An experimental investigation on performance of converted CNG engine by varying piston bowl geometry: A case study

ABSTRACT

Using natural gas as an alternative fuel will be reduced exhaust emissions and dependence on petroleum. The practical solution for Vietnam’s conditions is to convert diesel engines into natural gas engines. Understanding the effect of piston geometry on the performance of converted CNG engines helps domestic manufacturers come up to spark ignition natural gas engines with high performance and low emission. In this work, three different types of combustion chamber were made from milling piston head and varying thickness of cylinder head gasket. The experiment results showed that torque and power of ε = 11.5 were more significant than ε = 12.5. During experiment processes, the stable operation state of the converted engine with ε = 12.5 was challenging to control, and the noise levels were intense. The results of mass fraction burned were presented the effect of squish gap on performance was more substantial than that of the squish area. It can be concluded that for modifying the original diesel engines to run on CNG fuel, the maximum compression ratio should be set to 12.5 and must be redesigned accordingly to use the natural results.

Implications: This manuscript shows the new technology to convert traditional engines into engines fueled by renewable energy. It contributes to reducing the fossil fuel crisis and environmental emissions.

Introduction

Traditional energy, such as gasoline and diesel, is the most critical energy resource used in the transport sector nowadays. Due to the rapid development of vehicles, many environmental and economic issues have emerged. Using natural gas as an alternative fuel for internal combustion engines is carried out to solve these critical problems. The development of natural gas fuel is of paramount importance in Vietnam as a rise in its usage can simultaneously cut down air pollution and dependence on conventional fossil fuel (Korakianitis, Namasivayam, and Crookes 2011). Natural gas has been known as a promising alternative fuel for internal combustion engines because of its main component being Methane (CH₄) and plentiful supply such as natural gas mine, petroleum distillation, shale excavation (Skorek and Włodarczyk 2018). Under the same conditions, when burning natural gas, the emission component can be reduced by 30% for CO₂, 95% for CO, 87% for NO_x, and particulate matter emission is dramatically smaller than non-methane hydrocarbons (Bag 2009; Jahirul et al. 2007). Oppositely, the development of natural gas engines to meet new emission standards must face two difficulties: low burning speed and low density (Chala, Aziz, and Hagos 2018). According to its low density, the natural gas is compressed under high pressure of approximately 250 bar, so CNG fuel and engine can operate longer. The combustion efficiency is one of the important parameters to assess engine performance, and this parameter is affected by burning rate and burning speed.

While compressed natural gas (CNG) has long been used in stationary engines, its use as a transportation fuel has advanced significantly in the last decade due to the development of lightweight high-pressure storage cylinders (Bianchi et al. 2014). Numerous researchers investigated compressed natural gas as an alternative fuel due to the economic, environmental, and strategic benefits of alternative fuels (Aslam et al. 2006; Cho and He 2007; Fino et al. 2006; Hill 2000; Najafi et al. 2009; Srinivasan et al. 2006). Multiple alternative fuels have been identified as having the potential to significantly reduce overall pollutant emissions compared to gasoline and diesel fuel. Natural gas, primarily composed of Methane, has been identified as the leading candidate for transportation applications among these fuels due to its availability, environmental compatibility, and compatibility with conventional diesel and gasoline engines.

Many previous studies investigated the natural gas combustion fundamentals utilizing experimental setups such as Particle Image Velocimetry (PIV) and Proper Orthogonal Decomposition (POD) as the optical measurement technique. The gained results from these studies focused on the relationship between the variation of cycle-to-cycle, HC, and CO emissions and varying swirl ratios (Vester, Nishio, and Alfredsson 2019; Zhang et al. 2019; Zhuang and Hung 2016). The analysis of A. Kalpakli Vester, Myoung-Seok Jie, and Bambang Wahono asserted that cylinder head geometry on swirling motion in the cylinder was strong; meanwhile, the value of opened valve lift was one of the parameters that affect the cycle-to-cycle variation (Wahono, Jwa, and Lim 2019; Yang et al. 2019). The research of Mohammed El-Adawy illustrated that at low valve lifts, the high-velocity vectors were concentrated behind the intake valves. At approximately 5 mm, the valve lift of the velocity distribution was symmetrical, resulting in no tumble motion. At a high valve lift, the air is intaken into the cylinder with a large amount, and it is directed toward the exhaust side, forming a strong jet from the left side of the valve seat. The interaction of this air jet with the left cylinder wall and then with the flat piston led, finally, to a strong tumbling motion within the cylinder (El-Adawy et al. 2017). In the case of a high tumble engine, brake-specific fuel consumption and HC were decreased, and NO_x was increased at part load; meanwhile, the BMEP and combustion peak pressure were increased at full load (Tanov et al. 2018). The research on tumble motion illustrated that tumble motion is needed to improve combustion efficiency by controlling the main burn speed in case of a high tumble ratio. It will increase flame burning rate, turbulence kinetic energy before ignition, brake specific fuel consumption. Meanwhile, HC emissions decreased, NOx increased at part load, and combustion peak pressure increased at full load (Benajes et al. 2016; Han and Reitz 1997; Liu and Dumitrescu 2018).

Ideally, converting a gasoline engine to a CNG engine is more manageable than a diesel engine. The original spark ignition system can ignite the CNG/air mixture without changing engine parameters (Sarkar and Bhattacharyya 2012). However, a converted CNG engine from gasoline engines does not take full advantage of natural gas characteristics with high-octane numbers. In addition, the low density of natural gas is the leading cause of the low power; its power is low even under stoichiometric operation (Tahir et al. 2015; Yang et al. 2017). As previously mentioned, the conversion of the diesel engine to operating by natural gas fuel is more attractive and suitable to the condition in Vietnam. In this research, the technical method to ignite the mixture of natural gas/air was spark ignition. The spark plug superseded the location of the diesel injector, and the CNG injector was inserted in the intake pipe. There are two parameters affecting combustion efficiency: combustion chamber geometry and compression ratio (Wang et al. 2019). Ordinarily, the cylinder head of the diesel engine is flat, so the combustion chamber geometry musters up all piston heads. Additionally, the combustion processes influence the inside and outside of bowl combustion by generating squish phenomena (Johansson and Olsson 1995; Kaplan 2019; Martinez et al. 2017).

This paper aims to increase the fundamental knowledge and understanding of the effect of bowl depth and center-bowl on the performance of converted CNG engines. The researched engine is a single-cylinder diesel engine for agriculture, modified and replaced with a natural gas spark-ignition engine. Three-piston bowl geometries were used and investigated at the same experimental conditions to achieve the experimental results and then to analyze and evaluate. Based on the experimental conditions and achieved results, the effect of three bowl-in-pistons on engine performance is investigated and evaluated. The calculated data from in-cylinder pressure according to the crank angle will be used to predict the flammability of the air/fuel mixture into the cylinder. The results infer from this study are one of the first steps to research and develop a CNG engine with high thermal efficiency and low emission in Vietnam.

Experimental setup and method

Experimental apparatus

Figure 1 shows an overview schematic diagram of the experimental setup. Initially, natural gas is stored in a high-pressure bomb around 150 bar and then goes through two pressure regulators to decompress the pressure until 50 bar and 4 bar. An MFC/MFM unit controls the natural gas supply to the engine, and a manometer measures the airflow rate. The converted CNG engine is a single-cylinder diesel engine for agriculture and is modified appropriately for the experiment. The converted engine specifications are shown in Table 1, and Figure 2 shows the test equipment, the location of the converted engine, and subsystems in the test room. The CNG injector is installed on the intake pipe and close to the intake valve. In addition, the location of the diesel injector is replaced by the spark plug.

Table 1. The specifications of the converted engine

Items	Specifications
Engine type	Port injection CNG engine
Bore x stroke (mm)	100 x 115
Compression ratio, ε	11.5 & 12.5
The shape of the piston head	Bowl in piston head (Piston bowl)
Cooling system	Water-cooled with pump
Fuel	CNG

Figure 1. Schematic diagram of the experimental setup.

Figure 2. Photograph of CNG converted engine on the testbed.

Another modification is the cooling system, an electrical pump used to supply the coolant water to the cylinder head and cylinder block separately. Two coolant flow valves were installed between the electrical pump and the converted engine to keep the temperature of coolant water out of the engine at approximately 70°C during the experiment. The air intake temperature is remained at around 27°C during the investigation. The dynamometer used in this study is AVL APA 100, corresponding to the maximum values of torque and power of about 849 Nm and 200 kW, respectively. A particular sensor is inserted in the cylinder head to record the cylinder pressure variation according to crank angles to calculate the specific parameters, as seen in Figure 2. This sensor includes a high accuracy AVL piezoelectric pressure transducer and a crank angle encoder model Kistler 2613B – GU12P. Furthermore, different sub-systems such as an oil system, a controlling engine unit, intake and exhaust systems, and the data acquisition system have been used to settle and control the engine performance.

Test procedure

To evaluate the effect of piston bowl geometry on the performance of the converted CNG engine, the original piston head was modified with the bowl depth (H_b = 10 and 17.5) and the position of the Center-bowl = 4. It is tantamount to compression ratios of converted CNG engine such as: ε = 11.5 (Center-bowl), ε = 12.5 (Center-bowl) and ε = 12.5 (Bowl-offset), as seen in Figure 3. As mentioned, to understand the effect of piston bowl geometry on the performance of the converted CNG engine, the experiments were performed at different engine speeds (1000, 1400, 1800, and 2000 rpm) for each piston model. The lambda is fixed at λ = 1 to obtain complete combustion and increases laminar flame speed (Huang et al. 2007; Vipavanich, Chuepeng, and Skullong 2018; Altın, Bilgin, and Çeper 2017; Mohammed et al. 2019).

Figure 3. The photo of three-piston bowl tops.

Results and discussions

Effect of piston bowl geometry on torque and power

Figure 4 shows the obtained results from experiments at varying fuel and spark ignition timing to achieve the maximum torque at each engine speed. The torque of the three piston-bowl types was similar when engine speed increased from 1000 rpm to 2000 rpm. The torque tended to increase in the speed range from 1000 to 1400 rpm; the maximum torque appears at 1400 rpm; however, the torque decreased when the engine operated at a higher speed. Firstly, in this case, the torque variation can be explained by a physical phenomenon called back-pressure; it appears inside the engine’s intake pipe. This phenomenon is growing up strongly inside the intake pipe of the CNG engine with a port injection when working at high speed. Since occurring back pressure, the volumetric efficiency decreases and results in engine torque decrease, as seen in the figure. However, the shortage of volumetric efficiency is compensated by the conformation of the piston bowl and the compression ratio. According to the previous studies, the increase in compression ratio will be enhanced thermal efficiency (Vester, Nishio, and Alfredsson 2019).

Figure 4. Variations of torque according to engine speed.

Meanwhile, the research on internal combustion engines has shown that increasing compression ratio enhances thermal efficiency and heat loss and the working portion of the compression process. Corresponding to the optimal compression ratio, improving thermal efficiency is ineligible to compensate for the operative part of the compression process and heat loss. The experimental results prove the effectiveness of this compression ratio as in the figure, the torque of ɛ = 12.5 (Center-bowl) is too smaller compared to ɛ = 11.5 (Center-bowl) at each engine speed. The effect of heat loss is more evident when comparing torque of ɛ = 12.5 (Center-bowl) to of ɛ = 12.5 (Bowl- offset), and the torque of ɛ = 12.5 (Bowl-offset) is smaller than that of other piston bowl geometries for all engine speeds. To the apparent effect of heat loss on engine torque, the subsequent experiments were performed at 1400 rpm and λ = 1. The results in Figure 5 indicated that both torque and power variations were similar when compression ratio (ɛ) and center-bowl (OB) positions changed. In the case of OB = 0 (the area of center-bowl coincided with centerline cylinder), increasing compression ratio will enhance both heat loss and the working portion of the compression process. Practically, the operative part of the compression process was improved insignificantly as the compression ratio increased from ɛ = 11.5 to ɛ = 12.5. The heat loss was the main factor impacting torque and power. The effect of heat loss on torque and power was evident as the results of ɛ = 12.5 (Offset-Bowl) are always smaller than ɛ = 12.5 (Center-bowl) even torque and power of ɛ = 11.5 (Center-bowl). From experimental results, one can generalize that the cause of the reduction in torque and power was heat loss and the work portion of the compression process. Still, the main factor predicted in this study was squish generation inside the cylinder. This study needs to use calculated parameters through the values of in-cylinder pressure according to crank angle for enquiring the prediction. In this research, the PUMA software will control and save data during the measurement process. Before the experiment, the user must be established the initial conditions for measuring systems, and a measurement step was 15 seconds. For this reason, the values of torque and power as seen in Figures 4 and 5 are confident. The research on lower compression ratios such as ε = 10 will be conducted in future research.

Figure 5. Variations of power and torque following piston bowl geometry.

Effect of piston bowl geometry on flammability

Figure 6 shows the comparison of in-cylinder pressure with the crank angle for the three-piston bowl geometries. The result again demonstrates the heat loss phenomenon in peak pressure of ɛ = 12.5 (Bowl-offset) is smaller than that of ɛ = 12.5 (Center-bowl) and ɛ = 11.5 (Center-bowl). The above result has affirmed that the OB increase will reduce the flammability of the mixture into the cylinder. However, flammability is significantly improved when the center-bowl is aligned with the cylinder’s centerline, even though the peak pressure of = 12.5 (Bowl-offset) is slightly greater than = 11.5 (Center-bowl). The main parameter that created this difference is the bowl depth (H_b); as modified piston bowl geometry, the Hb of ɛ = 12.5 (Bowl-offset) is smaller, so the squish velocity at near TDC is generated larger. The swirl motion has been dramatically improved by more significant squish velocity inside the cylinder and, as a result, is better flammability. The next step will consider the change in mass fraction burned to clarify the effect of swirl motion on mixing flammability. The interaction of the squish flow with the swirl motion was highlighted as having a significant impact on the axial flow configuration in the cylinder. In the presence of a swirl in the cylindrical bowl, the swirl formed by squish flow is reversed, and an extra vortex is generated in a reentrant bowl. It was proven that a double vortex is formed by the interaction between the squish flow and the swirl motion. Furthermore, the injection pressure affects both these two vortices’ size and relative intensity.

Figure 6. Variations of in-cylinder pressure according to crank angle.

Based on the obtained results from the experiment, this research presents the compression ratio limit of a horizontal single-cylinder converted engine. In addition, the research results were shown that the piston bowl geometry, such as bowl-offset type was unable to be used to spark ignition in natural gas engines. The cycle pressure is 100 cycles for ε = 11.5. However, ε = 12.5 must be adjusted to 50 cycles with the desire to catch the stable or knock pressures. The results were presented in Figures 7 and 8 (because there is no significant difference between the two results of the center-bowl and offset-bowl, and thus, we only showed a two-piston bowl configuration). The results were calculated from the in-cylinder pressure as shown in Figure 6, reflected according to the experimental conditions. The main difference here is that the cycle pressure of ε = 11.5 was longer than that of ε = 12.5.

Figure 7. Mass fraction burned at a varying crank angle.

Figure 8. Heat release rate at a varying crank angle.

Figure 7 shows mass fraction burned as functions of crank angle for two-piston bowl geometries. In the range of 0 to 0.9, the combustion duration of ɛ = 12.5 (Center-bowl) is shorter than that of ɛ = 11.5 (Center-bowl). In addition, it is longer than that in the range of 0.9 to 1. This variation is due to increasing squish when the compression ratio is enhanced; this result plays a vital role in confirming that the mixture inside the cylinder is more flammable when the swirl motion is increased. However, the increase in large squish generation impacted the combustion processes because the mass fraction burned of ɛ = 12.5 is longer in the range of 0.9 to 1. This result can be predicted that the heat loss of ɛ = 12.5 is more significant than that of ɛ = 11.5. For high-performance SI engines, squish is frequently useless as a turbulence augmentation strategy on its own. High degrees of swirl mixed with squish produce turbulence and a toroidal vortex in the bowl, both of which help mix the fuel and improve combustion.

Figure 8 shows the variation of heat release rate as a function of crank angle for two-piston bowl geometries. Observing the figure, one can see the variation of heat release rate was different. For ɛ = 12.5 (Center-bowl), the peak value of the heat release rate appeared near TDC, which indicated that the heat loss for the combustion chamber was quite significant. The reason for this increase in heat loss was the large squish generation.

Conclusion

In this study, the effect of piston bowl geometry on the performance of converted CNG engines has been investigated. Based on the experiment and calculation results, it is observed that the bowl depth and the position of the center bowl have strongly affected engine performance. For this converted CNG engine, the limitation of compression ratio is lower than that of ε = 12.5, and the location of the center-bowl must coincide with the centerline of the cylinder. The modified piston bowl geometry has improved both flammability and the heat loss of the air/fuel mixture inside the cylinder. The study shows that the improvement in flammability inside the cylinder is due to increased swirl velocity, but the increase in heat loss is due to large squish generation.

Nomenclature

Table

cp	Constant pressure specific heat [kJ/kg.K]
cv	Constant volume-specific heat [kJ/kg.K]
dQn/dθ	Heat release rate with crank angle [J/deg]
dQw/dθ	Heat transfer coefficient [J/m^2.K]
hc	Heat transfer rate with crank angle [J/deg]
m	Mass of cylinder gases [kg]
p	Cylinder gas pressure [Mpa]
T	Mean gas temperature [K]
R	Gas constant [J/kg.K]
Tw	Wall temperature [K]
V	Cylinder volume [m^3]
N	Wiebe form factor
a	Wiebe efficiency factor

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Quoc Dang Tran

Quoc Dang Tran received Ph.D. in Engineering (Thermal and Fluid Engineering) from Sungkyunkwan University, Korea in 2013. I currently work as a lecturer at the Faculty of Vehicle and Energy Conversion Engineering, School of Mechanical Engineering, Hanoi University of Science and Technology, Hanoi, Vietnam. Fields of interest: Alternative fuel for an internal combustion engine, Research and Development such as lean combustion system, internal combustion engine with gas fuels and Stirling engine.

Tam Thanh Tran

Tam Thanh Tran received a B.S. degree in mechanical engineering from Ho Chi Minh City University of Technology and Education, Vietnam, in 2004. He received an M.s degree in mechanical and automotive engineering at Military Technical Academy, Vietnam, in 2006. Since January 2019, he has been a lecturer at the Faculty of Mechanical Engineering, Nam Dinh University of Technology Education, Vietnam. Fields of interest: Alternative fuel for an internal combustion engine, Research and Development such as lean combustion system, internal combustion engine with gas fuels and Stirling engine

Vinh Nguyen Duy

Vinh Nguyen Duy received the B.S. and M.S. degrees in mechanical engineering from the Hanoi University of Science and Technology, Hanoi, Vietnam, in 2007 and 2011, respectively. He received a Ph.D. degree in mechanical and automotive engineering at Inje University, Gimhae, Republic of Korea, in 2016. From 2016 to 2018, Dr. Vinh worked as a Research Professor at the Power System & Sustainable Energy Lab, Inje University. Since January 2019, he has been a researcher in the Faculty of Vehicle and Energy Engineering, Phenikaa University, Hanoi, Vietnam. He has been the Team Leader of Sustainable Energy for Automobile Laboratory. His research interests include Internal combustion engines, power systems, fuel cells, electric vehicles, renewable energy and engine emissions, Kinetic energy harvesting, intelligent materials.