Effects of Various Lubricants and Fuels on Pre ... - ACS Publications

Subscriber access provided by UNIVERSITY OF LEEDS

Article

Effects of Various Lubricants and Fuels on Pre-Ignition in a Turbocharged Direct Injection Spark Ignition Engine Sangki Park, Seungchul Woo, Heechang Oh, and Kihyung Lee Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b01052 • Publication Date (Web): 18 Oct 2017 Downloaded from http://pubs.acs.org on October 24, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Effects of Various Lubricants and Fuels on PreIgnition in a Turbocharged Direct Injection Spark Ignition Engine Sangki Park†, Seungchul Woo†, Heechang Oh‡, Kihyung Lee*† †

Department of Mechanical Engineering, Hanyang University, 1271 Sa1-dong, Sangrok-gu,

Ansan-si, Gyeonggi-do 426-791, Korea ‡

Hyundai Motor Company, 150 Hyundaiyeonguso-ro, Namyang-eup, Hwaseong, Gyeonggido, 18280, Korea

KEYWORDS Low-speed pre-ignition (LSPI); Turbocharger; Direct injection spark ignition (DISI); Knocking intensity; Downsizing

ABSTRACT Recently, downsized turbocharged direct injection engines with high efficiency and power have drawn attention worldwide. However, abnormal combustion can occur when these engines are driven at low speeds with a high load, causing potentially fatal damage to these engines. This phenomenon pre-ignition is known to be caused by lubricant droplets, and the calcium, zinc, and molybdenum components of the lubricant affect the pre-ignition. We performed several experiments to determine the effects of these components, various fuels, and different coolant temperatures on the occurrence of pre-ignition. We also carried out a separate analysis to investigate and compare how each lubricant component affects the frequency and intensity of pre-ignition with each fuel.

1. Introduction 1

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 40

The global automotive market is facing the challenge of reducing emissions and fuel consumption while increasing power. In order to achieve these targets, research institutes and automotive companies the world over have studied various technologies such as electric cars, hybrid systems that combine a combustion engine and an electric motor, and new fuels. However, the most practical and quickly implemented method is engine downsizing technology using a turbocharger system that forcefully supplies air to the cylinder by rotating a turbine using an exhaust system. The introduction of turbochargers that use the highcompressive-ratio technology of direct injection allows downsized engines to generate the same or greater power levels as older, larger engines. One of the merits of downsizing a turbocharged engine is higher power generation compared with naturally aspirated engines with the same displacement volume. In addition, because the target torque is generated at lower speeds than in a naturally aspirated engine of the same power, the exhaust emissions can be reduced. Furthermore, weight reduction through exhaust reduction, friction loss reduction through a reduction in piston surface area, and pumping loss reduction can be achieved. Given these merits, automobile makers worldwide have introduced downsized turbocharger engines. The use of this technology in cars is growing, and this trend is expected to continue.1,2 In a downsized engine, the combustion pressure generated at low speeds is generally on the higher side of the average value. This high combustion pressure can cause ignition earlier than the normal timing. This abnormal phenomenon, called pre-ignition, is combustion that occurs outside the specified ignition timing, and it can instantaneously generate abnormal combustion and combustion pressure, causing fatal damage to the engine. According to Zahdeh et al.3 and Dahnz et al.4, pre-ignition occurs owing to floating lubricant particles. The experimental and theoretical approach was undertaken to understand the phenomenon of preignition and to assess parameters to improve or even eliminate it completely. Amer et al.5 and Chapman et al.6 explained that the octane number and fuel characteristics also affect the preignition in direct injection spark ignition engine. In addition, emissions and fuel effects on pre-ignition need to be considered along with knock in designing optimum fuels for future spark ignition engines. According to Amann et al.7, pre-ignition could be affected by the engine operating conditions. Fuel blends with high levels of aromatics increase the frequency at which low-speed pre-ignition occurs somewhat where as oxygenated fuels and low 2


Page 3 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

aromatic blends reduced the low-speed pre-ignition frequency. Okada et al.8 observed the actual movement of the lubricant toward pre-ignition in a visualized experiment. Among the various possible causes, the lubricant appears to be the most probable direct cause of preignition, and the pre-ignition occurs through the process depicted in Figure 1. In the figure, the fuel injected into the cylinder through the injector accumulates on the cylinder wall, piston crown, or crevice, and the carbon particles that remain after the combustion come into contact with oil to form a mixture. Any mixture that accumulates in the crevice or on the piston crown irregularly falls during combustion and is vaporized and oxidized inside the combustion chamber, eventually auto-igniting. Eventually, pre-ignition occurs, resulting in fatal damage to the engine. Different types of knocking combustion can be observed on some engines. This phenomenon is called super knock, mega knock or extreme knock. It features a much higher pressure amplitude than a normal knock. The super knock is also stochastic and can not be controlled in the same way as a normal knock.9 Generally, pre-ignition can be identified as a type of knocking, which is classified according to the time and cause of its occurrence. When a piston approaches top dead center (TDC) in the normal combustion process (Figure 2), combustion is started by the spark plug, as shown in Figure 2 (a). However, as shown in Figure 2 (b), when combustion is started at the spark plug, the pressure increases at one end owing to the combustion and causes auto-ignition at the other end before the normal flame reaches that point, and thus, knocking occurs. In most cases, this type of knocking can be controlled by controlling the ignition timing. Figure 2 (c) shows the ignition during the compression stroke before the piston reaches TDC, which is known as pre-ignition. Preignition has several causes, self-ignition of a mixture of lubricant and fuel particles inside the combustion chamber or ignition at a hot spot. There also exists post-ignition, which occurs when the piston has passed TDC. Because all these phenomena can seriously damage engines, they must be restricted during combustion. Furthermore, according to Kuboyama et al.10, preignition of lubricant particles can cause self-ignition and induced ignition, and therefore, additional study on the role of lubricants is required. In this study, we focused on lubricants and fuels and measured the frequency and intensity of the pre-ignition. Our goal was to provide the information required to manufacture the next generation of downsized turbocharger engines and lubricants by determining the combustion 3


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 40

characteristics of each condition and the relationships between them.

2. Experimental equipment and methods 2.1 Construction of experimental apparatus In order to accurately analyze the experimental data, we used an engine dynamometer to control the power and speed of the test engine, a combustion analyzer to collect data from the combustion pressure sensor and measure the combustion pressure, an ECU controller to maintain constant experimental conditions, a thermo-hygrometer to measure the humidity and temperature, and an exhaust gas analyzer to measure the exhaust products. We also conducted engine break in work to ensure that the engines performed consistently. After constructing the equipment required for our experiments, we selected our conditions. The frequency and intensity of pre-ignition differed depending on the lubricant additives and fuel characteristics. As the residual lubricant, fuel, or carbon particles inside an engine after an experiment could affect future experimental results obtained with the same engine, it was important to minimize this interference using pre-conditioning between experimental stages. After selecting the experimental conditions and methods, we conducted the main experiment after a validation process. The test engine was a 1.6-L engine with a turbocharger and was developed to replace an existing 2.4-L engine. The specifications of the test engine are given in Table 1. We installed various devices on the test engine to perform our main experiment. Table 2 lists the appearance, model name, and resources of the major equipment. The engine dynamometer was an EC-80 model from MEIDEN with a maximum power of 220 kW, rotation limit of 7000 rpm, and absorption torque of 620 Nm. The engine used for our experiment can be driven within that limit range; thus, it can be driven under conditions suitable to the dynamometer. The exhaust gas analyzer was a MEXA-7100DEGR from HORIBA and was used to measure the THC, NOx, CO, and CO2, EGR. The combustion analyzer, which was used to measure and analyze the combustion pressure generated from each cylinder, was an INDIMICRO from AVL. We used the measured combustion pressure, timing of the preignition, mass fraction burned, and knocking intensity to determine the pre-ignition intensity and trends. The mass flow meter, a MICROMOTION from EMERSON, was used to measure 4


Page 5 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

the amount of fuel supplied from the fuel tank to the high-pressure pump. We measured the experimental abnormality in the fuel consumption by comparing the amount of fuel supplied to the engine in each experiment. The thermo-hygrometer, an HMT333 from VAISALA, was used to measure the humidity and temperature and maintain the temperature of the fuel–air mixture that flowed into the engine. Generally, in an engine with a turbocharger, the amount of fuel–air mixture that flows into the cylinder is greater than that in a naturally aspirated engine; thus, the injected fuel amount increases proportionally, and the combustion pressure and combustion temperature also rise. Thus, the temperature near the engine also increases, and the temperature of the fuel–air mixture that flows into the cylinder through the air cleaner also increases. Therefore, the fuel–air mixture inside the combustion chamber becomes unstable, which can lead to knocking. To prevent this problem, we installed an intercooler and initiated heat exchange by exposing the intercooler to the ambient air, along with the radiator, to reduce the temperature of the fuel–air mixture inside the combustion chamber and improve the combustion stability. As we performed the experiments indoors, using only atmospheric air as is the case with ordinary cars would impair the efficiency of the heat exchanger. Therefore, we installed a shower system to spray water directly to the intercooler and thereby initiate forced cooling. The oil separator filters oil from the blow-by gas, which re-enters the intake manifold through the bypass. If oil from the gas flows into the cylinder here, it can cause pre-ignition; thus, we used an oil separator to prevent such an occurrence. Figure 3 shows a schematic of the equipment used for the experiment. After installing the equipment as shown in the figure and breaking-in the engine, we conducted the main experiment.

2.2 Selection of experimental method and conditions Before starting the main experiment, we were required to assign a low-speed, high-load zone. Though it is not included as a routine drive condition, we selected the zone that applies the lowest speed and highest load based on the engine source. In this way, we obtained test results from the most extreme zone to identify improvement items without jeopardizing the safety of the engine. Before selecting the test conditions and executing the actual test, we tuned the original performance of the engine. 5


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 40

Figure 4 shows the torque and power graph by engine speed zone for our test engine. In the graph, the test engine continuously generated a maximum torque of 1500 rpm to 4500 rpm. Thus, the slowest high-load condition in the test engine was 1500 rpm with a load of approximately 250 Nm (20 bar). Based on this information, we defined the engine load condition as 1500 rpm/250 Nm.11 For the low-speed pre-ignition test, we selected three parameters: lubricant, fuel, and coolant temperature. Table 3 presents the type, octane number, and distillation temperature of each tested fuel: RON-90, RON-92, and RON-95. RON-90 is the main fuel used in Korea. RON-92 contains additives to improve performance. Although RON-95 has the highest octane number, it also has a large amount of impurities. The distillation temperature is the lowest for RON-92 and highest for RON-95. RON-92 has the highest volatility, and RON-95 has the lowest volatility. We set the coolant temperature at 50 °C or 90 °C for the fuel–air mixture that flows into the cylinder to consider its effect on the combustion stability and to observe the frequency of pre-ignition. Table 4 shows the conditions in the actual experiment. After injecting RON-92 or RON-95 fuel based on each lubricant, we conducted the experiment at coolant temperatures of 50 °C and 90 °C. We did not use RON-90 in the main experiment because we wanted to determine whether RON-92 and RON-95 presented significantly different risks than the currently market-available fuel. After selecting the conditions, we established the experimental procedure for minimizing the interference in the test results from incorrect test procedures and changing lubricants or fuels. To ensure fair test results, we ensured that the experimental procedures were applied equally in each condition. For example, when an experiment is conducted using lubricant A and then using lubricant B, any residual lubricant A inside the engine would deteriorate and fresh lubricant would be mixed. However, this would change the characteristics of the lubricant, which could produce unexpected results and degrade the fairness and reliability of the results. Therefore, we conducted experiments only after cleaning the inside of the injectors by using an ultrasonic cleaner in order to minimize the interference.12 We performed each experiment after performing the process shown in Figure 5. Before starting the main experiment for each condition, the engine was subjected to a warm-up process. In the cold state, the engine has a torque limit that is used to prevent abrupt power 6


Page 7 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

changes and engine damage. Furthermore, because several parameters have performance limits, such as valve timing, a normal experiment can be performed only after a sufficient engine warm-up. The warm-up was followed by pre-conditioning because residual fuel, lubricant, and fuel deposits remained inside the combustion chamber after each experiment. These residues could affect the next experiment and impair the reliability of the study. The pre-conditioning process lasted approximately 15 min at an engine speed of 3000 rpm and engine load of 100 Nm. Therefore, it can be concluded that such experiments will remove some of the lubricating oil and combustion deposits in the combustion chamber. The next stage was the stabilization and accumulation of deposits, which created a suitable environment for pre-ignition. Then, we performed each experiment for approximately 35 min at an engine speed of 1500 rpm and engine load of 250 Nm. Pre-ignition does not occur regularly; thus, we conducted each experiment for a relatively long time in conditions that promote pre-ignition. We planned our experiments using the unit of cycles and not hours. By obtaining the combustion pressure-change data for 25,000 cycles, we were able to observe the frequency of pre-ignition for the chosen duration. We also performed flushing and injector cleaning to minimize interference in the test results.

2.3 Data analysis method To investigate the occurrence, intensity, and frequency of pre-ignition, we examined our data in various ways. We coded the data obtained from the equipment for use in an analysis program. To analyze the occurrence and intensity of the pre-ignition, we investigated the mass fraction burned and knocking strength. The mass fraction burned can give the timing of the combustion. The investigation on the pre-ignition intensity and pre-ignition timing produced the most accurate information regarding the knocking intensity. When combustion occurs, the fuel–air mixture inside the combustion chamber burns in a sequence through flame spreading. The mass fraction burned is normally expressed as 0–100% (or 0–1). Generally, the zone of 0–10% (0–0.1) is called the flame development angle. Within that angle, the zone 0–5% (0–0.05) is considered as the ignition delay period. The flame development angle is defined as the beginning of combustion. The formula for the mass fraction burned is as follows:

7


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

dQ mdu PdV

where dT

(1)

can be substituted for dT in eq 2, from PV mRT. Using the gas state

equation, the mass fraction burned

is defined as the ratio of the generated calories at any

random point to the total heat generated. Therefore,

can be written as follows:

1 !"

where

(2)

(3)

&

%$#& $& '

(4)

& $% '

#& ( $&

is the crank angle at the beginning of the combustion, and

is the crank angle at

the end of the combustion. A functional form often used to represent the mass fraction burned versus crank angle curve is the Wiebe function:

!" 1 ) exp -). where

is the arbitrary crank angle,

combustion, and

/' 1 ∆

2

(5)

is the crank angle at the beginning of the

is the crank angle (from

= 0 till

= 1) for the total combustion

period. The variables ɑ and m are control parameters to match the shape of the curve. In practice, for the mass fraction burned curves, ɑ = 5 and m = 2 are adequate.13 Based on the formula in eq 5, a 5% mass fraction burned can be considered to indicate the start of the combustion. Therefore, in our data, we considered cases in which combustion occurred in advance of the normal ECU timing to be pre-ignition cases, and we compared these cases with the knocking intensity, the value of which indicates the combustion pressure between cycles and can be expressed as follows:

8


Page 9 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

345 6785 ) 6785/1

(6)

We coded these mass fraction burned and knocking intensity formulas using an analysis program to analyze the combustion conditions for each cycle and the differences for the lubricants, fuels, and coolant temperatures during the occurrence of pre-ignition.14

3. Results and discussion 3.1 Pre-experimental results We performed our first experiment to measure pre-ignition by adopting the same procedure and conditions as given for the main experiment while using a pure lubricant made for the commercially available test engine. During this process, we used normal gasoline along with RON-92 and RON-95, which were the experimental conditions, and then measured the preignition frequency for the three fuel types along with the coolant temperature, knocking intensity, and mass fraction burned. Based on the obtained measurement results, we evaluated the differences in pre-ignition occurrence between each fuel and the base lubricant at the coolant temperature of 50 °C or 90 °C. The results of these experiments are the 5% mass fraction burned and knocking intensity for 25,000 cycles. Figure 6 shows the results of the experiments conducted with RON-90 gasoline at coolant temperatures of 50 °C and 90 °C. At 50 °C, almost no pre-ignition occurred. However, several abnormal combustions occurred with changes in knocking intensity at 90 °C. Nonetheless, abnormal combustion that can be regarded as pre-ignition was rare. In the comparison of the 5% mass fraction burned and knocking intensity, the combustion pressure increased abruptly even during the normal combustion timing, which indicated a general knocking phenomenon, and not pre-ignition as expected at a low-speed, high-load condition with the relatively low-octane RON-90 fuel. Figure 7 shows that no abnormal combustion, including pre-ignition, occurred with RON92 gasoline, irrespective of the coolant temperature. Figure 8 shows the pre-ignition graph for RON-95 gasoline. Unlike with the other fuels, the incidence of pre-ignition was relatively high. Furthermore, as the coolant temperature decreased, the frequency of pre-ignition increased. Therefore, our preliminary experiments showed that the various fuels used affected the frequency of pre-ignition differently. 9


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 40

Furthermore, depending on the octane number, spark knocking occurred but pre-ignition did not. For RON-92 and RON-95 gasoline used in the main experiment, other than the preignition, knocking was not found. Therefore, we confirmed that our main experiment could proceed without concern for any abnormal combustion other than pre-ignition.

3.2 Comparison of pre-ignition frequency Throughout the main experiment, we measured the frequency of pre-ignition for 25,000 cycles according to the lubricant, fuel, and coolant temperature. The frequency of pre-ignition provides information regarding the characteristics of pre-ignition affected by the lubricant components, fuel, and coolant temperature. Table 5 shows the lubricant additives used for our main experiment. The additive contents are displayed as relative amounts, with the highest setting of 1. The measurement of pre-ignition is the accumulated value of the start of combustion and higher knocking intensity than normal. However, when pre-ignition occurred within five cycles of an earlier pre-ignition, it was not judged as a new pre-ignition and was excluded from the accumulated count. Figures 9 and 10 are comparison graphs of the pre-ignition according to the type of lubricant at coolant temperatures of 50 °C and 90 °C with RON-92: pre-ignition was restricted irrespective of the coolant temperature and lubricant type. Figures 11 and 12 are pre-ignition occurrence comparison graphs by lubricant type with RON-95 gasoline. Preignition occurred more frequently with all lubricants at 50 °C than at 90 °C. Although the frequency did not exhibit a clear trend by lubricant type, overall, the lubricants that had few pre-ignitions at 50 °C also had few pre-ignitions at 90 °C. Generally, as the temperature inside the combustion chamber decreased, the combustion stability increased, thus preventing abnormal combustion. Several methods can be used to lower the temperature inside the combustion chamber, including controlling the coolant temperature. When we used a general gasoline in this method, we obtained the results shown in Fig. 6. The temperature inside the combustion chamber affected the combustion, thus increasing the knocking resistance and thereby reducing the knocking. However, in the main experiment with RON-92, pre-ignition was rare, which made a direct comparison difficult. With RON-95, we observed a prominently opposite trend. For a more detailed observation, we categorized the knocking intensity distribution for 25,000 cycles of an experiment. 10


Page 11 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Most of the knocking intensity was distributed in the range of 0–10 bar, and the following characteristics were observed in higher zones, as shown Table 6. The pressure zone distribution shows some differences in the combustion pressure during combustion. However, when compared with the 5% mass fraction burned, these differences occurred in the normal combustion zone and could not be regarded as pre-ignition. Therefore, we used these differences as an index for evaluating the combustion stability. A 5% mass fraction that burned within the zone in which the pressure was greater than 20 bar was advanced beyond the actual ignition timing; this was judged to be caused by pre-ignition. First, we observed the knocking intensity at an average of 250–400 per cylinder. When the coolant temperature was 90 °C, the knocking intensity occurred every 700–900 cycles on average. As shown in the two graphs, the high knocking intensity occurred more often at pressures between 10–20 bar when the coolant temperature was 90 °C than when it was 50 °C. On the contrary, in the zone in which the pressure was greater than 20 bar, the opposite trend was observed. In all the zones in which knocking occurred at a coolant temperature of 50 °C, knocking also occurred at a coolant temperature of 90 °C, but the frequency of knocking was considerably less at 90 °C than at 50 °C. We expected a higher knocking intensity at 90 °C, but we observed the opposite trend. The stability or knock-resistance of combustion may be superior with a coolant temperature of 50 °C, but the fuel characteristics significantly increased the frequency of pre-ignition. RON-95 fuel has low vaporization. One representative cause of pre-ignition is, as we explained in our introduction, the combination of fuel and lubricant particles that undergo evaporation and vaporization to become glowing particles that cause abnormal combustion. The relatively low vaporization of RON-95 suggests a high possibility that it will remain in the liquid state, especially at low temperatures. In fact, with RON-95, the timing was delayed, exposing the combustion chamber to the liquid state for a longer period of time and creating more opportunities for the liquid fuel to combine with lubricant particles deposited in the crevice or on the piston crown. Consequently, the pre-ignition frequency rose. Our experiment thus clarified which characteristics affect pre-ignition and allowed us to analyze differences in pre-ignition frequency by lubricant type.

3.3 Characteristics of pre-ignition We performed multiple pre-ignition experiments for various lubricants, fuels, and coolant 11


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 40

temperature conditions, and the results show that the pre-ignition changed with the conditions. Though our experiment could not accurately predict the occurrence of pre-ignition, it did show that the frequency of pre-ignition varied significantly with the variation in conditions. By analyzing each experimental result from various angles, we were also able to compare the characteristics of pre-ignition and the effects of the lubricant additives. As the knocking intensity increased, the frequency of pre-ignition became greater. Figures 13–15 show the knocking intensity comparison graphs versus the ignition timing for the representative conditions with the lubricant types B, C, and I. Each condition has its own frequency of pre-ignition, but some common points are shown. As the ignition timing advanced, the knocking intensity and combustion pressure increased. In addition, pre-ignition generally occurred from -20°. Therefore, we assume that the pre-ignition can be explained as follows: when the lubricant and fuel particles float inside the cylinder during the compressive stroke, the compressive rates of the fuel–air mixture, temperature, and oxygen density increase, thus causing evaporation and oxidation to occur abruptly; the resulting temperature increase in the particles ultimately causes pre-ignition.

3.4 Effect analysis of pre-ignition according to lubricant additive The frequency of pre-ignition is most affected by the fuel and coolant temperature. However, the cause of pre-ignition can largely be attributed to the effect of the lubricant. Various studies have shown that some lubricant additives affect pre-ignition. The representative additives are calcium and molybdenum. When Ca and CaCO3 particles in the lubricant combust, they can be converted into solid CaO particles, which can contact nonevaporated fuel and oxidize, thus generating heat that results in pre-ignition.10 Ca in the lubricant can be a basic cause of pre-ignition, and indeed the frequency of preignition varies with the Ca content of the lubricant. Zn, Mo, and other lubricant additives are known to affect the occurrence of pre-ignition.11 It would be preferable to avoid the addition of elements that can affect pre-ignition to lubricants, but each additive also has a unique positive role. As other side effects can occur if the additives are arbitrarily reduced or removed to prevent pre-ignition, an optimum composition ratio is required, which in turn requires an understanding of the effect of each component on pre-ignition. Therefore, we investigated these effects (shown in Table 5). For the analysis, we used a signal-to-noise ratio 12


Page 13 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

based on the loss function in the experimental design. Figure 16 compares the degree to which each component affects pre-ignition. As has been demonstrated through other experiments, Ca, Zn, and Mo significantly affect pre-ignition. In addition, boron, phosphorus, and magnesium indirectly affect pre-ignition. Given the limits of our experimental rounds, we could not conduct a completely accurate comparison. If the effect could be numerically analyzed through additional experiments, we could suggest suitable contents for lubricants.

4. Conclusions The abnormal combustion that occurs in a downsized turbocharged engine at low-speed, high-load conditions is called pre-ignition. To analyze the lubricant composition (already known to be a major cause of pre-ignition), we performed an engine experiment using various types of lubricants with various additives. We also varied the fuel and coolant temperature to investigate their effects on pre-ignition. Our conclusions are summarized as follows.

(1) We compared the pre-ignition characteristics by fuel type with the performance of a pure lubricant. We used three market-available fuels for our experiment: RON-90, RON-92, and RON-95. With the normal fuel (RON-90), we observed abnormal combustion but no pre-ignition. With RON-92, abnormal combustion was not observed. With RON-95, we did observe pre-ignition particularly at a low coolant temperature and possibly from impaired volatilization that allowed fuel particles to combine with the lubricant particles. (2) Our main pre-ignition experiment was conducted with RON-92 and RON-95 at various coolant temperatures and with various lubricants. With RON-92, pre-ignition did not occur with most of the lubricants, and when it did occur, the frequency was extremely low. With RON-95, several incidents of pre-ignition occurred, more often with a low coolant temperature, just as with the pure lubricant. Furthermore, we found a large variation in pre-ignition frequency depending on the lubricant type. (3) We investigated the knocking intensity at different ignition timings to determine the characteristics of pre-ignition. As the ignition timing advanced, the knocking intensity increased in all the experiments. We assume that the oxidation of the floating particles inside the combustion chamber caused temperature increases and pre-ignition. Therefore, 13


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 40

preventing the occurrence of the blast of floating matter and the fuel–air mixture at that advanced timing should reduce the occurrence of pre-ignition. (4) We analyzed the signal-to-noise ratio to compare the relationship between the lubricant additives and pre-ignition. We found that Ca, Zn, and Mo had the greatest effect on preignition. Further experiments could numerically explain this relationship and predict a suitable composition for lubricant manufacturers. (5) We compared the characteristics of pre-ignition for the various types of lubricants and fuels in a downsized direct injection turbocharged engine. The frequency and restriction of pre-ignition could be more clearly elucidated by adding other parameters to our results in future research.

14


Page 15 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ASSOCIATED CONTENT

Supporting Information AUTHOR INFORMATION

Corresponding Author *

Corresponding author: Department of Mechanical Engineering, Hanyang University, 1271

Sa1-dong, Sangrok-gu, Ansan-si, Gyeonggi-do 426-791, Korea. E-mail: [email protected]

Present Addresses †

Department of Mechanical Engineering, Hanyang University, 1271 Sa1-dong, Sangrok-gu,

Gyeonggi-do 426-791, Korea ‡

Hyundai Motor Company, 150 Hyundaiyeonguso-ro, Namyang-eup, Hwaseong, Gyeonggido, 18280, Korea

Author Contributions Sangki Park†, Seungchul Woo†, Heechang Oh‡, Kihyung Lee*† ACKNOWLEDGMENT 15


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 40

This work was supported by the research fund of Hanyang University (HY-2017-P).

REFERENCES (1)

Wang, Z.; Qi, Y.; He, X.; Wang, J.; Shuai, S.; Law, C. Fuel 2015, 144, 222-227.

(2)

Park, S.; Kim, H.; Lee, C. Energy Fuels 2016, 30, 810-818.

(3)

Zahdeh, A.; Rothenberger, P.; Nguyen, W.; Anbarasu, M.; Schmuck-Soldan, S.;

Schaefer, J.; Goebel, T. SAE Technical Paper 2011-01-0340; Society of Automotive Engineers (SAE) International: Warrendale, PA, 2011. (4)

Dahnz, C.; Han, K.; Spicher, U.; Magar, M.; Schiessl, R.; Maas, U. SAE Int. J.

Engines 2010, 3, 214-224. (5)

Amer, A.; Babiker, H.; Chang, J.; Kalghatgi, G.; Adomeit, P.; Brassat, A.; Güntehr,

M. SAE Technical Paper 2012-01-1634; Society of Automotive Engineers (SAE) International: Warrendale, PA, 2012. (6)

Chapman, E.; Davis, R.; Studzinski, W.; Geng, P. SAE Technical Paper 2014-01-

1226; Society of Automotive Engineers (SAE) International: Warrendale, PA, 2014. (7)

Amann, M.; Mehta, D.; Alger, T. SAE Technical Paper 2011-01-0342; Society of

Automotive Engineers (SAE) International: Warrendale, PA, 2011. (8)

Okada, Y.; Miyashita, S.; Izumi, Y.; Hayakawa, Y. SAE Technical Paper 2014-01-


Dahnz, C.; Spicher, U. International Journal of Engine Research 2010, 11, 485-498. 16


Page 17 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(10)

Kuboyama, T.; Moriyoshi, Y.; Morikawa, K. SAE Technical Paper 2015-01-0761;

Society of Automotive Engineers (SAE) International: Warrendale, PA, 2015. (11)

Hirano, S.; Yamashita, M.; Fujimoto, K.; Kato, K. SAE Technical Paper 2013-01-


Takeuchi, K.; Fujimoto, K.; Hirano, S.; Yamashita, M. SAE Technical Paper 2012-

01-1615; Society of Automotive Engineers (SAE) International: Warrendale, PA, 2012. (13)

Heywood, J. Internal Combustion Engine Fundamentals; McGraw-Hill Education:

New York, 1988. (14)

Yeom, K.; Bae, C. Energy Fuels 2007, 21, 1942-1949.

17


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 40

FIGURE CAPTIONS

Figure 1. Mechanism of low-speed pre-ignition. Figure 2. Characterization of knocking. Figure 3. Schematic of the equipment used for the engine experiment. Figure 4. Performance graph for our 1.6-L turbo-GDI engine. Figure 5. Dynamometer test procedure. Figure 6. Knocking intensity versus 5% mass fraction burned with RON-90 at coolant temperatures of 50 °C and 90 °C.

Figure 7. Knocking intensity versus 5% mass fraction burned with RON-92 at coolant temperatures of 50 °C and 90 °C.


Figure 9. Comparison of pre-ignition events by lubricant type with RON-92 and a coolant temperature of 50 °C.




Figure 13. Knocking intensity with RON-95 and lubricant B. 18


Page 19 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 14. Knocking intensity with RON-95 and lubricant C. Figure 15. Knocking intensity with RON-95 and lubricant I. Figure 16. Effects of lubricant components on pre-ignition.

Figure 1. Mechanism of low-speed pre-ignition.

19


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 40

Figure 2. Characterization of knocking.

20


Page 21 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 3. Schematic of the equipment used for the engine experiment.

21


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 40

Figure 4. Performance graph for our 1.6-L turbo-GDI engine.

22


Page 23 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 5. Dynamometer test procedure.

23


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 40


24


Page 25 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels


25


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 40


26


Page 27 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels


27


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 40


28


Page 29 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels


29


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 40


30


Page 31 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 13. Knocking intensity with RON-95 and lubricant B.

31


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 40

Figure 14. Knocking intensity with RON-95 and lubricant C.

32


Page 33 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 15. Knocking intensity with RON-95 and lubricant I.

33


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 40

Figure 16. Effects of lubricant components on pre-ignition.

34


Page 35 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

TABLES

Table 1. Specifications of the engine. Description

Specification

Engine type

L4 DOHC

Bore × stroke (mm)

77 × 85.44

Displacement volume (cc)

1591

Compression ratio (-)

9.5:1

Ignition order

1-3-4-2

Open timing

ATDC 8° / BTDC 42°

Close timing

ABDC 69° / ABDC 19°

Open timing

BBDC 40° / BBDC 0°

Close timing

ATDC 3° / ATDC 43°

Intake valve

Exhaust valve

35


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 40

Table 2. Experimental equipment for the 1.6-L T-GDI engine. Instrument

Manufacturer

Specifications 220 kW 7000 rpm 620 Nm THC, NOx, CO, CO2, O2, EGR 8 Channel 0.1 CAD 40,000 Cycle

Engine dynamometer

MEIDEN EC-80

Exhaust gas analyzer

HORIBA MEXA-7100DEGR

Combustion analyzer

AVL INDIMICRO

Flow meter

EMERSON MICRO MOTION

MAX. 108 kg/h

Thermo-hygrometer

VAISALA HMT333

−40–120 °C

Intercooler shower system

Self-production

± 5 °C

Fuel chamber

Self-production

Capacity 55 L

Oil separator

Self-production

Capacity 1 L

36


Page 37 of 40

Table 3. Distillation characteristics of the tested fuels. Fuel type RON-90

RON-92

RON-95

RON (-)

89.4

92

94.8

Residual fuel (ml)

1.1

1.3

1.1

196.1

183.6

210.8

151.6

119

167.7

78

80.8

101.6

44.4

46.6

54.8

End distillation

Distillation characteristics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

temperature (°C) 90% distillation temperature (°C) 50% distillation temperature (°C) 10% distillation temperature (°C)

37


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 40

Table 4. Table of experimental conditions. Lubricant type

Fuel type

Coolant temperature [°C] 50

RON-92 90 A–G 50 RON-95 90 50 H and I

RON-95 90

38


Page 39 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 5. Properties of lubricants. Lubricant Grade

A

B

C

D

E

F

G

H

I

5W-30 0W-30 5W-30 0W-20 0W-20 0W-20 0W-20 0W-20 0W-20

Ca (%)

1.00

0.95

0.88

0.96

0.75

0.47

0.60

0.39

0.53

Mg (%)

0.00

0.00

0.31

0.00

0.00

0.61

0.00

1.00

0.72

Zn (%)

0.78

1.00

0.98

0.67

0.80

0.81

0.72

0.60

0.65

P (%)

0.88

1.00

8.84

0.68

0.71

0.71

0.71

0.58

0.63

B (%)

0.00

0.14

1.00

0.30

0.30

0.28

0.00

0.67

0.23

Mo (%)

0.78

0.00

0.00

0.20

0.20

0.20

0.75

0.07

1.00

39


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 40

Table 6. Pre-ignition events with knocking intensity and lubricant type for RON-95. KI (bar) Type and

10–20

20–50

50–100

100–150

150–

50

1186

6

6

22

28

90

3745

2

4

0

7

50

1136

11

6

18

34

90

3810

3

2

4

8

50

1109

4

7

11

49

90

2640

1

0

1

4

50

1073

10

13

20

34

90

3116

2

5

4

6

50

855

12

5

26

31

90

2591

1

0

2

6

50

1391

7

7

8

15

90

3479

2

0

1

1

50

1559

5

3

6

17

90

3189

3

0

1

1

50

1916

5

2

4

4

90

3569

2

0

2

2

50

1607

8

3

2

3

90

3215

0

0

0

0

coolant temperature (°C) A

B

C

D

E

F

G

H

I

40


Effects of Various Lubricants and Fuels on Pre ... - ACS Publications

Recommend Documents