DESCRIPTION

The object of the present finding is a series of provisions for the transformation of a conventional diesel engine for stationary applications into a gas-fed engine.

From the known literature on the subject it can be seen that internal combustion engines fed with gaseous fuels, like for example methane or gas deriving from gasification, require lower propagation speeds of the flame inside the combustion chamber compared to those observed when feeding the engine conventionally with petrol or Diesel oil.

Therefore, it can be deduced that the combustion chamber and more specifically the selection of the design thereof is without doubt one of the components that should be designed and manufactured with maximum care in order for the engine itself to work optimally when fed with gas.

In and the authors tested a gas engine by installing three different combustion chambers: the first one flat, the second with the bowl cylindrical and the third with the bowl parallelepiped. By "bowl" we mean the cavity formed in the space located between the roof of the cylinder and the top part of the cylinder

¹ Jesper Ahrenfeldt, Torben Kvist Jensen, Ulrik Henriksen and Jesper Schramm, Experiments with Wood Gas Engines, SAE Paper 2001-01-3681.

² Olsson K. "On Combustion Chambers For Natural Gas SI Engines". Lund Institute of Technology, 1995. itself, typical of direct injection engines and that constitutes the combustion chamber itself. The fuel is injected into such a cavity, generally through an injector (in the case of conventional Diesel engines). Going back to the three types of combustion chamber tested by the authors of ^l and ² , the first (flat shape), being the simplest of the three, does not disturb the swirl motion generated by the intake ducts and, therefore, provides low levels of turbulence and therefore longer-lasting combustion. The chamber with the cylindrical bowl, on the other hand, increases both the intensity of the swirl motion since it forces the flow into a duct with a smaller radius and the squish effect that derives from the clearance height between cylinder head and piston at TDC producing a high level of turbulence and therefore a larger flame surface; the combustion speed is, in this case, greater than the previous case. Finally, the last chamber, the one with the bowl with square section, generates the greatest level of turbulence, but at the same time it is the most difficult to make and above all generates problems in terms of excessive production of NO _x promoted by the high degree of turbulence.

In ^[3] , ^[4] , ^[5] it is highlighted, as already stated, how important the

³ P. G. Tewari, J. P. Subrahmanyam and M. K. Gajendra Babu, Experimental Investigations on the Performance Characteristics of a Producer Gas Fuelled Spark Ignition Engine, SAE Paper 2001-01-1189.

⁴ Shashikantha, Parikh P.P., Spark Ignition Producer Gas Engine and Dedicated Compressed Natural Gas Engine- Technology Development and Experimental Performance Optimisation, SAE Paper 1999-01-3515. geometry of the combustion chamber is, as well as how important the position of the spark plug for igniting the combustion itself is. The spark plug must be placed in a position as central as possible, in order to ensure that all of the area of the cylinder are reached by the front of the flame in the shortest possible time. Moreover, again in ^t4] , it is emphasised how important the value of the optimal compression ratio is for gaseous fuels, which is somewhere around a value of 11.5 for gases deriving from gasification. Such a value is justified by the need on the one hand to ensure a high efficiency of the engine, and on the other hand to avoid detonation from occurring. The characteristics that have emerged up to now from the quoted studies and publications individually do not provide a clear solution to the problem of adaptation and optimisation of a Diesel engine for stationary applications in a gas- fed engine (whether using lean gas or not).

Together with the selection of the shape of the combustion chamber most suitable for the operation of a gas-fed engine for stationary applications, starting from the modification of a pre- existing conventional petrol or Diesel engine, other factors of undoubted importance come into play, which shall emerge from the detailed description that shall be provided hereafter.

⁵ Francisco V. Tinaut, Andrέs Melgar, Alfonso Horrillo, Ana Dϊez de Ia Rosa, Method for predicting the performance of an internal combustion engine fuelled by producer gas and other low heating value gases, Fuel Processing Technology 87 (2006) 135 - 142. The purpose of the present finding, therefore, is the transformation of a conventional diesel engine for stationary applications into a gas-fed engine through a series of provisions the most important of which is the modification of the shape of the combustion chamber.

The wide availability of diesel engines for stationary applications of all sizes has channelled research for the development of gas engines (using lean gas or not) in the direction of the adaptation of such units, so as to minimise the costs linked to the ex-novo design of the components.

In particular, of the aforementioned conventional Diesel engines the engine block, drive shaft and cylinders remain unaltered, whereas the characteristics that undergo modifications, on the other hand, are: - Compression ratio: the compression ratio is lowered from a value of about 17:1 (typical of Diesel engines) to an optimal value within the range 9.5 - 11.5:1.

Motion of the feed: Diesel engines are built so as to have high swirl motion values (rotation motion of the gases around the axis of symmetry of the bowl) to allow the optimal mixing between injected fuel and air. Whereas in the case of gas-fed engines (similarly to petrol engines), the main goal is to bring the fresh mixture close to the spark plug and avoid the formation of areas masked from the flame front, all nevertheless in high turbulence operating conditions. Combustion chamber: to obtain the motion characteristics just described a W-shaped chamber is used, having a greater volume of the "bowl" inside the piston compared to the case of Diesel to allow a reduction of the compression ratio. The diameter of the chamber is also increased to decrease the swirl speed.

Ignition: the ignition of the mixture is carried out by a spark plug instead of the conventional fuel injector in the case of a Diesel engine. Due to the poor quality of the gas used as fuel and the presence of inerts in the mixture, the propagation speed of the flame is much lower than in the case of other more valuable fuels. This means that to deal with the greater combustion time it is necessary to foresee an adequate advance ignition. Valve diagram: the valve opening and closing diagram has a large influence upon the yield of the engine. For a gas-fed engine it is necessary to avoid the backflow of the exhaust gas in the intake duct during zero-crossing, since there could be an explosion upstream of the engine. An increased performance is obtained by modifying the lifting curve of the valves obtaining faster lifting movements, which allow a greater average passage section thus increasing the filling coefficient. Dilution ratio: such gas-fed engines usually operate with a greater mixture ratio than the stechiometric one, usually between 1 and 3.

Intake and exhaust ducts: to improve the yield of the gas-fed engine it is also necessary to increase the diameter of the ducts, to decrease the length of the intake manifold, to increase the length of the exhaust manifold and to give it a certain shape.

These and other innovative characteristics shall become clearer from the following detailed description that shall refer to the following attached tables 1/1, 2/5, 3/5, 4/5, 5/5 where: • fig. 1 comprises a 3D section view of the piston of the engine with the combustion chamber highlighted;

• fig. 2 is a cross section view of the cylinder highlighting the position of the intake and exhaust valves and of the ignition spark plug; • fig. 3 shows the 3D model of the intake duct;

• fig. 4 shows a 3D view of the geometry of the intake duct relative to the conventional diesel engine;

• fig. 5 shows the experimental results relative to the mass taken in by the individual cylinders in the case of the intake duct shown in fig. 4;

• fig. 6 shows some 3D views of the modified geometry of the intake duct relative to the conventional gas-fed diesel engine;

• fig. 7 shows the comparison between the masses taken in by the individual cylinders in the case of a conventional duct (fig. 4) and a modified duct (fig.6); • fig. 8 shows the diagram of the feed-ignition unit;

• fig. 9 shows the diagram of the ignition order of the cylinders.

For the purposes of a conversion of the feed of a conventional Diesel engine for stationary applications towards a gas feed, we start from the analysis both from the point of view of the components and from the point of view of the experiments on the performance of a pre-existing Diesel engine. For example, in the case under consideration and in any case not in a limiting way, we shall refer to a 12 litre engine, 6 cylinders in line with forced feed through exhaust gas turbocompressor with intercooler. The modes of adaptation either outlined earlier or discussed later can be applied to units of this type. In the logic of the transformation of the engine one of the substantial modifications is that of eliminating the injector, typical of the Diesel engine, replacing it with a spark plug, in this way making the engine perform an Otto cycle and no longer a Diesel cycle. Such a solution allows the need for liquid fuel to be completely removed, but it requires substantial modifications to the engine itself. However, in order to exploit such a modified engine within a possible local energy generation programme, such an approach is without doubt the best. Other characteristics of the Diesel engine to be adapted are: S work at a constant number of revolutions, which can be selected between 1500 and 1800 rpm, according to whether the generation of current is required at 50 or 60 Hz; S delivery of power respectively equal to 187 and 220 kW peak and 155 and 180 kW continuous;

S engine equipped with just two valves per cylinder with shaft and rocker actuation;

S camshaft housed inside the engine block;

^• S intake valve with diameter of 53 mm;

S exhaust valve with diameter of 51 mm.

As a first step for the adaptation of the combustion chamber a CAD reproduction thereof (fig. 1) was made based upon the original measurements of the diesel engine and direct measurements made after the work carried out to reduce the compression ratio. In particular, the characteristic dimensions of the cylindrical bowl, identified with 1 , were taken to a diameter d = 88 mm and to a height h = 35.4 mm, in order to reduce the compression ratio to a value equal to 9.55. Knowing the piston displacement and the compression ratio the volume of dead space is then worked out. Indicating the piston displacement of the individual cylinder with "C", the volume of the combustion chamber with "Vc", the volume of the dead space with "Vm" and the compression ratio with V ₅ we get the following equation:

cwcλVm which gives us Vrn = 37 ≤2≤,3 mm? μ Vm+Vc

Fromm the result obtained, knowing the bore "a" of the piston, the distance of the piston from the head at the top dead center is worked out:

Vm h = T = 3.16 mm nar ~T ^~

The volume of the combustion chamber thus consists of that of the bowl and of a very small portion of cylinder, corresponding to the space between the head of the piston and the plane of the valves. From fig. 2 it can be seen that the spark plug (identified with 2) is not perfectly centred and is housed in what in the conventional diesel engine is the injector seat; the head of the engine, on the other hand, is flat. Again from figure 2 it is possible to identify the position of the intake valve 3 and the exhaust valve 4. With regard to the feed-ignition system shown in fig. 8, another substantial modification is that concerning the carburettor to obtain the gaseous fuel mixture. In the case of such engines, adapted to be fed with gas, said component usually consists of a Venturi tube, indicated in the diagram with 5, arranged between the outlet for the gas available to be fed to the engine and the inlet of the engine itself. The motion inside this component is ensured by the depression generated downstream of the engine or by the turbocompressor. In the throat section some radial holes are formed from which outside air is taken in and the dilution ratio is varied by acting upon the flow rate of air taken in through a valve 5'. Such a carburettor takes care of adjusting the flow rate of gas entered into the air flow so as to control the dilution ratio and it is positioned upstream of the turbocompressor group to be able to exploit the Venturi effect with the gas at relatively low pressure. The aforementioned choke valve, which adjusts the feed of the engine, is arranged downstream of the entire feed system and is controlled by an electronic system. The feed- ignition system as a whole is in any case coupled with an electronic adjustment system with actuation through servomotors, considering the particular function required of the engine, i.e. to operate at a constant number of revolutions responding rapidly to the variations in load. A further characteristic of the engine is that of consisting of a monoblock drive shaft with phase displacement between the cylinders of 120°, where the particular order of ignition of the cylinders and the duration of the intake step, equal to about 230° of crankshaft rotation, are such as to bring about an overlapping between the intake steps of the cylinders, as can be seen from fig. 9. The ducts (6) that connect the combustion chamber to the intake manifold and to the exhaust manifold, are practically identical for the two valves, having a spiral-shaped progression as shown by fig. 3, starting from a rectangular section at the sides of the head to then wrap around the valve seat. A configuration of this type promotes the formation of whirling motion inside the combustion chamber, particularly swirling. On the other hand this solution involves greater load losses than a duct with a less articulated shape, however such a drawback is less noticeable in the case of forced intake engines, since the filling driving force is substantial.

Fig. 4 shows the CAD model of the intake duct 7 referring to the Diesel engine to be adapted: geometrically it is made up of a large cylinder with the ends closed, having the function of a lung, on which the opening for the inlet and outlet of air are formed. Again from fig.4 it can be seen that the channels that carry the air to the cylinders have a rectangular section that joins up with the central cylinder tangentially, assuming a volute-shaped progression, whereas the inlet duct has a square section, is connected to the lung radially and directly faces onto the duct relative to the cylinder 5. The most significant geometric parameters of such a duct are summarised in the following table:

Tab. 1 - Geometric parameters of intake duct

After having concluded the step of modelling such a duct, a simulation stage relative to the operation of the cylinders of the engine was carried out in order to work out the progression of the dynamic pressure on the intake duct and to work out the necessary modifications for an optimal adaptation of such a duct to gas feeding. Then, starting from the known surrounding conditions (average pressure in the intake duct and in the exhaust manifold) the simulation of the operation of the cylinder (made without taking into account the combustion) provided in output the progression of the dynamic pressure at the interface between cylinder and intake duct (such an interface is represented by the connection surface between manifold and head). After a first step of discretization of the volume with different types of mesh with gradually increasing elements, we moved on to a second step in which the progression of the mass-flow in the various cylinders was analysed as the mesh varied, with the aim of identifying the most suitable mesh to obtain plausible data. Once the appropriate mesh was selected the results of the simulation were analysed in terms of behaviour of the duct. In terms of mass taken in the graph shown in fig. 5 was obtained, in which it is possible to see how, predictably, the cylinder 5 (the one facing directly onto the intake duct) has a mass taken in clearly greater than the other cylinders. Another result worth noting is relative to the different mass taken in by cylinders no. 4 and no. 6, which, whilst being the same distance from the intake duct, have different masses taken in, greater for cylinder no. 4. This behaviour can be explained with the overlapping of the intake steps of cylinders no. 2 and no. 4: part of the mass pulled out by cylinder no. 4 is drawn in by cylinder no. 2, whereas such a phenomenon does not occur in the case of cylinder no. 6 that, being at the end of the duct, is not affected by the other intake steps (in this particular case that of cylinder no. 3).

In light of such results, we went ahead and modified the intake duct in terms of displacement of the intake flange 8 to the centre of the duct as shown by the 3D model resulting from the transformation in fig. 6. Similarly to the original duct, such a duct was firstly discretized with mesh analogous to the case described previously and, setting the same surrounding conditions and the same parameters as the previous simulation, the same data was worked out in output relative to the standard duct, i.e. the progression of the mass-flow in the various cylinders. The comparison between the masses taken in overall by the cylinders in the case of a standard duct and a modified duct are summarised in the graph of fig. 7, in which on the x-axis for each cylinder the left-hand column refers to the mass-flow of the modified duct, whereas the right-hand one refers to the standard duct. The benefits deriving from such a choice are clear, since compared with a definite loss of mass on cylinders 5 and 6, all of the other cylinders obtain clear benefits, from the modification made. Furthermore, such a modification is also simple to make and implement, requiring just the lengthening of the tube that connects the throttle to the intake duct.