Printer Friendly

An investigation on the ignition characteristics of lubricant component containing fuel droplets using rapid compression and expansion machine.

INTRODUCTION

In recent years, global warming and depletion of oil resources as well as air quality problems have become more serious. For automotive engines, further reduction of C[O.sub.2] emissions and fuel consumption by achieving higher efficiency is rigorously required. With this motivation, the concept of engine downsizing has been the focus of much attention in recent years and has become more widespread.

The concept of downsizing is to achieve higher efficiency by reducing engine displacement and simultaneously making up for the loss of torque/power via turbocharging. Generally, downsized engines adopt gasoline direct injection (GDI) for mitigating knocking by the mixture cooling effect so that the downsizing ratio and compression ratio can be set higher.

With the development of highly turbocharged direct injection gasoline engines, low-speed pre-ignition (LSPI) has been observed more frequently as an abnormal combustion phenomenon. LSPI is known to induce autoignition of decisively stronger intensity compared with normal knocking. Thus, it is regarded as an important technical challenge to reconcile high torque output with sufficient suppression of LSPI under conditions of low engine speed and high boost pressure.

Two typical types of LSPI behavior have been reported and are well known. First, it occurs suddenly under normal steady-state operation. Second, once an abnormal LSPI cycle occurs, subsequent LSPI cycles will follow intermittently and finally combustion will return to normal cycles [1].

For the second type of behavior, a mechanism has been reported and clarified by several researchers. It has been reported based on precise visualization that solid particles generated or exfoliated by pressure oscillations in the first abnormal cycle can subsequently pre-ignite due to self-heating in a normal cycle [2-3]. Artificial triggering of subsequent LSPI by a significant spark timing advance to prove the mechanism has also been reported [2, 3, 4]. Further, the authors confirmed the dependency of subsequent LSPI events on normal operation time [5]. These results confirmed that subsequent LSPI is induced and triggered by continuous operation and the first LSPI cycle.

On the other hand, many possible mechanism hypotheses have been reported for the sudden occurrence of the first LSPI cycle: droplets [1], deposits [4, 6], and blow-by oil re-flux [7], among others. To validate these mechanism hypotheses, fundamental investigations on the ignitability of pre-ignition sources have been conducted and reported [8, 9, 10].

In this work, we focused on lubricant oil containing droplets as a candidate of pre-ignition sources. The mechanism hypothesis examined is shown in Figure 1. The objective of this investigation was to observe and clarify the mechanism causing droplet ignition and subsequent flame propagation. LSPI sensitivity to the base oil or metallic additives as lubricant oil components in particular has been reported [11, 12, 13]. We tried to understand the fundamental mechanism of this sensitivity by separating the effects of each parameter.

EXPERIMENTAL METHODOLOGIES

Constant Volume Chamber (CVC)

Figure 2 shows the experimental CVC apparatus used in this investigation. A vacuum pump and supply lines of [C.sub.2][H.sub.4] (ethylene), [O.sub.2] and [N.sub.2] from cylinders are connected to the chamber. Each gas can be charged under controlled pressure. Pre-combustion of a lean premixed [C.sub.2][H.sub.4]/air mixture can raise the temperature and pressure of the ambient gas with residual [O.sub.2]. The sample, lubricant oil and fuel mixture are directly injected with a single-hole diesel injector into the pre-heated and pre-compressed chamber. Ignition and combustion of the fuel spray were observed and recorded by a high-speed camera (MEMRECAM GX-8, NAC Image Technology) through a glass window.

Although ambient conditions should be equivalent to those of the pre-ignition timing (before compression TDC) in highly turbocharged downsized engines, it was impossible to set exactly the same conditions from the standpoint of safety. Thus, the temperature and [O.sub.2] density, known as the dominant parameters of ignition, were selected to be equivalent to those of actual engines, as shown in Figure 3. These conditions are equivalent to -45/-30/-20/-10 deg. ATDC in the compression stroke under turbocharging. In addition to these basic conditions, a/b/c/d conditions with different O2 density or temperature were set to supplement condition sensitivities.

Ignitability tests of fuel spray with various mixture ratios of lubricant oil and fuel were conducted as shown in Table 1. In the CVC experiments, only commercial oil was tested. Under all conditions, the density in the chamber was set to 12 kg/[m.sup.3]. The pressure difference between the sample and ambient gas was kept constant at 30 MPa. The amount of fuel injected was controlled to 1 mg.

Rapid Compression and Expansion Machine (RCEM)

CVC experiments enabled us to evaluate the ignitability of fuel/oil droplets under pre-heated ambient gas with residual [O.sub.2].

However, the actual pre-ignition event was expected to be ignition of fuel/oil droplets in the fuel/air mixture. For that reason, we decided to use a rapid compression and expansion machine (RCEM) to observe the ignition of fuel/oil droplets and subsequent flame propagation in the fuel/air mixture.

Figure 4 shows a schematic of the RCEM. The basic RCEM specifications are shown in Table 2. This RCEM can simulate the piston motion of a reciprocating engine's compression and expansion strokes with its cam profile. The descending cam (total weight = 180 kg) moves the piston along the cam profile one time. Compression ratio settings can be changed easily by changing the cylinder head position.

A transparent cylinder head made from polished acrylic was used to facilitate visualization of in-cylinder phenomena. The ignition kernels of fuel/oil droplets and subsequent flame propagation were observed using images obtained with the high-speed camera mentioned earlier. Both the visualized images and pressure measurements allow quantitative evaluation of ignition delay times.

For the injection of the oil/fuel mixture, we prepared a special six-hole, solenoid-driven DI injector for gasoline engines. To avoid wetting of the spray stream on the piston crown surface, the injector was mounted at an inclined angle. Targets of the spray streams were also optimized. One stream for observation was designed to penetrate freely in the visualization area without wetting any wall, while the other five streams penetrated toward the outside of the visualization area. Figure 5 shows a schematic of the cylinder head and injector mounting design. Figure 6 shows the geometrical relation between the injector streams and the visualization area.

Table 3 shows the initial conditions and compression ratio settings of the RCEM tests. The difference in the compression ratio settings between methane (120 RON) and the primary reference fuel (PRF, 90-100 RON) was derived from the autoignitability of the premixed fuel/air mixture. Due to the piston speed limitation of the RCEM (800 rpm), there was a risk of compression ignition of the entire mixture. To avoid that, the base conditions were set on the safe side at lower temperature and pressure levels.

Lubricant Oil Properties and Test Conditions

Table 4 shows the properties of commercial oil A that was used for both CVC and RCEM experiments. Table 5 shows the properties of base oil B that was used only for RCEM experiments. In addition to the commercial oil and base oil, the effects of calcium sulphonate and ZnDTP, typical metallic additives, on fuel droplet ignition characteristics were also examined. Fuels for the premixed mixture were methane and PRF. Iso-octane alone was used as the fuel for droplets. Various combinations of ambient conditions and droplet compositions were tested in the CVC and RCEM experiments. Table 6 summarizes all the combinations and the test conditions.

RESULT AND DISCUSSION

Test 1: Market Oil Ignitability in CVC

The ignitability of the commercial oil and fuel mixture with various mixture ratios was first experimentally evaluated in the CVC. The commercial oil or oil/fuel droplets were injected into the pre-heated ambient gas with residual [O.sub.2] in the CVC.

Figure 7 shows examples of the visualized ignition and subsequent combustion of oil/fuel mixture droplets under a condition equivalent to -10 deg. ATDC of a downsized engine. The results for only iso-octane (oil ratio = 0%) were excluded because no ignition was observed at all. The ignition and combustion of oil containing droplets produced a bright red light, which became brighter with a higher oil ratio. Presumably, it was caused by soot radiation or flame reaction of metallic detergents in the lubricant oil.

Table 7 shows the obtained ignition delay times. A higher oil ratio led to a shorter ignition delay. The absolute values of the extremely short ignition delay times are thought to be relevant for the pre-ignition source of LSPI. Assuming 1600 rpm as a typical engine speed for the occurrence of LSPI, a short ignition delay time of around 1.0 ms is equivalent to roughly 10 deg. CA. This is probably short enough for ignition soon after droplets begin flying in the compression stroke [5].

Test 2: Market Oil Ignitability in RCEM

The ignitability of the commercial oil and fuel mixture with various mixture ratios was experimentally evaluated in the RCEM. The oil/ fuel droplets were injected into a premixed methane/air mixture in the compression stroke. Ignitability tests were conducted five times for each condition.

Figure 8 shows the ignition delay times of oil/fuel droplets with various mixture ratios. While no ignition was observed with the lower oil ratio, higher ignition stability and shorter ignition delay times were observed with higher oil ratios.

Figures 9 and 10 show the visualized ignition and subsequent blue flame propagation for different oil ratios. With a higher oil ratio (Figure 10), the initial flame kernels with a luminous red light (presumably ignition of lubricant oil derived components) look wider. The subsequent blue flames of the methane/air premixed mixture also propagated faster with such wider flame kernels.

Test 3: Ignitability of Base Oil only in RCEM

The high ignitability of the commercial oil was confirmed by the results of Tests 1 and 2. However, the commercial oil actually contained so many components. Separation of the sensitivity to these parameters was needed to understand the fundamental mechanism.

In Test 3, the ignitability of the base oil and fuel mixture without any additives was evaluated in the RCEM. The droplets of the base oil and fuel were injected into a premixed methane/air mixture in the compression stroke. Figure 11 shows the obtained ignition delay times of the droplets for various mixture ratios. Compared with the commercial oil results in Figure 8, although different base oils were used, the absolute values and the tendency of the ignition delay times were comparable.

These results suggested that the base oil was the dominant contributor to the ignitability of the commercial oil. In other words, the base oil itself had high enough ignitability without any promotion by other components.

Figures 12 and 13 show the visualized ignition and subsequent blue flame propagation for different mixture ratios of the base oil. Compared with the Test 2 results (Figures 9 and 10), no significant difference was observed in the color and brightness of the flame kernels.

Test 4: Ca Effect with Base Oil in Methane Mixture

Previously, calcium additives have been reported as a promoter of LSPI [11, 12, 13]. The effect of calcium additives on the ignitability of droplets had to be evaluated separately.

In Test 4, the ignitability of the base oil and fuel with various amounts of calcium additives was evaluated in the RCEM. The droplets of base oil and fuel with calcium additives were injected into a methane/air mixture. Figure 14 shows the obtained ignition delay times of the droplets for various amounts of calcium additives. Contrary to expectations, the ignition delay times did not show any significant sensitivity to the calcium additives in the methane/air mixture.

Figures 15 and 16 show the visualized ignition and subsequent blue flame propagation for different amounts of calcium additives. Compared with the results for the base oil only without any calcium additives, shown in Figures 12 and 13, flame kernels with a brighter orange color were observed with the calcium addition. It was probably caused by calcium's own flame reaction as a metallic element.

Test 5: Ca Effect without Base Oil in Methane Mixture

To confirm the less sensitivity seen for the calcium additives in the methane/air mixture, the ignitability of fuel droplets with calcium additives and without the base oil was evaluated in the methane/air mixture in Test 5. No ignition or even flame kernels were observed under all the conditions used.

The results of Tests 4 and 5 revealed that the base oil was an essential and dominant component for the short ignition delay time of the lubricant oil. On the other hand, the ignition delay time did not show any significant sensitivity to the calcium additives so long as the mixture was composed of methane and air.

Test 6: Ca Effect without Base Oil in PRF Mixture

From Tests 2 to 5, methane was consistently used as the fuel for the premixed mixture. However, in actual downsized engines where LSPI events occur, the fuel for the premixed mixture is basically gasoline, which has a higher carbon number than methane. For this reason, we decided to check the sensitivity to calcium additives of the ignitability of fuel droplets without the base oil in a PRF/air mixture in Test 6.

Figure 17 shows the obtained ignition delay times of fuel droplets for various amounts of calcium additives in a PRF (95 RON)/air mixture. Previously, we defined the ignition delay time as the duration from the injection timing to the time for the emergence of the initial flame kernel. However, there were many cases that had the same ignition delay time, but different subsequent combustion behavior in terms of both the time for the emergence of the initial flame and also the initial flame growth speed that contributed to the subsequent combustion speed. Thus, we defined a new ignition delay time as the duration from the injection timing to the time of mass burnt 10%.

The results in Figure 17 clearly show ignition stabilization, shorter ignition delay times and faster flame kernel growth rates with higher amounts of calcium additives. As previously indicated in Table 3, the experimental conditions were preliminarily confirmed to avoid complete autoignition of the PRF/air mixture. It was also confirmed that ignition did not occur in the cases without calcium additives. These results definitely showed that calcium additives played a role in enhancing the droplet ignition and subsequent flame kernel growth in the surrounding premixed PRF/air mixture.

Figure 18 shows the enhanced ignition and combustion indicated by in-cylinder pressure profiles. Figures 19 and 20 show the visualized enhancement of droplet ignition and combustion due to the calcium additives for different amounts of additives. Wider spatial spreading of the flame was clearly observed with higher amounts of calcium additives (Figure 20).

Test 7: Mixture RON Effect (with Ca) in PRF Mixture

In Test 6, by changing the fuel for the premixed mixture from methane to PRF, the enhancing effects of calcium additives on droplet ignition were clearly observed. In other words, marked changes were induced by the differences in the properties of the fuels used for the premixed mixture. To confirm the sensitivities to the chemical properties of the fuels for the premixed mixture, we conducted Test 7 in which the RON values of the PRF/air mixture were intentionally varied.

Figure 21 shows the ignition delay times obtained for various RON values of the premixed mixture with calcium additives of 0.784 wt % in the droplets. The results clearly show ignition stabilization, shorter ignition delay times and faster flame kernel growth rates with lower RON values.

Figure 22 shows the enhanced ignition and combustion indicated by in-cylinder pressure profiles. Figures 23 and 24 show the visualized enhancement of droplet ignition and combustion by changing the RON values from 95 to 90, similar to the sensitivities to the calcium additives observed in Test 6.

Test 8: ZnDTP Effect (with Ca) in PRF Mixture

In addition to calcium additives, ZnDTP, known as an additive combining anti-wear and anti-oxidant functions, has also been reported to be an inhibitor of LSPI f 12-131. We decided to evaluate the sensitivity to zinc additives of the ignitability of droplets in a PRF/air mixture in Test 8.

Figure 25 shows the obtained ignition delay times for various amounts of zinc additives and a constant amount of calcium additive (= 0.784 wt %) in droplets in a premixed PRF (90 RON)/air mixture. The results clearly show ignition destabilization, longer ignition delay times and a slower flame kernel growth rate with higher amounts of zinc additives.

Figure 26 shows the inhibited ignition and combustion indicated by in-cylinder pressure profiles. Figures 27 and 28 show the visualized inhibition of the droplet ignition and combustion for different amounts of zinc additives. Zinc additives clearly show the opposite sensitivities to the calcium additive (Test 6) and to lower RON fuel for the premixed mixture (Test 7).

DISCUSSION

The experimental evaluations of droplet ignitability with various combinations of fuel compositions for the premixed mixture and droplet components revealed the following findings. All these findings account for the sensitivity of LSPI to various parameters, which have been already reported.

* The high ignitability of the lubricant oil was re-confirmed. It is further inferred that the base oil itself has high enough ignitability without any promotion by additives (Tests 1-3).

* No significant effect of calcium additives on promoting droplet ignition in the methane/air mixture was observed (Tests 4-5).

* The effects of calcium additives on promoting droplet ignition were observed in the PRF/air mixture (Test 6).

* The enhancement of droplet ignition with lower RON values of PRF fuels for the premixed mixture was observed in the presence of calcium additives (Test 7).

* The effects of zinc additives on inhibiting droplet ignition were observed with the PRF/air mixture and in the presence of calcium additives (Test 8).

The difference between the methane/air mixture and the PRF/air mixture was unique. Although the compression ratio was set even higher (as shown in Table 3). leading to higher temperature and pressure levels in the compression stroke. calcium additives never showed any promotion effect in the methane/air mixture. though they did in the PRF/air mixture. Presumably. the promotion effects of the calcium additives depended on the properties of the fuels used for the premixed mixtures.

As is well known. methane (C[H.sub.4]) has no low temperature oxidation (LTO) characteristic due to its carbon number because five-membered rings are required for typical LTO characteristics. The oxidation behavior of methane is rather unique. Its oxidation chemistry is initiated by hydrogen abstraction (R1). but recombination (R2 and R3) and the generation of formaldehyde (C[H.sub.2]O) (R4) follow and shift to the chemistry of [C.sub.2] (hydrocarbons with a carbon number of 2). Finally. it transitions to high temperature oxidation (HTO).

C[H.sub.4] + (M) = C[H.sub.3] + H + (M) (R1)

C[H.sub.3] + C[H.sub.3] = [C.sub.2][H.sub.4] + [H.sub.2] (R2)

C[H.sub.3] + C[H.sub.3] = [C.sub.2][H.sub.6] (R3)

C[H.sub.3] + 0 = C[H.sub.2]O + H (R4)

In contrast. PRF fuels have typical LTO characteristics because they have a sufficient carbon number by themselves. The oxidation chemistry of PRF fuels is initiated by the same hydrogen abstraction (R5) and followed by a sequence of 1st (R6. R7. R8) and 2nd oxygen addition reactions (R9. R10. R11). Finally. it transitions to HTO and chain branching reactions leading to autoignition via [H.sub.2][O.sub.2] chemistry.

RH + [O.sub.2] = R + H[O.sub.2] (R5)

R + [O.sub.2] = R[O.sub.2] (R6)

R[O.sub.2] = QOOH (R7)

QOOH = QO + OH (R8)

QOOH + [O.sub.2] = [O.sub.2]QOOH (R9)

[O.sub.2]QOOH = H[O.sub.2]QO + OH (R10)

H[O.sub.2]QO = OQO + OH (R11)

We hypothesized that the promotion effects of calcium additives are valid for the LTO chemistry (gas phase reaction in a premixed mixture) of hydrocarbons with a sufficiently high carbon number.

Based on the observation of the ignition delay times and the visualized images in Test 6 (Figures 17. 19 and 20). the gas phase reaction with the surrounding premixed mixture. rather than initial droplet ignition. is probably more important for the enhancement of droplet ignition and subsequent combustion in the cylinder. In Figure 17, a more significant change was observed in the ignition delay times defined by the time of mass burnt 10% than that defined by the initial flame kernel. This indirectly indicates the importance of the reaction of the premixed mixture over a spatially wide area. Figures 19 and 20 also show spatially wider flame kernels with a higher amount of calcium additives. Finally, the pressure profiles indicated in Figure 18 show heat release under a low temperature condition. Although detailed analysis (calculation of the temperature range and comparison with the visualized flame kernel area in the images) is necessary. we infer that this heat release is LTO heat release of the gas phase reaction.

The Test 7 results showing the sensitivity to the RON values also provided an important insight. As is well known. PRF fuels with lower RON values have stronger LTO characteristics due to the straight chain structure of n-heptane. Although the contribution of the increased laminar flame speed with a higher n-heptane mixture ratio was also expected. the observed promotion effects on droplet ignition and combustion with the premixed mixture were rather marked. This can be understand as the effect derived from LTO characteristics promoted by the calcium additives.

These observations support the hypothesis of "LTO promotion". However. the fundamental mechanism producing the effect of calcium additives on droplet ignition or LSPI has not been clarified yet. Moriyoshi et al. [14] have proposed the following hypothesis about the promotion effects of calcium additives on LSPI:

1. CaC[O.sub.3], which is contained in lubricant oil, is mixed with gasoline fuel and disperses into the combustion chamber.

2. In a combustion cycle, CaC[O.sub.3] undergoes thermal decomposition (endothermic reaction) and generates CaO and C[O.sub.2].

3. CaO and C[O.sub.2] cause inverse reactions (exothermic reactions) in the compression stroke of the next cycle.

4. CaO enhances the ignitability of the surrounding premixed gas due to its own heat release.

According to this hypothesis. CaC[O.sub.3] does not have any influence on ignition without first undergoing thermal decomposition (generation of CaO and C[O.sub.2]). However, in this study, although the lubricant oil and fuel mixture were injected only once in each ignitability test, the effects on droplet ignition were definitely observed. Thus, we can expect that CaC[O.sub.3] has an effect on droplet ignition even without undergoing decomposition.

Additionally, assuming local decomposition of CaC[O.sub.3], the effects of the generated CaO are thought to be thermal activation. If so, it should promote droplet ignition regardless of the fuel properties of the premixed mixtures. However, in this study, the results confirmed less promotion effect in the methane/air mixture and obvious promotion effects in the PRF/air mixture.

Based on the foregoing discussion, we can expect CaC[O.sub.3] itself to have a promotion effect that should be chemical, not thermal or physical. The different behavior observed in the methane/air and the PRF/air mixtures implies the differences between the two fuels, namely their molecular structure and the LTO characteristics, were key factors.

Historically, the catalytic promotion effects of CaC[O.sub.3] and CaO on hydrogen abstraction reactions have been well known in the research fields of thermal decomposition or gasification of coal [15-16]. The hypothesized reactions are shown below (R12-13). However, at present we are not able to judge the applicability of these reactions to the combustion phenomena examined in this research. We need to conduct further fundamental investigations in this area, including an examination of the detailed elementary reactions.

RH + (CaO) = R + H + (CaO) (R12)

RH +(CaC[O.sub.3]) = R + H + (CaC[O.sub.3]) (R13)

It is well known that ZnDTP functions to prevent oxidation of the lubricant oil by changing radicals or peroxides (like QOOH in (R7)) into stable substances, as shown in (R14) [17]. If such a reaction is effective and can be applied in the gas-phase reaction, the observed inhibition effects of zinc additives can be explained. More fundamental investigations are needed concerning the detailed mechanism of zinc additive effects as well as that of calcium additives.

QOOH + ZnDTP = QOH + ZnO + DTP (R14)

CONCLUSION

For the purpose of understanding the fundamental mechanism of LSPI, the ignition characteristics of lubricant component containing fuel droplets were investigated by using a constant volume chamber (CVC) and a rapid compression and expansion machine (RCEM).

Various combinations of fuel compositions for the premixed mixture and droplet components were tested. The results show:

(1). The significantly higher ignitability of the lubricant oil was confirmed. It is high enough and relevant for the pre-ignition source of LSPI.

(2). The high ignitability of the base oil itself, comparable to that of the lubricant oil, was confirmed. High ignitability of the lubricant oil is presumed to be primarily dependent on the base oil.

(3). The effects of calcium additives were confirmed to be ignition stabilization, shortening of ignition delay times and promotion of flame kernel growth in the surrounding fuel/air mixture.

(4). These effects of calcium additives were observed only in the premixed PRF/air mixture, especially with lower RON values, while no noticeable effect was observed in the premixed methane/air mixture.

(5). The effects of zinc additives were confirmed to be ignition destabilization, prolonging of the ignition delay time and inhibiting of flame kernel growth. These effects were observed in the premixed PRF/air mixture and with calcium additives in the droplets.

Based on the sensitivity to these parameters mentioned in (3) (4) and (5), the effects of both calcium and zinc can presumably be explained by the elementary reactions in the low temperature oxidation region.

More thorough investigation of the detailed mechanism of such chemical additives will be needed.

Masaharu Kassai, Taisuke Shiraishi, and Torn Noda Nissan Motor Co., Ltd.

Mamoru Hirabe, Yoshiki Wakabayashi, Jin Kusaka, and Yasuhiro Daisho Waseda University

REFERENCES

[1.] Dahnz, C., Han, K., Spicher, U., Magar, M. et al., "Investigations on Pre-Ignition in Highly Supercharged SI Engines," SAE Int. J. Engines 3(1):214-224, 2010, doi:10.4271/2010-01-0355.

[2.] Palaveev, S. et al. : Simulations and experimental investigations of intermittent pre-ignition series in a turbocharged DISI engine, 4th International Conference on Knocking in Gasoline Engines (Berlin), pp.414-442 (2013)

[3.] Lauer T. et al.: Model Approach for Pre-Ignition Mechanisms, MTZ worldwide, Vol. 75, Issue 1, pp.44-49 (2014)

[4.] Okada, Y., Miyashita, S., Izumi, Y., and Hayakawa, Y., "Study of Low-Speed Pre-Ignition in Boosted Spark Ignition Engine," SAE Int. J. Engines 7(2):584-594, 2014, doi:10.4271/2014-01-1218.

[5.] Kassai, M., Hashimoto, H., Shiraishi, T., Teraji, A. et al., "Mechanism Analysis on LSPI Occurrence in Boosted S. I. Engines," SAE Technical Paper 2015-01-1867, 2015, doi:10.4271/2015-01-1867.

[6.] Kuboyama, T., Moriyoshi, Y, and Morikawa, K., "Visualization and Analysis of LSPI Mechanism Caused by Oil Droplet, Particle and Deposit in Highly Boosted SI Combustion in Low Speed Range," SAE Int. J. Engines 8(2):529-537, 2015, doi:10.4271/2015-01-0761.

[7.] Schunemann E. et al.: Pre-ignition analysis on a turbocharged gasoline engine with direct injection, 4th International Conference on Knocking in Gasoline Engines, pp.380-393 (2013)

[8.] Ito, H. et al.: A Study on Ignition Characteristics of Droplets Containing Lubricant Using a Constant Volume Chamber and Multi-Dimensional CFD Code, JSAE Papers, Vol. 45, No.4, pp.671-676 (2014)

[9.] Echigo, R. et al.: An Investigation on LSPI Mechanism in a Turbocharged Direct Injection Gasoline Engine, JSAE Papers, Vol. 46, No.4, pp.763-768 (2015)

[10.] Ohtomo, M. et al.: Effect of a Single Oil Droplet on Ignition of Gasoline Mixture, Proceeding of JSAE Annual Spring Conference No.53-14, pp.5-10 (2014)

[11.] Zahdeh, A., Rothenberger, P, Nguyen, W., Anbarasu, M. et al., "Fundamental Approach to Investigate Pre-Ignition in Boosted SI Engines," SAE Int. J. Engines 4(1):246-273, 2011, doi:10.4271/2011-010340.

[12.] Takeuchi, K., Fujimoto, K., Hirano, S., and Yamashita, M., "Investigation of Engine Oil Effect on Abnormal Combustion in Turbocharged Direct Injection--Spark Ignition Engines," SAE Int. J. FuelsLubr. 5(3):1017-1024, 2012, doi:10.4271/2012-01-1615.

[13.] Hirano, S., Yamashita, M., Fujimoto, K., and Kato, K., "Investigation of Engine Oil Effect on Abnormal Combustion in Turbocharged Direct Injection--Spark Ignition Engines (Part 2)," SAE Technical Paper 201301-2569, 2013, doi:10.4271/2013-01-2569.

[14.] Moriyoshi, Y. et al.: A study of a Highly Boosted Gasoline Engine in Low Speed Region Aiming at Extension of the Limit of Downsizing Concept(2nd Report: Numerical Analysis of LSPI Mechanism and Auto-ignition Caused by CaO Formation, The 25th Internal combustion engine symposium (Tsukuba, Japan), Speech No.2 (2014)

[15.] Schachter, Y et al.: Calcium-Oxide-Catalyzed Reactions of Hydrocarbons and of Alcohols, Journal of Catalysis, Vol.11, Issue 2, pp. 147-158(1968)

[16.] Ellig, DL et al.: Pyrolysis of Volatile Aromatic Hydrocarbons and n-Heptane over Calcium Oxide and Quartz, Industrial & Engineering Chemistry Process Design and Development, Vol.24, Issue 4, pp.1080-1087 (1985)

[17.] Mortier, R.M. et al: Chemistry and Technology of Lubricants, 3rd Edition, Springer

CONTACT INFORMATION

Masaharu KASSAI

NISSAN MOTOR CO., LTD.

560-2, Okatsukoku, Atsugi-shi

Kanagawa 243-0192, Japan

m-kassai@mail.nissan.co.ip

DEFINITIONS/ABBREVIATIONS

ATDC--After Top Dead Center

CVC--Constant Volume Chamber

DI--Direct Injection

GDI--Gasoline Direct Injection

HTO--Hot Temperature Oxidation

LSPI--Low Speed Pre-Ignition

LTO--Low Temperature Oxidation

PRF--Primary Reference Fuel

RCEM--Rapid Compression and Expansion Machine

RON--Research Octane Number

SI--Spark Ignition

TDC--Top Dead Center

ZnDTP--Zinc Dialkyldithiophosphate

Table 1. Experimental conditions of CVC tests

Mixing Ratio of             [wt %]       0, 20,50. 60. 80. 100
  Lubricant Oil
Pres. Difference            [MPa]                 30
  (fuel [??] ambient)
Amount of Injection          [mg]                 1.0
Ambient Density         [kg/[m.sup.3]]            12
Camera Flame Rate           [fps]                10000

Table 2. Basic specifications of RCEM

Bore           [mm]   125
Stroke         [mm]   140
Displacement   [cc]   1718

Table 3. Experimental conditions of RCEM with different
fuels

Fuel Type                  [-]    Methane     PRF
                                  Mixture   Mixture

Initial Gas Pressure      [kPa]     25        25
Initial Gas Temperature    [K]    383.15    383.15
Compression Ratio          [-]     14:1     10.5:1
Engine Speed              [rpm]     800       800
Equivalence Ratio          [-]      1.0       1.0

Table 4. Properties of commercial oil A

Name                                  [-]         Commercial Oil A
Grade                                 [-]             SM/GF-4
Base Oil Group                        [-]              Gr. m
SAE Viscosity Grade                   [-]              5W-30
Density (15 deg. C)              [kg/[m.sup.3]]        0.8414
Kinetic Viscosity   100 deg. C   [[mm.sup.2]/s]        9.808
                    40 deg. C    [[mm.sup.2]/s]        47.15
Viscosity Index                       [-]               200

Table 5. Properties of base oil B

Name                                  [-]         Base Oil B
Base Oil Group                        [-]           Gr. m
Density (15 deg. C)              [kg/[m.sup.3]]     0.886
Kinetic Viscosity   100 deg. C   [[mm.sup.2]/s]      10.8
                    40 deg. C    [[mm.sup.2]/s]      93.9
Viscosity Index                       [-]             97

Table 6. Experimental conditions of CVC and RCEM:
Combinations of ambient conditions and droplet
composition

Case No.   Method.         Ambient Conditions

                        Fuel Type       Comp. Ratio

[-]          [-]           [-]              [-]

Test 1       CVC         Air Only           --
Test 2      RCEM         Methane          14 : 1
Test 3      RCEM         Methane          14 : 1
Test 4      RCEM         Methane          14 : 1
Test 5      RCEM         Methane          14 : 1
Test 6      RCEM       95 RON (PRF)      10.5 : 1
Test 7      RCEM     90-100 RON (PRF)    10.5 : 1
Test 8      RCEM       90 RON (PRF)      10.5 : 1

Case No.            Droplet Composition

               Fuel       Com. Oil A   Base Oil B

[-]           [wt %]        [wt %1       [wt %]

Test 1                      0-100         N/A
Test 2                       0-8          N/A
Test 3                       N/A          1-4
Test 4       100 RON         N/A           2
Test 5     (Iso-octane)      N/A          N/A
Test 6                       N/A          N/A
Test 7                       N/A          N/A
Test 8                       N/A          N/A

Case No.        Droplet        Droplets
              Composition      Injection
                               Settings

           Calcium   ZnDTP      Timing

[-]        [wt %]    [wt %]   [deg. ATDC]

Test 1       N/A      N/A     See Table 1
Test 2       N/A      N/A         -28
Test 3       N/A      N/A         -28
Test 4       0-2      N/A         -28
Test 5     0-0.39     N/A         -28
Test 6     0-1.57     N/A         -22
Test 7      0.784     N/A         -22
Test 8      0.784    0-0.3        -22

Case No.    Droplets Injection
                 Settings

           Inj. Pres.   Amount

[-]          [MPa]       [mg]

Test 1
Test 2        3.8        0.5
Test 3        3.8        0.5
Test 4        3.8        0.5
Test 5        3.8        0.5
Test 6        3.8        0.5
Test 7        3.8        0.5
Test 8        3.8        0.5

Table 7. Ignition delay times of oil/fuel mixture (CVC)

Ignition                          In-cylinder Condition

                       -45 deg.   -30 deg.   -20 deg.   -10 deg.

Lub. Ratio [%]   20       --        2.1        0.9        0.5
                 50       --        2.3        1.0        0.6
                 60      3.6        1.4        0.6        0.4
                 80      3.5        1.5        0.7        0.4
                 100     3.2        1.0        0.8        0.4

Ignition                  In-cylinder Condition

                        a     b     c     d

Lub. Ratio [%]   20    0.9   1.6   1.3   1.0
                 50    1.1   1.5   1.3   0.9
                 60    0.8   0.9   0.8   0.7
                 80    0.8   1.0   0.8   0.6
                 100   0.7   0.8   0.8   0.6
COPYRIGHT 2016 SAE International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kassai, Masaharu; Shiraishi, Taisuke; Noda, Torn; Hirabe, Mamoru; Wakabayashi, Yoshiki; Kusaka, Jin;
Publication:SAE International Journal of Fuels and Lubricants
Article Type:Report
Date:Nov 1, 2016
Words:5577
Previous Article:Compositional effects of gasoline fuels on combustion, performance and emissions in engine.
Next Article:Advanced fuel formulation approach using blends of paraffinic and oxygenated biofuels: analysis of emission reduction potential in a high efficiency...
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters