Experimental Research of High-Temperature and High-Pressure Water Jet Characteristics in ICRC Engine Relevant Conditions

Abstract

The internal combustion Rankine cycle (ICRC) concept provides a potential solution for future high thermal efficiency and low emission powertrains, and direct water injection (DWI) proved to be the key parameter for ICRC optimization. This paper was dedicated to investigating the fundamental mechanisms of water spray characteristics under different water injection control parameters. In order to do so, an experimental test system was carefully designed and built based on the Bosch and Schlieren methods: the Bosch method is utilized to measure the effect of injection and ambient pressure on water injection characteristics, and the Schlieren method is utilized to investigate the impact of water injection and ambient temperature on water spray and evaporation processes. The experimental results indicate that both control parameters show important effects on water injection and spray characteristics. The water injection and ambient pressure show significant impacts on steady-state flow quantity and cyclic water injection quantity, and the water injection and ambient pressure affect the evaporation ability of water vapor within the spray which leads to a different variation trend during the initial, developing, and developed water spray stages. The results of this work can be used as fundamental supplements for ICRC, steam assistant technology (SAT), and DWI-related ICEs experimental and numerical researches, and provide extra information to understand the DWI process within engine-relevant conditions.

1. Introduction

As the dominant power source in automotive industry, internal combustion engines (ICEs) show tenacious vitality when facing the challenges of battery electric vehicles (BEVs) and fuel cell vehicles (FCVs) in the modern transportation sector. Compared to BEV and FCV, ICEs show superiority in driving mileage, refueling infrastructure, flexibility, and environment adaptability [1]. Although traditional ICEs suffer from relatively low tank-to-wheel efficiency and exhaust emissions [2], the accelerated evolvements of ICE electrification have paved the way for the research and development of high-efficiency and low-emission hybrid dedicated ICEs [3]. The ICEs within hybrid electric vehicles (HEVs) operated under relatively narrow working conditions compared to traditional ones [4]—this alteration brings huge benefits when facing the challenges of efficiency and emissions optimization in ICEs.

On this technical basis, to further enhance the thermal efficiency of hybrid-dedicated ICEs and to eliminate the exhaust emissions thoroughly, a novel ICE concept named “internal combustion Rankine cycle” (ICRC) can be utilized [5]. The idea of ICRC can be dated back to 1999 when there was strong demand from the power generation industry for high efficiency and zero emission facilities [6]. The basic principle of the ICRC engine is to utilize oxygen instead of air as its intake to directly oxidize the injected fuel: as the nitrogen is absent, the generation of nitric oxide can be fully avoided. As the oxidizer within the combustion chamber is pure oxygen, the combustion intensity within the combustion chamber is relatively strong, in order to control and optimize the combustion process, direct water injection (DWI) is implemented to regulate the combustion process. On the other hand, the direct injected water can further absorb combustion heat and evaporate rapidly: the phase transformation of liquid water into high-pressure steam further increases the expansion work during engine operation, and, benefitting from the above factors, the thermal efficiency of ICRC engine can be improved significantly [7,8]. As well as the efficiency enhancement, the ICRC concept also contributes to the decrement of exhaust emissions. According to the working principle of ICRC, the combustion products in the ICRC engine are mainly carbon dioxide and steam. By utilizing a specially designed condenser, the steam is cooled to under dew point, so the carbon dioxide and water within the exhaust gases can be easily separated. The gaseous carbon dioxide are further cooled into liquefied carbon dioxide by exchanging heat with liquefied oxygen for better storage, and water is recycled and heated by exhausted gases for less dependence in external water supply. A schematic diagram of the ICRC engine is illustrated in Figure 1. Compare to BEV and FCV, the design and implementation of the ICRC engine is based on conventional ICEs and mature technologies utilized among transportation and power generation industries, so it is reasonable to expect lower realization difficulty and production cost [8]. It is also reasonable to utilize the ICRC engine into HEV, especially in the range extension electric vehicle (REEV), because of much simpler working conditions which further decrease the implementation difficulties of ICRC engine.

Since the proposal of ICRC concept, the authors have conducted extensive works related to its theoretical and experimental researches. The theoretical thermal efficiency boundaries and its optimization direction are determined based on the establishment of theoretical thermodynamic model, the best thermal efficiency achieved in theoretical thermodynamic model is 62% [9]. The combustion and performance characteristics under different intakes of oxygen concentration [10], engine operation parameters [11], water injection strategies [12], and in-cylinder combustion feedback control [13] within spark ignition ICRC (SI-ICRC) engine were investigated, and all of these experimental results show a promising tendency of ICRC engine utilization in the modern transportation sector. Despite the proved efficiency enhancement in the SI-ICRC engine, the knocking combustion phenomenon within SI-ICRC resulted in end gas auto-ignition restraint the further thermal efficiency improved [14]. Although the DWI can be utilized to mitigate the knocking tendency, the obstacles between practical and theoretical thermal efficiency are difficult to overcome. In order to solve this problem, compression ignition ICRC (CI-ICRC) utilizing both homogeneous and heterogeneous charge compression ignition is proposed to eliminate flame propagation so the knocking combustion can be avoided, as illustrated in a schematic diagram of CI-ICRC engine test bench in Figure 2. Similarly to the SI-ICRC engine, the combustion process and performance characteristics of CI-ICRC under different oxygen concentration [15], engine operation parameters [16], and water injection strategies [17] were investigated experimentally and numerically. It is proved that the ICRC concept shows great potential in improving thermal efficiency and lowering exhaust emissions, and the DWI process is a key control parameter in ICRC engine. It is important to conduct comprehensive experimental research to simulate and understand the detailed DWI and its evaporation process.

On the other hand, with the demand for higher power density and torque response, the intake pressure in the modern advanced turbocharged gasoline direct injection (TGDI) engine is increasing aggressively, which results in higher risk for knock and super knock [18,19]. The former impacts engine efficiency and performance due to retarded minimum spark advance (SA) for maximum torque (MBT) and enriched air-fuel ratio (AFR) [20], and the latter has much stronger in-cylinder pressure peak and fluctuation, which directly affect engine functionality and durability [21]. Water injection (WI) also plays an important role in mitigating knocking combustion phenomenon in highly intensified TGDI engine [22]. Bozza et al. conducted experimental researches to verify the potential of port water injection (PWI) in optimizing combustion phase and control knocking combustion, and a simulation model was established to predict the fuel economy potential under worldwide harmonized light vehicles test procedure (WLTP) [23]. Naber et al. conducted experimental related to PWI in a TGDI engine to test the control ability in knocking mitigation, and a 30% enhancement in thermal efficiency was realized through optimizing MBT and AFR [24]. The PWI system mentioned in the literatures above was mainly modified based on port fuel injection (PFI), so the industrialization of PWI is much easier compare to DWI. BMW [25], Robert Bosch GmbH [26], FEV [27], and Nostrum Energy [28] all reported commercialized PWI products for OEM and aftermarket application. Despite the high cost performance of PWI, there are still drawbacks of PWI such as water incomplete evaporation, slower transient response, and relatively high water consumption rate [29]. To overcome the above-mentioned problems, DWI shows an irreplaceable position in WI technology for modern ICEs, and many automotive corporations and universities show great interests in DWI. Hoppe et al. conducted experimental and numerical researches in DWI implementation within a TGDI engine, and an elevated compression ratio 13.5:1 was realized under 1 MPa indicated mean effective pressure (IMEP), which can further increase to 14.7 with variable valve actuation (VVA) technology [27]. The enhancement results under optimum fuel consumption rate and full load condition are 3.3~3.8% and 16% with 1 L/100 km water consumption. The overall best brake specific fuel consumption (BSFC) achieved is 201 g/kWh [30]. Bae et al. implemented DWI into a PFI gasoline engine and conducted experimental research related to the effect of DWI strategies under different engine operation parameters on compression ratio and fuel economy. The optimized compression ratio was increased to 13.5, and fuel economy was enhanced around 7~17% without performance deterioration [31]. DWI also contributes to a reduction of exhaust emissions. Miyamoto et al. conducted experimental researches about DWI in an in-direct injection (IDI) diesel engine to test the NO_X reduction ability [32]. Chybowski et al. presented a comprehensive overview in NO_X reduction potential by using WI [33,34]. The authors also observed this phenomenon by utilizing DWI within a diesel engine to optimize NO_X emissions and performance simultaneously [35]. Tauzia et al. also reported that particulate emission was reduced by using DWI in a diesel engine [36]. It is obvious that DWI, and especially its water injection process, plays a vital role in the future of high efficiency and performance hybrid dedicated engines from the reported literatures, so the proposed experimental research related to direct water injection characteristics is also beneficial for the future DWI industrialization and calibration.

There is another kind of DWI application in high efficiency ICEs field which is related to steam assistant technology (SAT) [37]. The fundamental idea of SAT is utilizing high-pressure and -temperature steam generated through waste or combustion heat absorption by direct injected water—the massive expansion work enhancement during the phase transformation of water contributes to the significant thermal efficiency improvement [38]. Szybist et al. proposed a six-stroke engine conception by adding DWI into an extra stroke to recover waste heat in exhaust gases and improve fuel economy [39]. Fu et al. also presented a similar concept to utilize DWI to establish an open steam cycle for thermal efficiency enhancement [40]. Azmi et al. applied DWI into a hydrogen-fueled compression ignition engine to optimize the combustion process and improve engine efficiency [41]. Hewavitarane et al. proposed a superheated liquid flash boiling (SLFB) cycle which is implemented within the engine exhaust system to recover waste heat and generate steam to increase engine efficiency [42]. It is obvious to conclude that DWI and its injection process plays a key role in SAT, and the experimental researches in establishing knowledge about high-temperature and -pressure direct water injection characteristics also contribute to the fundamental researches in SAT.

Since DWI, especially high-temperature and high-pressure water jet, occupy an important place in future extreme high-efficiency and low-emission ICEs as described above, this paper proposed to utilize Bosch method to measure and investigate the water injection characteristics under different injection and ambient pressure which simulate the water injection process in engine-relevant conditions, and the water spray characteristics under different injection and ambient temperature is also investigated using the Schlieren method within a constant volume vessel (CVV) to explore the impact effect of temperature on water jet atomization and evaporation, besides, the experimental results provided in this work also provide extra information for future DWI simulation in real engine working conditions.

2. Experimental Setup

The main objective of this study is to investigate the high-temperature and high-pressure water jet characteristics under different injection and ambient conditions, the results of this study can be beneficial in providing detailed injection and spray design or for calibrating information for ICRC and other advanced DWI technologies in modern ICEs. An injection process test system based on the Bosch method is utilized in order to study the effect of injection pressure and ambient pressure on high-pressure water injection characteristics, on the other hand, a CVV with Schlieren method is established to investigate the high-temperature water spray and evaporation process. The detailed experimental setups for different objectivities are introduced below.

2.1. Bosch Method for High-Pressure Water Injection Characteristics Measurement

The high-pressure water injection characteristics test system based on the Bosch method is shown in Figure 3. As shown in the figure, an air-fluid booster with a compression ratio 60:1 was equipped in the system to pressurize the water to the target injection pressures up to 48 MPa, and an air compressor with maximum 0.8 MPa high-pressure air supply was utilized to provide pressure source for water pressurization. The water was stored in a water tank before entering the air-fluid booster and a water filter is installed to prevent impurities. The pressurized water was delivered into a specially-designed common rail with a maximum allowed pressure of 80 MPa. A heating rod coupled with inner thermocouple was installed within the common rail to simulate the waste heat recovery and supply extra energy to heat up the pressurized water. The highest heating temperature was fixed at 160 °C under continuous water injection condition. The water injector utilized during high-pressure water injection characteristics measurement is the same one used during the ICRC engine bench test. The water injector with 6 holes was specially-designed and manufactured by the Wuxi fuel injection equipment research institute to ensure preferable durability and reliability during long term WI operation. The high-precision piezoelectric pressure sensor was installed on the connecting block to measure the pressure fluctuations in the long tube during WI, a charge amplifier is used to amplify the charge generated within the pressure sensor and transfer to data acquisition system (DAQ) for data storage. The connecting block was utilized to connect the water injector with long tube which provided enough reflection time during pressure waves measurement. During the measurement of high-pressure water injection characteristics, the throttle valve and back pressure valve installed at the end of the long tube was used to adjust the ambient pressure within the long tube, and also to adjust the number of reflection waves after water injection to ensure measurement precision [43]. The control circuit used in the system is CompactRIO platform with cRIO-9024 real-time controller which is capable of providing a 4 GB storage and highly timed real-time operation system, cRIO-9118 FPGA back board with 40 MHz clock timing, which means control iteration less than 25 ns, the C series module NI-9751 and NI-9205 were also used to trigger direct water injection and to collect rail pressure information.

The detailed experimental conditions are presented in

Table 1. Experimental conditions for high-pressure water injection characteristics measurement.

Parameter	Value
Water Injection Pressure/MPa	30, 35
Water Injection Duration/ms	0.4, 0.6, 0.8 1.2, 1.6, 2, 2.5, 3
Ambient Pressure/MPa	0.1, 1, 3, 5, 7
Water Injection Temperature/°C	25

. During the measurement process, 30 and 35 MPa water injections, pressures which represent actual ICRC working conditions, are acquired by adjusting the air compressor outlet pressure. The water injection duration varies from 0.4 ms to 3 ms and was chosen to provide sufficient information for ICRC water injector discharge coefficient calculation and ICRC engine calibration. An ambient pressure range from 0.1 MPa to 7 MPa is established by adjusting the back-pressure valve. The 0.1 MPa ambient pressure stands for water injection quantity measurement under atmosphere condition using weighing method, which represent conventional water injection quantity calibration method before engine experiment.

The high-pressure water injection characteristics measurement based on Bosch method is developed derived from the hydraulic pulse theory [44], the injection rate can be obtained from the manipulation below:

$$q=\frac{d{V}_b}{dt}=Au$$

(1)

In the Equation (1), $A$ stands for sectional area of the long tube, $u$ stands for the flow velocity of water within the long tube, and the $u$ can be obtained from original wave data $p\left(t\right)$:

$$p\left(t\right)=\alpha \rho u$$

(2)

With $\alpha$ stands for sound velocity within water and $\rho$ is the density of the liquid. By combing Equations (1) and (2), the injection rate $q\left(t\right)$ can be described as below:

$$q\left(t\right)=Ap\left(t\right)/\alpha \rho$$

(3)

By simply calculation the integral of $q\left(t\right)$, the injection quantity can be obtained where $d$ is the diameter of the long tube:

$$Q={{\displaystyle \int }}_{t_1}^{t_2}\frac{\pi d^{2}p\left(t\right)dt}{4\alpha \rho }$$

(4)

Of course, there are drawbacks in the Bosch method when measuring the high-pressure liquid injection rate. The system errors of the pressure sensor and system design can show cumulative phenomenon that affect the measurement precision and credibility. To compensate this negative effect, a numerical integration is also added during injection rate calculation as introduced in literature [45].

Figure 4 illustrates the definition of the injection characteristic parameters during measurement. Figure 4a is an example of original wave and reflection waves during experiment: the $t_1$ is start of original wave and $t_2$ is end of original wave, and the $t$ stands for the interval between two waves, while enlarging the $t$ needs longer tube which is better for multi-injection rate measurement [46]. The calculated water injection rate and water injection quantity is illustrated in Figure 4b.

2.2. Schlieren Method for High-Temperature Water Spray Characteristics Measurement

The schematic of water spray characteristics test system based on Schlieren method is shown in Figure 5. The experiments were conducted in a cubic CVV with quartz windows on each side. A specially designed and manufactured water injector was mounted into an adaptor installed on the top of the CVV, compared to the high-pressure water injection characteristics measurement, and a single hole water injector was utilized to replace the 6-hole water injector for better macroscopic diagnostic. To establish an adjustable ambient temperature within the CVV, a thermocouple and four PID-controlled heaters were placed at each corner inside the CVV. These heaters provided sufficient energy to heat the gases within the CVV up to 200 °C. The utilized high-temperature and high-pressure water injection system was the same one as introduced in the high-pressure water injection characteristics measurement. The water injection pressure was set at 35 MPa and the water injection temperature varied from 25 °C to 160 °C. The visualization of both the liquid and gas phase of high-temperature water spray was realized by using the Schlieren method, the experimental transient spray and atomization images were captured by a Phantom V7.3 high speed camera, and the sampling rate of high speed camera was set at 12,012 frame per second and a spatial resolution of 0.317 mm per pixel.

The image processing procedure is illustrated in Figure 6. Firstly, the spray images were captured and recorded by a high-speed camera, then the image was transformed to grey image, in order to extract spray information. The grey image was further binarized with a threshold calculated through the Ostu method [47]. The edge of the liquid phase spray was then obtained through Canny approximation. The spray penetration can be defined as the distance between injector tip and end of the spray. There are different definitions of spray angle, but the fundamental principles can be concluded as the angle between two narrowest straight lines which envelope the outer boundary of the captured spray image. The definition of the spray angle utilized in this study is an amendment based on the above theory which contains two calculation conditions: (1) if 0.1 × spray penetration ≥ 10 pixels, the spray angle equals the largest angle between the injector tip and spray outer boundary from 0.1 × spray penetration to 0.75 × spray penetration; (2) if 0.1 × spray penetration < 10 pixels, the spray angle equals the largest angle between injector tip and spray outer boundary from 10 pixels to 0.75 × spray penetration [48]. Each test was repeated several times to guarantee the accuracy and repeatability of the experimental results. The experimental conditions for high-temperature water spray characteristics measurement are presented in

Table 2. Experimental conditions for high-temperature water spray characteristics measurement.

Parameter	Value
Water Injection Temperature/°C	25, 80, 120, 160
Ambient Temperature/°C	25, 105, 160
Water Injection Pressure/MPa	35
Ambient Pressure/MPa	0.1
Water Injection Duration/ms	3

. The water injection parameters were selected based on actual engine operation conditions, and the ambient parameters were designed according to the heating capability of the CVV.

3. Results and Discussion

3.1. High-Pressure Water Injection Characteristics under Different Water Injection and Ambient Pressure

3.1.1. Water Injection Characteristics under Different Water Injection Pressure

The water injection rate and water injection quantity under different water injection pressures are illustrated in Figure 7a. The slope of water injector rate indicates the opening speed of the water injector nozzle is relatively slower compared to its closing speed. The maximum steady state injection rate increased significantly from 33.7 mg/ms to 42.0 mg/ms as the water injection pressure changed from 30 MPa to 35 MPa, and the water injection quantity also increased from 105 mg to 126 mg. The increased water injection rate and water injection quantity is caused by the enhancement of water injector nozzle opening speed and water injection initial kinetic energy. These two factors contribute together to optimize the water injector rate and quantity of the tested water injector.

Figure 7b shows the water injection steady-state flow quantity variation trend under different water injection durations. These results are crucial for the ICRC engine bench test. As it can be seen in the figure, compared to the steady-state flow quantity under 30 MPa water injection pressure, the steady-state flow quantity increased about 20% while a 35 MPa water injection pressure is occupied, which means for the same cyclic water injection quantity demand, the water injection duration can be significantly shortened under 35 MPa water injection pressure. According to the ICRC bench test, it is crucial to inject high-pressure and high-temperature water into the combustion chamber with an injection duration as short as possible [49]. The elevated water injection pressure can be helpful in supporting this requirement of the ICRC engine concept, and a 35 MPa water injection strategy will be much better compared to 30 MPa or some other even lower water injection pressure.

3.1.2. Water Injection Characteristics under Different Ambient Pressure

In ICRC engine concept, high-temperature and high-pressure water needs to be injected into the combustion chamber at firing top dead center (TDC), at this crank angle, the ambient pressure within the combustion chamber is significantly higher compared to traditional water injection conditions (during intake stroke or early compression stroke) in a conventional TDGI engine. The literature [50] shows that the maximum in-cylinder pressure of SI-ICRC and CI-ICRC is normally 7 MPa. In this section, the ambient pressure in the long tube is adjusted by continuous injection and modulation of the back-pressure valve. The target ambient pressures varies from 1 MPa to 7 MPa. The experiment results are shown in Figure 8.

Figure 8a presents the water injection-rate and -quantity under different ambient pressures, the experimental results show that an increment of ambient pressure has a negative effect on water injection rate. Compared to a 41.7 mg/ms maximum steady state injection rate under 1 MPa ambient pressure, the maximum steady state injection rate under 7 MPa ambient pressure dropped to 36.7 mg/ms, which is a decrement around 12.0%. The deceased maximum steady state injection rate leads to a lower water injection quantity when higher ambient pressure is deployed. The decrement of water injection quantity is around 10% (129 mg water injection quantity under 0.1 MPa ambient pressure; 116 mg water injection quantity under 7 MPa ambient pressure). The reason for water injection-rate deterioration under higher ambient pressure is the decrement of pressure difference between injection and ambient pressure: as the pressure difference decreases, the discharge coefficient of the injector nozzle is influenced, and the reduced discharge coefficient directly leads to a lower water injection rate and therefore a decrement in water injection quantity.

A similar conclusion can be drawn from the water injection steady-state flow quantity under different ambient pressures shown in Figure 8b. As the ambient pressure increased from 0.1 MPa (atmospheric pressure) to 7 MPa (firing TDC in ICRC engine), the water injection steady-state flow quantity dropped about 10.3%, which indicates the actual water injection quantity decreased significantly under the same injection duration. This phenomenon is very important in ICRC and SAT engine concepts, because all these novel engine concepts are established on DWI under an elevated ambient pressure environment. There are very few literatures reporting this phenomenon and the results presented in this study can be beneficial in providing extra information for the ICRC and SAT engine experimental and numerical investigation.

3.2. High-Temperature Water Spray Characteristics under Different Water Injection and Ambient Temperature

The high-pressure water spray and atomization processes are vital parameters for thermal efficiency optimization of ICRC, SAT, and DWI in modern ICEs. To analyze its spray characteristics and provide supplementary information for further numerical researches in engine relevant conditions, high-temperature water spray characteristics under different water injection and ambient temperatures ranging from 25 °C (room temperature) to 160 °C are investigated in this section.

3.2.1. Water Spray Characteristics under Different Water Injection Temperature

The Schlieren images of water spray at 0.8 ms (a to d) and 0.2 ms (e to h) after start of injection (ASOI) are presented in Figure 9. In order to investigate the effect of water injection temperature on spray characteristics, the ambient temperature is coordinated at 160 °C by the PID controlled heaters within the CVV, the water injection temperature is varied from 25 °C to 160 °C by controlling the heaters installed in the common rail. As it can be seen in the figure, the water spray penetration is significantly shortened while the injection temperature increased from 25 °C to 160 °C. The detailed water spray penetration data at 0.8 ms ASOI are illustrated in Figure 10a. The results show that water spray is fully developed after 0.8 ms ASOI under all the test conditions, and the water spray penetration under 25 °C at 0.8 ms ASOI reaches 110 mm compare to 88 mm water spray penetration under 160 °C, which is a prominent 20% reduction. The detailed water spray angles under different water injection temperature are also calculated and compared. From the water spray angle results presented in Figure 10b, the water injection temperature shows little impact on the water spray angle, the water spray angle under 25 °C water injection temperature at 0.8 ms ASOI is 45.3° and 47.4° under 160 °C water injection temperature, but from another perspective, the water spray angle during the initial and developing stage of the high-temperature water injection shows linear enlargement relationship as the water injection increased. The water spray angle under 25 °C water injection temperature at 0.2 ms ASOI is 40.6° and the water spray angle under 160 °C water injection temperature increased significantly to 61.3°, which is a 50.1% increment.

The reason for water spray penetration decrement with an increasing water injection temperature is enhanced water evaporation rate, as the temperature of water droplet carried by water spray increased, the evaporation rate of water droplet is enhanced which leads to smaller sauter mean diameter (SMD) and larger energy dissipation, therefore leading to a shortened spray penetration as the water injection temperature increased.

Unlike the relatively simple effect of water injection temperature on spray penetration, the spray angle is a selection to enhance the evaporation rate between the water injection temperature and ambient temperature within the CVV. During the early stage of high-pressure water jets (e.g., 0.2 ms ASOI), there is relatively shorter time for the injected water spray to contact with surrounding hot air, so the evaporation rate of water droplet within spray is mainly affected by the water injection temperature which lead to a larger spray angle with higher water injection temperature, on the other hand, as the water spray entering a fully developed stage (e.g., 0.8 ms ASOI), the injected water droplet within the spray boundary contact with the surrounding hot air directly and the ambient temperature becomes dominant factor which influence the evaporation rate, with an unchanged ambient temperature (especially temperature above boiling temperature), the spray angle within fully developed waster spray remains similar even though the water injection temperature has varied.

3.2.2. Water Spray Characteristics under Different Ambient Temperature

As discussed above, the ambient temperature shows a significant effect on the water spray characteristics. In order to investigate the detailed effect mechanism of ambient temperature on water spray and atomization, the macroscopic water spray images are captured by varying the temperature setting in the PID controlled heaters, by doing so, the ambient temperature within the CVV can be set at 25 °C, 105 °C, and 160 °C, and in order to eliminate the interference, the water injection temperature is fixed at 160 °C.

The Schlieren images of water spray under different ambient temperature at 0.9 ms (a to c) and 1.6 ms (d to f) ASOI are illustrated in Figure 11. As it can be seen in the figure, the effect of ambient temperature and water spray penetration is different compared to the water injection temperature and the spray penetration shows an incremental trend as the ambient temperature increase. A detailed water spray penetration can be seen in Figure 12a. Compared to the 110 mm spray penetration under 160 °C ambient temperature, the water spray penetration decreased to 101 mm under 25 °C ambient temperature which is an 8.2% decrement. On the other hand, the water spray angle under a different ambient temperature also shows a different pattern. The increment in ambient temperature has a positive effect in enhancing water spray angle, as shown by the experimental data illustrated in Figure 12b. The fully-developed water spray angle at 0.8 ms ASOI under 160 °C ambient temperature is around 50° while this value under 25 °C ambient temperature dropped to 40.8°, which is a 18.4% decrement. The Schlieren images of water spray at 1.6 ms ASOI are also presented in this section. The objective of these images is to explain the effect of ambient temperature on water spray evaporation. As it can be seen in the figure, the ambient temperature shows a significant impact on the width and area of the water spray at 1.6 ms ASOI, compared to the 26.1 mm spray width under 160 °C ambient temperature, and the spray width decreased to 13.1 mm under 25 °C ambient temperature, which is a 50% decrement.

The reason for water spray penetration increment in the elevated ambient temperature can be attributed to the variation of ambient density. As the ambient temperature increases, while the ambient pressure and the volume of the CVV remain constant, according to the ideal gas law, the ambient density will be decreased. Then the water spray penetration is a result of competence between ambient temperature and ambient density. As the ambient temperature increases, the water spray and atomization is enhanced which leads to a decrement in spray penetration. On the other hand, decreasing ambient density leads to a lower resistance and viscosity in spray penetration, and therefore enlarges the result of water spray penetration. Apparently, the variation in ambient density shows a heavier impact compared to ambient temperature in water spray penetration determination.

The incremental rise in ambient temperature shows a positive impact on water spray evaporation. The water spray width and area are both enhanced under an elevated ambient temperature, and this result is beneficial for providing fundamental support for ICRC and SAT concepts which are established on steam generation and flash boiling spray. Limited by the heating capability of the CVV experimental system, the maximum ambient temperature is 160 °C, which is adequate for providing supporting data for future numerical researches at near engine condition.

4. Conclusions

In this paper, a brief literature review of ICRC, DWI, and SAT were first conducted to demonstrate the necessity of current research work, then water injection and spray characteristics were investigated by establishing an experimental test systems based on the Bosch and Schlieren methods. The effects of water injection pressure, water injection temperature, ambient pressure, and ambient temperature were discussed detailed. The conclusions of this work are presented below:

(1) The water injection pressure shows a positive effect in enhancing the steady-state flow quantity of water injection process. Compared to 30 MPa water injection pressure, the maximum steady state injection rate increased to 42.0 mg/ms under 35 MPa water injection pressure, which is around a 20% increment. It is beneficial in proving higher water injection pressure in ICRC and SAT concepts because these concepts require a water injection duration as short as possible.

(2) The increment of ambient pressure shows negative impact on steady-state flow quantity enhancement, the water injection quantity under same water injection pressure can be dropped around 10.3% while the ambient pressure increased from 0.1 MPa to 7 MPa. This is an important discovery which is rarely reported in current literatures, as the water injection process within ICRC and SAT are operated under elevated in-cylinder pressure, and the effect of ambient pressure on water injection characteristics should be considered in experimental or numerical researches.

(3) The increment in water injection temperature shortens water spray penetration caused by an enhanced water spray evaporation rate. The reduction rate reaches 20% under the selected experimental conditions. The water spray angle under different water injection temperatures shows complex variation. During the initial and developing stage of water spray, the increase in water injection temperature leads to a larger spray angle caused by the enhanced water spray evaporation rate itself, while within a fully developed water spray, the spray angle is mainly affected by ambient temperature as there is sufficient time for the water droplets within the spray boundary to make contact with the surrounding hot air directly.

(4) The elevated ambient temperature is beneficial in providing a faster water evaporation, and due to the reduction in ambient density as the ambient temperature increases, the water spray penetration is shortened. As a limitation of the established CVV test system, the highest ambient temperature reached during experiment is 160 °C which will be further investigated in a detailed numerical CVV model with near-engine conditions.

The uncertainty and sensitivity analysis, high-fidelity simulations, and particle behaviors will be addressed in future experiments by using multi-data comparisons [51], commercial computational fluid dynamics software [52], and a Malvern particle size analyzer [53] to further analyze the fundamental mechanisms of interactions between flow, water vapor, and evaporation.