Issuu on Google+

Experimental Thermal and Fluid Science 76 (2016) 12–23

Contents lists available at ScienceDirect

Experimental Thermal and Fluid Science journal homepage: www.elsevier.com/locate/etfs

Macroscopic spray characteristics of kerosene and diesel based on two different piezoelectric and solenoid injectors Wenbin Yu, Wenming Yang ⇑, Kunlin Tay, Balaji Mohan, Feiyang Zhao, Yunpeng Zhang Department of Mechanical Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117575, Singapore

a r t i c l e

i n f o

Article history: Received 17 September 2015 Received in revised form 3 March 2016 Accepted 7 March 2016

Keywords: Spray characteristics Kerosene Diesel Piezoelectric injector Constant volume spray chamber

a b s t r a c t In this study, the behaviours like injection momentum and spray characteristics for piezoelectric and solenoid injectors were compared. With piezoelectric injector, the macroscopic spray characteristics of kerosene and diesel were investigated. The characteristic like spray penetration, spray velocity, spray angle and air entrainment were used to understand the fuel spray behaviours. The result shows that the piezoelectric injector has the advantage of faster injection response and needle opening than the conventional solenoid injector, which will enable precise and rapid injection control and must be beneficial to air–fuel mixing before ignition in a real engine. Due to the lower value of viscosity in kerosene, the injection durations for kerosene are longer than diesel. Also longer injection intervals will be needed for kerosene to realize multiple injections. Kerosene has shorter spray penetration and lower spray velocity compared with diesel, while the spray angle of kerosene is larger than diesel. Entrained air mass is a typical trade-off result of comprehensive influence of spray penetration, spray angle and charge density. Kerosene sprays get more entrained air than diesel, which means the usage of kerosene has advantage in fuel–air mixing. Ó 2016 Elsevier Inc. All rights reserved.

1. Introduction After the concepts of low temperature combustion (LTC), premixed charge compression ignition (PCCI) with diesel [1–3] and partially premixed combustion (PPC) with gasoline [4], researchers have proposed to integrate the advantages of gasoline and diesel fuels to get lower emissions while maintaining high thermal efficiency, dual-fuel reactivity controlled compression ignition (RCCI) [5,6] and blend fuel premixed compression ignition [7,8] are typical representatives. Most recently, a concept called wide distillation fuel (WDF) was proposed, which refers to fuels with distillation range from initial boiling point of gasoline to final boiling point of diesel. As one kind of WDF, kerosene is getting more and more researchers’ attention. The measured physical and chemical properties of kerosene indicate that the density, viscosity and cetane number vary between those of diesel and gasoline, while the heat value is higher than both diesel and gasoline fuel. Also, after the Second World War, the idea of using a single military fuel was conceived to simplify the logistic supply chain for petroleum products, which has been called single fuel concept (SFC). Kerosene, which had been used as an aviation fuel, has been considered as the SFC. Actually, quite a number of researchers have ⇑ Corresponding author. http://dx.doi.org/10.1016/j.expthermflusci.2016.03.008 0894-1777/Ó 2016 Elsevier Inc. All rights reserved.

investigated the effect of kerosene in diesel engines [9–13], and it was found that kerosene, as a substitution for diesel, showed no critical limitation [9–13]. However, most studies performed so far have focused on emission measurements, which give little information about macroscopic spray characteristics for kerosene, especially by a piezoelectric injector. In this paper, based on a piezoelectric injector, the spray characteristics of kerosene are studied and compared to the conventional diesel fuel. The characteristic like spray penetration, spray velocity, spray angle and air entrainment were used to understand the fuel spray behaviours.

2. Experimental setup Fig. 1 shows experimental set up, including constant volume spray chamber (CVSV), fuel supply and injection system, visualization system and an intelligent control system. In order to reproduce the condition in the engine combustion chamber, very high ambient density in constant volume spray chamber can be realized for spray characteristics study. For spray dynamic measurement, the constant volume spray chamber is developed to allow optical access to the whole injected fuel sprays. The ambient pressure can be up to 6 MPa under room temperature and inert nitrogen gas was used to pressurize the chamber. By changing the ambient


W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

13

Fig. 1. Schematic of spray visualization experimental setup.

pressure, the charge density inside the chamber can be varied in accordance with the will. The chamber has a volume of 6 L. The diameter and the thickness of the optically polished quartz glass window are 150 mm and 60 mm. A built-in-house control system for spray test was established, which is with the embedded FPGA real-time technology. Based on the control system, the excitation current and voltage can be flexibly and accurately adjusted to adapt most kinds of injectors (driven by solenoid or piezoelectric valve) and high pressure fuel pumps to realize adjustable injection timing, injection duration, injection rate, common rail pressures and multiple injections. A multifunctional data acquisition system was also established to collect experiment data such as transient and averaged spray momentum, injection rate, fuel pressure, fuel injection quantity, injection duration, and energizing current profile. The fuel supply and injection system consists of fuel injectors, common rail, high pressure fuel pump, electric motor for driving fuel pump and high pressure fuel lines. The highest common rail pressure can be adjusted up to 200 MPa. In this study, a piezoelectric injector was used for spray study. And also the comparison between piezoelectric injector and conventional solenoid injector

Table 1 Specification of injectors. Energizing type

Solenoid

Piezoelectric

Number of holes Angle of the spray (°) Orifice diameter (lm)

7 154 ± 5 138 ± 5

8 154 ± 5 138 ± 5

was carried out. The specifications of the injectors are shown in Table 1. A Photron Fastcam SA5 high speed camera was used for spray visualization. The camera sensor is 12 bit monochrome with a spatial resolution of 20 lm pixel with a minimum exposure time of 1 ls. The images are captured at 20,000 frames per second with a maximum spatial resolution of 832  448 pixels and temporal resolution of 50 ls. A Nikon lens (Nikkor AF 28-85 mm f/1:3:54:5) is installed in the high speed camera with C-mount adapter. A 400 W Hydrargyrum medium-arc iodide (HMI) lamp is used to provide illumination and also a high speed electronic ballast is equipped to limit current and voltage. In order to quantify the macroscopic spray characteristic like spray penetration, spray velocity and spray angle, a post processing for the raw spray image was conducted in this study. The processing of spray image is similar to the one which authors used before [14–16]. Fig. 2(a) and (b) shows the processing of spray image. Firstly, the background was subtracted from the original spray images to eliminate some uniform illumination and isolate the spray. Then, the isolated images are coloured for further processing. After that, the isolated and coloured spray images are subjected to thresholding. Finally, the while/black coloured thresholding images are used for edge detection. Similar method was used by Delacourt et al. [17]. The spray momentum measurement was similar to Payri et al. [18], and the method is shown in Fig. 3. Based on principle of conservation of momentum, the momentum flux from injector nozzle can be obtained from the force which is measured by force sensor (Kistler 9207). The force sensor is piezoelectric type capable of measuring force between ±50 N with sensitivity of <±0.05 N. On


14

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12â&#x20AC;&#x201C;23

(a) Steps of image processing

(b) Flowchart of image processing Fig. 2. Processing of spray image.


15

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

Trigger Signal (V)

6

Piezo Injector Solenoid Injector

4 2 0 Piezo Injector Solenoid Injector

Current (A)

30 20 10 0 -10

Piezo Injector Solenoid Injector

Fig. 3. Schematic of spray momentum measurement.

the top of the sensor, a small but big enough round block is installed to suffer spray impingement in order to collect the force signal. The force sensor is installed very close to the nozzle of injector and the range between the top of force sensor and the nozzle exit can be adjusted from 0 to 12 mm along the spray axial direction. Besides, a Kistler multi-channel charge amplifier is used to transfer the electric charge signal from force sensor to the analog signal, which will be automatically collected and post processed by the data acquisition system. In this study, ultra-low sulphur diesel (ULSD) and ultra-low sulphur kerosene (ULSK) were used for experiments. All the pure diesel and kerosene fuels were obtained from Singapore. Fuel properties such as density, kinematic viscosity and flash point were listed in Table 2. 3. Results and discussion 3.1. Behaviours comparison between piezoelectric and solenoid injectors As the key part of a combustion system in an engine, fuel injection equipment (FIE) is always keeping improving to realize lower engine exhaust emissions and higher thermal efficiency. Fig. 4 shows the trigger signal, energizing current and spray momentum of two injectors with different energizing mode: piezoelectricdriven and solenoid-driven. In this study, the real injection duration is defined as the period when the fuel is delivered through the injector which is indicated by the injection momentum. In addition, the injection delay defines the time interval between the start of energizing and the initial increase in the injection momentum. In the experiment, diesel fuel was used for the behaviour comparison between piezoelectric and solenoid injectors. Due to different energizing mode, the energizing time (ET) was adjusted

Table 2 Fuel properties. Properties

ULSD (diesel)

ULSK (kerosene)

Density at 20 °C (kg/m3) Viscosity at 40 °C (mm2/s) Cetane number Low Calorific value (MJ/kg) Sulphur content (ppm) Flash point (°C)

840 3 52 42.7 10 55

830 1.3 45 45 15 50

Momentum (N)

2 Injection pressure: 100 MPa 60 MPa

1

0.45 ms 0

0.55 ms 0.0

0.5

1.0

1.5

2.0

2.5

3.0

Time (ms) Fig. 4. Performance of piezoelectric and solenoid injectors.

differently to maintain the same injection duration. The ambient temperature was 305 K and the ambient pressure was atmospheric pressure. In the experiment, the injection pressures of 60 and 100 MPa were chosen. From Fig. 4 it is clear that, the piezoelectric injector shows faster injection response and needle opening than the conventional solenoid injector. Firstly, compared with solenoid injector, the piezoelectric injector reduced injection delay from 0.55 ms to 0.45 ms, which will enable precise and rapid injection control. Secondly, the momentum comparison also shows that, in the initial period of spray, the spray momentum from piezoelectric injector is higher than that from solenoid injector, which means the spray from piezoelectric injector grows faster at the beginning of injection. This is mainly because with faster needle lifting in piezoelectric injector, more fuel will be injected out to realize higher spray momentum. And this phenomenon is more obvious when the injection pressure is lower. This must be beneficial to air–fuel mixing process in a real engine. Fig. 5(a) and (b) shows the spray penetration and spray velocity under two injection pressures (60 MPa, 100 MPa) and three charge densities (22 kg/m3, 45 kg/m3, 67 kg/m3) for test injectors. In the experiment the ambient temperature was fixed at 305 K. It is shown in the figure that, under both injection pressures, due to faster needle opening, the spray injected from the piezoelectric injector grows faster at the beginning of injection. When the charge density is lower, due to smaller resistance for initial spray, the difference between two injectors in terms of spray penetration and spray velocity are more obvious. The processed spray evolution for different injectors is shown in Fig. 6. It is clear that the spray penetrations are longer for piezoelectric injector when ASOI (after start of injection) timing is at 0.2 ms, which means the spray penetration grows faster for piezoelectric injector in the initial period of injection. With injection pressures of 60 and 100 MPa, when ASOI timing is at 0.55 ms,


16

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

Injection Pressure: 60MPa

Injection Pressure: 100MPa

50 40

Spray tip penetration (mm)

Spray tip penetration (mm)

60

Injector_Density (kg/m 3) Piezo_22 Piezo_45 Piezo_67 Solenoid_22 Solenoid_46 Solenoid_67

30 20 10

Injector_Density (kg/m3) 60 Piezo_22 Piezo_45 50 Piezo_67 Solenoid_22 Solenoid_46 40 Solenoid_67 30 20 10 0

0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0

1.4

0.2

0.4

Time (ms)

0.6

0.8

1.0

1.2

1.4

Time (ms)

(a) Spray penetration Injection Pressure: 100MPa Injector_Density (kg/m3) Piezo_22 Piezo_45 Piezo_67 Solenoid_22 Solenoid_46 Solenoid_67

Spray Velocity (m/s)

80

60

40

Injection Pressure: 60MPa

100

Injector_Density (kg/m3) Piezo_22 Piezo_45 Piezo_67 Solenoid_22 Solenoid_46 Solenoid_67

80

Spray Velocity (m/s)

100

60

40

20

20

0

0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.0

0.2

0.4

Time (ms)

0.6

0.8

1.0

1.2

1.4

Time (ms)

(b) Spray velocity Fig. 5. Macroscopic spray characteristics for different injectors.

the spray penetration of piezoelectric injector becomes obviously longer than that of the conventional solenoid injector. The difference in spray penetration between two injectors shows larger with the reduction of charge density. 3.2. Spray momentum for test fuels based on the piezoelectric injector In order to better understand spray behaviours of the tested fuels, a distinction has to be made between the stationary and transient part of the injection momentum shape. The spray momentum comparison for kerosene and diesel can be seen in Fig. 7. In Fig. 7(a), with the same energizing time of 1.2 ms, results corresponding to three different injection pressures are presented: 60, 100 and 140 MPa. Fig. 7(b) shows the spray momentum comparison with different energizing times under the same injection pressure of 100 MPa. All the experiments are carried out with the piezoelectric injector. From the figure it is clear that, for each injection pressure and energizing time at stationary condition, kerosene has similar momentum flux to diesel, which is in accord with the expectations. The reason is that the spray momentum at the outlet is only slightly affected by the fluid properties, theoretical analysis from Payri et al. [18] shows that it is only affected by operation conditions and geometrical parameters.

In Fig. 7, the influence of fuel properties on the injector’s needle dynamic action can be observed. As can be seen at the end transient period of the injection momentum for each injection pressure and energizing, the needle closing for kerosene is slower than that for diesel, which means with the same energizing time, kerosene owns longer injection duration than diesel. The injection durations and their difference for kerosene and diesel are shown in Fig. 8. In this investigation, the results corresponding to three injection pressures (60, 100, 140 MPa) and 6 energizing times (0.4, 0.6, 0.8, 1.0, 1.2, 1.4 ms) are presented. As can be seen that the injection durations for kerosene are longer than that for diesel. For each energizing time, the injection duration differences for test fuels can be observed clearly, especially under higher injection pressure. The cause for this phenomenon can be explained as follows. Once the energizing process finishes, the needle descends to its initial position and the viscous force assists the closing of the injector. The higher viscous force the more assistance can be obtained for the closing of the injector. Since the viscosity of kerosene is lower than that of diesel, more time is needed to close the injector needle due to lack of viscous force assistance. Similar investigation result was also reported by Park et al. [19]. As the development of engine technology, in order to realize ideal air–fuel stratified distribution, multiple injections are becoming


17

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12â&#x20AC;&#x201C;23

Injector Charge density (kg/m3)

Piezoelectric 22

45

Solenoid 67

22

45

67

0.2ms ASOI

0.55ms ASOI

Injection pressure: 100MPa

(a) Injection pressure: 100MPa Injector Charge density (kg/m3)

Piezoelectric 22

45

Solenoid 67

22

45

67

0.2ms ASOI

0.55ms ASOI

Injection pressure: 60MPa

(b) Injection pressure: 60MPa Fig. 6. Spray evolution with different injectors under different injection pressures and charge densities.

more and more popular. From the study above, it can be deduced that the effect of fuel properties on injection durations will definitely influence the dwelling time between injections in multiple injections. Fig. 9 shows split injections momentum for kerosene and diesel, in which the injection pressure is 100 MPa and injection interval is 2 ms. It can be seen that, for kerosene, each one in the split injections has longer injection duration time. And in accordance

with expectations, due to longer injection duration, the injection dwelling time for kerosene (L1: 1.13 ms) is shorter than that for diesel (L2: 1.28 ms). As can be deduced that for some fuels with low viscosity, longer injection intervals will be needed to realize multiple injections. The comparison for kerosene and diesel with five pulses injection can be seen in Fig. 10. In the experiment, the injection pressure is fixed


18

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12â&#x20AC;&#x201C;23

Energizing Time: 1.2 ms Diesel Kerosene

3.5

Injection Pressure: 100 MPa Diesel Kerosene

2.8

3.0

2.0

Momentum (N)

Momentum (N)

2.4

Pin: 140MPa

2.5

Pin: 100MPa

1.5 Pin: 60MPa

1.0

1.6 1.2

0.5

1.0

1.5

2.0

2.5

3.0

Enegizing Time: 1.2 ms

1 ms 0.6 ms

0.8

Longer Duration

Longer Duration

0.5 0.0 0.0

2.0

0.4 0.0 0.0

3.5

0.5

1.0

1.5

2.0

Time (ms)

Time (ms)

(a)

(b)

2.5

3.0

3.5

Fig. 7. Spray momentum comparison for kerosene and diesel.

0.6

1.5 0.4 1.0 0.2 0.5

0.4

0.6

0.8

1.0

1.2

(b) 2.0

0.6

1.5 0.4 1.0 0.2

0.5

1.4

0.0 0.4

0.6

0.8

1.0

1.2

1.4

Energizing Time (ms) Injection Pressure:60MPa 1.0 Duration Difference

Injection Duration 3.0 Diesel Kerosene

0.8

2.5

(c)

0.6

1.5 0.4 1.0 0.2 0.5

Duration Difference (ms)

Injection Duration (ms)

Energizing Time (ms)

2.0

0.8

2.5

0.0

0.0

0.0

Injection Duration (ms)

(a) 2.0

Injection Pressure: 100MPa 1.0 Duration Difference

Injection Duration 3.0 Diesel Kerosene

Duration Difference (ms)

0.8

2.5

Duration Difference (ms)

Injection Duration (ms)

3.0

Injection Pressure: 140MPa 1.0 Duration Difference

Injection Duration Diesel Kerosene

0.0

0.0 0.4

0.6

0.8

1.0

1.2

1.4

Energizing Time (ms) Fig. 8. Injection duration for kerosene and diesel.

at 100 MPa and the injection interval for each pulse is kept at 1 ms. It is very clear in the figure that, five injections with diesel can be ideally realized. But, with the same control signal, if the fuel was

changed to kerosene, this multiple injection strategy is destroyed: due to longer injection duration for kerosene, if injections are close enough to each other, the needle will be lifted again even though


19

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12â&#x20AC;&#x201C;23

Charge density: 22 kg/m3

Injection pressure: 100 MPa

9

Diesel 2 ms

60

Kerosene

Spray tip penetration (mm)

Trigger Signal (V)

Split Injections

6 3 0

Current (A)

10

0

50 40 30

Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

20 10 0

-10 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

2

Charge density: 45 kg/m 3 1

60

L2 (1.28ms) L1 (1.13ms)

0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Time (ms) Fig. 9. Split injections for kerosene and diesel.

9

Current (A)

40 30 20 10

Kerosene

0.0

6

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) 3

Charge density: 67 kg/m 3 60

0 10

0

-10

Momentum (N)

50

Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

0

Diesel

1 ms

Spray tip penetration (mm)

Trigger Signal (V)

Multiple Injections-5 Pules Injection Pressure: 100 MPa

Spray tip penetration (mm)

Momentum (N)

Time (ms)

2

50 40

Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

30 20 10 0

1 0.0 0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) 0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

Time (ms)

Fig. 11. Spray penetration for kerosene and diesel.

Fig. 10. Multiple injections for kerosene and diesel.

3.3. Spray characteristics for test fuels based on piezoelectric injector the previous injection hasnâ&#x20AC;&#x2122;t finished. The unstable injections caused by incompletely closed needle will result in uncontrolled combustion. This is also one of the reasons that recalibration of engine control unit is very important when the fuel for an engine is changed.

To further understand the characteristics of fuel spray, the macroscopic parameters like spray penetration, spray velocity and spray angle are investigated under three injection pressures (60 MPa, 100 MPa, 140 MPa) and three charge densities (22 kg/ m3, 45 kg/m3, 67 kg/m3). The ambient temperature is fixed at


W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

   1 3 h 1 þ 2 tan 2h 2 VðtÞ ¼ pS ðtÞ tan  3 2 1 þ tan h 3 2

Charge density: 22 kg/m3 Fuel_Pi Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

Spray velocity (m/s)

120 100 80 60 40 20 0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) Charge density: 45 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

120

Spray velocity (m/s)

305 K. Fig. 11 shows spray penetrations for kerosene and diesel. The results indicate that, for each fuel, the spray penetration increases with the increase of injection pressure. This is mainly due to the fact that spray momentum is increased as injection pressure increasing. As the charge density increases, the spray penetration shows reduction trend, which is mainly because of the fact that the spray loses momentum when spray particles collide with high density charge molecules. This result is also well in line with other researchers [20–22]. Under charge density of 22 kg/m3, the spray penetration for kerosene is obviously shorter than that for diesel, especially for the lower injection pressure. With the increase of injection pressure, the difference in spray penetration between test fuels is slightly reduced. Under charge density of 67 kg/m3, the difference in spray penetration between test fuels shows much smaller than lower charge density. The shorter spray penetration for kerosene is due to two reasons: (i) the decrease in fuel viscosity enhances the breaking of injected spray, resulting in a reduction in the size of the spray droplets. The smaller the spray droplet size, the lower the momentum and the larger the resistance [23,24]; (ii) the decrease in discharge coefficient from the nozzle exit [25] caused by increasing cavitation due to less viscosity [26], which the authors have studied before [16]. The spray velocity for test fuels is shown in Fig. 12. For each test fuel, spray velocity increases with increasing injection pressure, especially at the initial period of spray development. As charge density increases, the spray velocity is reduced which can give support to spray momentum reduction. Due to lower spray momentum with lower viscosity, the spray velocity of kerosene is lower than that of diesel, especially under low charge density. Fig. 13 shows the spray angle for test fuels. It is clear that the spray angle is relatively stable with time and almost insensitive to injection pressure. As the charge density increases, the spray angle increases. This is mainly due to the fact that, as the charge density increases, the probability of collision for fuel and air molecule is increased, which results in greater resistance to spray development. It should be noted that the spray angle of kerosene is larger compared to diesel, which is due to higher fuel density of diesel [19,27–30] and higher spray momentum caused by higher viscosity of diesel [16,31–33]. Fig. 14 shows the processed spray evolution for the test fuels under different injection pressures and charge densities, for which the ASOI is 0.55 ms. In this figure, the trend of macroscopic spray characteristics can be shown intuitively. Fig. 15 shows spray volume comparison between kerosene and diesel. The fuel spray is assumed to be a cone with a hemisphere, and the spray volume is calculated as described by following relation [17]

100 80 60 40 20 0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) Charge density: 67 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

120

Spray velocity (m/s)

20

100 80 60 40 20

ð1Þ

0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) where V(t) is the spray volume over time t in mm3, S(t) is the spray penetration over time t in mm, and h is spray angle. It can be known from Eq. (1) that spray volume is significantly influenced by the spray penetration and spray angle. It is clear from Fig. 15 that, for each test fuel and under all injection pressures, the spray volume shows reduction trend with the increase of charge density. This is mainly due to the reduction of spray penetration, and secondly by spray angle. It should be noticed that under all conditions, kerosene shows larger spray volume than diesel, which is due to larger spray angle of kerosene. Fig. 16 shows the mass of entrained air for kerosene and diesel, which can give insight into fuel air mixing quality. The calculation

Fig. 12. Spray velocity for kerosene and diesel.

of mass of entrained air was using the model proposed by Rakapoulos et al. [34] as given below

ma ðtÞ ¼

p 3

tan2

  h 3 S ðtÞqa 2

ð2Þ

where ma(t) is the mass of entrained air over time t in kg, S(t) is the spray penetration over time t in mm, h is spray angle and qa is the


21

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

Charge density: 22 kg/m3

30

Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

Spray Angle (o)

27

24

Fuel Injection Pressure (MPa)

Diesel 140

100

K er os e ne 60

140

100

60

Charge Density (kg/m3)

67

21

18

15 45

12 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) Charge density: 45 kg/m3

30

Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

Spray Angle (o)

27

24

22

Fig. 14. Spray evolution for test fuels under different injection pressures and charge densities (ASOI: 0.55 ms).

21

18

15

12 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) Charge density: 67 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

30

Spray Angle (o)

27

24

21

18

15

12 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms) Fig. 13. Spray angle for kerosene and diesel.

charge density in kg/m3. As can be seen from Eq. (2) that the mass of entrained air is determined by the spray penetration, spray angle and charge density. It should be noticed that, compared with charge density of 22 kg/m3 which has longer spray penetration and 67 kg/m3 which has larger charge density and spray angle, more entrained air mass can be obtained under 45 kg/m3, which is a typical trade-off result among spray penetration, spray angle and charge density. It is also

obvious in the figure that under all conditions, spray of kerosene gets more entrained air than diesel, which is mainly due to larger spray angle of kerosene. This also means kerosene has advantages in fuel–air mixing. In addition, lower cetane number of kerosene will result in longer ignition delay to get more time to promote fuel–air mixing, which is finally good for reduction in both PM and NOx emissions [7,8,35]. 4. Conclusions In this study, behaviours of piezoelectric and solenoid injectors were compared based on an intelligent electronic control system. With piezoelectric injector, experiment for macroscopic spray characteristics was carried out with kerosene and diesel. The conclusions of the present study are summarized as below: (1) The piezoelectric injector shows faster injection response and needle opening than the conventional solenoid injector, which will enable precise and rapid injection control and is beneficial to air–fuel mixing before ignition in a real engine. (2) Due to the lower value of viscosity in kerosene, the injection durations for kerosene are longer than diesel. For kerosene, longer injection intervals will be needed to realize multiple injections. (3) The spray penetration for kerosene is obviously shorter than diesel, especially for the lower injection pressure. Under charge density of 67 kg/m3, the difference in spray penetration between test fuels is much smaller than lower charge density. Due to lower spray momentum with lower viscosity, the spray velocities of kerosene are lower than diesel, especially under low charge density. Spray angle is relatively stable with time and almost insensitive to injection pressure. The spray angle of kerosene is larger than diesel, which is due to higher fuel density and spray momentum caused by higher viscosity of diesel.


22

W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

4500

Spray volume (mm3)

4000 3500 3000

Charge density: 22 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

200 180

Mass of entrained air (mg)

5000

2500 2000 1500 1000 500

160 140 120

Charge density: 22 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

100 80 60 40 20

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0

1.4

0.0

0.2

Time (ms)

4500

Spray volume (mm3)

4000 3500 3000

Charge density: 45 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

200 180

2500 2000 1500 1000 500 0.2

0.4

0.6

0.8

1.0

1.2

3500 3000

1.2

1.4

160 140 120

Charge density: 45 kg/m3

100 80 60 40

0

1.4

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms)

Charge density: 67 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

200 180

Mass of entrained air (mg)

Spray volume (mm3)

4000

1.0

Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

Time (ms)

4500

0.8

20 0.0

5000

0.6

Time (ms)

Mass of entrained air (mg)

5000

0.4

2500 2000 1500 1000 500

160 140 120

Charge density: 67 kg/m3 Fuel_Pi (MPa) Kero_140 Kero_100 Kero_60 Dies_140 Dies_100 Dies_60

100 80 60 40 20

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms)

0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (ms)

Fig. 15. Spray volume for test fuels. Fig. 16. Mass of entrained air for test fuels.

(4) Kerosene shows larger spray volume than diesel, which is due to larger spray angle of kerosene. As a factor which can give insight into fuel air mixing quality, entrained air mass is a typical trade-off result of comprehensive influence of spray penetration, spray angle and charge density. Compared with diesel fuel, kerosene sprays get more

entrained air. This is mainly due to larger spray angle of kerosene, which also means kerosene has advantage in fuel–air mixing. In addition, lower cetane number of kerosene will result in longer ignition delay to get more time to promote fuel–air mixing, which is finally good for reduction in both PM and NOx emissions.


W. Yu et al. / Experimental Thermal and Fluid Science 76 (2016) 12–23

Acknowledgement This work is supported by the MOE of Singapore research grant R-265-000-529-112.

References [1] W.L. Hardy, R.D. Reitz, An Experimental Investigation of Partially Premixed Combustion Strategies Using Multiple Injections in a Heavy-Duty Diesel Engine, SAE Paper 2006-01-0917, 2006. [2] Wenbin Yu, Bin Liu, Yang Li, Qingpeng Su, Yiqiang Pei, Wanhua Su, A Hybrid Combustion Control Strategy for Heavy Duty Diesel Engines Based on the Technologies of Multi-Pulse Injections, Variable Boost Pressure and Retarded Intake Valve Closing Timing, SAE Paper 2011-01-1382, 2011. [3] W.H. Su, W.B. Yu, Effects of mixing and chemical parameters on thermal efficiency in a partly premixed combustion diesel engine with near-zero emissions, Int. J. Engine Res. 13 (3) (2012) 188–198. [4] Gautam T. Kalghatgi, Per Risberg, Hans-Erik Ångström, Partially pre-mixed auto-ignition of gasoline to attain low smoke and low NOx at high load in a compression ignition engine and comparison with a diesel fuel, SAE Paper 2007-01-0006, 2007. [5] Sage L. Kokjohn, Reed M. Hanson, Derek A. Splitter, Rolf D. Reitz, Experiments and Modeling of Dual-Fuel HCCI and PCCI Combustion Using In-Cylinder Fuel Blending, SAE Paper 2009-01-2647, 2009. [6] Wenbin Yu, Zhiqiang Kang, Wei Guo, Jichun Liu, Weibo Zhao, Zhiyong Zhang, Effects of Mixing and Chemical Parameters on Homogeneous Charge Induced Ignition Combustion Based on a Light-Duty Diesel Engine with Ultra-Low NOx and Soot Emissions and High Thermal Efficiency, SAE Paper, 2013-01-0914, 2013. [7] D. Turner, G.H. Tian, H.M. Xu, et al., An experimental study of dieseline combustion in a direct injection engine, SAE Paper 2009-01-1101. [8] D. Han, A.M. Ickes, D.N. Assanis, et al., Attainment and load extension of highefficiency premixed low-temperature combustion with dieseline in a compression ignition engine, Energy Fuels 24 (6) (2010) 3517–3525. [9] E.C. Owens, M.E. LePera, S.J. Lestz, Use of aviation turbine fuel jp-8 as the single fuel on the battlefield, SAE Paper 892071, 1989. [10] S.J. Lestz, M.E. LePera, Technology demonstration of US army ground material operating on aviation kerosene fuel, SAE Trans. J. Fuels Lubr. 112 (4) (1992) 204–225 (SAE Paper 920193). [11] D.A. Kouremenos, C.D. Rakopoulos, D.T. Hountalas, Experimental investigation of the performance and exhaust emissions of a swirl chamber diesel engine using jp-8 aviation fuel, Int. J. Energy Res. 21 (12) (1997) 1173–1185. [12] P.I. Lacey, S.J. Lestz, Effect of low-lubricity fuels on diesel injection pumps – part I: field performance, SAE Paper 920823, 1992. [13] G. Fernandes, J. Fuschetto, Z. Filipi, D. Assanis, H. Mckee, Impact of military jp8 fuel on heavy-duty diesel engine performance and emissions, J. Automot. Eng. 221 (D8) (2007) 957–970. [14] B. Mohan, W. Yang, W. Yu, Effect of internal nozzle flow and thermo-physical properties on spray characteristics of methyl esters, Appl. Energy 129 (2014) 123–134.

23

[15] B. Mohan, W. Yang, K.L. Tay, W. Yu, Macroscopic spray characterization under high ambient density conditions, Exp. Thermal Fluid Sci. 59 (2014) 109–117. [16] B. Mohan, W. Yang, K.L. Tay, W. Yu, Experimental study of spray characteristics of biodiesel derived from waste cooking oil, Energy Convers. Manage. 88 (2014) 622–632. [17] E. Delacourt, B. Desmet, B. Besson, Characterisation of very high pressure diesel sprays using digital imaging techniques, Fuel 84 (2005) 859–867. [18] R. Payri, J. Garcia, F. Salvador, J. Gimeno, Using spray momentum flux measurements to understand the influence of diesel nozzle geometry on spray characteristics, Fuel 84 (2005) 551–561. [19] S.H. Park, J. Cha, S.H. Kim, C.S. Lee, Effect of early injection strategy on spray atomization and emission reduction characteristics in bioethanol blended diesel fueled engine, Energy 39 (2012) 375–387. [20] J.D. Naber, D.L. Siebers, Effects of gas density and vaporization on penetration and dispersion of diesel sprays, SAE Paper 960034, 1996. [21] D.L. Siebers, Liquid-phase fuel penetration in diesel sprays, 1998. [22] E. Delacourt, B. Desmet, B. Besson, Characterisation of very high pressure diesel sprays using digital imaging techniques, Fuel 84 (2005) 859–867. [23] H. Hiroyasu, M. Arai, Structures of fuels sprays in diesel engines, SAE Paper 900475, 1990. [24] Yuan Gao, Jun Deng, et al., Experimental study of the spray characteristics of biodiesel based on inedible oil, Biotechnol. Adv. 27 (2009) 616–624. [25] J.-G. Nerva, C.L. Genzale, S. Kook, J.M. García-Oliver, L.M. Pickett, Fundamental spray and combustion measurements of soy methyl-ester biodiesel, Int. J. Engine Res. 14 (2013) 373–390. [26] J.-M. Shi, M. Arafin, CFD investigation of fuel property effect on cavitating flow in generic nozzle geometries, ILASS-Europe 2010, 2010. [27] S.H. Park, J. Cha, C.S. Lee, Effects of bioethanol-blended diesel fuel on combustion and emission reduction characteristics in a direct-injection diesel engine with exhaust gas recirculation (EGR), Energy Fuel 24 (2010) 3872–3883. [28] S.H. Park, I.M. Youn, C.S. Lee, Influence of two-stage injection and exhaust gas recirculation on the emission reduction in an ethanol-blended diesel fueled four-cylinder diesel engine, Fuel Process. Technol. 91 (2010) 1753–1760. [29] S.H. Park, S.H. Kim, C.S. Lee, Mixing stability and spray behavior characteristics of diesel-ethanol-methyl ester blended fuels in a common-rail diesel injection system, Energy Fuel 23 (2009) 5228–5235. [30] S.H. Park, I.M. Youn, C.S. Lee, Influence of ethanol blends on the combustion performance and exhaust emission characteristics of a four-cylinder diesel engine at various engine loads and injection timings, Fuel 90 (2011) 748–755. [31] M.H. Al-Abdullah, G.T. Kalghatgi, H. Babiker, Flash points and volatility characteristics of gasoline/diesel blends, Fuel 153 (2015) 67–69. [32] A. Alajbegovic, R. Tatschl, M. Volmajer, E. von Berg, L.C. Ganippa, W. Edelbauer, et al., Coupled simulations of nozzle flow, primary fuel jet breakup, and spray formation, J. Eng. Gas Turbines Power 127 (2005) 897–908. [33] F. Payri, V. Bermúdez, R. Payri, F.J. Salvador, The influence of cavitation on the internal flow and the spray characteristics in diesel injection nozzles, Fuel 83 (2004) 419–431. [34] C. Rakopoulos, D. Rakopoulos, E. Giakoumis, D. Kyritsis, Validation and sensitivity analysis of a two zone diesel engine model for combustion and emissions prediction, Energy Convers. Manage. 45 (2004) 1471–1495. [35] Jinwoo Lee, Choongsik Bae, Application of JP-8 in a heavy duty diesel engine, Fuel 90 (2011) 1762–1770.


Yu2016