) 6 x 6 5RPM WOT (PFI-E & DI-Gasoline) % DI 5% DI % DI Figure 5: Pressures and MFB signals (WOT) Figure 5 illustrates that the peak in-cylinder pressure signals decrease as DI gasoline fueling percentage increases; and the MFB signals show that the combustion slows down when the DI fueling percentage increases. This correlates to the IMEP curve presented in Figure 2, that is, the IMEP decreases as the DI gasoline fueling percentage increases. Figure 6 shows the calculated to percent burn duration for the engine operated at 5 RPM with 3.3 bar IMEP load and WOT. The solid line with circles is for 5 RPM with 3.3 bar IMEP; and dashed line with diamonds is for 5 RPM with WOT. As discussed before, the spark timing was adjusted to engine s MBT timing which corresponds to the IMEP curves shown in Figure 2. It can be observed that the to percent burn duration increases significantly at 5 RPM with 3.3 bar IMEP load, but at WOT, the burn duration remains relatively unchanged. This correlates well with the IMEP signal shown in Figure 2 since at WOT, IMEP reduces moderately as the DI gasoline percentage increases. 8

11 to % Burn Duration (deg) RPM 3.3bar & WOT (PFI-E & DI-Gasoline) 3.3bar IMEP WOT Percentage of DI Fueling Figure 6: Burn Duration vs. gasoline DI % (case C) For the case of PFI E fueling and DI gasoline fueling, in general the engine IMEP and peak cylinder pressure signals decrease as the DI gasoline fueling percentage increases. In addition, the reduction of IMEP is more significant at light load than at WOT. The to percent burn duration increases as the DI gasoline fueling percentage increases sharply at light load, but only moderately at WOT. CASE DISCUSSION dominant role. However, at heavy load (WOT), the percentage of DI fueling, regardless of the percentage of ethanol fueling, is the controlling factor RPM 3.3bar IMEP E-PFI/Gasoline-DI Gasoline-PFI/E-DI 5RPM WOT Percentage of DI Fueling Figure 7: IMEP comparison of cases B and C Table 4 summarizes the variations of IMEP and BD (Burn Duration) when DI fueling percentage increases from to percent. In almost all cases, the IMEP reduces except in case B with light load that the IMEP increases. The BD increases more at light load condition than at WOT. In this section, we compare the test results among different test cases. Particular attention is devoted to compare the results between cases B and C. Figure 7 compares how engine IMEP changes as a function of percentage DI fueling. The top graph in this figure is related to the engine operated at 5 RPM with 3.3 bar IMEP load and the bottom one is for the engine operated at 5 RPM with WOT, where the solid lines with circles are for case C (E PFI and gasoline DI); and the dashed lines with diamonds are for case B (gasoline PFI and E DI). It is interesting to see that at WOT condition, the IMEP reduces in a similar way in both cases, but at light load condition (3.3 bar IMEP) the situation is quite different. For case B, the engine IMEP increases slightly as the DI ethanol fueling percentage increases. But for case C the IMEP reduces sharply, as shown in the top graph of Figure 7. As illustrated in Figure 7, the percentage of ethanol fueling increases while the total fuel energy injected increases for a given air-to-fuel ratio. For case B, as DI gasoline fueling percentage increases, the sharp decrement of engine IMEP could be due to both total fueling energy reduction and DI charge cooling effect increment. While for case C, as DI ethanol (E) fueling increases, the fact that engine IMEP remains almost unchanged might be due to the combined effect of the total fueling energy increment (increment of IMEP) and increased DI charge cooling effect (decrement of IMEP). We can conclude that at light load, the percentage of the ethanol (E) plays a Table 4: Summary table Variations from to % DI fueling 5 RPM Case A Case B Case C IMEP (3.3bar) -2% +2% -% IMEP (WOT) N/A -4% -2.5% BD (3.3bar) +2 (deg) +2 (deg) +6 (deg) BD (WOT) N/A + (deg) + (deg) What we learned from the results shown in Table 4 is that at WOT, there is no significant combustion characteristic deviation for both cases B and C. In general, the engine IMEP reduces slightly as the DI fueling percentage increases, while the to percent burn duration remains almost unchanged. Therefore, both dual-fuel system configurations (gasoline- PFI/ethanol-DI and ethanol-pfi/gasoline-di) provide similar combustion characteristics at WOT. In contrary, at light load condition, case B (gasoline-pfi/ethanol-di) results in improved engine IMEP as DI fueling percentage increases and in case C (ethanol- PFI/gasoline-DI), the engine IMEP decreases significantly. Therefore, one needs to pay special attention to the light load operating conditions while designing a mixed dual-fuel system to optimize engine performance. For future work, further experiments will be performed to study the in-cylinder pressure and combustion duration 9

12 along with the emissions data to provide a more comprehensive investigation of the dual-fuel injection combustion process. CONCLUSIONS This paper presents the combustion characteristics of a single-cylinder dual-fuel injection SI engine with different configurations of its dual-fuel system that include gasoline PFI and DI (case A), gasoline PFI and E DI (case B), and E PFI and gasoline DI (case C). For each case, DI fuel percentage was varied from to percent while the engine AFR remained constant. It has been shown that in all cases that the IMEP decreases by as much as % as the DI fuel injection percentage increases, except for case B at light engine load that the IMEP increases by 2%. The combustion burn duration increases significantly as the DI fuel injection percentage increases at light load, but only moderately at WOT. Specifically, at light load conditions, DI fueling increment leads to a sharp decrease (-%) of engine IMEP for case B, but only a slight increase (2%) for case C. Also, test result shows that as the percentage of ethanol fueling increases, the total fuel energy injected increases for a given air-to-fuel ratio. For case B, as DI gasoline fueling percentage increases, significant decrease (%) of engine IMEP was observed due to both the reduction of total fueling energy injected and the increase of the DI charge cooling effect. While for case C, as DI ethanol (E) fueling increases, a relative steady engine IMEP increase (2%) might be caused by the combined effect of the total fueling energy increment (increment of IMEP) and a more pronounced DI charge cooling effect (decrement of IMEP). As a result, the percentage of the ethanol (E) plays a dominant role in the combustion process at light load. However, at heavy load (WOT) conditions, the percentage of DI fueling, regardless of the percentage of ethanol, is the controlling factor. ACKNOWLEDGMENTS The authors gratefully acknowledge the support for this work from the U.S. Department of Energy, National Energy Technology Laboratory, Energy Efficiency and Renewable Energy Division, Samuel Taylor, Project Manager and from the Michigan Economic Development Corporation. REFERENCES. Takuya Ikoma, Shizuo Abe, Tukihiro Sonoda, Hisao Suzuki, Yuichi Suzuki, and Masatoshi Basaki, Development of V liter Engine Adopting New Direct Injection System, SAE Jerry E. Sinor and Brent K. Bailey, Current and Potential Future Performance of Ethanol Fuels, SAE Liguang Li, Zhimin Liu, Huiping Wang, Baoqing Deng, Zongcheng Xiao, Ahengsuo Wang, Changming Gong, and Yan Su, " Combustion and Emissions of Ethanol Fuel (E) in a Small SI Engine," Keshav S. Varde and Christopher P. Clark, A Comparison of Burn Characteristics and Exhaust Emissions From Off- Highway Engines Fueled By E and E, SAE Koichi Nakata and Shintaro Utsumi, The Effect of Ethanol Fuel on a Spark Ignition Engine, SAE Takashi Tsunooka, Yohei Hosokawa, Shintaro Utsumi, Takashi Kawai, and Yukihiro Sonoda, High Concentration Ethanol Effect on SI Engine Cold Startability, SAE Satoshi Taniguchi, Kaori Yoshida, and Yukihiro Tsukasaki, "Feasibility Study of Ethanol Applications to A Direct Injection Gasoline Engine," SAE Paul E. Kapus, Alois Fuerhapter, H. Fuchs, Guenter K. Fraidl, Ethanol Direct Injection on Turbocharged SI Engines - Potential and Challenges, SAE Yan Zhang, Bangquan He, Hui Xie, and Hua Zhao, The Combustion and Emission Characteristics of Ethanol on a Port Fuel Injection HCCI Engine, SAE Leslie Bromberg, Daniel Cohn, and John Heywood, Optimized Fuel Management System for Direct Injection Ethanol Enhancement of Gasoline Engines, US patent application 26/236.. Min Xu, Dave L. Porter, Chao F. Daniel, Gus Panagos, Jim, Winkelman, K. Munir, Soft Spray Formation of a Low-Pressure High-Turbulence Fuel Injector for Direct Injection Gasoline Engines, SAE D.L.S. Hung, Jeff, Mara, Jim Winkelman, Tailoring the Spray Pattern of Multi-hole Fuel Injectors for Gasoline DI Engine Homogeneous-Charge Combustion, ILASS-America, 8 th Annual Conference on Liquid Atomization and Spray System, Irvine, CA, May, Guoming G. Zhu, Chao F. Daniels, and James Winkelman, MBT Timing Detection and its Closed- Loop Control Using In-Cylinder Pressure Signal, SAE SAE J275, Gasoline Fuel Injector Spray Measurements and Characterizations, (27), SAE International, Warrendale, Pennsylvania. 5. David L.S. Hung, Guoming G. Zhu, Jim Winkelman, Tom Stuecken, Harold Schock, Andrew Fedewa, A high speed flow visualization study of fuel spray pattern effect on mixture formation in a low pressure direct injection gasoline engine, SAE

13 CONTACT Guoming (George) Zhu, Ph.D., Mechanical Engineering, Michigan State University, 48 ERC-South, East Lansing, MI 48824, USA. ( ) DEFINITIONS, ACRONYMS, ABBREVIATIONS AFR: Air-to-Fuel Ratio BD: Burn Duration CT: Center Timing DBTDC: Degrees Before Top Dead Center DI: Direct-Injection HCCI: Homogenous Charge Compression Ignition IC: Internal Combustion IMEP: Indicated Mean Effective Pressure E: Fuel blended with 5% gasoline and % ethanol by volume EGR: Exhaust-Gas-Recirculation EOI: End of Injection LHV: Lower Heating Value LPDI: Low Pressure Direct Injection MBT: Minimal advance for the Best Torque MFB: Mass Fraction Burned PCP: Peak Cylinder Pressure PFI: Port Fuel Injection RPM: Revolution per Minute SI: Spark Ignited SMD: Sauter Mean Diameter ST: Spark Timing TDC: Top Dead Center UEGO: Universal Exhaust Gas Oxygen WOT: Wide Open Throttle