1 Performance and Emissions of the 1999 LS1 Engine Edward Froehlich Eric Tribbett Lex Bayer Mechanical Engineering Department Stanford University
2 2 ABSTRACT In this study we examine the performance and emissions of the 1999 GM LS1 engine. First we analyze the variation of performance with throttle position, engine speed, equivalence ratio and spark timing. Then we analyze emission trends with changing equivalence ratio and spark timing. Lastly, we discuss the challenges in trying to optimize both performance and emissions simultaneously. Varying manifold pressure, optimal performance was found to be at wide-open throttle. Increasing engine speed also resulted in enhanced performance. An equivalence ratio of 1.14 provided greatest BMEP. Optimal spark timing depended on engine speed. MBT at 2 RPM was found to be roughly 33? BTDC. Emission reduction suggests slightly different optimal settings for equivalence ratio and spark timing. These optimal settings are different for different pollutants. In general, precatalyst emissions are lowest at lean equivalence ratios and less advanced spark timing. The effects of equivalence ratio and spark timing on emissions must be incorporated into engine design in order to meet current pollution regulations. INTRODUCTION One of the greatest challenges faced by the automotive industry has been reducing emissions in order to comply with state and federal regulations. The first emissions requirements were established by the state of California in Since then, regulations have become commonplace worldwide and are continually becoming more stringent. The challenge of meeting these regulations has been especially difficult for high-performance cars. One engine that has met this challenge is the LS1 engine by General Motors. The LS1 is a 5.7L, V8, all aluminum, push-rod engine. It is the latest in a line of engines dating back to Chevrolet s classic cast iron small block LT1. The LS1 was first introduced into the Corvette in This engine meets customer demands while simultaneously satisfying legislative requirements for emissions. This is particularly impressive considering that GM decided not to include EGR for marketing reasons. The LS1 succeeds in providing increased power and torque while delivering better fuel economy. While the data derived from this study is specific to the LS1, it provides us with a feel for the key concepts in optimizing performance. Engine speed, throttle position, equivalence ratio and spark timing are all relevant. While engine speed and throttle are controlled by the driver, equivalence ratio and spark timing are predetermined by the manufacturer. Performance, however, is just one factor that must be taken into consideration in determining proper equivalence ratio and spark timing. Emissions regulations for automobile engines focus on hydrocarbons, carbon monoxide and nitrogen oxides. The abundance of these pollutants is largely a function of the two parameters essential for optimizing performance equivalence ratio and spark advance. Examining the effects of these paramaters allows for the simultaneous optimization of performance and emissions. As emissions regulations continue to tighten, it will become increasingly important to understand how to maximize performance while satisfying emissions standards. The study of the General Motors LS1 engine makes us aware of the challenges automobile engine designers will face in the future.
3 3 APPARATUS AND METHOD In order to measure the performance and emissions of the LS1 engine, it was mounted to a dynamometer and fitted with a series of sensors and controllers. Depicted in Figure 1, the system can be best understood by following the energy flows. The air first passes through a filter and laminar flow element. A laminar flow meter, a pressure sensor, and a thermocouple were used to record air intake conditions. A computer controlled throttle meters air into the intake manifold where a second pressure sensor is located. Computer-controlled injectors dispense fuel into the air flow. After combustion, the burned gases enter the exhaust manifold where a second thermocouple records temperature. Further down the exhaust line, a small sample of the exhaust gases is removed for emissions analysis, after which it is returned to the exhaust stream. Finally, the exhaust gas passes through a three-way catalyst before exiting the laboratory. Crankshaft and camshaft angle encoders were used to feed the computer controller with the necessary information for controlling the spark timing and fuel injection. 1 Intake Manifold Thermocouple Intake Manifold Pressure Sensor FUEL Computer Controller Throttle Spark Injection Dynamometer Controller Air Meter Pressure Sensor Air Meter Thermocouple Air Flow Rate Sensor AIR Exhaust Manifold Thermocouples Camshaft Position Sensor Crankshaft Angle Encoder HC Exhaust Analyzers CO CO 2 O 2 NO Exhaust Fig. 1: Line diagram of the experimental setup for the GM LS1 engine . Refrences denoted by square brackets.
4 4 The sample gases that were removed from the exhaust line were diverted to a series of emissions sensors which measured the concentrations of carbon monoxide (CO), carbon dioxide (CO 2 ), oxygen (O 2 ), nitric oxide (NO), and hydrocarbons (HC). The first four species were measured under dry conditions. The equipment relies on the absorption of light by particles in the exhaust stream. Condensation of water in the analyzer tubes would have skewed the test results, and was therefore removed prior to testing the samples. HC was measured under wet conditions to prevent the hydrocarbons from condensing out along with the water. The dynamometer joined to the LS1 is a 6 horsepower, eddy-current dynamometer capable of maintaining the engine at speeds up to 6 RPM. The engine speed was set using the dynamometer controller. The dynamometer readings measured both torque and power for the LS1 while providing a means for the engine to dissipate energy. For the purpose of this study four sets of data were recorded. Each set examined the effect of one operating variable: manifold pressure (throttle position), engine speed, equivalence ratio, and spark timing. RESULTS AND DISCUSSION PERFORMANCE The first set of performance data was taken varying throttle position. By changing the position of the throttle, the intake manifold pressure was varied from 4 kpa to 1 kpa corresponding to wide-open throttle. Spark timing was adjusted to MBT for each throttle position. Speed was held at 2 RPM and equivalence ratio was held at unity. 1. Volumetric Efficiency Intake Manifold Pressure (kpa) Fig. 2: The relationship between volumetric efficiency and intake manifold pressure. Equivalence ratio of unity; engine speed of 2 RPM; spark timing at MBT. Figure 2 shows volumetric efficiency versus intake manifold pressure. At higher pressure, the air in the intake manifold has greater density. This results in a greater mass of air entering the cylinder per intake stroke. Volumetric efficiency is the ratio of the mass of air
5 5 inducted into the cylinder to the mass of ambient air that would theoretically fill the cylinder. It follows that volumetric efficiency increases with increasing intake manifold pressure. For this set of data the air-fuel ratio was held constant. Since we stay at a stoichiometric air-fuel ratio, more fuel also enters the cylinder BMEP (kpa) 75 5 BMEP 2 1 Brake Power (kw) 25 Brake Power Intake Manifold Pressure (kpa) Fig. 3: The relationship between BMEP and brake power versus intake manifold pressure. Equivalence ratio of unity; engine speed of 2 RPM; spark timing at MBT. Figure 3 shows BMEP and brake power versus intake manifold pressure. BMEP is indicative of the work produced by the engine for a given cycle and tends to increase with intake manifold pressure. This is primarily a result of the increasing volumetric efficiency. At a stoichiometric air-fuel ratio and MBT timing, the combustion is generally limited by the amount of mixture available to burn, not the ability to burn it. More air and fuel are available to burn per cycle at higher manifold pressures, and so the work for a given cycle increases. Brake power is work per unit time produced by the engine. It is the product of work per cycle and the number of cycles per unit time (engine speed). The work per cycle (BMEP) increases linearly and the engine speed is held constant. As a result, the brake power increases linearly as well.
6 6.15 Brake Thermal Efficiency.4 BSFC (mg/j).1.5 BSFC Brake Thermal Efficiency Intake Manifold Pressure (kpa). Fig. 4: The relationship between BSFC and brake thermal efficiency versus intake manifold pressure. Equivalence ratio of unity; engine speed of 2 RPM; spark timing at MBT. Figure 4 shows brake thermal efficiency and brake specific fuel consumption (BSFC) versus intake manifold pressure. Thermal efficiency is lower at lower intake manifold pressures. This is a result of the increased pumping work that must be done to induct fresh charge into the cylinder. Under throttled conditions, the pressure in the intake is pulled below atmospheric as the piston falls during the intake stroke. Since the pressure below the piston (in the crankcase) is at atmospheric, work must be done to induct the fresh charge. At lower intake manifold pressures (more throttled) this work is greater. As a greater percentage of the gross power output goes to pumping and less to net power output, the efficiency drops. BSFC is the inverse of thermal efficiency divided by the lower heating value of the fuel. It is clear from this inverse relationship that as efficiency rises BSFC should fall.
7 7 15 Exhaust Temperature (K) Intake Manifold Pressure (kpa) Fig. 5: The relationship between exhaust temperature and intake manifold pressure. Equivalence ratio of unity; engine speed of 2 RPM; spark timing at MBT. Figure 5 shows exhaust temperature versus intake manifold pressure. Exhaust temperature increases over the range of intake manifold pressures studied. At higher intake manifold pressures more air-fuel mixture is inducted into the cylinder and burned. The energy released by the chemical bonds of the fuel goes into producing work as well as increasing the enthalpy of the exhaust gas. However, at higher intake manifold pressures, there is also more mixture present to absorb the additional energy released. These two factors balance out. The element responsible for the observed increase in the exhaust temperature results from the experimental setup. The thermocouple used to measure exhaust temperature was located approximately ten centimeters along the exhaust manifold. The manifold is cold compared to the exhaust gas. As the gas travels through the manifold it is cooled due to heat loss to the walls. At higher manifold pressures there is a greater mass of exhaust gas per cycle. This greater mass effectively acts as a larger thermal capacitor. This means that the temperature drop from the initial exhaust to the thermocouple is smaller at higher intake manifold pressures. The second set of performance data was taken varying engine speed from 2 to 5 RPM. Spark timing was adjusted to MBT for each speed. Throttle position was held wide open (corresponding to an intake manifold pressure of 1 kpa). Equivalence ratio was set at LBT.
8 8 1. Volumetric Efficiency Engine Speed (RPM) Fig. 6: The relationship between volumetric efficiency and engine speed. Equivalence ratio at LBT; wide-open throttle; spark timing at MBT. Figure 6 shows volumetric efficiency versus engine speed. Volumetric efficiency as a function of engine speed is affected by two major factors. The first of these is inertial effects in the intake and exhaust flows. At high engine speeds, fresh charge is inducted into the cylinder and exhaust gases leave the cylinder with greater velocity. This greater velocity means that the inertia in the flow becomes more important. At the beginning of the intake stroke the air in the intake manifold begins to move. This flow should continue until the pressures in the cylinder and the intake manifold have equilibrated. At high speeds, however, there is enough momentum in the flow that it does not stop instantaneously. The additional air that enters the cylinder increases the in-cylinder pressure above the intake manifold pressure. An analogous effect in the exhaust stream pulls the in-cylinder pressure down below atmospheric pressure after the exhaust stroke. This causes the general increasing trend in volumetric efficiency as a function of engine speed. The second factor influencing volumetric efficiency is the presence of tuning points in the intake manifold (and possibly exhaust as well). The intake manifold is tuned to cause a slight increase in volumetric efficiency around 44 RPM. As the intake valves open, an expansion wave is sent down the intake manifold runner. At the opening of this runner, the wave reflects back as a compression wave. The lengths of the intake manifold runners are designed such that, at a certain engine speed, this compression wave will push more fresh charge into the cylinder, increasing volumetric efficiency around that speed. While the exhaust on our experimental setup was not intentionally tuned, there is an expansion from the manifold into the exhaust pipe as well as a reflection point at the entrance to the catalyst. This may cause a similar tuning effect, increasing volumetric efficiency at some specific engine speed.
9 BMEP BMEP (kpa) 75 5 Brake Power 2 1 Brake Power (kw) Engine Speed (RPM) Fig. 7: The relationship between BMEP and brake power versus engine speed. Equivalence ratio at LBT; wide-open throttle; spark timing at MBT. Figure 7 shows BMEP and brake power versus engine speed. Initially BMEP increases due to increasing volumetric efficiency as it did in the first set of performance data. At high engine speeds, however, BMEP begins to decrease. As volumetric efficiency increases at higher engine speeds, there is more mass of mixture in the cylinder. This causes the in-cylinder pressure to generally be higher. This increases the work that must be done to compress the mixture and overcome friction. Another effect has to do with the amount of time it takes for combustion to take place. At higher engine speeds the piston spends less time at top dead center. Even though the spark timing was adjusted to MBT for each speed, at higher engine speeds, more of the combustion takes place too early or too late in the cycle. This results in decreased power output. At higher engine speeds these effects begin to dominate over the additional fuelair mixture that can be burned, decreasing BMEP. Brake power increases much faster than BMEP over the range studied. As discussed earlier, brake power is work per unit time produced by the engine, and is the product of the work per cycle and the number of cycles per unit time (engine speed). Even though BMEP increases quite slowly and, in fact, begins to fall off, brake power continues to increase due to the increasing engine speed.
10 1.15 Brake Thermal Efficiency.4 BSFC (mg/j).1.5 BSFC Brake Thermal Efficiency Engine Speed (RPM). Fig. 8: The relationship between BSFC and brake thermal efficiency versus engine speed. Equivalence ratio at LBT; wide-open throttle; spark timing at MBT. Figure 8 shows thermal efficiency and brake specific fuel consumption (BSFC) versus engine speed. Neither seems to be correlated with engine speed. BSFC is mass flow rate of fuel divided by the power output. At greater engine speeds the mass flow rate of fuel is higher, but the power output (work per unit time) is higher as well. These two increase at essentially the same rate, so there is little change in BSFC. Brake thermal efficiency is a scaled inversion of BSFC, and so it follows that brake thermal efficiency also does not change with engine speed. 15 Exhaust Temperature (K) Engine Speed (RPM) Fig. 9: The relationship between exhaust temperature and engine speed. Equivalence ratio at LBT; wide-open throttle; spark timing at MBT.
11 11 Figure 9 shows exhaust temperature versus engine speed. Our measurements show that the exhaust temperature increases with engine speed. The dominant factor causing the trend is, most likely, the increased thermal capacitance of the exhaust flow. At higher engine speeds the mass flow rate of exhaust gas through the exhaust manifold is clearly much greater. The heat lost to the manifold walls has less of an effect on the overall temperature of the flow at high flow rates (high engine speed). The third set of performance data was taken varying equivalence ratio from.84 to 1.24 by varying the fuel injection duration. This moderate range was chosen according to the limitations of our exhaust analyzer. Throttle position was held constant such that the intake manifold pressure was 6 kpa. Engine speed was held at 2 RPM and spark timing was set to 32.7? BTDC. 5 4 BMEP (kpa) Equivalence Ratio Fig. 1: The relationship between BMEP and equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. The BMEP decreases as the mixture becomes very lean or very rich. At very low equivalence ratios the available fuel limits the combustion process. At equivalence ratios that are too high the combustion process is limited by the availability of oxygen. The second-order polynomial fitted to the data has a maximum at an equivalence ratio of Maximum performance is expected to be slightly on the rich side of stoichiometric. In real combustion, there is an equilibrium concentration of secondary products, such as carbon monoxide. This means that there is oxygen left in the exhaust past and equivalence ratio of unity. This allows additional fuel to be added and oxidized, increasing power and BMEP.
12 BSFC (mg/j).1.5 Brake Thermal Efficiency BSFC Brake Thermal Efficiency Equivalence Ratio Fig. 11: The relationship between BSFC and brake thermal efficiency versus equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. Figure 11 shows that thermal efficiency reaches a maximum, and BSFC reaches a minimum, just slightly lean of stoichiometric. Thermal efficiency is work-out divided by energy-in. In this case the energy-in is the product of the mass of fuel and the lower heating value. As the mixture becomes richer a greater amount of the fuel does not burn. The energy contained in the bonds of this fuel is counted in the energy-in term but does not contribute to the work-out since it does not burn thus the thermal efficiency decreases. A similar effect is encountered on the lean side. Less fuel is burned at lean conditions and the temperature is generally lower. The lower temperature does not allow all of the fuel to burn. Once again, a portion of the fuel that is counted in the energy-in term does not contribute to the work-out. These two factors cause there to be a maximum value of thermal efficiency at an intermediate equivalence ratio. This maximum is achieved just slightly lean of stoichiometric. Since BSFC is inversely proportional to thermal efficiency, it reaches its minimum at the same point thermal efficiency reaches its maximum just lean of stoichiometric. The fact that the critical point occurs slightly on the lean side of stoichiometric can be explained by examining the effects of equivalence ratio on the specific heat ratio (k=c p /C v ). At richer mixtures k tends to be smaller due to the higher percentage of large fuel molecules in the mixture. If the expansion process is modeled as isentropic, the following equations hold: k-1 T 2 V 1 ; W=mC T 1 V v (T 1 -T 2 ). 2 From these relationships we see that increasing k (lower equivalence ratio) causes T 2 to decrease and work to increase, thus affecting the thermal efficiency and giving us a critical point slightly lean of stoichiometric..
13 13 1. Volumetric Efficiency Equivalence Ratio Fig. 12: The relationship between volumetric efficiency and equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. It is evident that there is no significant correlation between volumetric efficiency and equivalence ratio. We would expect to find a very weak increase in volumetric efficiency as one moves to values of the equivalence ratio that are either very rich or very lean. At very lean conditions there is little fuel in the mixture. This means that the cylinder takes in a greater amount of air for a given amount of mixture. At very high equivalence ratios the effects of charge cooling may be significant. This decrease in temperature increases the density of the mixture, thus increasing volumetric efficiency. The fact that volumetric efficiency does not notably change with equivalence ratio indicates that these effects are too weak to be detected. Thus for practical purposes we can regard volumetric efficiency as constant with varying equivalence ratio.
14 14 15 Exhaust Temperature (K) Equivalence Ratio Fig. 13: The relationship between exhaust temperature and equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. Figure 13 indicates that there is a weak trend between exhaust temperature and equivalence ratio. The exhaust temperature decreases toward the rich side as excess fuel remains unburned and cools the charge. At lean conditions, there is less combustion than occurs nearer stoichiometric, and thus the temperature of the mixture is not as high. The maximum exhaust temperature occurs at an intermediate equivalence ratio. While the above-mentioned trends are apparent in Fig. 13, we can see that the effects are only marginal over the range studied.
15 15 For the last set of performance data the spark advance was varied between 17? and 43? BTDC. The engine speed was held at 2 RPM and the equivalence ratio set at unity. The throttle was held in a position so as to create an intake manifold pressure of 6 kpa. 5 4 BMEP (kpa) Spark Timing ( BTDC) Fig. 14: The relationship between BMEP and spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. The results show that BMEP tends to increase with spark advance between 17? and 43? BTDC. It is expected that BMEP should increase with spark advance to a point, and then drop off. Best performance will be achieved when the greatest portion of the combustion takes place just after top dead center. If the spark is not advanced enough, the piston will already be moving down when much of the combustion takes place. In this case we loose the ability to expand this portion of the gas through the full range, decreasing performance. If the spark is too advanced, too much of the gas will burn while the piston is still rising. The work that must be done to compress this gas will decrease the net work produced. These competing effects cause there to be a maximum in the BMEP as a function of spark advance. The second-order polynomial fitted to the data has a maximum BMEP at an ignition timing of 38.4? BTDC. Minimum advance for best torque (MBT) is defined as the smallest advance that achieves 99% of the maximum power. Using the equation of the trend line, we obtained a value of 34.2? BTDC for MBT under these conditions. It should be noted that MBT will vary with both throttle position and engine speed. Under more throttled conditions, the density of charge in the cylinder will be lower. Since the flame propagates more slowly in less dense mixtures, a larger spark advance will be required. As the piston moves faster at greater engine speeds, the flame must be given a larger head start in order for most of the combustion to occur near top dead center.
16 BSFC (mg/j).1.5 Brake Thermal Efficiency BSFC Brake Thermal Efficiency Spark Timing ( BTDC). Fig. 15: The relationship between BSFC and brake thermal efficiency versus spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. Figure 15 shows that brake thermal efficiency and BSFC tend to improve over the range studied. Thermal efficiency is work-out divided by energy-in. In this case the energy-in is the product of the mass of fuel and the lower heating value which was held constant. The work-out increases proportionally to the BMEP increase. Thus the thermal efficiency increases. BSFC follows inversely. 1. Volumetric Efficiency Spark Timing ( BTDC) Fig. 16: The relationship between volumetric efficiency and spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. Figure 16 makes it evident that volumetric efficiency is unaffected by spark advance. Volumetric efficiency is the amount of air inducted into the cylinder divided by the theoretical
17 17 amount of ambient air that could occupy the cylinder. We expected that volumetric efficiency would increase slightly with spark advance due to the presence of cooler residual gases in the chamber (see Fig. 17) that would lower the density of the air. This effect, however, is not significant and cannot be seen in our data. Exhaust Temperature (K) BMEP Exhaust Temperature BMEP (kpa) Spark Timing ( BTDC) Fig. 17: The relationship between exhaust temperature and BMEP versus spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. Figure 17 shows that the exhaust temperature decreases between 17? and 43? BTDC. This is because more energy is transformed into work as the spark advance is increased. Since the same amount of energy is released by the combustion, the enthalpy of the exhaust gas must decrease in order to conserve energy. This trade off between work and decreasing exhaust gas enthalpy can be seen by the inverse relationship between exhaust temperature and BMEP in Fig. 17.
18 18 EMISSIONS The first set of emissions data was taken varying equivalence ratio from.84 to Intake manifold pressure was kept at 6 kpa. Engine speed was held at 2 RPM. MBT spark timing was used. 3 6 HC Concentration (PPMC, wet) 2 1 O 2 HC 4 2 O2 Concentration (%, dry) Equivalence Ratio Fig. 18: The relationship between HC and O 2 concentrations versus equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. Figure 18 shows hydrocarbon and oxygen concentrations in the exhaust gas as a function of equivalence ratio. HC emissions are the result of incomplete combustion. This can be caused by an insufficient supply of oxygen or by flame quenching. The flame can be quenched along walls and in crevices in the combustion chamber such as the region between the piston crown and the cylinder wall. At higher equivalence ratios there will be a greater number of fuel molecules in the crevices that escape combustion. This will cause the HC concentration to be higher at richer equivalence ratios. A much greater effect in this case, however, is the availability of oxygen. In mixtures that are rich there is not enough oxygen to combust all of the fuel. This leads to large amounts of unburned hydrocarbons. Figure 18 shows hydrocarbon concentration rising and oxygen concentration dropping at higher equivalence ratios. There also appears to be an increase in hydrocarbon emissions at very lean conditions. At lean conditions, less fuel is burned and the incylinder temperature is generally lower. We speculate that this lower temperature is not sufficient to allow all of the hydrocarbons to be oxidized. The HC concentration would increase dramatically were the equivalence ratio lowered to the point of lean misfire.
19 19 6 CO 2 15 CO and O2 Concentration (%, dry) 4 2 O 2 CO 1 5 CO2 Concentration (%, dry) Equivalence Ratio Fig. 19: The relationship between CO 2, CO, and O 2 concentrations versus equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. Figure 19 shows carbon monoxide, carbon dioxide, and oxygen concentrations in the exhaust versus equivalence ratio. In idealized combustion, hydrocarbons react completely to form CO 2 and water. If there is not enough oxygen available for all of the carbon in the fuel to react to form CO 2, some of the carbon is converted to CO instead. Figure 19 shows that at equivalence ratios less than unity, there is excess oxygen and thus very little CO. CO does increase slightly under lean conditions due to partial burn. The concentration of CO 2 increases with equivalence ratio on the lean side because putting in more fuel means there is more carbon available to convert to CO 2. Rich of stoichiometric there is a shortage of oxygen. This means that not all of the carbon in the fuel can react to form CO 2. As a result the concentration of CO 2 trails off and the concentration of CO increases dramatically.
20 2 5 NO Concentration (PPM, dry) Equivalence Ratio Fig. 2: The relationship between NO concentrations and equivalence ratio. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; spark timing at MBT. Figure 2 shows nitric oxide concentration in the exhaust versus equivalence ratio. This concentration shows a pronounced peak around stoichiometric, trailing off on either side. Nitric oxide formation is a strong function of temperature. A temperature difference of a few hundred Kelvin can cause a difference of a factor of ten in the NO concentration. Lean of stoichiometric there is not enough fuel for there to be much combustion. As a result, the in-cylinder temperature is lower. This causes the NO concentration to drop off dramatically. On the rich side there is significant charge cooling as large fuel molecules absorb much of the thermal energy. Again, the in-cylinder temperature is reduced and NO formation goes down. An additional factor contributing to reduced NO formation at rich conditions is the availability of oxygen. At rich conditions there is less oxygen available and thus it is harder to form products with oxygen in them such as NO.
21 21 The second set of exhaust data was taken varying spark timing from 16.9? to 43.2? BTDC. Intake manifold pressure was fixed at 6 kpa. Engine speed was held at 2 RPM. The equivalence ratio was set at unity. 3 6 HC Concentration (PPMC, wet) 2 1 HC 4 2 O2 Concentration (%, dry) O Spark Timing ( BTDC) Fig. 21: The relationship between HC and O 2 concentrations versus spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. Figure 21 shows HC and O 2 concentrations in the exhaust as a function of spark timing. In the case of variable equivalence ratio, the significant factor in HC emissions was the availability of oxygen and the two lines showed an inverse relationship. In the case of variable spark timing this does not appear to be true. Unburned hydrocarbons in small crevices in the combustion chamber seem to be the more significant contributor here. Advanced spark timing causes higher in-cylinder peak pressures. This higher pressure pushes more of the fuel-air mixture into crevices (most significantly the space between the piston crown and cylinder walls) where the flame is quenched and the mixture is left unburned. Additionally, the temperature late in the cycle, when the mixture comes out of these crevices, is lower at more advanced spark timing (see Fig. 17). The lower temperature means that the hydrocarbons and oxygen do not react. This increases the concentration of unburned hydrocarbons and oxygen in the exhaust.
22 22 6 CO 2 15 CO and O2 Concentration (%, dry) 4 2 O CO2 Concentration (%, dry) CO Spark Timing ( BTDC) Fig. 22: The relationship between CO 2, CO, and O 2 concentrations versus spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. Figure 22 shows that oxygen, carbon monoxide, and carbon dioxide concentrations change very little with spark advance in the range studied. In the previous set of emissions data (variable equivalence ratio), CO concentration increased and CO 2 concentration decreased when there was not enough oxygen. In this case, the equivalence ratio was held at unity, so there was enough oxygen to react most of the carbon to CO 2. Some carbon monoxide does appear in the exhaust due to frozen equilibrium concentrations of O 2, CO and CO 2. We might expect spark advance to affect the relative concentrations of CO and CO 2 in two ways. The higher temperatures achieved at higher spark advance might cause more CO 2 to disassociate and increase the CO concentration. Conversely, the higher pressures achieved at higher spark advance could tend to make the equilibrium favor CO 2 over CO and O 2. The fact that neither of these trends appear in our data would suggest that neither is significant or that the two effects offset each other.
23 23 5 NO Concentration (PPM, dry) Spark Timing ( BTDC) Fig. 23: The relationship between NO concentrations and spark timing. Engine speed of 2 RPM; intake manifold pressure of 6 kpa; equivalence ratio of unity. Figure 23 shows NO concentration in the exhaust gas as a function of spark timing. As discussed earlier NO formation is a strong function of temperature. As the spark is advanced, the in-cylinder peak pressure increases. The ideal gas law tells us that this increase in peak pressure must correspond to an increase in peak temperature. This higher temperature causes the NO concentration to be higher. An additional effect comes from the fact that much of the combustion takes place earlier in the cycle at more advanced spark timing. The products from this early combustion are exposed to high temperatures for a longer period of time. We speculate that this additional time at high temperature does not allow the NO to be destroyed. PERFORMANCE AND EMISSIONS CONSIDERATIONS While we have examined engine performance and emissions separately, in reality the two are coupled. In order to optimize performance we need to consider BMEP and brake thermal efficiency. The equivalence ratio for optimal performance (highest BMEP) was found to be at At this equivalence ratio, HC and CO concentrations are rising and NO emissions are close to their maximums. Thus an equivalence ratio of 1.14 is not optimal for emissions. Brake thermal efficiency is optimized just lean of stoichiometric. At this equivalence ratio, emissions are better, but NO concentrations are still high. Thus a slightly lean equivalence ratio does not provide an ideal solution either. It is difficult to discern an optimal equivalence ratio for all emissions, since the different pollutant species behave differently. In general emissions are better under lean conditions. In reality, however, attaining low emissions and high performance is made much easier by the use of a catalyst. Modern three-way catalysts reduce CO, NO x, and HC emissions considerably. These catalysts require an equivalence ratio that hovers around unity so the best equivalence ratio for emissions reduction is, in fact, determined by the catalyst. While performance is not maximized at an equivalence ratio of unity, the performance is the best possible that, together with a catalyst, will meet modern emissions standards.
24 24 Another parameter that is pertinent to both performance and emissions is spark timing. Thermal efficiency and BMEP reach maximums around 38.4? BTDC for 2 RPM and an intake manifold pressure of 6 kpa. MBT was found to be at 34.2? BTDC. At this spark timing NO and HC emissions are considerably higher than at less advanced spark timing. Despite these increased emissions, however, cars generally run at MBT timing. HC emissions are, instead, reduced by the use of a catalyst, while NO emissions can be reduced with a catalyst or with EGR. It is clear that the use of a catalyst is essential in modern-day engine design. Two common strategies include the use of a three-way catalyst with an equivalence ratio hovering around unity or an oxidation catalyst with a lean equivalence ratio. A further study in emissions concentrations after the catalyst would compliment this study well. CONCLUSION Throttle position, engine speed, equivalence ratio and spark timing were all found to significantly influence performance in the LS1 engine. Performance increased as the throttle was opened, as engine speed increased, at an equivalence ratio of 1.14, and at MBT timing. Spark timing for optimal performance depends on engine speed and was found to be at 32.4? BTDC for 2 RPM. Reducing emissions suggests slightly different optimal settings for equivalence ratio and spark timing. These optimal settings vary for different pollutants. In general, pre-catalyst emissions are lowest at lean equivalence ratios and less advanced spark timing. In order to optimize performance while simultaneously meeting emissions regulations, engines are set to run at MBT timing and an equivalence ratio which hovers around unity. This equivalence ratio allows for the use of a three-way catalyst. ACKNOWLEDGEMENTS This research was performed at the Engine Research Laboratory belonging to the Mechanical Engineering Department at Stanford University. The research was facilitated through the ME13 Internal Combustion class. The authors would like to thank Prof. Chris Edwards, Mr. A.J. Simon and Mr. Nalu Kaahaaina for helpful discussion and assistance during the course of this work. Thank you also to Matthew Svreck for editorial advice. A special thanks also to A.J and Nalu for providing much needed sustenance during lab sessions in the form of gummi, tootsie rolls, and root beer. REFERENCES 1. Heywood, J. B., Internal Combustion Engine Fundamentals, McGraw-Hill, New York, Amann et al, 1997 GM 5.7L LS1 V8 Engine, General Motors Corporation, Cengel and Boles, Thermodynamics, An engineering Approach, McGraw-Hill, Boston, McFarland, J, Old Bottle, New Wine, Popular Mechanics, September 1996