MIXER DESIGN FOR A HIGH PERFORMANCE BIOGAS
SI ENGINE CONVERTED FROM AN DIESEL ENGINE
Bui Van Ga, Tran Van Nam
The University of Danang
41 Le Duan, Danang, Vietnam
The performance and pollution emission of biogas engines strongly depend on the fuel-air equivalence ratio of the mixture. Thus an appropriate design of the mixer is the main issue in converting an existing diesel engine into a biogas spark ignition engine.
This paper presents some results of simulation and experiment research on a venturi type mixer for a biogas SI engine converted from a ZH1115 diesel engine. The results show that the fuel-air equivalence ratio of the mixture is less dependent on the opening of the butterfly valve which controls the mixture but it sharply depends on CH4 concentration in the biogas and/or on section of the biogas supplying pipe. At full load, the fuel-air equivalence ratio is slightly changed in relation to the engine speed but at partial load, it strongly depends on the engine speed, particularly at low regime. The dimensionless diameter of the biogas supplying pipe can be expressed by a power relationship with CH4 concentration in biogas with an exponent of -0.515.
Biogas is an attractive renewable source of energy for rural areas. It can be produced from organic wastes, such as dung of animals, plant matter and other wastes of agriculture production. Approximately two-thirds of biogas (in volume) is methane and the rest is mostly carbon dioxide. As a fuel, biogas has a low energy density on the volume because of its high CO2 content. The burning velocity of biogas is low, just at 25 cm/s as against 38 cm/s for LPG, due to the reason that carbon dioxide may change the combustion behavior of the air–fuel mixture. A large quantity of CO2 present in biogas lowers its calorific value, burning velocity and flammability range compared with those of natural gas. The self-ignition temperature of biogas is high and hence it resists knocking which is desirable for engines with a relatively high compression ratio to maximize thermal efficiency.
Power and thermal efficiency of biogas engines reached their highest values with the RAFR between 1.05 and 0.95 . Under these conditions, HC and CO emissions were relatively low but the NOx values were relatively high. Power and thermal efficiency were reduced for leaner mixtures, particularly, though engine speed was increased, emissions were all reduced . Mixtures richer or leaner than this optimal point will cause incomplete combustion or slow down the burning rate and hence lead to a drop in thermal efficiency. Chulyoung Jeong et al. observed that the maximum values of generating efficiency, cylinder pressure, and NOx emissions were obtained at an EAR of around 1.2 , which is slightly higher than the values reported by .
A high compression ratio spark ignition engine for biogas can be built by replacing the injectors of a diesel engine by a spark plug and modifying the pistons. It is necessary to maintain a proper ratio between the fuel gas and air in order to attain good combustion . However, since NOx production in this condition is relatively high, two alternative approaches, slight retardation of spark timing or a little leaner operation which can be applied to control NOx emission without sacrificing considerable thermal efficiency, were suggested as an optimum and practical operating point for use in an actual biogas site in the future .
The supply of the right mixture of air and fuel is therefore of utmost importance for the performance of a biogas spark ignition engine . Further, the engines operating close to the stoichiometric air/fuel ratio display lower levels of emissions of toxic gases. Enhanced methane concentration in biogas (as in methane enriched biogas) significantly improves the engine performance and reduces emissions of hydrocarbons .
Thus it is necessary to design appropriate mixers in order to ensure the right mixture with various biogas composition and pressure. In Vietnam, the research team GATEC of the University of Danang  has carried out a lot of studies on biogas engine. The results of researches allowed application of biogas on engines in rural areas which is very helpful for climate change mitigation . As the original engines are diverse in structure and dimensions, it is difficult and costly to carry out experiments for determination of basic parameters of appropriate mixers [8-9]. The simulation of mixers will be useful to predict characteristics of the flow under different operation conditions so that we can identify basic dimensions of the mixer corresponding to the size of each engine.
In this research, we study the characteristics of a mixer designed for a biogas spark ignition engine converted from a Jandong ZH1115 diesel engine. The objective of the research is to identify the section of biogas supplying pipe to ensure normal operation of the engine fueled with different components biogas. The experiences taken from this research can be applied on other kinds of diesel engine.
II. Method of Study
1. Mixer Design
The ZH1115 diesel engine with bore of 115mm, stroke of 115mm, and compression ratio of 17 reaches power of 24HP at rated speed of 2,200 rpm. The engine is converted into a biogas spark ignition engine by replacing injection systems with an electronic spark ignition system and a mixer mounted on the intake manifold.
Figure 1 shows the longitudinal sectionof the mixer system for this ZH1115 engine. A venturi injector of 33mm interior diameter is mounted on a 2.5mm diameter pipe at the narrowest section of biogas supply. The mixer is disposed of for 2 valves: a ball valve for biogas flow control with flow diameter of 18mm and a butterfly valve for mixture flow control. The open angle of the biogas ball valve (compared to the vertical axis) changes from b= 0 (completely closed) to b= 90 (fully open). The open angle of the butterfly valve (compared to the center line of the mixer) changes from a= 0 (fully open) to a= 70°(fully closed). Relationships between open angle aand the flow section of the intake manifold are shown in Table 1.
2. Experimental setup
The experimental setup is introduced in Figure 2. This experimental testing is conducted with the Froude dynamometer on site of biogas production. The ZH1115 engine is converted into the spark ignition engine with the compression ratio of 12. The biogas supplying pipe with variable diameter in accordance with the CH4 concentrations of biogas. Biogas from the digester is conducted by two different filtration systems. The first system removes only H2S by bentonite. The second system removes simultaneously H2S and CO2 by means of NaOH solution. Biogas mixture coming from these two sources with different concentrations of CH4 is supplied to the engine. Compositions of biogas are measured by a biogas analyzer GFM 435. Air mass flow is measured by ABB a flow meter. Biogas mass flow is measured by a Sigma flow meter.
Experimental data are transferred to computer via A/D card and Labview software.
Density of mixture is supposed to be r=1,293kg/m3. Average mass flow rate of mixture during the intake process is (kg/s). The vacuum pressure on the intake manifold is . Mass flow rate of mixture is also given as m=krSV. In other words, mass flow rate m is proportional to speed V whereas is proportional to V2, so it is proportional to m2.
In order to exclude the coefficient of proportion in the FLUENT calculation, we firstly suppose boundary conditions p_mix_out as p_mix_out_propos. The calculation results will give us the supposed mass flow rate: mmix_out_propos (kg/s). Hence the pressure at mixer outlet with known mass flow rate of the mixture during intake process mn will be identified by the following expression:
Pressure at the mixer outlet p_mix_out when the open angle of the butterfly valve is 30°, the open angle of the biogas ball valve is 75°, and the biogas fuel contains 70 CH4 is illustrated in Table 2.
With mass flows of air and biogas given by simulation of each case, we can then calculate the equivalence ratio of mixture supplied to the engine.
Figures 4a, 4b, 4c and 4d introduce calculation results of velocity field, contour of dynamic pressure, contour of O2 and CH4 concentrations on the symmetrical surface of the mixer with open angle of the butterfly valve at 30°. We can predict homogeneity of mixture through the mixer with help based on these results.
III. Results and Discussions
1. Simulation Prediction
Figure 5a introduces variations of fuel-air equivalence ratio fof the mixture versus speed of an/the engine fueled with biogas containing 60% of CH4. The biogas ball valve is fully opened (90°). The butterfly valve is opened at positions of 34%, 72%, and 100%, respectively. The engine speed in each case ranges between 1,000 rpm and 2,200 rpm. The calculation results show that when an engine operates on full load curves (with butterfly valve being 100% open), fof mixture is almost stable (fchanges from 1.3 to 1.4). When the engine operates on partial load curves, the curve of the fuel-air equivalence ratio varies in function of engine speed: the steeper the engine speed is, the smaller the butterfly valve open level becomes. The change in concentrations at high-speed positions is less than at low-speed positions. At any opening levels of the butterfly valve, when the engine runs at rated speeds between 1,800 rpm and 2,200 rpm, fuel-air equivalence ratio of the mixture changes narrowly from 1.02 to 1.10.
Figure 5b introduces the same results with biogas fuel containing 70% of CH4 and an opening level of the biogas ball valve at 75°. Figure 5c introduces the same results with biogas fuel containing 90% of CH4, with the opening level of the biogas ball valve at the position of 60°. The results manifest the changing principle of fin terms of n similar to the case in which biogas contains 60% of CH4. With biogas containing 90% of CH4, if the opening level of the biogas ball valve is at 60°, the mixture is poor. In order to increase fuel-air equivalence ratio fof the mixture in these cases, we can increase the opening levels of the biogas ball valve. On the contrary, in the case that the fuel contains 60% - 70% of CH4, we can reduce the opening level of the biogas ball valve so as to reduce fuel-air equivalence ratio fof the mixture.
The above results show that, when opening levels of the biogas valve and the butterfly valve are given, the fuel-air equivalence ratio fof the mixture is slightly reduced as decreasing of the engine speed slows. The bigger the opening level of the butterfly valve, the lower the changing degree of fbecomes. When CH4 concentration in biogas is varied, we can adjust the biogas ball valve to achieve the best fuel-air equivalence ratio f. This adjustment can be made once for each kind of fuel. Table 3 summarizes the results of calculations on the opening levels of the biogas ball valve corresponding to the biogas containing different percentages of CH4 concentration. The results show that with given a CH4 concentration, we can choose an appropriate opening level of biogas ball valve so that the fuel-air equivalence ratio fis in optimal range according to  at any opening level of butterfly valve.
Figures 6a, 6b and 6c introduce the effect of engine speed on fin relation to the opening level of the butterfly valve with biogas containing 60%, 70% and 90% of CH4 with opening levels of the biogas ball valve shown in Table 3. The results show that when engine speed and position of the biogas ball valve are fixed, fuel-air equivalence ratio fof mixture is reduced as the opening levels of the butterfly valve are increased. When the butterfly valve is fully open (the engine operates on full load curves), the fuel-air equivalence ratio fof the mixture is almost unaffected by the engine speed. Therefore the mixer must surely supply the best mixed components when the engine operates on full load curves. Under partial load operation, fslightly increases when the opening level of the butterfly valve is reduced. The results show that with 34% opening of the butterfly valve at 1,000 rpm engine speed, fis about 1.25 compared with its value of approximately 1 a fully opened butterfly valve with biogas containing 60-70% of CH4. Even when the mixture is richer as the engine runs on partial load curves, fis within combustible limit range.
However, as the biogas ball valve position, butterfly valve position and engine speed are fixed, fchanges considerably in accordance with the concentration of CH4 in biogas fuel. Figure 8 shows that at engine speed of 2,200 rpm, the biogas ball valve is open up to 75°and the butterfly valve is open 72%, freaches 0.85 and 1.65 with biogas containing 60% and 90% CH4, respectively. Therefore, to obtain an appropriate fuel-air equivalence ratio fof the mixture as CH4 concentration in biogas changes, we must change the opening levels of the biogas ball valve.
Figures 8a, 8b and 8c introduce the effect of biogas fuel and opening levels of the biogas ball valve on change of fin function of opening levels of the butterfly valve. With a given engine speed, the tangent of curves are almost similar, independent of CH4 concentration in biogas fuel. So if we adjust the position of the biogas ball valve so that for a given opening level of the butterfly valve we obtain the same fof mixture, then we can represent a linear relationship between fand the opening level of the butterfly valve.
Figure 9 illustrates the variation of fin function of butterfly valve opening levels with biogas containing 60%, 70%, 80%, and 90% of CH4 and opening levels of the biogas ball valve of 90°, 74°, 67°, and 61°, respectively. At 34% opening level of the butterfly valve, fis in range between 1.06 and 1.08. When the butterfly valve is fully opened foscillates from 1.02 to 1.05. These results show that the tangent of the curves is -0.0006 (if the opening level of the butterfly valve is in percentage).
As the tangent of the curves is very slight, we can consider fis unchanged in relation to opening levels of the butterfly valve. Contrarily, fof the mixture changes sharply in relation to the opening levels of the biogas ball valve and CH4 concentrations in biogas fuel. This means that for a given biogas, we need to determine the size of the pipe that supplies biogas to the mixer in relation to the size of the air admission pipe so that fis in optimal range observed by Huanga et al.  or by Jeong et al.  at any level of butterfly valve opening. In this case, we need not equip the biogas ball valve with the supplying pipe.
Figure 10 introduces the variation of dimensionless diameter y=deq/dad of the biogas supplying pipe in accordance with the concentration of CH4 in the fuel in case of ZH1115 biogas engine. When CH4 concentration in biogas increases, the amount of biogas supplied to the mixture must decrease to ensure that fof the mixture is unchanged. Because mass flow rate m is proportional to the flow section S, in other words, it is proportional to the square of the dimensionless diameter of the biogas-supplying pipe y, or the diameter y is in proportion to m0,5. Otherwise, to keep constant f,when the CH4 concentration in fuel increases, the mass flow rate of fuel decreases. This means that dimensionless diameter of biogas supplying pipe is proportional to x-0,5. Figure 10 shows that the exponent of the curve is -0.515. The absolute value of the exponent is slightly higher than 0.5. This is reasonable because when the biogas mass flow rate changes, the air mass flow rate is also changed to ensure the constant value of f.
2. Experiment Valuation
Figures 11a, 11b, and 11c introduce the variation of equivalence ratio versus engine speed at full load regime. Biogas contains 60%, 73%, and 87% of CH4 concentrations. Biogas supplying pipes are selected with diameters of 18mm, 16mm, and 14.5mm, respectively, corresponding to the relationships shown in Figure 10. During experimentation, the butterfly valve is fully open. The results show that equivalence ratios given by simulation are fitted well to their values given by experiment with different CH4 concentration in biogas. This confirms that the relationship in Figure 10 is reasonable.
Figure 12 introduces full load characteristic curves of ZH biogas engine fueled with biogas containing 60%, 73%, and 87% CH4 concentrations. The results show that at the speed of 2,500rpm, the maximum power of engine run by the biogas containing 87% of CH4 is 21 HP. Calculated power via proportions of CH4 passing into a/the cylinder of the engine fueled with biogas containing 73% and 60% of CH4 will be 20.58HP and 20.03HP, respectively. The experimental data are suitable for the first case, but as for the final case (biogas containing 60% of CH4), the real power is much lower than that of the calculation. This is because of incomplete combustion due to high concentration of CO2 in biogas. The suitability of the power proportions when the ZH1115 engine is run by biogas containing different concentrations of CH4 affirms that the relationships between the biogas supplying pipes with the CH4 concentrations in biogas shown in Figure 7 are accurate.
The results of this research are very useful to convert an existing engine running on diesel that is largely used in rural areas into biogas engine. Previously for converting a diesel engine into biogas engine we have to conduct a lot of experimental tests to determine appropriate parameters of the mixer. This takes a lot of time and money. Now thanks to this new method of simulation we can orient the technology of conversion. This way can be applied generally to any kind of diesel engine. This is very helpful in reducing the cost of conversion that will encourage numerous farmers to use biogas in their machines. It is an effective contribution to climate change mitigation.IV. Conclusion