- Email: [email protected]
Activities for high-efficiency small gas engines
Applied Thermal Engineering xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Activities for high-efficiency small gas engines Yoshitane Takashima a,⇑, Satoshi Katayama a, Takahiro Sako a, Masahiro Furutani b a b
Osaka Gas Co., Ltd., 5-11-61 Torishima, Konohana-ku, Osaka 554-0051, Japan Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, Aichi 466-8555, Japan
h i g h l i g h t s Our research is on various methods of improving the efficiency of small gas engines. In the short term, turbocharged stoichiometric combustion with EGR will be the best technology. In the medium-to-long term, it will be more effective to develop lower-temperature combustion. Turbocharged HCCI and D-EGR are the options with increased EGR ratio.
a r t i c l e
i n f o
Article history: Received 28 March 2016 Revised 5 October 2016 Accepted 26 October 2016 Available online xxxx Keywords: Forced induction Multi-point ignition Pre-chamber spark plug HCCI Dedicated EGR system
a b s t r a c t Interest in the potential of natural gas is growing, owing to its low CO2 emissions per unit of heat produced, and to the development of techniques for the exploitation of shale gas. Research and development relating to natural gas is also being pursued in view of its potential to increase Japan’s energy security, because deposits are found around the world. Osaka Gas is pursuing the medium-to-long term development of gas engines. Increasing compression ratio, improving combustion under lean combustion conditions, and increasing specific output will be key to improving thermal efficiency. This paper summarizes the results of tests Osaka Gas to date with a view to improving the efficiency of small gas engines. First, tests with downsizing through forced induction were conducted. A naturally aspirated gas engine with a displacement of 3318 cm3 was fitted with a high-efficiency turbocharger, and performance tests were conducted at brake mean effective pressure (BMEP). It was found that thermal efficiency under lean conditions reached 40%, but that NOx emissions exceeded 1500 ppm. When exhaust gas recirculation (EGR) was applied with a view to reducing NOx emissions, at the EGR limit (EGR ratio 18%), thermal efficiency was around 39% (NOx 500 ppm). With stoichiometric combustion, at the EGR limit, thermal efficiency reached a maximum of 39%. Next, the effectiveness of multi-point ignition, pre-chamber spark plugs and homogeneous charge compression ignition (HCCI) in improving lean/diluted combustion were studied, and each was found to improve stability, extend the lean limit and increase thermal efficiency under lean combustion conditions. Studies were also conducted in relation to the dedicated EGR system proposed by Southwest Research Institute (SwRI) as a means of improving EGR. It was determined that around 9% H2, which has a combustion-promoting effect, was produced under conditions where the equivalence ratio was 1.5. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Micro-cogeneration originally spread in Japan with a view to achieving energy and cost savings, and reducing peak power demand. Furthermore, in recent years, the emphasis has been on convenience, including the development of systems able to ensure ⇑ Corresponding author.
a power supply during outages, using LPG fuel and air to supply electricity for a short time in an emergency, or running on biogas fuels. Generating efficiency has also improved, with 50 kW-class systems achieving an efficiency of 34% [1]. These improvements are largely due to improvements in the efficiency of gas engines, in which lean combustion has played a particularly important role. However, further improvements will be required to promote wider use of natural gas in the future. For instance, technical guidelines issued by the Japan Gas Association call for the efficiency of small
E-mail address: [email protected] (Y. Takashima). http://dx.doi.org/10.1016/j.applthermaleng.2016.10.166 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
2
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx Table 1 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed Maximum pressure
4-cylinder water-cooled 110 mm 98 mm 13:1 3318 cm3 1500 rpm 10 MPa
gas engines with an output of 100 kW or less to be increased to 45% by 2030 [2]. Osaka Gas is pursuing the research and development of gas engines with a view to achieving this target. This paper summarizes the results of Osaka Gas’s work on improving the efficiency of small gas engines to date, and sets out directions for future technical development.
2. Technologies to improve the efficiency of small gas engines Improving thermal efficiency will depend on, among other measures, increasing compression ratio, improving combustion under lean combustion conditions, and increasing specific output (downsizing). This section describes the results of Osaka Gas’s research on these technologies to date.
2.1. Downsizing through forced induction This technology aims to improve thermal efficiency by forced induction of fuel-air pre-mixture using a turbocharger, and increasing the specific output of the cylinder. This study examined the performance of a naturally aspirated gas engine with a displacement of 3318 cm3 (equivalent to 25 kW generated output) fitted with a high-efficiency turbocharger. Stoichiometric and lean combustion were compared, and the effect of exhaust gas recirculation (EGR) on each was examined. Table 1 shows the specifications of the test engine and maximum pressure limit. Fig. 1 shows the experimental setup. Engine speed was 1500 rpm and brake mean effective pressure (BMEP) was 1.33 MPa (equivalent to 50 kW generated output). In every case, maximum brake torque (MBT) ignition timing was used, and the fluctuation of maximum in-cylinder pressure and indicated mean effective pressure was within the specified range.
Fig. 2 shows the relationship between change in thermal efficiency and NOx concentration. Lean combustion showed greater thermal efficiency than stoichiometric combustion, reaching 39.7%. However, although the thermal efficiency of stoichiometric combustion is lower, it allows the reduction of NOx emissions using a three-way catalyst. Since three-way catalysts are unsuitable for lean combustion, it is necessary to use other NOx reduction technology. Urea selective catalytic reduction is a possible way to reduce NOx emissions, but issues such as the method of providing the supply of urea make it impractical. It is also possible to use storage and reduction catalysts, but these use fuel as the reducing agent, which would lead to a fall in thermal efficiency. EGR was identified as a potentially simple and convenient means of NOx reduction. Increasing the EGR ratio caused a fall in NOx concentration, but even at the EGR limit of 18%, the NOx concentration was 500 ppm, and there is scope for further research. The use of EGR also reduced thermal efficiency. EGR was identified as a potential means of increasing thermal efficiency with stoichiometric combustion, because three-way catalysts can be used for NOx reduction. The Figure shows that NOx emissions fell as EGR ratio increased, as with lean combustion, but thermal efficiency increased. This was largely due to a reduction in cooling loss, caused by the fall in gas temperature in the combustion chamber. Thus, downsizing through forced induction increases thermal efficiency. However, with highly efficient lean combustion, NOx reduction becomes an issue, while with stoichiometric combustion, NOx reduction is unnecessary, but thermal efficiency is low compared with lean combustion. It is therefore likely to be necessary to adopt a combination of stoichiometric combustion and three-way catalysts in the short term, and to pursue NOx reduction technology compatible with lean combustion in the medium-tolong term. In these tests, output was limited to 50 kW. Increasing output further would make it possible to increase thermal efficiency, but it would be necessary to avoid knocking in engine.
2.2. Improving lean/diluted combustion As stated in the previous section, lean combustion will be the best option for increasing thermal efficiency, but there are issues with NOx reduction. If lean combustion can be further improved, NOx post-processing will become unnecessary, but stable ignition and combustion will be necessary if the lean limit is to be extended. For this reason, the use of multi-point ignition, pre-
Fig. 1. Schematic diagram of experimental setup.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
3
(Lambda > 1.7) and extended the lean limit. During stoichiometric combustion, when combustion is stable, increasing the number of ignition points produced no difference in indicated thermal efficiency, but as the mixture became leaner, indicated thermal efficiency rose as the number of ignition points increased.
Fig. 2. Effect of lean burn and stoichiometric combustion on thermal efficiency and NOx under supercharged.
Table 2 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed
Single-cylinder water-cooled 106 mm 110 mm 9.5:1 1007 cm3 1200 rpm
chamber spark plugs and homogeneous charge compression ignition (HCCI) was studied. 2.2.1. Multi-point ignition Multi-point ignition technology is designed to stabilize lean combustion. Mounting multiple spark plugs in the cylinder head, etc. to ignite the gas mixture, increases ignition probability and reduces flame propagation distance. The tests used a single-cylinder horizontal water-cooled diesel engine, with the cylinder head modified to allow it to run on natural gas. The specifications of the test engine are shown in Table 2. As shown in Fig. 3, up to 9 spark plugs were fitted in the bore center and around the liner [3]. First, the effect of increasing the number of ignition points on the combustion characteristics and performance of the natural gas engine was examined. Tests were conducted with 1, 5 and 9 ignition points, placed as shown in Fig. 4. Fig. 5 shows the relationship between air excess ratio, coefficient of variance of indicated mean effective pressure (COV(IMEP)) and indicated thermal efficiency. Increasing the number of ignition points from 1 to 5 or 9 reduced combustion fluctuation under lean conditions
2.2.2. Pre-chamber spark plugs Medium and large gas engines operate stably even under highcompression forced induction and very lean conditions. This is because these engines use a pre-combustion chamber, and the flame jet from the pre-combustion chamber ignites a lean airfuel pre-mixture in the main combustion chamber. This combustion method is also effective in small gas engines, but problematic in layout terms because, for instance, if a relatively large-capacity pre-chamber is fitted, the system required for the supply of fuel to the pre-chamber becomes complex. For this reason, it was decided to examine the use of a pre-chamber spark plug, which was expected to give similar benefits to those of a pre-chamber, with a relatively simple mechanism. As shown in Fig. 6, the mechanism involved was very simple, with the pre-chamber tip covering the spark plug electrode [4]. During the intake stroke, natural gas pre-mixture is aspirated and combustion exhaust gas is scavenged from the pre-chamber. During the compression stroke, piston compression supplies pre-mixture to the pre-chamber, and ignites it. Injecting the high-temperature gas from the pre-chamber into the main chamber through the orifice ignites the pre-mixture. The specifications of the test engine are shown in Table 3. The operating conditions were 1200 rpm and a constant fuel flow rate (0.64 N m3/h: equivalent to an indicated mean effective pressure of around 500 kPa). The basic characteristics of a pre-chamber plug were determined using a standard spark plug, as shown in the left-hand drawing of Fig. 6, and a pre-chamber plug, as shown in the right-hand drawing. Fig. 7 shows the cumulative heat release (Qt), at the same air excess ratio and ignition timing, for the prechamber plug and the standard plug. With the pre-chamber plug, the increase in heat release rate began earlier, and combustion duration, defined as the period between 10% of total heat release and 90% of total heat release (h10-90), was shorter. Fig. 8 shows the relationship between air excess ratio and COV(IMEP). At the same air excess ratio, the pre-chamber plug showed a smaller COV(IMEP) than the standard plug and the lean limit was extended. Residual gas, flow, gas temperature and other factors influence ignition in the pre-chamber, but it was thought likely that ambient temperature in the pre-chamber in particular contributed to ignition and had an effect on combustion stability in lean conditions. It was therefore decided to evaluate this factor using prechamber plugs made of three materials with different thermal conductivity: stainless steel, steel and aluminum. Table 4 shows the thermal conductivity of the three materials used in this test. The
Fig. 3. Schematic diagram of multi-ignition engine.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
4
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Fig. 4. Placement of ignition points in the cylinder.
Standard plug Pre-chamber plug
One-point ignition Five-point ignition Nine-point ignition
Fig. 7. Effect of pre-chamber plug on cumulative heat release.
Fig. 5. Effect of multi-ignition on COV(IMEP) and thermal efficiency.
Standard plug Pre-chamber plug Miss fire
Fig. 8. Effect of pre-chamber plug on COV(IMEP). Fig. 6. Schematic diagrams of standard plug and pre-chamber plug. Table 4 Pre-chamber material.
Table 3 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed
Single-cylinder water-cooled 96.9 mm 85 mm 12:1 550 cm3 1200 rpm
specifications of the pre-chamber plugs were as shown in Fig. 9 and Table 5. Fig. 10 shows the rate of heat release profile with for prechamber plugs made of stainless steel, steel and aluminum, at the same air excess ratio and ignition timing. Stainless steel and steel showed similar heat release rates, but the rate of increase in heat release was slightly slower with aluminum. However, when
Material
Type
Heat conductivity
Stainless steel Steel Aluminum
SUS301S S45C A7075
16.3 W/(mK) 45 W/(mK) 130 W/(mK)
the graphs were aligned to so that the beginning of heat release coincided, the shape of the heat release profiles was largely similar. Fig. 11 shows the distribution of duration between ignition and the timing of 5% of total heat release plotted against crank angle. The distribution in the direction of advance of crank angle was in the order stainless steel, steel, aluminum. This is likely to be because pre-chamber material affects heat release in the initial stage of combustion. Fig. 12 shows the relationship between air excess ratio and COV(IMEP) for each material. Like the order of distribution of the
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
5
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Stainless steel Steel Aluminum Miss fire
Fig. 9. Schematic of pre-chamber plug. Fig. 12. Effect of pre-chamber material on COV(IMEP). Table 5 Pre-chamber specifications. Number of orifices (n) Diameter of orifice (dn) Angle of orifice axis (un) Pre-chamber volume (Vc)
5 1.2 mm 60° 400 mm3
Stainless steel Steel Aluminum
Fig. 10. Effect of pre-chamber material on heat release ratio.
λ
θ Stainless steel Steel Aluminum
θ Fig. 11. Effect of pre-chamber material on frequency of duration between ignition timing and timing of 5% of total heat release.
initial stage of heat release, this differed significantly according to the material, aluminum giving a much earlier increase in combustion fluctuation than the other materials. This is likely to be due mainly to the fact that aluminum produces the greatest fall in temperature in the pre-chamber. The results of temperature measurements in the pre-chamber confirm that the lower the heat transfer rate of the pre-chamber material, the higher the average temperature and the faster the heat release in the initial stage.
The issue for the future will be to identify the factors that extend the lean limit with a pre-chamber plug, and use this information in designing plugs. 2.2.3. Homogeneous charge compression ignition Homogeneous charge compression ignition (HCCI) may reduce NOx emissions in diesel engines and improve the efficiency of gasoline engines in the low output range [5–7]. Because HCCI combustion is suitable for a wide range of different fuels, its use is also being considered with alternative fuels such as natural gas [8–10] and methanol [11]. Table 6 compares a spark ignition engine and a HCCI engine. In HCCI combustion, adiabatic compression of the pre-mixture raises the gas in the cylinder to auto-ignition temperature and causes bulk combustion. The advantage of this method is that local density distribution and temperature distribution in the cylinder are narrow, while compression ratios are higher than in spark ignition engines, making it possible to achieve lean combustion beyond the flame propagation limit and to combine high efficiency with low NOx emissions. On the other hand, a tendency to misfiring, combustion fluctuation and knocking tends to limit the load range under which HCCI engines can operate, and research has suggested that it is difficult to achieve HCCI combustion under a wide load range [12]. Cogeneration systems developed in recent years have been connected to electric power systems through inverter links, allowing them to operate virtually at rated output, and have a very narrow range of load variance. HCCI engines therefore have great potential as power sources for co-generation. Moreover, the main component of natural gas is methane (CH4), the characteristics of which make it an ideal fuel for HCCI combustion. In other words, using an HCCI engine fueled by natural gas as the power source for co-generation is likely to make it possible to achieve economy equivalent to that of diesel, together with reduced pollutant emissions. The discussion in this report will not cover combustion characteristics in turbocharged HCCI engines, or their tendencies, or the prototype small high-efficiency turbocharger that was built [13], and will be confined to thermal efficiency. Fig. 13 compares the performance of a naturally aspirated HCCI engine, a turbocharged HCCI engine, and a spark ignition engine. The generating efficiency of the power generation system was estimated at 95%. The operating conditions for the naturally aspirated HCCI engine were an Table 6 Compares a spark ignition engine and a naturally aspirated HCCI engine. Engine type Power generating efficiency (power generation output) Heat recovery efficiency (Heat output) Overall efficiency
SI 33.0% (25 kW)
HCCI 37.0% (25 kW)
52.0% (39 kW)
41.0% (28 kW)
85.0%
78.0%
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
6
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Fig. 13. Compares the performance.
Table 7 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed
2.3.1. D-EGR system This system is proposed by Southwest Research Institute (SwRI) [14], and involves dedicating one of multiple cylinders to EGR. Combustion using this cylinder under fuel-rich conditions cylinder produces H2 and CO. These are distributed to each cylinder, to increase combustion speed even in a high-EGR atmosphere [15], with a view to achieving stable operation. Osaka Gas has identified this system as being of interest and is studying its use with natural gas engines also by using forced induction condition to enhanced efficiency. The specifications of the test engine are shown in Table 7. Table 8 shows the components produced in a fuel-rich atmosphere in the single-cylinder test engine. In an atmosphere with an equivalence ratio of 1.5, it was found that around 9% H2 was emitted. Osaka Gas plans to continue studying the effect of using the H2 obtained to improve combustibility and thermal efficiency in a high-EGR atmosphere. 3. Summary
Single-cylinder water-cooled 106 mm 110 mm 12:1 1007 cm3 1200 rpm
Table 8 Gas composition in a fuel-rich atmosphere. Component
Emission rate
H2 CO
8.84% 8.97%
engine compression ratio of 25.9 and an engine speed of 1700 rpm; the conditions for forced induction operation were an engine compression ratio of 21, a forced induction compression ratio of 1.9 and an engine speed of 1900 rpm. The generating efficiency of the naturally aspirated HCCI engine was around 5 points higher than that of a conventional gas engine. The turbocharged HCCI engine had very high potential. Increasing air density roughly doubled BMEP (equivalent to specific output in output per unit displacement), and generating efficiency was around 40%. HCCI engines are more expensive because of the need for combustion control systems, but have the advantage that increasing engine output through turbocharging directly reduces initial costs, and that devices to raise intake air temperature in order to ignite the fuel are not required. Thus, HCCI engines have great potential toward customers of a low thermoelectric ratio are demanding cogeneration systems with high generation efficiency, but their commercial development will require the establishment of low-cost combustion status detection techniques, operation control systems including engine start and load input method, verification of durability, etc. 2.3. Improving EGR Low-temperature combustion is being studied with reference to automobile engines. This combustion method reduces heat loss by inducting a large volume of exhaust gas (EGR) and reducing flame temperature. The issue with this technology is to achieve stable combustion of a lean mixture composed of a large volume of exhaust gas, air and fuel. This section describes studies of technologies to improve combustion stability.
These are the results of our research on various methods of improving the efficiency of small gas engines. In the short term, turbocharged stoichiometric combustion with EGR will be the best technology for achieving thermal efficiency combined with low emissions. In the medium-to-long term, however, it will be more effective to develop lower-temperature combustion options with increased EGR ratio, such as turbocharged HCCI and D-EGR. Osaka Gas will continue to pursue the development of gas engines capable of attaining the targets set out by the Japan Gas Association. References [1] 35 kW Micro-Cogeneration (In Japanese), Osaka Gas Co., LTD., (accessed 201505-31). [2] Efforts of the city gas industry in light of the direction of the future of energy polic (in Japanese), The Japan Gas Association, (accessed 2015-05-31). [3] Y. Takashima, H. Tanaka, T. Sako, Evaluation of the Effects of Combustion by Multi-Ignition in Natural Gas Engines, SAE Technical Paper 2012-32-0065, 2012. [4] Y. Takashima, H. Tanaka, T. Sako, Evaluation of engine performance and combustion in natural gas engine with pre-chamber plug under lean burn conditions, SAE Int. J. Eng. 8 (1) (2015) 221–229. [5] P.M. Najt, D.E. Foster, Compression-ignited Charge Combustion, SAE Paper 830264, 1983. [6] T. Thring, Homogeneous-Charge Compression-Ignition (HCCI) Engines, SAE Technical Paper 892068, 1989. [7] M. Christensen, B. Johansson, P. AmnJus, F. Mauss, Supercharged Homogeneous Charge Compression Ignition, SAE Paper 980787, 1998. [8] J. Deasu, N. Iida, A Study of High Combustion Efficiency and Low CO Emission in a Natural Gas HCCI Engine, SAE Paper 2004-01-1974, 2004. [9] T. Sako, S. Nakai, K. Moriya, N. Iida, Performance and exhaust emission in a natural-gas fueled homogeneous charge compression ignition engine, Trans. JSME B 70 (694) (2004) 1583–1589 (In Japanese) 2004-06-25. [10] S. Morimoto, Y. Kawabata, T. Sakurai, T. Amano, Operating characteristics of a natural gas-fired homogeneous charge compression ignition engine (Performance Improvement Using EGR), SAE Paper 2001-01-1034, 2001. [11] N. Iida, Combustion Analysis of Methanol-Fueled Active Thermo-Atmosphere Combustion (ATAC) Engine Using a Spectroscopic Observation, SAE Technical Paper 940684, 1994. [12] T. Ishiyama, M. Shioji, H. Tanaka, H. Nakagawa, S. Nakai, Performance and exhaust emissions of a premixed charge compression ignition engine fueled with natural gas employing direct injection, JSAE Trans. 34 (3) (2003) 29–34 (In Japanese), 2003-07-15. [13] T. Sako, S. Morimoto, N. Iida, Performance improvement for natural gas fueled HCCI engine with turbocharger (second report), Trans. Soc. Autom. Eng. Jpn. 38 (5) (2007) 131–136, 2007-09-25. [14] T. Alger, B. Mangold, Dedicated EGR: a new concept in high efficiency engines, SAE Int. J. Engines 2 (1) (2009) 620–631. [15] K. Ozaki, T. Sako, S. Lee, N. Iida, Potential for a high efficiency of natural gas SI engine by ‘‘Dedicated EGR”, JSAE Trans. 45 (5) (2014) 769–774 (In Japanese), 2014–09.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Activities for high-efficiency small gas engines Yoshitane Takashima a,⇑, Satoshi Katayama a, Takahiro Sako a, Masahiro Furutani b a b
Osaka Gas Co., Ltd., 5-11-61 Torishima, Konohana-ku, Osaka 554-0051, Japan Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, Aichi 466-8555, Japan
h i g h l i g h t s Our research is on various methods of improving the efficiency of small gas engines. In the short term, turbocharged stoichiometric combustion with EGR will be the best technology. In the medium-to-long term, it will be more effective to develop lower-temperature combustion. Turbocharged HCCI and D-EGR are the options with increased EGR ratio.
a r t i c l e
i n f o
Article history: Received 28 March 2016 Revised 5 October 2016 Accepted 26 October 2016 Available online xxxx Keywords: Forced induction Multi-point ignition Pre-chamber spark plug HCCI Dedicated EGR system
a b s t r a c t Interest in the potential of natural gas is growing, owing to its low CO2 emissions per unit of heat produced, and to the development of techniques for the exploitation of shale gas. Research and development relating to natural gas is also being pursued in view of its potential to increase Japan’s energy security, because deposits are found around the world. Osaka Gas is pursuing the medium-to-long term development of gas engines. Increasing compression ratio, improving combustion under lean combustion conditions, and increasing specific output will be key to improving thermal efficiency. This paper summarizes the results of tests Osaka Gas to date with a view to improving the efficiency of small gas engines. First, tests with downsizing through forced induction were conducted. A naturally aspirated gas engine with a displacement of 3318 cm3 was fitted with a high-efficiency turbocharger, and performance tests were conducted at brake mean effective pressure (BMEP). It was found that thermal efficiency under lean conditions reached 40%, but that NOx emissions exceeded 1500 ppm. When exhaust gas recirculation (EGR) was applied with a view to reducing NOx emissions, at the EGR limit (EGR ratio 18%), thermal efficiency was around 39% (NOx 500 ppm). With stoichiometric combustion, at the EGR limit, thermal efficiency reached a maximum of 39%. Next, the effectiveness of multi-point ignition, pre-chamber spark plugs and homogeneous charge compression ignition (HCCI) in improving lean/diluted combustion were studied, and each was found to improve stability, extend the lean limit and increase thermal efficiency under lean combustion conditions. Studies were also conducted in relation to the dedicated EGR system proposed by Southwest Research Institute (SwRI) as a means of improving EGR. It was determined that around 9% H2, which has a combustion-promoting effect, was produced under conditions where the equivalence ratio was 1.5. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Micro-cogeneration originally spread in Japan with a view to achieving energy and cost savings, and reducing peak power demand. Furthermore, in recent years, the emphasis has been on convenience, including the development of systems able to ensure ⇑ Corresponding author.
a power supply during outages, using LPG fuel and air to supply electricity for a short time in an emergency, or running on biogas fuels. Generating efficiency has also improved, with 50 kW-class systems achieving an efficiency of 34% [1]. These improvements are largely due to improvements in the efficiency of gas engines, in which lean combustion has played a particularly important role. However, further improvements will be required to promote wider use of natural gas in the future. For instance, technical guidelines issued by the Japan Gas Association call for the efficiency of small
E-mail address: [email protected] (Y. Takashima). http://dx.doi.org/10.1016/j.applthermaleng.2016.10.166 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
2
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx Table 1 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed Maximum pressure
4-cylinder water-cooled 110 mm 98 mm 13:1 3318 cm3 1500 rpm 10 MPa
gas engines with an output of 100 kW or less to be increased to 45% by 2030 [2]. Osaka Gas is pursuing the research and development of gas engines with a view to achieving this target. This paper summarizes the results of Osaka Gas’s work on improving the efficiency of small gas engines to date, and sets out directions for future technical development.
2. Technologies to improve the efficiency of small gas engines Improving thermal efficiency will depend on, among other measures, increasing compression ratio, improving combustion under lean combustion conditions, and increasing specific output (downsizing). This section describes the results of Osaka Gas’s research on these technologies to date.
2.1. Downsizing through forced induction This technology aims to improve thermal efficiency by forced induction of fuel-air pre-mixture using a turbocharger, and increasing the specific output of the cylinder. This study examined the performance of a naturally aspirated gas engine with a displacement of 3318 cm3 (equivalent to 25 kW generated output) fitted with a high-efficiency turbocharger. Stoichiometric and lean combustion were compared, and the effect of exhaust gas recirculation (EGR) on each was examined. Table 1 shows the specifications of the test engine and maximum pressure limit. Fig. 1 shows the experimental setup. Engine speed was 1500 rpm and brake mean effective pressure (BMEP) was 1.33 MPa (equivalent to 50 kW generated output). In every case, maximum brake torque (MBT) ignition timing was used, and the fluctuation of maximum in-cylinder pressure and indicated mean effective pressure was within the specified range.
Fig. 2 shows the relationship between change in thermal efficiency and NOx concentration. Lean combustion showed greater thermal efficiency than stoichiometric combustion, reaching 39.7%. However, although the thermal efficiency of stoichiometric combustion is lower, it allows the reduction of NOx emissions using a three-way catalyst. Since three-way catalysts are unsuitable for lean combustion, it is necessary to use other NOx reduction technology. Urea selective catalytic reduction is a possible way to reduce NOx emissions, but issues such as the method of providing the supply of urea make it impractical. It is also possible to use storage and reduction catalysts, but these use fuel as the reducing agent, which would lead to a fall in thermal efficiency. EGR was identified as a potentially simple and convenient means of NOx reduction. Increasing the EGR ratio caused a fall in NOx concentration, but even at the EGR limit of 18%, the NOx concentration was 500 ppm, and there is scope for further research. The use of EGR also reduced thermal efficiency. EGR was identified as a potential means of increasing thermal efficiency with stoichiometric combustion, because three-way catalysts can be used for NOx reduction. The Figure shows that NOx emissions fell as EGR ratio increased, as with lean combustion, but thermal efficiency increased. This was largely due to a reduction in cooling loss, caused by the fall in gas temperature in the combustion chamber. Thus, downsizing through forced induction increases thermal efficiency. However, with highly efficient lean combustion, NOx reduction becomes an issue, while with stoichiometric combustion, NOx reduction is unnecessary, but thermal efficiency is low compared with lean combustion. It is therefore likely to be necessary to adopt a combination of stoichiometric combustion and three-way catalysts in the short term, and to pursue NOx reduction technology compatible with lean combustion in the medium-tolong term. In these tests, output was limited to 50 kW. Increasing output further would make it possible to increase thermal efficiency, but it would be necessary to avoid knocking in engine.
2.2. Improving lean/diluted combustion As stated in the previous section, lean combustion will be the best option for increasing thermal efficiency, but there are issues with NOx reduction. If lean combustion can be further improved, NOx post-processing will become unnecessary, but stable ignition and combustion will be necessary if the lean limit is to be extended. For this reason, the use of multi-point ignition, pre-
Fig. 1. Schematic diagram of experimental setup.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
3
(Lambda > 1.7) and extended the lean limit. During stoichiometric combustion, when combustion is stable, increasing the number of ignition points produced no difference in indicated thermal efficiency, but as the mixture became leaner, indicated thermal efficiency rose as the number of ignition points increased.
Fig. 2. Effect of lean burn and stoichiometric combustion on thermal efficiency and NOx under supercharged.
Table 2 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed
Single-cylinder water-cooled 106 mm 110 mm 9.5:1 1007 cm3 1200 rpm
chamber spark plugs and homogeneous charge compression ignition (HCCI) was studied. 2.2.1. Multi-point ignition Multi-point ignition technology is designed to stabilize lean combustion. Mounting multiple spark plugs in the cylinder head, etc. to ignite the gas mixture, increases ignition probability and reduces flame propagation distance. The tests used a single-cylinder horizontal water-cooled diesel engine, with the cylinder head modified to allow it to run on natural gas. The specifications of the test engine are shown in Table 2. As shown in Fig. 3, up to 9 spark plugs were fitted in the bore center and around the liner [3]. First, the effect of increasing the number of ignition points on the combustion characteristics and performance of the natural gas engine was examined. Tests were conducted with 1, 5 and 9 ignition points, placed as shown in Fig. 4. Fig. 5 shows the relationship between air excess ratio, coefficient of variance of indicated mean effective pressure (COV(IMEP)) and indicated thermal efficiency. Increasing the number of ignition points from 1 to 5 or 9 reduced combustion fluctuation under lean conditions
2.2.2. Pre-chamber spark plugs Medium and large gas engines operate stably even under highcompression forced induction and very lean conditions. This is because these engines use a pre-combustion chamber, and the flame jet from the pre-combustion chamber ignites a lean airfuel pre-mixture in the main combustion chamber. This combustion method is also effective in small gas engines, but problematic in layout terms because, for instance, if a relatively large-capacity pre-chamber is fitted, the system required for the supply of fuel to the pre-chamber becomes complex. For this reason, it was decided to examine the use of a pre-chamber spark plug, which was expected to give similar benefits to those of a pre-chamber, with a relatively simple mechanism. As shown in Fig. 6, the mechanism involved was very simple, with the pre-chamber tip covering the spark plug electrode [4]. During the intake stroke, natural gas pre-mixture is aspirated and combustion exhaust gas is scavenged from the pre-chamber. During the compression stroke, piston compression supplies pre-mixture to the pre-chamber, and ignites it. Injecting the high-temperature gas from the pre-chamber into the main chamber through the orifice ignites the pre-mixture. The specifications of the test engine are shown in Table 3. The operating conditions were 1200 rpm and a constant fuel flow rate (0.64 N m3/h: equivalent to an indicated mean effective pressure of around 500 kPa). The basic characteristics of a pre-chamber plug were determined using a standard spark plug, as shown in the left-hand drawing of Fig. 6, and a pre-chamber plug, as shown in the right-hand drawing. Fig. 7 shows the cumulative heat release (Qt), at the same air excess ratio and ignition timing, for the prechamber plug and the standard plug. With the pre-chamber plug, the increase in heat release rate began earlier, and combustion duration, defined as the period between 10% of total heat release and 90% of total heat release (h10-90), was shorter. Fig. 8 shows the relationship between air excess ratio and COV(IMEP). At the same air excess ratio, the pre-chamber plug showed a smaller COV(IMEP) than the standard plug and the lean limit was extended. Residual gas, flow, gas temperature and other factors influence ignition in the pre-chamber, but it was thought likely that ambient temperature in the pre-chamber in particular contributed to ignition and had an effect on combustion stability in lean conditions. It was therefore decided to evaluate this factor using prechamber plugs made of three materials with different thermal conductivity: stainless steel, steel and aluminum. Table 4 shows the thermal conductivity of the three materials used in this test. The
Fig. 3. Schematic diagram of multi-ignition engine.
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
4
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Fig. 4. Placement of ignition points in the cylinder.
Standard plug Pre-chamber plug
One-point ignition Five-point ignition Nine-point ignition
Fig. 7. Effect of pre-chamber plug on cumulative heat release.
Fig. 5. Effect of multi-ignition on COV(IMEP) and thermal efficiency.
Standard plug Pre-chamber plug Miss fire
Fig. 8. Effect of pre-chamber plug on COV(IMEP). Fig. 6. Schematic diagrams of standard plug and pre-chamber plug. Table 4 Pre-chamber material.
Table 3 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed
Single-cylinder water-cooled 96.9 mm 85 mm 12:1 550 cm3 1200 rpm
specifications of the pre-chamber plugs were as shown in Fig. 9 and Table 5. Fig. 10 shows the rate of heat release profile with for prechamber plugs made of stainless steel, steel and aluminum, at the same air excess ratio and ignition timing. Stainless steel and steel showed similar heat release rates, but the rate of increase in heat release was slightly slower with aluminum. However, when
Material
Type
Heat conductivity
Stainless steel Steel Aluminum
SUS301S S45C A7075
16.3 W/(mK) 45 W/(mK) 130 W/(mK)
the graphs were aligned to so that the beginning of heat release coincided, the shape of the heat release profiles was largely similar. Fig. 11 shows the distribution of duration between ignition and the timing of 5% of total heat release plotted against crank angle. The distribution in the direction of advance of crank angle was in the order stainless steel, steel, aluminum. This is likely to be because pre-chamber material affects heat release in the initial stage of combustion. Fig. 12 shows the relationship between air excess ratio and COV(IMEP) for each material. Like the order of distribution of the
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
5
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Stainless steel Steel Aluminum Miss fire
Fig. 9. Schematic of pre-chamber plug. Fig. 12. Effect of pre-chamber material on COV(IMEP). Table 5 Pre-chamber specifications. Number of orifices (n) Diameter of orifice (dn) Angle of orifice axis (un) Pre-chamber volume (Vc)
5 1.2 mm 60° 400 mm3
Stainless steel Steel Aluminum
Fig. 10. Effect of pre-chamber material on heat release ratio.
λ
θ Stainless steel Steel Aluminum
θ Fig. 11. Effect of pre-chamber material on frequency of duration between ignition timing and timing of 5% of total heat release.
initial stage of heat release, this differed significantly according to the material, aluminum giving a much earlier increase in combustion fluctuation than the other materials. This is likely to be due mainly to the fact that aluminum produces the greatest fall in temperature in the pre-chamber. The results of temperature measurements in the pre-chamber confirm that the lower the heat transfer rate of the pre-chamber material, the higher the average temperature and the faster the heat release in the initial stage.
The issue for the future will be to identify the factors that extend the lean limit with a pre-chamber plug, and use this information in designing plugs. 2.2.3. Homogeneous charge compression ignition Homogeneous charge compression ignition (HCCI) may reduce NOx emissions in diesel engines and improve the efficiency of gasoline engines in the low output range [5–7]. Because HCCI combustion is suitable for a wide range of different fuels, its use is also being considered with alternative fuels such as natural gas [8–10] and methanol [11]. Table 6 compares a spark ignition engine and a HCCI engine. In HCCI combustion, adiabatic compression of the pre-mixture raises the gas in the cylinder to auto-ignition temperature and causes bulk combustion. The advantage of this method is that local density distribution and temperature distribution in the cylinder are narrow, while compression ratios are higher than in spark ignition engines, making it possible to achieve lean combustion beyond the flame propagation limit and to combine high efficiency with low NOx emissions. On the other hand, a tendency to misfiring, combustion fluctuation and knocking tends to limit the load range under which HCCI engines can operate, and research has suggested that it is difficult to achieve HCCI combustion under a wide load range [12]. Cogeneration systems developed in recent years have been connected to electric power systems through inverter links, allowing them to operate virtually at rated output, and have a very narrow range of load variance. HCCI engines therefore have great potential as power sources for co-generation. Moreover, the main component of natural gas is methane (CH4), the characteristics of which make it an ideal fuel for HCCI combustion. In other words, using an HCCI engine fueled by natural gas as the power source for co-generation is likely to make it possible to achieve economy equivalent to that of diesel, together with reduced pollutant emissions. The discussion in this report will not cover combustion characteristics in turbocharged HCCI engines, or their tendencies, or the prototype small high-efficiency turbocharger that was built [13], and will be confined to thermal efficiency. Fig. 13 compares the performance of a naturally aspirated HCCI engine, a turbocharged HCCI engine, and a spark ignition engine. The generating efficiency of the power generation system was estimated at 95%. The operating conditions for the naturally aspirated HCCI engine were an Table 6 Compares a spark ignition engine and a naturally aspirated HCCI engine. Engine type Power generating efficiency (power generation output) Heat recovery efficiency (Heat output) Overall efficiency
SI 33.0% (25 kW)
HCCI 37.0% (25 kW)
52.0% (39 kW)
41.0% (28 kW)
85.0%
78.0%
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166
6
Y. Takashima et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Fig. 13. Compares the performance.
Table 7 Engine specifications. Engine type Stroke Bore Compression ratio Displacement Engine speed
2.3.1. D-EGR system This system is proposed by Southwest Research Institute (SwRI) [14], and involves dedicating one of multiple cylinders to EGR. Combustion using this cylinder under fuel-rich conditions cylinder produces H2 and CO. These are distributed to each cylinder, to increase combustion speed even in a high-EGR atmosphere [15], with a view to achieving stable operation. Osaka Gas has identified this system as being of interest and is studying its use with natural gas engines also by using forced induction condition to enhanced efficiency. The specifications of the test engine are shown in Table 7. Table 8 shows the components produced in a fuel-rich atmosphere in the single-cylinder test engine. In an atmosphere with an equivalence ratio of 1.5, it was found that around 9% H2 was emitted. Osaka Gas plans to continue studying the effect of using the H2 obtained to improve combustibility and thermal efficiency in a high-EGR atmosphere. 3. Summary
Single-cylinder water-cooled 106 mm 110 mm 12:1 1007 cm3 1200 rpm
Table 8 Gas composition in a fuel-rich atmosphere. Component
Emission rate
H2 CO
8.84% 8.97%
engine compression ratio of 25.9 and an engine speed of 1700 rpm; the conditions for forced induction operation were an engine compression ratio of 21, a forced induction compression ratio of 1.9 and an engine speed of 1900 rpm. The generating efficiency of the naturally aspirated HCCI engine was around 5 points higher than that of a conventional gas engine. The turbocharged HCCI engine had very high potential. Increasing air density roughly doubled BMEP (equivalent to specific output in output per unit displacement), and generating efficiency was around 40%. HCCI engines are more expensive because of the need for combustion control systems, but have the advantage that increasing engine output through turbocharging directly reduces initial costs, and that devices to raise intake air temperature in order to ignite the fuel are not required. Thus, HCCI engines have great potential toward customers of a low thermoelectric ratio are demanding cogeneration systems with high generation efficiency, but their commercial development will require the establishment of low-cost combustion status detection techniques, operation control systems including engine start and load input method, verification of durability, etc. 2.3. Improving EGR Low-temperature combustion is being studied with reference to automobile engines. This combustion method reduces heat loss by inducting a large volume of exhaust gas (EGR) and reducing flame temperature. The issue with this technology is to achieve stable combustion of a lean mixture composed of a large volume of exhaust gas, air and fuel. This section describes studies of technologies to improve combustion stability.
These are the results of our research on various methods of improving the efficiency of small gas engines. In the short term, turbocharged stoichiometric combustion with EGR will be the best technology for achieving thermal efficiency combined with low emissions. In the medium-to-long term, however, it will be more effective to develop lower-temperature combustion options with increased EGR ratio, such as turbocharged HCCI and D-EGR. Osaka Gas will continue to pursue the development of gas engines capable of attaining the targets set out by the Japan Gas Association. References [1] 35 kW Micro-Cogeneration (In Japanese), Osaka Gas Co., LTD.,
Please cite this article in press as: Y. Takashima et al., Activities for high-efficiency small gas engines, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j. applthermaleng.2016.10.166