GB2069594A - A compound expansion internal combustion engine - Google Patents

Abstract

At least one cylinder C receives exhaust gas from at least one cylinder A, B in which a four-stroke ignition cycle takes place and the gases are expanded to substantially atmospheric pressure with the addition of heat from coolant flowing around the ignition cylinder. Additional heating may be provided by an electric element and the engine may be supercharged. The engine may housed within insulation (64), Fig. 2 (not shown). <IMAGE>

Description

SPECIFICATION An internal combustion engine and operating method The present invention relates to internal combustion engines.

A conventional Otto cycle internal combustion engine may operate on a four stroke cycle, the first stroke being an induction stroke where the size of the combustion chamber is increased, inducing a fuel/air mixture therein. The second stroke is a compression stroke where the size of the combustion chamber is decreased to compress the fuel/air mixture; the third stoke is a power stroke in which the size of the combustion chamber is again increased after combustion of the compressed fuel/air mixture; and the fourth stroke is an exhaust stroke in which the size of the combustion chamber is again decreased to expel exhaust gases from the chamber It will be noted that there is only one power stroke in every cycle of operation.

A major disadvantage of the conventional internal combustion engine lies in the fact that the power and exhaust strokes are the same length as the induction and compression strokes, thus limiting the thermal efficiency of the engine to approximately 20%. In addition, the combustion products are exhausted from the combustion chamber at high temperatures resulting in a considerable waste of energy. The calorific value of the fuel supplied to a typical engine is used approximately as follows: 40% of the calorific value of the fuel is rejected to atmosphere by the engine cooling system and 35% is also rejected to atmosphere during an exhaust stroke as a result of insufficient expansion of the combusted gas when the heat in the combusted gas is converted to mechanical work during the combustion stroke. Approximately 25% only of the fuel calorific value is converted to mechanical work.In addition a significant proportion of this percentage must be used by the engine in driving ancillary equipment such as a water pump and cooling fan.

The present invention seeks to provide an improved internal combustion engine.

The present invention provides an internal combustion engine comprising at least one primary cylinder, and an associated secondary cylinder operably coupled to said primary cylinder for enabling further expansion of a fuel/air mixture ignited in said primary cylinder; and means for applying heat to said secondary cylinder; and wherein the pistons of said cylinders are coupled to a common crankshaft.

The present invention also provides a method of operating an internal combustion engine comprising igniting a compressed fuel/air mixture in a primary cylinder of the engine to generate a power stroke in said cylinder, subsequently enabling further expansion of the ignited fuel/air mixture in an associated secondary cylinder to generate a further power stroke, and apply heat to said secondary cylinder during operation of the engine.

Preferably the ratio of the working volumes of the or each primary cylinder and the associated secondary cylinder are such that the gas exhausted from the primary cylinder expands into the secondary cylinder substantially to atmospheric pressure before being exhausted from the secondary cylinder to atmosphere.

Conveniently the length of the strokes of the pistons of the primary and secondary cylinders are substantially the same.

In a preferred embodiment of the present invention the engine has a closed loop cooling system which transfers heat generated in the primary cylinder to the secondary cylinder to maintain the temperature of the latter as high as possible. Such a cooling system may obviate the need for a radiator and fan with associated valves, thermostat and piping, or at least reduce the radiator size, thus effecting a reduction in engine costs. Additional insulation can be provided around either the secondary cylinder or all of the cylinders to reduce as much as possible the heat loss to the surrounding environment. Thus a significant proportion of the 40% of the fuel calorific value which is rejected to atmosphere in conventional engines may be converted to mechanical work in an engine according to the present invention.The power output obtained from each unit cube swept volume of the engine may thus be increased.

Supercharging of an engine involves the supplying of air or a combustion mixture of fuel and air to be engine cylinders at a pressure greater than atmospheric. In conventional engines this provides an increase in the engine power output but also increases the fuel consumption per horse power. The main advantage of supercharging is to enable the rate and volume of an engine of a given power output to be reduced but unfortunately the reduction in thermal efficiency of conventional engines which results from supercharging has restricted the use ofsupercharging in those applications where high fuel economy is important, the design of conventional engines rendering the obtaining of a higher specific power by supercharging and a reduction in the amount of fuel consumed per horse power opposing objectives.

However, by suitably choosing the ratio of the working volumes of the primary and secondary cylinders of a supercharged internal combustion engine according to the present invention to enable expansion substantially to atmospheric pressure of the combusted gases in the secondary cylinder both an increase in specific power and a reduction in fuel consumption per horse power can be obtained. Here. in contrast with conventional engines the combusted gas is not exhausted to atmosphere at a relatively high temperature and pressure.

Since the full expansion of the fuel/air mixture ignited in a primary cylinder of an internal combustion engine according to the present invention is not dependent solely upon the stroke of the primary cylinder piston but is also dependent upon the volume of the secondary cylinder the stroke of both the primary and secondary cylinder pistons may be considerably shortened. Use of a shorter piston stroke reduces the relative velocity between the piston and the cylinder thus permitting an increase in the normal working speed of the engine. Since the actual power output of an internal combustion engine is proportional to the engine r.p.m. (revolutions per minute) the enabled increase in permissible engine working speed allows an improvement in specific engine power.

It will be noted that the secondary cylinder of an internal combustion engine according to the present invention operates on a two-stroke cycle. It is advantageous therefore to provide one secondary cylinder fed alternately from each of two primary cylinders, the secondary cylinder performing two two-stroke cycles during the four-stroke cycle of each primary cylinder. the primary cylinders being 3606 out of phase one with respect to the other. Thus a three cylinder engine of the present invention is the equivalent of a four cylinder engine of the conventional four stroke or Otto design.

An internal combustion engine according to the present invention may operate by spark ignition or by compression ignition. Auxiliary services for the engine may be driven from the crankshaft in a conventional manner, such services being pumps for fuel oil, lubricating oil, generators and the like.

One form of engine according to the present invention has a non-return inlet valve in the head of the or each primary cylinder for induction of fuel/air mixture into said cylinder, and a valve controlling the exhausting of gas from the or each primary cylinder to the associated secondary cylinder and also the exhausting of the exhaust gas from the secondary cylinder.

The control valve is conveniently a rotary valve although it may alternatively be provided by a suitable arrangement of poppet valves in known manner.

The present invention is further described hereinafter, by way of example, with reference to the accompanying drawings, in which: Figures la to 7d are schematic diagrams showing the principle of operation of an internal combustion engine according to the present invention; Figure 2 is a diagrammatic plan view of an internal combustion engine according to the present invention; Figure 3 is a section along the line Ill-Ill of Fig. 2 showing the cylinders thereof; Figure 4 is a diagrammatic view of a conventional four cylinder internal combustion engine modified to operate according to the present invention; and Figures 5A- 1 show the value timing cycles for the cylinders of the engine of Fig. 4 over two crankshaft revolutions; and Figures 6-17 relate to an analysis of an engine according to the present invention.

Figs. 1 a to 1 d show in schematic form an engine 10 comprising a single thermodynamic assembly of two primary cylinders A and B and a single secondary cylinder C. Valves 1 2 and 1 4 control inlet of fuel/air mixture to cylinders A and B respectively. Valve 1 6 controls passage of combustion gases from cylinder A to cylinder C, and valve 1 8 controls passage of combustion gases from cylinder B to cylinder C. Valve 20 controls exhaust of spent gases from cylinder C.

The pistons associated with the cylinders A, B and C are connected to a common three-throw crankshaft (not shown in the drawings).

In Fig. 1 a the cylinder A has just reached T.D.C. with valve 12 closed and valve 16 open, combustion gas being transferred from cylinder A to cylinder C, valves 20 and 1 8 being closed.

At this point in time cylinder B has also reached T.D.C. and the fuel/air mixture therein has been ignited, the valve 14 being closed. The piston of cylinder C has been driven down to B.D.C. by the exhaust gas from cylinder A. The pistons of cylinders A and B now move downwardly and the piston of cylinder C upwardly until the position shown in Fig. 1 b is reached, cylinder B moving under its power stroke and cylinder A moving under its induction stroke.

Referring now to Fig. 1 b the piston of cylinder A has reached B.D.C., valve 12 having been open during its downstroke with valve 16 closed thus allowing fuel/air mixture to be drawn into cylinder A. and at the point shown valve 1 2 has just closed. During the downstroke of the piston of cylinder B both valves 1 8 and 14 have been closed and at the point shown valve 18 is just about to open. During the upstroke of piston of cylinder C the valve 20 has been open and at the point shown has just closed. spent gas being exhausted through valve 20 to the atmosphere. The pistons of cylinders A and B again start to move upwards and the piston of cylinder C starts to move downwards, the gas in cylinder A being compressed and that in cylinder B being transferred to cylinder C, combustion of the fuel/air mixture in cylinder B, if incomplete. continuing during the transfer.

In Fig. 1 cthe piston of cylinder A has reached the end of its compression stroke at T.D.C. and the gas therein is ignited. Piston of cylinder B has also reached T.D.C. and the gas therefrom has been transferred to cylinder C. During the upstroke of cylinder A the valves 1 2 and 1 6 have been closed. During the upstroke of the piston of cylinder B valve 14 has been closed and valve 1 8 open and during the downstroke of cylinder C the valve 20 has been closed. At the point shown in Fig. 1 C valves 1 4 and 20 are about to open, and valve 1 8 about to close, valves 1 2 and 1 6 being closed.The piston of cylinder A is driven down under its power stroke, the piston of cylinder B moving down in its induction stroke, fueld/air mixture being drawn in through valve 14. The piston of cylinder C moves upwardly exhausting the spent gas through valve 20.

In Fig. 1 dthe piston of cylinders A and B have both reached B.D.C. The valve 16 of cylinder A is about to open to transfer gas therefrom into cylinder C pushing down the piston thereof with valves 20 and 1 8 closed. The piston of cylinder B is about to start its compression stroke with valves 1 4 and 1 8 closed. The pistons of cylinders A and B therefore move upwardly and the piston of cylinder C moves downwardly until the position shown in Fig. 1 a reoccurs.

The above described cycle of operation is then repeated.

Figs. 2 and 3 diagrammatically show one form of internal combustion engine according to the present invention, comprising two of the thermodynamic assemblies shown in Figs. 1 a to 1 d. A pair of primary cylinders Al and B1 are operatively linked to a secondary cylinder C1 by means of valves 12, 14, 16, 1 8 and 20 which correspond to the valves of the engine shown in Figs.

1 a to 1 d. The pistons Al, B1 and C1 are linked to a crankshaft 30, by connecting rods 31, 33 and 35. A second pair of primary cylinders A2 and B2 are operatively linked with a second secondary cylinder C2 by means of valves 42, 44, 46, 48 and 50 which also correspond to the valves, 12, 14, 16, 1 8 and 20 in the engine shown in Figs. 1 a to 1 d. The pistons of cylinders A2, B2 and C2 are linked to the crankshaft 30 by means of connecting rods 41, 43 and 45, the latter all being slave connecting rods co-operating with the crankshaft 30 and also with the connecting rods 31, 33 and 35 which are the master connecting rods operating in known manner. Bearings 60 are provided between each crank of the crankshaft 30.Operation of the engine is similar to that shown in Figs. 1 a to 1 d, the set of cylinders Al, B1 and C1 being 90 out of phase with the cylinders A2, B2 and C2.

The valves shown schematically in Figs. 1 a to 1 d and Fig. 3 are preferably provided by poppet valves in the case of valves numbers 12, 14, 42 and 44 the remainder of the valves being preferably rotary sleeve valves or alternatively poppet valves. The engines shown in the figures may be made of any suitable materials particularly metal.

The engine shown in Fig. 2 also has a cooling system for the primary cylinders. The system has a pump 61 which circulates a coolant around both the primary cylinders and the secondary cylinder in the general directions indicated by the arrows 62. Cooland fluid is circulated past the primary cylinders which are here shown located one at each end of the engine block, and around the wall of the secondary cylinder to transfer heat from the primary cylinder walls to the secondary cylinder wall. The fan and radiator of a conventional engine may therefore be dispensed with or, at least, considerably reduced in size. Additional insulation 64 may advantageously be provided around all or part of the engine and cooling system to reduce as much as possible heat loss to the surrounding environment.It is believed that it is necessary to maintain the temperature of the secondary cylinder as high as possible to realise the maximum improvement in thermal efficiency of the engine and additional heat sources such as an electrical heating element powered by an alternator may be used to provide heat to the secondary cylinder.

Auxiliary services for the engines shown in the figures may conveniently be driven by the crankshaft, such services being pumps for fuel and lubrication etc.

In the engine shown in Fig. 1 each combustion cylinder supplies a power impulse to the crankshaft once per two revolutions and the expansion cylinder supplies a power impulse once per revolution. Thus the engine provides two power impulses per crankshaft revolution.

In the engine of Figs. 2 and 3 the engine provides four power impulses per revolution and is equivalent to a conventional eight cylinder engine. The duration or time of application of each power stroke to the crankshaft is doubled and in a practical engine the demand for flywheel effect is reduced in proportion.

An engine according to the present invention provides a simplified structure over the conventional engine and is therefore potentially less costly.

An engine according to the present invention may also be capable of accepting supercharging without a significant reduction in thermal efficiency provided the supercharging is at the level dictated by the ratio in cross-sectiona! areas between each primary cylinder and the secondary cylinder specified in the engine design.

An engine according to the present invention may also provide a greater specific power (here specific power is defined as the power delivered at a preselected r.p.m. of the crankshaft by an engine of specific capacity). For a given power output an engine according to the present invention has a reduced capacity. It is therefore physically smaller than equivalent conventional engines. It may also accept supercharging without substantial reduction in thermal efficiency.

The engine stroke can therefore be shortened allowing the maximum r.p.m. of the crankshaft to be raised.

In a modification (not shown in the drawings) of the illustrated engine a charge of fuel may be injected during transfer of gas from a primary cylinder to the secondary, expansion cylinder to provide an increase in power output for such short periods of time as may be required. for example where steep gradients are encountered by a vehicle being fitted with an engine according to the present invention or. where the engine is fitted in an aircraft, during takeoff.

The additional charge of fuel is injected at the most suitable location to aid mixing with the gas transferred from the primary to the secondary cylinder and is advantageously injected at or adjacent to the gas entrance to the secondary cylinder. Here the gas flow is at a relatively high speed and provides a thorough mixing of the injected fuel with free oxygen in the gas. The heat of the combustion gas being transferred is sufficient to ignite the fuel charge. As will be appreciated. in addition to increasing the power output of the engine up to the limits set by the quantity of excess oxygen present, the injection of a further charge of fuel also serves to reduce considerably the noxious contaminents such as carbon monoxide and nitrous oxide which would normally be expelled to atmosphere in the exhaust gas.

Fig. 4 is a schematic illustration of a conventional four cylinder internal combustion engine operating on the Otto cycle which has been modified to operate in accordance with the present invention. The engine 70 has four in-line cylinders 72. 74, 76 and 78 with respective gas ports 80-86 and 88-94 which are designated in the conventional engine respectively as exhaust and inlet ports. In the modified engine the cylinders 72 and 78 operate as primary, combustion cylinders while the two cylinders 74 and 76 are combined to serve as a single secondary.

expansion cylinder. The conventional exhaust manifold is removed and a new exhaust manifold connected to ports 82 and 84 to conduct to atmosphere gases exhausted from the expansion cylinder 74, 76. A carburettor 96, 98 is connected to each of the ports 80, 86 which now serve as inlet ports of the modified engine for the fuel/air mixture. Of course. although two carburettors are shown a single carburettor may conveniently supply the fuel/air mixture to both ports 80 and 86.

The conventional inlet manifold is replaced by a gas transfer manifold 100 which interconnects all of the inlet ports to enable transfer of gas from each of the combustion cylinders 72, 78 to the expansion cylinder 74, 76. The transfer manifold 100 also includes transfer manifold valves 102,104 to avoid the possibility that pressure in the end firing cylinder, on "exhaust" transfer valve opening, would force open the opposite end cylinder valve. In addition. the inlet valves are removed from the cylinders 74 and 76 since these are no longer required. The conventional camshaft is also modified to enable operation of the various inlet and exhaust valves in the required sequence.

The operation of the engine is substantially as previously described with reference to Fig.

1-3, the gas flow or fuel/air mixture flow being indicated by arrows 102. Figs. 5a-1 show the valve timing cycles for the cylinders at 60" intervals over two crankshaft revolutions.

In Figs. 5a-1 the following legends are used: Q = crankshaft angular position INO = inlet valve opens INC = inlet valve closes EXO = exhaust valve opens EXC = exhaust valve closes TVO = transfer valve opens TVC = transfer valve closes If we consider the piston of cylinder 72 having completed its exhaust stroke it now commences its induction stroke with the valve controlling port 80 opening at TDC to allow induction of fuel/air mixture from the carburettor 96. The valve controlling port 88 is closed during this induction stroke. The piston then completes its compression and combustion strokes with the valves controlling the ports 80 and 88 both being closed.At approximately 5" before the piston reaches BDC the valve controlling the port 88 opens to allow the combustion gas in the cylinder 72 to expand into the combined cylinders 74 and 76 during the exhaust stroke of the piston of cylinder 72. Of course, as is described earlier, the pistons of the cylinders 74 and 76 are effecting an expansion stroke during the exhaust of the piston of cylinder 72. When the pistons of the combined cylinders 74, 76 reach BDC the valve controlling port 88 closes and the valves controlling ports 82, 84 open to allow the gases in the combined expansion cylinders 74, 76 to exhaust to atmosphere.

The primary combustion cylinder 78 co-operates with the combined expansion cylinder 74, 76 in a similar manner but of course the operating cycle of the primary cylinder 78 is 180 out of phase with that of the cylinder 72.

Finally, as will be appreciated from the description with reference to Figs. 1-3, the coolant flow path of the conventional engine is modified to ensure that as much of the heat as possible generated in the primary cylinders 72 and 78 is transferred to the cylinders 74 and 76, with, if necessary, additional insulation such as the insulation 64 being provided.

An internal combustion engine according to the present invention, therefore, by providing a secondary cylinder which allows additional expansion of combustion gases increases both the time and the volume available for the expansion of the combustion gases, thus converting to work more of the heat generated in the combustion gases. An engine according to the present invention enables an increase in thermal efficiency to be obtained together with a corresponding reduction in fuel consumption per horse power. In addition, the advantage of increased specific power by supercharging without the increase in fuel consumption obtained with conventional engines is here possible by suitably choosing the ratio of the working volumes of the primary and secondary cylinders to enable full expansion of the combustion gases.This advantage is not, of course, available with conventional engines converted to operate in accordance with the present invention since the ratio of the working volumes of the primary and secondary cylinders are to all intents and purposes fixed.

The following is a basic theoretical comparison between a conventional four-stroke engine cycle and an operating cycle of an internal combustion engine according to the present invention. For purposes of brevity the latter is referred to as a MEMS engine (maximum expansion minimum stroke engine).

MAIN THERMODYNAMIC CALCULATIONS The following cycle calculations are based on the Newhall-Starkman Combustion Chart methods, ref. 1.

In reality, the MEMS cycle would be operating in the 'throttled' condition as shown by Fig. 7 diagrams 1-4. This condition would occur because the expansion cylinder, transfer port and end cylinder are all at roughly the same pressure and temperature at transfer, valve closing and T.D.C., point 5 in Fig. 6 diag 2. As intake charge is admitted to the end cylinder the incoming charge would become considerably degraded due to the end gas remaining in the combustion space above the piston, mixing with the fresh charge. This situation could be avoided as suggested under 'Modifications to Cycie', described later. It is therefore more representative to assume an un-throttled condition for the following MEMS calculations.

See Figs. 6-9 Of the three sets of calculations, part (ii) is the most typical of a current small to medium sized road car engine with air cleaners on the intake and full silencer exhaust system. Appendix A calculations show the likelyhood of the MEMS cycle achieving a very near atmospheric pressure (P6) condition even with a full silencer system as the initial blow-down pressure is less than half that of the conventional cycle. Hence the part (i) calculations show the effect of blow-down to atmosphere of both conventional and MEMS cycles and part (ii) shows the more realistic blowdown, in the conventional cycle only, to just above atmosphere. A comparison of parts (i) and (ii), therefore, shows the effect on net work output of the inability of the conventional cycle to achieve blow-down to atmospheric pressure in the the exhaust manifold.

The following 3 calculations each show a balance of the residual fraction f, and cylinder charge temperature T, for their particular sets of input conditions. These are the result of repeated sets of calculations assuming initial valves for f, and T1 until the correct balance was achieved.

(i) INPUT Compression ratio = 8:1 Inlet manifold temperature = 550tR Inlet manifold pressure = 1 3 Ibf/in2 ABS Exhaust manifold pressure = 14.7 Ibf/in2 ABS Ambient temperature = 520"R Lower Heating Value of fuel = 1 9020 Btu/lb ref. 2 p46 Air/fuel ratio = 1 2.8:1 (typical) Assume F1 = 0.039 T1 = 660"R CA LCULA TONS actual fuel:air ratio For A/F = 12.8, < p = 1.2 = equivalence ratio = chemically correct fuel:air ratio ie 20% rich which is typical in practice.

For octane-air mixture, the fuel:air ratio for 9 = 1 is 0.0653 by weight.

ie for # = 1.2, fuel:air ratio = 1.2 x 0.0653 = 0.0784 = f 1 and A/F ratio = = 12.8:1 0.0784 Consider 1 lb of air + 0.0784 Ib fuel, From Fig. 6. (charts) and T1 = 660R U1 = 25 Btu / chart quantity nR@T1 From gas laws, V1 = where R@ = universal gas constant P1 = 1545 ft lbf / R/lb.

In an engine, the mass fraction of burnt combustion products. f remains in the next charge as a residual. For 41 = 1.2, can be shown that the number of moles, n, of mixture in the engine cylinder per chart quantity (CQ) = 0.0352 + 0.004f = 0.0352 + (0.004 x 0.0784) #N = 0.0355 0.0355 x 1545 x 660 #V1 = = 19.34 ft /CQ 144X 13 From Fig. 5, at T1 = 660-R and # = 1.2

and (MR)# = 1.2 = 0.06993 (MR)#= 1.2 loger = 0.06993 loge 8 = 0.145

#T2 = 1255 R From gas laws.

T, 1255 P2 = r@P1 = 8 x 13 x = 198 lbf/in2 T1 660 and V@ 19.34 V= = =2.42 ft3/CQ 8 From Fig. 11, at T@ = 1255 R U = 168 Btu/CO From Fig. 8, at U@ = (1 - F1)( - 55.162) + F1( - 1379.24) = (1 - O.039)( - 55.162) + 0.039( - 1379.24) = - 107 Btu/CQ U3 = U2 + U@ = 168 - 107 = 61 Btu/CQ anc V3 = V2 = 2.42 ft3/CO S3 = 2.31 Btu/CQ R T3 = 5223 R P3 = 892 Ibf/in2 Normal Cycle expansion to exhaust opening, From Fig. 8, V4 = V1 = 19.34 ft3/CQ U4 = - 605 Btu/CQ T4 = 3205 R P4 = 70 Ibf/in2 Further expansion in MEMS cycle From Fig. 8, V6 = 2 x 19.34 = 38.68ft3/CQ U5 = - 760 Btu/CQ T5 = 2700 R P5 = 33 lbf/in2 Final blow-down to atmosphere normal and MEMS cycles same From Fig. 8, S6 = S5 = S4 = S3 = 2.31 Btu/CQ R V6 = V4' = 61.0 ft3/CQ U6 = U4' = - 860 Btu/CQ T6 = T4' = 2350 R P6 = P4' = 147 Ibf/in2 Check of temperature T, and residuals f, balance 2.42 fi = - = 0.0397 (0.039 assumed) 61.0 From Fig. 12, at T, = 550 'R, HSa = 9 Btu/CQ From Fig. 8. at U6 (= U4,) = - 860 Btu/CQ H56 = 655 Btu/CQ Hs1 = (1 = f1)HSa + f1(HS6) = (1 - 0.039)9 + 0.039(655) = 34 Btu/CQ From Fig. 12, at HS1 = 34 Btu/CQ T, = 660 R (660 R assumed) ie. Balance is achieved for both F, and T,.

CYCLE PERFORMANCE For derivation of pumping losses calculation for both Normal and MEMS cycles, see Appendix B.

'Normal' Engine Cycle Compression work, Wc = U1 - U2 = 25 - 168 = - 143 Btu/CQ Expansion work, WE = U3 - U4 = 61 - (-605) = 666 Btu/CQ 144 144 Pumping work, WP = (P4' - P1)(V2 - V1) =(14.7 - 13)(2.42 - 19.34) 778 778 = 5 Btu/CO WNET = - 143 + 666 - S = 518 Btu/CQ Indicated Mean Effective Pressure, WNET 518 x 778 IMEP = = = 166 lbf/in2 V1 - V2 (19.34 - 2.42) 144 Fuel: air cycle Indicated Efficiency, WNFT x 102 A = (FUEL LHV)(F/A RATIO)(1 - f1) 518 x 102 = = 36.2% 19020 x 0.0784(1 - 0.039) Indicated Specific Fuel Consumption.

2545 Btu/hr ISFC = ~~~~~~~~~~~~~~~~ #F-A(FUEL LHV) 2545 0.362 x 19020 = 0.370 lb/IHP hr.

MEMS Engine Cycle W@ = - 143 Bt

144 144 Wp = (2P6 - P,)(V, - V,) = [(2 x 14.7) - 1 3](2.42 - 19.34)- 788 778 = - 51 Btu/CQ WNET = - 143 + 744 - 51 = 550 Btu/CQ 550 x 778 IMEP = - = 176 Ibf/in.2 (19.34-2.42)144 NOTE IMEP based on equivalent end (normal cycle) cylinder mean effective pressure.

550 x 10 -=38.4% 19020 x 0.784(1 - 0 039) 9545 ISFC = - = 0.348 Ib/lHP hr.

0.384 x 19020 There is therefore, an improvement in net work output. and hence IMEP, #F-A and ISFC of

AT THE SAME EXHAUST PRESSURE.

Air Standard Efficiency,

For air at intake conditions used, y = 1.4 rc = 8

Relative Efficiency, #F-A X 107 #REL #A-S NORMAL CYCLE: 36.2X 102 #REL = - = 64.0% 56.5 MEMS CYCLE: 38.4 x 102 #REL= - = 68.0% 56.5 into the MEMS cycle shows a 4% improvement in relative efficiency over the normal cycle.

See Fig. 10 for distribution of temperatures, pressures and volumes in the real cycle.

(ii) INPUT As in (i) but exhaust manifold pressure = 16.7 lbf/in2 CALCULATIONS Assume f, = 0.045 T, = 660 H Following the same procedure as in the part (i) calculations, for the normal cycle only: U, = 25 0.0355 X 1545 x 660 V1 = - = 19.34 144X 13

(MR)@=1 2logerc = 0.145 V4 = 19.34

T2 = 1255 1255 S4' = S4 = S3 = 2.308 P2 = 8 x 13 x = 198 V4' = 55 660 U4' = - 845 T4' = 2400 19.34 P4' = 16.7 V2 = - = 2.42 8 U2 = 168 Uc = ( 1 = 0.043)( - 55.162) + 0.043( - 1379.24) = - 112 U3 = 168 - 112 = 56 V3 = V2 = 2.42 Check of T7 and f, 2.42 f1 = = 0.0440 (0.045 assumed) 55 HSa = 9 U4' = - 845 HS6 = 683 HS1 = (1 - 0.045)9 + 0.045(683) = 39 #T1 = 660 (660 assumed) ie. T, and f, balance.

CYCLE PERFORMANCE 'Normal' Cycle (only) Wc = 25 - 168 = - 143 WE = 56 - ( - 610) = 666 144 WP = (16.7 - 13)(2.42 - 19.34)- = - 12 799 WNET = - 143 + 666 - 12 = 511 511 x778 IMEP =163 (19.34 - 2.42)144 511 x 102 #F-A = = 35.9% 19020 x 0.0784(1 = 0.045) 2545 ISFC = - = 0.373 0.359 x 19020 35.9 x 102 #REL = = 63.5% 56.5 ie. the effect of the increased exhaust pressure (typical case) in the conventional cycle in to lower the net work output, and hence the rest of the calculated spotifich, by

It can be seen from ref, 3 P128 Fig. 7.3, that the actual net work loss (&alpha; to BHP) nt P1 = 13 and P4' = 14.7 raised to 16.7, and read off this graph for a 'typical engine', is approx.

This is somewhat higher than the calculated valve of 1.4% and will be considered further under the 'LOSSES' section.

(iii) INPUT Within (i) but compression ratio = 9.5:1 CALCULATIONS Assume f, = 0.0323 T1 = 620 Following the same procedure as in part (i) calculations for both cycles.

U1 = 23 0.0355 x 1545 x 620 V1 = = 18.17 144 x 13

T2 = 1260 1260 P = 9.5 x 13 x - = 251 620 18.17 V,- = 1.91 9.5 U2 = 170 Uc = (1 - 0.0323)( - 55.162) + 0.0323( - 1379.24) = - 98 U3 = 1 70 - 98 = 72 V3 = V2 = 1.91 S3 = 2.29 T3 = 5240 P3 = 1120 Normal Cycle Expansion to exhaust opening, V4 = V1 = 18.17 U4 = - 650 T4 = 3090 P4 = 72 Further expansion in MEMS cycle V5 = 2 x 18.17=36.34 U5= -810 T5 = 2510 P5 = 33 Final blow-down-Normal and MEMS S6 = S6 = S4 = S3 = 2.29 V6 = = V4' = 60.0 U6 = U4' = - 890 T6 = T4' = 2215 P6 = P4' = 14.7 Check of T, and f, 1.91 f, = - = 0.0318 (0.0323 assumed) 60.0 HSa = 9 U4' = - 890 HS6 = 610 HS1 = (1 - 0.0323)9 + 0.0323(610) = 28 #T1 = 625 (620 assumed) ie. T, and f1 balance.

CYCLE PERFORMANCE 'Normal' Cycle only Wc = 23 - 170 = - 147 WE = 72 - ( - 650) = 722 144 WP = (14.7 = 13)(1.91 - 18.17) - = - 5 778 WNET = - 147 + 722 - 5 = 570 570 x 778 IMEP = = 189 (18.17 - 1.91)144 570 X 102 #F-A = = 39.0% 19020 X 0.0784(1 - 0.032) 2545 ISFC = = 0.343 0.39 X 19020 MEMS Cycle Wc = - 147

144 WP = [(2 X 14.7) - 13](1.91 - 18.17) = - 49 778 WNET = - 147 + 802 - 49 = 606 606 X 778 IMEP = = 201 (18.17 - 1.91)144 606 X 102 DF-A ""41.5% 19020 X 0.0784(1 - 0.032) 2545 ISFC = ~~~~~~~~~~~~ = 0.322 0.415 X 19020 Hence improvement in net work output, etc. of MEMS over normal cycle

AT THE SAME EXHAUST PRESSURE.

NORMAL CYCLE: 39.0 X 102 #REL = = 65.7% 59.4 MEMS CYCLE: 41.5X 102 #REL = = 69.9% 59.4 ie. MEMS cycle is 4.2% better than normal cycle.

The improvement of 6.2% appears to be unchanged by an increase in compression ratio and this part (iii) calculation serves also as an accuracy check for the part (i) calculations.

ESTIMATION OF LOSSES These losses are additional to the calculated compression and pumping losses etc. shown in the Main Calculations, and can be summarized as follows: 1. Time loss due to motion of piston during combustion and further expansion in the MEMS cycle.

2. Exhaust blow-down loss due to early opening of exhaust valve.

3. Piston blow-by loss.

4. Heat loss during the working cycle: (i) Convection (+ radiation) heat loss to coolant.

(ii) Incomplete combustion.

(iii) 'Slow' burning.

5. Internal friction.

6. Transfer phase losses in MEMS cycle only: (i) Loss due to expansion through transfer port from end to expansion cylinder.

(ii) 'Dead' volume loss during initial expansion through transfer port.

7. Exhaust system loss.

The following is a breakdown and estimation of these losses in both the conventional and MEMS cycles.

The preceding Figs. 11-14 depict the following: Figure 11-normal and MEMS pressure-volume diagrams showing the idealized and real diagrams.

Figure 12-shows Fig. 11 effectively plotted over an "unfolded" volume to clarify the normal and MEMS expansion and exhaust blow-down phases.

Figure l3shows actual P-V diagrams with the 'normal' and MEMS expansion volumes separated to show the breakdown of losses in the expansion cylinder.

Figure shows rough layout of one normal (end) cylinder and the expansion cylinder without valves etc. with approximate dimensions for the transfer port for calculations shown later.

These figures are referred to in the text.

1. TIME LOSS This is described in ref. 2 P108 Figs. 5-18, horizontal hatched lines area.

Normal Cycle From ref. 2 P122 and P123 Fig. 5-11, the apparent time loss is about 6%. ie. ratio of MEP fuel-air:actual = 80% (full load).

MEMS Cycle Similarly. the apparent time loss for the normal cycle cylinder will be 6%.

A small additional time loss will occur in the expansion cylinder due to the piston nearing T.D C in this cylinder as the exhaust valve closes and before the transfer valve in one end cylinder has opened enough for the transferred charge to do useful work on the expansion cylinder piston as it starts descending again after T.D.C. Fig. 9 diagrams 1, 2, 3 and 4 phases show this. Fig. 1 3 diagrams 1 and 3 show this time loss which is the work gained (vertical hatching) minus the work lost (horizontal hatching) ie. same as ref. 2 P113 (Time Loss).

This loss is essentially the same as with the time loss that takes place in a normal 4-stroke cycle during the combustion phase and is as shown above to be 6% of the total cycle efficiency, ie. equivalent to 30% of 20% total loss between the actual and fuel-air cycles. We can, therefore, make a reasonable assumption that the expansion cylinder's apparent time loss will be the same (6%) proportion of the expansion cylinder's contribution to the total work output, when related to the actual cycle efficiency.

ie. from r, 8:1 calculations: Expansion work output from end cylinder = 666 Btu/CQ Expansion work output from expansion cylinder = 744 Btu/CQ 744 - 666 X x x 6% = 0.7% contribution to total loss 666 Total MEMS time loss = 6 + 0.7 = 6.7% EXHAUST BLOW-DOWN LOSS This is described in ref. 2 P108 Fig. 5-1A point C to 1.

Normal Cycle From ref. 2 P122 and P123 Fig. 5-11, exhaust blow-down loss is about 2% for 80% fuel-air cycle efficiency. See diagram 1 in Fig. 11 and 1 2.

MEMS Cycle This loss is essentially of the same nature as that of the normal cycle. Early exhaust valve opening before B.D.C. causes loss of work done on the piston due to a pressure drop as exhaust gas starts to expand to the exhaust system pressure before the piston reaches B.D.C.

In the MEMS cycle, the condition of the exhaust gas at the start of blow-down is as follows from previous fuel-air cycle calculations: Cycle Pressure (Ibf/in-) Temperature ( R) NORMAL 70 3205 MEMS 33 2700 It may be possible to have a later exhaust valve opening near B.D.C. in the expansion cylinder and still achieve efficient emptying as the initial pressure is much lower due to the further expansion work phase. In any case the MEMS expansion cylinder blow-down loss will be small and can be reasonably estimated as follows: U6 - U5 - 860 - ( - 760) %loss = x 2% x 2% = 0.392 x 2% = 0.8% U6-U4 -860-(-605) See diagram 2 Figs. 11 and 12.

3. PISTON BLOW-BY LOSS This is described in ref. 7 P109 under "Leakage".

In the normal 4-stroke cycle, the ring blow-by is usually very small, the same obviously applied to the MEMS firing cylinder. There will be an additional loss due to blow-by in the expansion cylinder, but as the peak pressure in the cylinder is approximately 1/13 of the firing cylinder pressure, this loss will be negligible. The loss due to blow-by in both the normal and MEMS cycles can therefore be ignored as in any case the actual loss for both cycles will be the same.

4. HEAT LOSS DURING THE WORKING CYCLE The nature of the (apparent) total heat loss is fully described in ref. 2 Chapter 8 P266 and P122-127 and also ref. 3 chapter 8.

(i) Convection, conduction and radiation loss.

Normal Cycle Fig. 5-11 in ref. 2 P123 shows that for low cylinder wall temperature (water cooled) engines, the typical total heat loss is about 12% of the total efficiency.

MEMS Cycle Ref. 2 P303 table 8-3 gives distribution of heat loss from the cylinder surfaces of a normal cycle engine and from chapter 8; P226 and P126, 127 and 302 also from ref. 2, a breakdown of the heat losses (only) can be approximated as follows: Compression + combustion + expansion = 17% barrel + 18% head = 35% (20-50%) Rest of loss = exhaust opening to end of blow-down, heat transfer = 65% consisting of 42% head exhaust passage + exhaust valve seat + 23% from exhaust port.

These are percentages of total heat loss of 12%.

In the MEMS cycle heat will also be rejected through the expansion cylinder walls and space above the piston. The increase in heat loss due to this can be approximated as follows, as heat loss is proportional to increased work output from the expansion cylinder: U6-U6 744-666 increase = - x 35% = - x 35 = 0.117 x 35 = 4.1% U6-U4 666 ie. total compression, combustion and expansion loss = 35 + 4.1 = 39.1 Similarly, the head heat passage + exhaust valve seat + port heat loss will be reduced, as "energy" has already been removed from the expansion cylinder in the same proportion as the work gain: ie. 0.392(42% + 23%) = 25.5%, total = 42 + 23 - 25.5 = 39.5%.

Heat rejection will also take place through the transfer port walls and transfer valve seat in the same manner as the normal cycle exhaust valve and port passage. This loss would be approximately: (Normal exhaust head passage + seat heat loss) - (heat loss from expansion cylinder exhaust head passage + seat) = 42-(0.397 x 42) = 25.5% Total MEMS heat loss = (39.1 + 39.5 + 25.5)12 = 12.5% In the MEMS cycle, it is intended to run the expansion cylinder 'hot' and "partial cooling' up to the thermo-mechanical limit of the expansion cylinder materials ie. running at the highest possible temperature in the expansion cylinder and cooling the 'end' cylinders normally. The end cylinders output heat flow will be passed to the expansion cylinder. See Fig. 15. The overall thermal efficiency should thus be raised.

It can be seen from ref. 2 P124 Fig. 5-12 and ref. 3 P138, that in an air-cooled engine that runs at a high cylinder wall temperature, the heat loss is reduced. The high temperature of the normal cycle 'end' cylinders would not be acceptable from a reliability standpoint, but the much lighter stressed further expansion cylinder could be run at a much higher temperature.

Thus we think it possible that by careful design resulting from research into the optimum cooling jacket system and head and block design etc.. that it will be possible to convert about 30% of the lost heat by conversion to useful work in the expansion cylinder. Summarizing, this would occur by: 1. Re-couping a proportion of the jacket heat loss to the coolant by re-direction, which in an efficient engine design, would be nearer 50% of the total heat loss.

2. Increasing the operating temperature and hence work output of the expansion cylinder by 1, above + 'partial cooling' of this cylinder, raising its temperature up to its maximum possible reliable operating value.

This saving of 30% represents 30% X 12.5% from previously, ie. 3.8%. Therefore the total MEMS heat loss becomes 12.5 - 3.8 = 8.7%.

(ii) Incomplete Combustion Normal and MEMS Cycles This phenomena is described in ref. 2 P109.

Combustion in the MEMS cycle is likely to be more complete than the normal cycle, as it is intended to run the expansion cylinder hot. The incompetely burned part of the exhaust gas, in the normal cycle. is due to quenching of the hot combustion gases on the cool surfaces of the combustion chamber. In the MEMS cycle there is likely to be more complete combustion as the gases are expanded down further in a hot' cylinder. Although this loss is very small in terms of lost work, a significant improvement should be achieved in harmful emission levels, at it is the unburnt products of combustion that are largely responsible for these emissions. See also ref. 4 P33.

In conclusion the incomplete combustion loss can be ignored for both cycles as the difference between them would be negligible. although the emissions aspect is an important difference between the two cycles.

(iii) 'Slow' Burning Normal and MEMS cycles Again. this is described in ref. 2 P109, 112 and Fig. 5-9, P121.

The IMEP and cycle efficiency are the same, using fuel-air cycle charts, with progressive (slow) burning as those where simultaneous burning takes place. The piston is assumed to remain at T.D.C. during combustion of all the charge before expansion takes place.

Although a very small loss exists, it will be essentially the same for both cycles and can be ignored for this analysis.

5. INTERNAL FRICTON Ref. 2 chapter 312 deals with internal friction in detail.

Normal Cycle An estimation of the fricton loss can be calculated with reference to ref. 2 P331 Fig. 9-10 and table P332, as follows: Assume as previously calculated that IMEP = 1661bf/in2 and engine mean piston speed, Vp"" = 1600 ft/min. (typical maximum torque figure).

.'. Mean gas pressure, PG = 0.25 X 1 66 = 41.5 Ibf/in2 From Fig. 9-10, .. motoring mean effective pressure, MMEP = 181bf/in2 Mechanical efficiency

ie. loss 10089.2 = 10.8% MEMS Cycle End cylinder is same as above: MMEP = 18 Ibf/in2 and 10.8% loss Expansion cylinder proportion of IMEP = 176 (from previously) - 166 = 10 Ibf/in2 At Vp= 1600 ft/min, equivalent pressure= 10 X 0.25 = 2.5 Ibf/in.

From Fig. 9-10, .MMEP"" = 15.3 Ibf/in2 The MEMS cycle has the following extra components including modification of extra and cylinder exhaust valve: Main frictional components Normal MEMS Pistons + rings 1 1 + 1 expansion = 2 Valves (+ bearings) 2 2 + 1 transfer + 1 extra exhaust = 4 From ref. 2, P359 27% of MMEP is due to bearins + valve gear, ie. 27% X 15.3 = 4.1 Ibf/in2.

In the MEMS engine, the component friction would be the same for the 'end' cylinder + 2 more cam lobes, rockers and valves (assuming the modification of the extra end cylinder 'exhaust' valve) + 1 more bearing. The increase in friction would be a bit less than twice the normal cycle as in practice, in a multi-cylinder engine, there would be the addition of only 1 bearing required. Also the extra end cylinder exhaust valve would be of small proportions compared with the rest of the valves due to its light duty requirements. Therefore it seems reasonable to assume that 80% more friction than standard would be about right, ie.

0.8 X 15.3 X 27% = 3.3 Ibf/in2 (compare with 4.1 Ibf/in2).

From ref. 2, P359, 73% of the total friction is due to the pistons and rings, ie.

73% X 15.3 = 11.2%.

In the MEMS engine, a lower ring pressure could be used in the expansion cylinder, due to the lower pressure and temperatures. This would reduce the friction loss. Although the piston is twice the diameter of its normal cylinder counterpart, there would therefore only need to be 1 dual purpose oil control and compression ring instead of the normal 2 or 3 rings.

A low friction silicon coated bore could probably be used successfully in the expansion cylinder to cut friction down still further and the piston could be made very light due to its light duty. A slipper piston would again reduce skirt friction with the non-thrust faces cut away and would last much longer due to its light loading, than in a normal engine cylinder. Therefore the piston + ring friction of the expansion piston would probably be approximately the same as an end cylinder piston + ring of half the diameter, and a fair bit lower still with good design. See ref. 2 P329 for lower friction in aircraft engines due to these factors.

From ref. 2 P335 and 336, at Vp = 1 600 ft/min, ring + piston friction is likely to be about equivalent to 4 Ibf/in2 lower in the expansion cylinder.

As 80% of the piston + ring friction is due to the rings (ref. 2 P335), it is reasonable to assume that the additional piston + ring fricton would be approximately 11.2 - 4"" 7.2 Ibf/in2.

ie. Equivalent expansion cylinder total friction MEP = 3.3 + 7.2 = 10.5 Ibf/in2 Total MMEP, end + expansion cylinders = 10.5 + 1 8 = 28.5 Ibf/in2 .-. % of total mechanical efficiency

ie. loss = 100 - 83.8"" 16.2% (compared with 10.8% normal cycle ie. 50% greater) 6. TRANSFER PHASE LOSSES IN MEMS CYCLE ONLY There will be a work loss in the MEMS cycle due to a pressure loss during the further expansion phase by transference of the gas from the end cylinder to the expansion cylinder through the transfer valve and port restriction. This loss would have 2 components as follows: (i) Loss due to quasi-steady flow through valve and port during further expansion. See Fig.

1 3 diagram 2 and Fig. 1 4 for approximate transfer port dimensions and layout.

The effect of the valve and port restrictions would be to increase the pressure in the end cylinder. which in turn will increase the work loss in this cylinder as the piston is rising. The pressure in the expansion cylinder would drop and hence the amount of useful work being done on the expansion cylinder piston would be decreased. It is therefore very important to effect the exhaust gas transference with as small a pressure drop as possible.

The flow between the end and expansion cylinders through the transfer valve and short port can be approximated very roughly to riozzle flow between 2 reservoirs constantly varying in volume in the ratio of 2:1. See Fig. 14 and 16.

Inefficiency occurs in the port due to degradation of energy in the devergent section as shown by diagram 2 in Fig. 16.

In reality the gas transfer through the nozzle also takes place through a valve, whose discharge coefficient and flow area is constantly varying and the compressible flow will probably produce wave effects in the transfer port. and the gas velocity will vary from zero to sonic and back to zero per cycle. Because of this very complex flow system it is adequate, for the purposes of this report. to factor the likely flow by analogy with a similar flow system such as a jet aircraft air intake. We can reasonably say that the MEMS cycle efficiency would normally be lower, due to the valve in the flow system and sonic exhaust gas velocity limiting mass flow at the throat, than a typical isentropic ram recovery efficiency of 90% is, say 80%.We can say that from the pressure drop normally experienced through poppet valves and ports, see ref. 2 P340-342 and 503-509, that this figure is probably reasonable. This loss could be predicted reasonably accurately by computer simulation of the basic flow system although wave effects in practice, may appreciably interfere with the flow and be much more of a problem to model.

We feel however that this 80% valve could be raised to 90% by using a large and well designed transfer port and valve or 2 of these to increase the area.

From the main calculations (i), Normal cycle expansion to exhaust opening gives internal energy U4 = - 605 Btu/CQ MEMS cycle further expansion to exhaust opening gives US = - 760Btu/CQ As As assumed work degradation gives -r1 = 90% #Work lost in transference between cylinders = [ - 760 - ( - 605)](100 - 90) = - 16 Btu / CQ

# WNET = - 143 + 736 - 51 = 542 Btu / CQ ie. net work output gain between 2 cycles is down to

from 6.2% ie. difference of loss = 1.6%.

(ii) The second component of this loss will occur due to the 'dead' volume in the 2 transfer ports and the space above the expansion cylinder piston. An initial 'volume loss' would occur as the high pressure end cylinder gas expands into the transfer ports and mixes with the residual low pressure gas trapped in the transfer ports and small volume above expansion cylinder piston.

The re-calculation of this work loss using the valves and combustion chart method of the main thermodynamic calculations is as follows: Likely practical valve of V6, from Fig. 14,

For typical 4 cylinder 2 litre engine, CR = 8:1 as calculation (i), Swept volume of 1 cylinder, 2000 SV = - = 500 cm3 4 SV 500 Clearance volume = - = = 71.4 cm3 CR-1 8 - 1 Expansion from V. to (V5 = 2V,') gives 'effective expansion ratio' = (500 + 71.4)+(2 x 32.18) = 1.11:1 (500+71.4) (end cylinder: 2 transfer ports) Original conditions for rc = 8 in calculation (i) are V4 = V1 = 19.34 U4= -605 T4 = 3205 P4 = 70 Expansion to 2 transfer port volumes From Fig. 8, V5' = 1.11 x 19.34 = 21.47 S5' = S3 = 2.31 U5' = - 630 T5' = 3140 P5' = 62 Further expansion From Fig. 3, V5 = (21.47 - 19.34) + (2 x 19.34) = 40.81 S5 = S3 = 2.31 U5 = - 780 T5= 2595 p5 = 29 Final blowdown and checks will be same as previously.

CYCLE PERFORMANCE From previously, Wc= -143 Wp= -51

# WNET = - 143 + 741 - 51 = 547 Btu / CQ ie. a loss on original figure of

The total loss from (i) and (ii) is therefore approximately = 1.6 + 0.5 = 2.1%.

7. EXHAUST SYSTEM LOSS A further loss occurs in practice in both normal and MEMS cycles due to the restrictive effect of a practically quiet exhaust system.

Normal Cycle It was stated in the main calculations section (ii), that the actual net work loss for a test engine exhaust system was about 8.3%. The calculated value was 1.4%. It therefore seems realistic to take a mean value of (8.3 + 1.4) = 4.9%.

2 With a good design this figure should be attainable and we have experience of a well designed system that gave a loss of 5.9%.

MEMS Cycle As the exhaust pressure at the start of blow-down is only about half of that of the normal cycle. as shown in "Exhaust Blow-down Loss" section and appendix A, atmospheric or very nearly atmospheric pressure should be obtainable in the exhaust manifold and as an approximation we can say that the loss would be in about the same ratio as the internal energy ratio at start of blow-down as shown under the same section.

i.e. 0 392 X 4.9% = 19% with the same exhaust system as the normal cycle.

SUMMARY OF NORMAL AND MEMS CYCLE LOSSES LOSS NORMAL MEMS Time loss 6.0 6.7 Exhaust blow-down 2.0 0.8 Convection, conduction and radiation 12.0 8.7 Internal friction 10.8 16.2 Transfer phase-valve and port flow loss - 1.6 Transfer phase port volume work loss - 0.5 Exhaust system 4.9 1.9 TOTAL LOSS 35.7% 36.4% Original increase in net work output and hence IMEP, #F-A nad ISPC = 6.2% # Predicted increase in net work output = 6.2 - (35.7 - 36.4) = 5.5% RE-CALCULATION OF CYCLE PERFORMANCE WITH LOSSES Normal cycle From previously, calculation (i), IMEP = 166 lbf/in2 IND#F-A = 36.2% ISFC = 0.370 Ib/lHP hr.

(100 - 35.7) BMEP = 166 = 106.7 lbf/in2 102 (100 - 35.7) # Overall brake thermal efficiency = 36.2 = = 23.3% 102 2545 # BSFC = = 0.574 lb/BHP hr.

0.233 x 19020 and Mechanical efficiency = 100 - 10.8"" 89.2% As a check on the practical validity of these figures, see for example P65, ref. 3-efficiency tests on a 1 litre petrol engine. Above calculated figures are in brackets.

2000 R.P.M. (near max. torque), BMEP, calculated from bore. stroke, no. cylinders, power and R.P.M.. 118.0 Ibf/in2 (106.7) BR.TH@ = 23.2% (23.3) BSFC = 0.584 lb/BHP hr. (0.674) MECH@ = 89 0% (89.2) ie Figures are realistic-it is interesting to note that CR = 8.9 compared with the calculated 8:1. hence drop in BMEP from 118.0 to 106.7 Ibf/in2.

MEMS Cycle From previously. calculation (i), IMEP = 176 Ibf/in2 IND#F-A = 38.4% ISFC = 0.348 Ib/BHP hr.

(100 - 36.4) . . BMEP = 176 = 111 9 Ibf/in2 102 (100-36.4) Overall gBRAIE THERMAL 38.4 = = 24.4% 102 2545 BSFC ""0.548 Ib/BHP hr 0.244 x 19020 and gMECH = 100-16.2 = 83.8% Note that the mechanical efficiency of the MEMS cycle is lower than the standard cycle due to the expansion cylinder mechanical components.

Note also that the MEMS cycle has a brake specific fuel consumption improvement over the conventional cycle of

The fuel consumption of a vehicle would therefore be improved by the same amount.

MODIFICATIONS TO BASIC MEMS CYCLE 1. It may be advantageous to place the transfer valves in the top of the expansion cylinder and make the transfer port part of an end cylinder combustion chamber. The advantages would then be reduced flow losses during transfer and no 'dead volume' loss. The detail design layout of this area needs studying in some depth to see if this would be possible without reducing the compression ratio or making the combustion chamber of the end cylinders poor.

2. A modification to the MEMS cycle, albeit at the expense of extra friction loss, would be to put an extra exhaust bypass valve in each end firing cylinder, with a bypass exhaust port connecting this valve throat to the expansion cylinder exhaust system as shown in Fig. 1 7. The four diagrams show the operating sequence of the extra valve during an end cylinder's final exhaust transference phase as the piston approaches T.D.C. and the further expansion gas transference is completed. As the exhaust transfer valve closes (diagram 2) the extra exhaust valve opens and the exhaust gas that would have been trapped at T.D.C. in the combustion chamber, is allowed to escape and will to some extent be extracted by the main stream exhaust gases flowing through the expansion cylinders exhaust system from both end cylinders.In this way the beneficial exhaust extraction and inlet charge flow inducement during overlap between the inlet and exhaust ports on a normal cycle engine, would be maintained. Otherwise the residual end cylinder exhaust gas would pollute the new intake charge causing poor combustion on the next firing stroke, and reduce the mass flow of gas through the engine. The volumetric efficiency would also be reduced with no overlap.

This modification would permit the design of a pulsing extraction bypass exhaust system taking advantage of the inertia of the main gas stream flowing from two expansion strokes per single four-stroke end cylinder cycle. This would further improve the cycle efficiency by being able to delay the closing of the transfer valve till the piston is nearing T.D.C. so that the further expansion is more complete, and still maintain adequate scavenging and intake charge promotion during the shortened overlap period.

CONCLUSIONS We do stress that the treatment of the losses here is an over-simplification, and that the difficulties of attempting to break down a very complex inter-related problem such as the engine cycle losses of a (practical) 4-stroke internal combustion engine, would result in only a rough estimate of the likely behaviour of the MEMS cycle compared to standard cycle.

Without being able to accurately quantify the expected losses, other than those shown in the main calculations such as pumping losses, we think it reasonable, at this stage to assume that these losses would be roughly balanced out by the gains. We think it possible, therefore, from a practical standpoint, to achieve the sort of gain shown by the comparative combustion chart calculations of around 6% increase in relative thermal efficiency. We would say, in conclusion, that even if only 3-4% advantage were obtainable this gain would be worthwhile in terms of improved fuel economy.

Some other points of note are: 1. Emission levels in the MEMS engine should be lower than the equivalent normal cycle for the reasons explained in 4 (ii) of 'losses' section of this report, and by virtue of the lower residual fraction shown in the main thermodynamic calculations, due to the likely ability of the MEMS cycle to blow-down to atmosphere more easily than the normal cycle.

2. The extra work output theoretically available from the expansion cylinder is relatively high. See main calculations. This is (for r, = 8),

This figure would be

if, during the further expansion phase, the expanding gas was not also acting on the end cylinder piston producing negative work, at the same time as the further expansion cylinder is producing positive work by virtue of its volume being double that of the end cylinder. There will be an optimum compromise between the necessarily large expansion: end cylinder volume ratio and excessive actual pumping loss. 2:1 was chosen as being a practical baseline for calculation.

3. It may be possible to fuel the expansion cylinder with a weak mixture and 'afterburn' this cylinder, thus releasing considerably more power. A gain in brake thermal efficiency may be possible by employing this technique with very little increase in engine weight.

APPENDIX A Estimafion of final exhaust pressures from different initial pressures.

In the combustion chart calculations it was assumed, in part(i) that atmospheric pressure (= 14.7 lbf/in2) was obtainable in the exhaust manifold after blow-down and part (ii) assumed this pressure to be 1 6.7 Ibf/in2, and the volumes and temperatures were hence read from the charts at these values. To try to determine what pressure the MEMS cycle pre-blow-down gas state would achieve in the exhaust system, it will be assumed that this process takes place isentropically and that the restrictive effect of the exhaust system can be approximated to the throat restriction In a nozzle.

From gas laws for nozzles: Throat pressure which would be proportional to exhaust manifold pressure, at critical pressure ratio,

Normal Cycle From Calculation (i), at START of blow-down, P1"" 70 y= 1.343 for exhaust gas (see ref. 2 P383)

MEMS Cycle From calculation (i), at START of blow-down, P. = 33 y= 1.343 # P1= 0.538 X 33 = 17.7 Ibf/in2 Note that these pressures would occur at the start of blow-down only and merely indicate that the exhaust manifold pressure, PE is LIKELY to be nearer atmospheric mean pressure in the MEMS cycle than in the normal cycle.

APPENDIX B Derivation of Pumping Work Loss See ref. 2 P79.

Normal Cycle Work done on piston during inlet stroke, W@= P1(V1 - V2) and work done on piston during exhaust stroke, We = Pe(V2 - V1) where Pj = inlet manifold pressure Pe = exhaust manifold pressure V1 = clearance volume above the piston V2 = swept + clearance volumes or total cylinder volume.

The pumping work is the algebraic sum of the inlet and exhaust work, Wp # WP = P,(V1 - V2) + Pe(V2 - V1) = P1V1 - P,V2 + PeV2 - PeV1 = P,V1 - PeV1 + PeV2 - P1V2 = V1(P1- - Pe) + V2( - P, + Pe) ""V1(P, - Pe) - V2(Pi - Pe) .Wp = (P, - Pe)(V1 - V2)2) MEMS Cycle The further expansion cylinder is assumed to be twice the colume of one end cylinder.

ie. W, = P,(V1 - V2) and We = Pe 2(V2 - V1) # WP = P,(V, - V2) + Pe2(V1 - V2) = P,V, - P,V2 + 2PeV2 - 2PeV1 = P1V1 - 2PeV1 + 2PeV2 - P1V2 = (P, - 2Pe)V, + (2Pe - P,)V2 = (P1 - 2Pe)V1 - (P, - 2Pe) V2 = (P, - 2Pe)(V, - V2) . WP = (2Pe - P1)(V2 - V1) The references referred to hereinbefore are as follows: "Ref. 1 "-"Thermodynamic Properties of Octane and Air for Engine Performance Calcula tions." Newhall and Starkman, SAE 633G.

"Ref. 2"-' 'The Internal Combustion Engine in Theory and Practice". Taylor vol. 1.

"Ref. 3"-' 'The Testing of Internal Combustion Engines". Greene and Lucas.

"Ref. 4"-' 'The Internal Combustion Engine in Theory and Practice". Taylor vol. 2.

Claims (8)

1. A method of operating an internal combustion engine characterised by the steps of igniting a compressed fuel/air mixture in a primary cylinder of the engine to generate a power stroke in said cylinder, subsequently enabling further expansion of the ignited fuel/air mixture in an associated secondary cylinder to generate a further power stroke and applying heat to said secondary cylinder during operation of the engine.

2. A method as claimed in claim 1 wherein the rafio of the working volumes of the or each primary cylinder and the associated secondary cylinder are such that the gas exhausted from the primary cylinder expands into the secondary cylinder substantially to atmospheric pressure before being exhausted from the secondary cylinder to atmosphere.

3. An internal combustion engine characterised in that it comprises at least one primary cylinder (A,B) and an associated secondary cylinder (C) operably coupled to said primary cylinder for enabling further expansion of a fuel/air mixture ignited in said primary cylinder; and means for applying heat to said secondary cylinder; and wherein the pistons of said cylinders are coupled to a common crankshaft.

4. An engine as claimed in claim 3 characterised in that the ratio of the working volumes of the or each primary cylinder and the associated secondary cylinder are such that said exhaust gas from said primary cylinder expands into said secondary cylinder substantially to atmospheric pressure.

5. An engine as claimed in claim 3 or 4 characterised in that the length of the strokes of the pistons of the primary and secondary cylinders are substantially the same.

6. An engine as claimed in any of claims 3 to 5 characterised in that there are provided two primary cylinders associated with said secondary cylinder and operably coupled to said common crankshaft such that said primary cylinders are 180 out of phase with one another and exhaust alternately into said secondary cylinder.

7. An engine as claimed in any of claims 3 to 6 characterised in that there is provided a non-return inlet valve (2. 12, 42, 4. 14, 44) in the head of the or each primary cylinder for induction of fuel, air mixture into said cylinder, and a valve (6, 8, 16, 18, 46, 48) controlling the exhausting of exhause gas from the or each primary cylinder to the associated secondary cylinder and also the exhausting of said exhaust gas from said secondary cylinder.

8. An engine as claimed in claim 7 wherein said controlling valve is a rotary sleeve.