GB2278155A

GB2278155A - Ignition timing control at starting of a mixed fuel engine

Info

Publication number: GB2278155A
Application number: GB9413310A
Authority: GB
Inventors: Yoichi Saito
Original assignee: Fuji Jukogyo KK; Fuji Heavy Industries Ltd
Current assignee: Subaru Corp
Priority date: 1991-07-19
Filing date: 1992-07-17
Publication date: 1994-11-23
Anticipated expiration: 2012-07-17
Also published as: GB2278156A; GB9413312D0; GB2278155B; GB9413311D0; GB2278157B; GB2278156B; GB9413310D0; GB2278157A

Abstract

In an engine supplied with a variable proportion of alcohol to petrol the ignition timing as starting is fixed for a period which is dependant on the engine temperature, e.g. coolant or fuel temperature. After the period of fixed timing the timing is variable dependant upon engine speed and knock, the intake air amount and the alcohol proportion.

Description

DESCRIPTION "A CONTROL METHOD AND AN APPARATUS FOR A FLEXIBLE FUEL VEHICLE" The present invention relates to a starting control method and a starting assist apparatus for an engine for an FFV which enhance starting characteristics.

Recently, for such reasons as the worse situation of fuel and a demand for the purification of emission, systems capable of simultaneously using alcohol as substitute fuel in addition to the conventional fuel of gasoline are coming into practical use as disclosed in, for example, the official gazette of Japanese Patent Application Laid-open No. 69768/1989.

A vehicle such as automobile (FFV; Flexible Fuel Vehicle) equipped with the system can be driven, not only with the gasoline, but also with mixed fuel consisting of the alcohol and the gasoline or with only the alcohol. The alcohol concentration (content) of the fuel for use in the FFV changes between 0% (gasoline only) and 100% (alcohol only) in dependency on user conditions in the case of fuel supply.

In general, the alcohol fuel has such properties as being more difficult to vaporize at low temperatures, exhibiting a larger amount of latent heat of vaporization and exhibiting a higher inflammation point, when compared with the gasoline fuel.

Accordingly, when the alcohol concentration changes, the starting characteristics vary largely in accordance with the temperature condition. In particular, when the alcohol concentration is high, the problem of degradation in the starting characteristics arises in a cold engine state.

There have hitherto been known techniques wherein, in order to cope with the problem, the vaporization of the fuel is promoted or enhancing the starting characteristics by a heater, a heating element or the like serving as a starting assist device. By way of example, the official gazette cf Japanese Patent Application Laid-open No. 52665/1982 discloses a technique wherein a heating device for heating an intake passage is controlled in accordance with the output of an alcohol concentration sensor, and wherein the heat generating quantity of the heating device is increased when the alcohol concentration is a preference value or above.In addiction, the official gazette of Japanese Patent Application Laid-open No. 35179/1980 discloses a technique wherein a distributing value is provided for controlling the distributive amounts of a mixture to flow into main and subsidiary intake passages, while a heat generating element is disposed in the subsidiary intake passage so as to vaporize fuel droplets gathering in this passage at starting the cold engine.

At starting the engine, however, the temperature of the combustion chamber of the engine needs to be raised by the heating means, besides the promotion of the vaporization of the fuel. The prior art therefore necessitates a fuel quantity increasing correction after starting, or the like. The fuel quantity increase after starting produces a deterioration in emission control and an increase in fuel cost.

Further, at starting the engine, the initial knocking which is unpleasant is sometimes caused by the residual fuel in a cylinder or an intake manifold. It has sometimes hampered the engine from smoothly shifting from the starting into the ordinary operation state thereof.

Besides, at the low temperatures, the alcohol is liable to separate from the gasoline in a fuel tank or in piping. It is accordingly understood that the starting mode control based on the output of the alcohol concentration sensor will become inaccurate on account of a nonuniform alcohol concentration distribution. Moreover, in promoting the vaporization of the fuel by the heating means such as heater, the vaporization depends upon the mounted position of the heating means, and the fuel is not always vaporized entirely. It is accordingly understood that the air fuel ratio of the mixture will become inappropriate due to the adhesion of the residual fuel to the inner wall of the intake port -of the engine and the subsequent evaporation of the adhering fuel, resulting in inferior starting of the engine and an increase of the fuel cost.

The present invention has been made in view of the above circumstances, and has for its object to provide a starting control method and a starting assist apparatus for an engine of an FFV according to which, at starting the engine, a precise control is realized to hold a mixture at an appropriate air-fuel ratio and to effectively promote the vaporization of fuel at low temperatures, wnereby the engine can be started smoothly and quickly.

In the first aspect cf the present invention, a control method for a flexible fuel vehicle having an engine mounted on the flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with the intake manifold, a temperature sensor mounted on the engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between a fuel tank and the injector for sensing a concentration of an alcohol in the fuel and for generating a concentration signal, comprises the steps of: setting an engine startable temperature in dependency on said concentration signal; judging whether it is possible to start said engine by comparing said engine startable temperature with said coolant temperature; heating a heater for a predetermined time before injecting said fuel into said cylinder when said coolant temperature is lower than said engine startable temperature; comparing said coolant temperature with a heating stop temperature when said coolant temperature is higher than said engine startable temperature; injecting said fuel when said coolant temperature is lower than said heating stop temperature without heating said heater; activating said heater until said coolant temperature reaches a predetermined temperature; and deactivating said heater when said coolant temperature reaches said heating stop temperature after repeating said comparing step so as to effectively start said engine as soon as possible.

Besides, in the second aspect, a control method for a flexible fuel vehicle having an engine mounted on said flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with said intake manifold, a starter motor mounted on said engine for starting said engine with electric power, a fuel pump for supplying said fuel from a fuel tank to said fuel injector, a temperature sensor mounted on said engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between said fuel tank and said injector or sensing a concentration of an alcohol in said fuel and for generating a concentration signal, comprises the steps of: prohibiting actuation of said starter motor for a predetermined time when starting said engine; returning said fuel from a pressure regulator to said fuel tank; and homogenizing a concentration of said alcohol and a gasoline in said fuel so as to contain an optimum control of said engine.

Further, in the third aspect, a control method for a flexible fuel vehicle having an engine mounted on said flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with said intake manifold, a starter motor mounted on said engine for starting said engine with electric power, a fuel pump for supplying said fuel from a fuel tank to said fuel injector, a temperature sensor mounted on said engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between said fuel tank and said injector for sensing a concentration of an alcohol in said fuel and for generating a concentration signal, comprises the steps of: setting an engine startable temperature in dependency on the concentration; judging whether it is possible to start the engine by comparing the engine startable temperature with the coolant temperature; and cranking the engine for a predetermined time without injecting the fuel when starting the engine is judged impossible, so as to enable an optimum starting of the engine.

Still further, in the fourth aspect, a control method for a flexible fuel vehicle having an engine mounted on the flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with the intake manifold, a starter motor mounted on the engine for starting said engine with electric power, a fuel pump for supplying the fuel from a fuel tank to the fuel Injector, a temperature sensor mounted on the engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between the fuel tank and the injector for sensing a concentration of an alcohol in the fuel and for generating a concentration signal, comprises the steps of:: setting a duration time for fixing ignition timing in dependency on the coolant temperature, and fixing an ignition timing at a predetermined ignition timing while the fixed ignition time elapses after starting the engine so as to effectively start the engine as soon as possible.

The fifth aspect of the present invention consists in that the activating step in the first aspect further comprises: fixing an adhesive ratio of the fuel on an intake port of the cylinder ead and an evaporating ratio of the fuel in the intake port between a present intake stroke and a next intake stroke per cylinder at predetermined values when activating the heater so as to attain an optimum control of the starting of the engine.

Yet further, in the sixth aspect of performance, a control method for a flexible fuel vehicle having an engine mounted on the flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with the intake manifold, a temperature sensor mounted on the engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between a fuel tank and the injector for sensing a concentration of an alcohol in the fuel and for generating a concentration signal, comprises the steps of: setting an engine startable temperature in dependency on the concentration; judging whether it is possible to start the engine by comparing the engine startable temperature with the coolant temperature; prohibiting a fuel injection and an ignition when it is decided to be impossible to start the engine; activating a heater; cranking the engine; and allowing the fuel injection and the ignition after a predetermined time so as to perform an effective and optimum engine starting.

In the seventh aspect of the present invention, a control system for G flexible fuel vehicle having an engine mounted on the flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with the intake manifold, a temperature sensor mounted on the engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between a fuel tank and the injector for sensing a concentration of an alcohol in the fuel and for generating a concentration signal, comprises: a heater included in a heater unit and interposed between the intake manifold and the cylinder head for effectively heating the fuel so as to attain an optimum starting of the engine.

With the control method for an FFV according to the first aspect of the present invention, In the first place, the engine startable temperature is set on the basis of the alcohol concentration of the fuel, and whether or not the engine is startable is judged by comparing the engine startable temperature with the coolant temperature. Herein, when the engine has been judged to be unstartable, the heating means for promoting the vaporization of the fuel is activated for the predetermined time period before the fuel injection.

In contrast, when the engine has been decided to be startable, the coolant temperature is compared with the heating stop temperature. When the coolant temperature is lower than the heating stop temperature as the result of the comparison, the fuel is injected without previously activating the heating means, and the heating means is thereafter activated until the coolant temperature reaches the predetermined value. On the other hand, when the engine temperature is not lower than the heating stop temperature, the heating means is held deactivated.

With the control method for an FFV according to the second aspect, at starting the engine, the actuation of the starter motor is prohibited for the predetermined time period, while the fuel pump is driven to feed the fuel in the fuel tank to the pressure regulator. Then, the fuel fed under pressure returns to the fuel tank via the pressure regulator, whereby the fuel is circulated and stirred. In consequence, the alcohol concentration distribution of the fuel is homogenized.

With the control method for an FFV according to the third aspect, the engine startable temperature is set on the basis of the alcohol concentration of the fuel, and whether or not the engine is startable is judged by comparing the engine startable temperature with the coolant temperature. Herein, when the engine has been judged to be unstartable, it is cranked for the predetermined time period with the fuel Injection prohibited.

With the control method for an FFV according to the fourth aspect, at starting the engine, the ignition timing is fixed to the predetermined timing and is held while the fixed ignition time period set in dependency on the coolant temperature elapses.

With the control method for an FFV according to the fifth aspect, when the heating means for promoting the vaporization of the fuel has been activated as starting the engine, the adhesive ratio of the injected fuel adhering on the inner wall of the intake port and the evaporating ratio of the fuel in the intake port between the intake stroke and the next intake stroke per cylinder are fixed to the predetermined values.

With the control method for an FFV according to the sixth aspect, the engine startable temperature is set on the basis of the alcohol concentration of the fuel, and whether or not the engine is startable is judged by comparing the engine startable temperature with the coolant temperature. Herein, when the engine has been judged to be unstartable, the engine is cranked under the conditions that the fuel injection and the ignition are prohibited and that the heating means for promoting the vaporization of the fuel is activated. Thereafter, when the present period has elapsed, the fuel injection and the ignition are allowed, whereby the injected fuel with its vaporization promoted by the heating means is fed to the engine and is ignited.

The control system for an FFV according to the seventh aspect comprises the heater included in the heater unit and interposed between the cylinder head and the intake manifold for each cylinder, whereby the fuel injected from the injector is vaporized.

By way of example only, a specific embodiment of the present invention will now be described, with reference to the accompanying drawings, in which: Figs. 1 - 4 are flow charts showing control steps in the mode of starting an engine in accordance with the present invention; Fig.5 is a schematic view of an engine control system in accordance with the present invention; Fig.6 is a detailed view of a heater mounting portion; Fig.7 is a sectional view taken along line A-A in Fig.6; Fig.8 is an explanatory diagram showing a fuel injection state in the vicinity of an intake port; Fig.9 is a front view of a crank rotor as well as a crank angle sensor; Fig. 10 is a front view of a cam rotor as well as a cam angle sensor; Fig.ll is a circuit arrangement diagram of a control unit;; Fig. 12 is an explanatory diagram showing a startable range and an unstartable range; Fig. 13 is a conceptual diagram of a coolant temperature map for judging if an engine is startable; Fig. 14 is a diagram of the characteristics of a heater; Fig. 15 is a conceptual diagram of an electric power map for judging if heating is to be finished; Fig. 16 is a conceptual diagram of a fixed ignition time map; Fig. 17 is a flow chart showing steps for controlling a starter motor; Fig. 18 is a flow chart showing steps for discriminating cylinder No. and for calculating engine R.P.M.; Figs. 19 - 21 are flow charts showing steps for setting a fuel injection amount and an ignition timing; Fig. 22 is a conceptual diagram of a map for a desired air-fuel ratio; Fig. 23 is a conceptual diagram of a map for the adhesive ratio of fuel adhering on a wall;; Fig. 24 is a conceptual diagram of a map for the evaporating ratio of the fuel; Fig. 25 is a conceptual diagram of a map for a basic ignition timing; Fig. 26 is a conceptual diagram of a map for an injection starting crank angle; Fig. 27 is a flow chart showing ignition control steps; Fig. 28 is a flow chart showing fuel injection control steps; and Fig. 29 is a time chart of fuel injection and ignition operations.

Now, embodiments of the present invention will be described with reference to the drawings.

[Construction of Engine Control System] Referring to Fig. 5, numeral 1 designates an engine for an FFV (a horizontal opposed type four-cylinder engine in the illustration), the cylinder head 2 of which is formed with intake ports 2a and exhaust ports 2b. An intake manifold 3 is held in communication with the intake ports 2a, a throttle chamber 5 is provided upstream of the intake manifold 3 through an air chamber 4 in communication therewith, and an air cleaner 7 is mounted upstream of the throttle chamber 5 through an intake pipe 6.

On the other hand, an exhaust pipe 9 is held in communication with the exhaust ports 2b through an exhaust manifold 8, and a catalytic converter 10 is disposed in the exhaust ppe 9. In addition, a throttle valve 5a is provided in the throttle chamber 5, and an intercooler 11 is dis?osec ifl the intake pipe 6 directly upstream of the throttle chamber 5. Further, a resonator chamber 12 is provided in the intake pipe 6 downstream of the air cleaner 7.

Besides, a bypass 13 is provided for bringing the resonator chamber 12 and the air chamber 4 into communication, thereby to bypass the upper stream side of the throttle valve Sa from the lower stream side thereof, and an idle speed control valve (ISCV) 14 which is an actuator for the operation of controlling an engine speed is mounted in the bypass 13. Further, a check valve 14a which is opened in response to a negative intake pressure Is mounted directly downstream of the ISCV 14.

The ISCV 14 is, for example, a rotary valve which is driven by a duty solenoid. In accordance with a valve opening degree which is determined by the duty ratio of a drive signal or the ISCV 14, the quantity of air in the bypass 13 is regulated to control the engine speed at engine idling. By the way, in this embodiment, the opening degree of the SCV 14 is enlarged with the duty ratio.

Numeral 15 indicates a turbocharger which is an example of a supercharger. The turbine wheel 15a of the turbocharger 15 is housed in turbine housing 15b which is formed midway of the exhaust pipe 9, while the compressor wheel 15d thereof connected to the above turbine wheel 15a through a turbine shaft 15c is housed in a compressor housing 15e which is formed in the intake pipe 6 downstream of the resonator chamber 12.

Besides, a wastegate valve 16 is provided in the inlet port of the turbine housing 15b, and a lever 17 connected to the wastegate valve 16 is connected to the diaphragm 18a of a diaphragm actuator 18 through a rod 19. Further, the pressure chamber 18b of the diaphragm actuator 18 is held in communication with the intake pipe 6 downstream of the turbocharger 15, through a pressure passage 20, a duty solenoid valve 21 as an example of an actuator for controlling a supercharging pressure is mounted midway of the pressure passage 20, and the valve body 21a of the duty solenoid valve 21 is disposed in opposition to the drain port of a pressure reducing passage 22 communicating with the resonator chamber 12.

The duty solenoid valve 21 is controlled by a duty signal applied from an electronic control unit (ECU) 41 to be described later, and it regulates a pressure to be supplied into the pressure chamber 18b of the diaphragm actuator 18. In this regard, a diaphragm spring 18c normally urges the diaphragm 18a of the diaphragm actuator 18 in the direction of retreating the rod 19 so as to close the wastegate valve 16 through this rod 19 as well as the lever 17. Thus the wastegate valve 16 is operated by the balance between the inner pressure of the pressure chamber 18b and the force of the diaphragm spring 18c. In turn, the opening area of the inlet port of the turbine housing 15b is controlled by the wastegate valve 16, thereby to control the maximum supercharging pressure.

Here in this embodiment, as the duty ratio of the duty signal is increased more, the pressure reducing passage 22 is opened longer per unit time by the valve body 21a of the duty solenoid valve 21, and the leaking quantity of a positive pressure downstream of the compressor wheel '5d to be supplied into the pressure chamber 18b of the diaphragm actuator 18 is enlarged more. Consequently, the supercharging pressure is relatively raised to heighten the maximum supercharging pressure based on the turbocharger 15.

Besides, an Intake-port heater unit 23 as a starting assist device is provided in each of the intake ports 2a of respective cylinders communicating with the intake manifold 3, and an injector 24 is confronted to a position opposing to the corresponding intake-port heater unit 23 directly upstream of the intake port 2a.

Further, a spark plug 40 whose fore end is exposed to a combustion chamber is mounted in each of the cylinders in the cylinder head 2.

As shown in Fig. 6, the intake-port heater unit 23 includes a heating element 23a facing the interior of an intake passage. A mounting portion configured of an insulator 23b and a flange 23c is held between the intake manifold 3 and the cylinder head 2, and is fixed to the cylinder head 2 with bolts or the like, not shown. The heating element 23a has a built-in heater 23d made of a PTC (positive temperature coefficient) pill on a side corresponding to the fuel injecting direction of the injector 24.

Besides, as shown in Fig. 7, the heating element 23a is formed to be cylindrical. It is supported by the flange 23c through stays 23e so as to face the interior of the intake passage, and is substantially thermally insulated from the intake manifold 3 and the cylinder head 2 by the insulator 23b.

Herein, when the heater 23d is energized through a terminal 23f, fuel injected from the injector 24 is vaporized by the heating element 23a, and it is distributed to two intake valves 2c as shown in Fig. 8.

In addition, the injectors 24 are held in communication with a fuel tank 26 through 2 fuel feed passage 25. The fuel tank 26 stores therein fuel consisting of gasoline only, fuel consisting of alcohol only, or mixed fuel consisting of gasoline and alcohol at a predetermined alcohol concentration, that is, fuel whose alcohol concentration changes between 0% and 100% in dependency on user conditions in the case of fuel supply.

Besides, a fuel pump 27 is provided in the fuel tank 26. The fuel from the fuel pump 27 is fed to the injectors 24 and a pressure regulator 30 via a fuel filter 28 and an alcohol concentration sensor 29 which are inserted in the fuel feed passage 25. Then, the fuel is returned from the pressure regulator 30 into the fuel tank 26, thereby to have its pressure regulated to a predetermined value.

The alcohol concentration sensor 29 is constructed of, for example, a pair of electrodes which are provided in the fuel supply passage 25, and which detect the alcohol concentration by detecting a current variation based on the electric conductivity variation of the uel. Incidentally, the alcchol concentration sensor 29 is not restricted to the type utilizing the electric conductivity variation, but one of resistance detection type, capacitance type or optical type may well be employed alternatively.

Also, an intake air quantity sensor (a hot wire type airflow meter in the illustration) 31 is provided in the intake pipe 6 directly downstream of the air leaner 7. A throttle opening-degree sensor 32a, and an idle switch 32b for detecting the full closure of the throttle valve 5a are connected to this throttle valve 5a. Further, a knock sensor 33 is mounted on the cylinder block la of the engine. A coolant temperature sensor 34 is confronted to a coolant passage (not shown) formed in the cylinder block la, while an 02sensor 35 is installed to the exhaust pipe 9.

Further, a crank rotor 36 is secured to a crankshaft 1b supported in the cylinder block la, and a crank angle sensor 37 is provided in opposition to the outer periphery of the crank rotor 36. Also, a cam angle sensor 39 made of a magnetic pickup or the like for discriminating cylinder Nos. is mounted in opposition to a cam rotor 38 which is secured to the cam shaft lc of the engine 1.

As shown in Fig. 9, the crank rotor 36 is formed with prcjections 36, 36b and 36c at its outer periphery.

By way of example, the respective projections 36a, 36b and 36c are formed at positions 61, e2 and 63 (for example, el = 970, 2 = 650 and 63 = 100) before the top dead centers (BTDC) of compression in the respective cylinders (#1, 2 and 3, '4).

More specifically, the projection 36a indicates a reference crank angle in the case of setting an ignition timing as well as a fuel injection timing. The rotational frequency = of the engine is calculated from a time period in which the section between the projections 36a and 36b passes. The projection 36c serves as a reference crank angle indicative of a fixed ignition timing.

Also, as shown in Fig. 10, the outer periphery of the cam rotor 38 is formed with projections 38a, 38b and 38c for discriminating the cylinder Nos. By way of example, the projections 38a are respectively formed at positions 64 (for example, 84 = 200) after the top dead centers (ATDC) of compression in the cylinders 3 and =4. Besides, the projection group 38b is configured of three projections, the firs= one of which is formed at a position 65 (for example, 65 = 50) after the top dead center (ATDC) of the cylinder -1. Further, the projection group 38c is confIgured of two projections, the first one of which is foxed at a position 66 (for example, 66 = 20e) after the top dead center (ATDC) of the cylinder 2.

Incidentally, each of the crank angle sensor 37 and the cam angle sensor 39 is not restricted to the magnetic sensor such as magnetic pickup, but it may well be an optical sensor or the like.

[Circuit Arrangement of Electronic Control Unit] Referring now to Fig. 11, numeral 41 designates an electronic control unit (ECU) constructed of a microcomputer or the like, in which a CPU 42, a ROM 43, a RAM 44, a backup RAM 44a and an I/O interface 45 are interconnected through a bus line 46, and in which each element is supplied with predetermined stabilized voltages by a voltage regulating circuit 47.

The voltage regulating circuit 47 is connected to a battery 49 through the relay contact of an ECU relay 48, and the relay coil of the ECU relay 48 is connected to the battery 49 through an ignition switch 50. Also, a starter motor 62 is connected to the above battery 49 through a starter switch 60 and the relay contact of a starter motor relay 61, while the fuel pump 27 is connected thereto through the relay contact of a fuel pump relay 51. Further, the intake-port heater unit 23 of each cylinder is connected to the same battery through the relay contact of a heater relay 52 and a current sensor 63.

In addition, the various sensors 29, 31, 32a, 33, 34, 35, 37, 39 and 63, the idle switch 32b and the starter switch 60 are connected to the input ports of the I/O interface 45, while the battery 49 is connected to the input port thereof so as to monitor a battery voltage. On the other hand, the igniter 40a of the spark plug 40 is connected to the output port of the I/O interface 45. Further, the ISCV 14, the duty solenoid valve 21, each injector 24, the relay coils (of the fuel pump relay 51, heater relay 52 and starter motor relay 61), and an ECS lamp 59 which is indIcation means for indicating the occurrence of any abnormality or for indicating the state of heater energization are connected to the output ports of the I/O interface 5 through a driver circuit 58.

The ROM 43 stores therein control programs, and the fixed data of various maps etc., while the RAM 44 stores therein data after processing the output signals of the various sensors and switches, and data arithmetically processed by the CPU 42. Besides, the backup RAM 44a stores therein a trouble code etc. at the occurrence of any fault, and the data are held even when the ignition switch 50 is OFF.

The CPU 42 sets various controlled variables such as a fuel injectIon quantity, an Ignition timing, and the duty ratio of a signal for actuating the duty solenoid valve 21, in accordance with the control programs stored in the ROM 43 and on the basis of the various data stored in the RAM 44. Then, it delivers corresponding signals to the injector 24 and the igniter 40a, thereby to perform an air-fuel ratio control and an ignition timing control, and it also delivers the actuation signal to the duty solenoid valve 21, thereby to control the maximum supercharging pressure based on the turbocharger 15.

[Operation] Next, the operation of the embodiment constructed as stated above will be described.

(Starting-moae Control Steps) Flow charts in Figs. 1 - 4 show the program of a starting-mode control which starts upon the closure of the power source of the ECU 41. First, at a step S101, the program is initialized to turn OFF the relays such as starter motor relay 61 and heater relay 52 and to set a timer and clear counters and flags.

Subsequently, at a step S102, the starter motor actuation prohibition flag FLAG1 is set (FLAG1 + 1) to prohibit the actuation of the starter motor 62. Then, at a step S103, the fuel pump actuation allowance flag FLAG2 is set (FLAG2 * 1) to allow the actuation of the fuel pump 27.

Subsequently, at a step S104, the fuel injection prohibition flag FLAG3 is set (FLAG3 i 1) to prohibit fuel injection. Then, at a step S105, the ECS lamp 59 is lit up. Besides, at a step 5106, a coolant temperature TW is read from the coolant temperature sensor 34, and whether or not the coolant temperature TW as the temperature of the engine is, at least, equal to a preset coolant temperature RCHE TW is decided.

When TW ) RCHE TW holds at the step S106, the control flow jumps to a step S110. On the other hand, when TW < RCHE TW holds, the control flow proceeds to a step S107, at which the timer is started counting. The loop of the next step 5108 is iterated until the counted time period TIMER of the timer reaches a preset time period T1. When TIMER > T1 has been met at the step S108, the loop is quitted to a step Sly9, at which the timer is cleared (TIbER c 0) and which is followed by the step SilO.

More specifically, in such a case where alcohol and gasoline In the fuel tank 26 or the fuel feed passage 25 are separated at a low temperature or where only alcohol (or only gasoline) is supplied in a condition in which an alcohol concentration M in the fuel tank 26 is low (high), the alcohol concentratIon M of fuel in the fuel feed passage 25 fluctuates greatl, and the alcohol content changes with time.

Accordingly, when the coolant temperature TW is lower than the preset coolant temperature RCHE TW, only the fuel pump 27 is actuated before cranking the engine so as to return the fuel from the pressure regulator 30 into the fuel tank 26, whereby the fuel in this fuel tank is circulated and stirred. Herein, the circulation of the fuel is continued for the preset time period T1 which is determined by the discharge capacity of the fuel pump 27 and the volume of the fuel contained between the alcohol concentration sensor 29 and the injector 24. Thus, the alcohol concentration distribution of the fuel is homogenized, and the temporal and special deviations of the alcohol concentration M between the mounted position of the alcohol concentration sensor 29 and that of the injector 24 actually feeding the engine with the fuel are eliminated, so that the control characteristics of the system are enhanced.

Subsequently, at the step Slit, a coolant temperature TWMET for judging if the engine is startable, is set in such a way that a startability judgement coolant temperature map MP TW is derived from an interpolative calculation by using the alcohol concentration M as a parameter. At a step S111, the coolant temperature TW is compared with the startability judging coolant temperature TWMET to judge if the engine is startable.

More specifically, as illustrated in Fig. 12, experiments etc. are conducted for specifying the temperature condition range of the alcohol concentration M in which the engine is startable without heating the fuel to be injected from the injector 24, by means of the heater 23d, and the temperature condition range thereof in which the engine is not startable in that condition. The startability judgement coolant temperature map M TW (refer to Fig. 13) corresponding to a series of addresses in the ROM 43 is prepared on the basis of the specified ranges, and the startability judging coolant temperature TWMET Is set from this map by using the alcohol concentration M as the parameter.

Then, whether or not the engine is startable can be judged by comparing the coolant temperature TW with the startability judging coolant temperature TWMET.

ncidentally, as the engine temperature for judging the startability, any of the temperatures of the fuel etc. may well be adopted instead of the coolant temperature TW from the coolant temperature sensor 34.

In consequence, when TW < DIMET holds at the step S111, it is judged that the engine is unstartable, and the control flow proceeds to a step S112. When TW > TWMET holds, it is judged that the engine is startable, and the control flow proceeds to a step S129.

Here shall be first explained steps in the case where the engine has been judged unstartable.

At the step S112 following the step S111 at which the unstartable condition has been judged, the fuel pump actuation allowance flag FLAG2 is cleared (FLAG2 c O) to stop the drive of the fuel pump 27. A flag FLAG4 for deciding a control for the unstartable condition is set (FLAG4 + 1) at a step S113, which is followed by a step 5114. The unstartable condition control deciding flag FLAG4 is referred to at starter motor control steps to be described later, and the corresponding steps are executed upon deciding the unstartable condition control.

Subsequently, at the step Slut, the ECS lamp 59 is changed-over from the lit-up state into a flashed state in order to indicate that the heater 23d is being warmed up. At a step S115, the heater relay 52 is turned ON to start the activation of the heater 23d and to warm up this heater.

Next, at a step S116, the timer is started counting so as to count the activation time period of the heater 23d. Besides, at a step S117, the counting operation of the timer is continued until a time period TIMER counted by the timer reaches, at least, a preset time period TSET (for example, TSET = 3 sec.).

Then, when TIMER > TSET has been met at the step S117, the control flow proceeds to a step S118, at which the timer is cleared (TIMER c 0). At the next step S119, heating finish judging power W1 is set in such a way that a heating finish judgment power map MP HW is derived from an interpolative calculation by using the coolant temperature TW and the alcohol concentration M as parameters. Thereafter, the control flow proceeds to a step S120.

At the step S120, heater consumption power W is calculated from a battery voltage VB and a heater consumptIon current I detected by the current sensor 63 (W # I x VB), whereupon at a step 5121, the heater consumption power W is compared with the heating finish judging power W1 having been set at the step S119.

More specifIcally, as Illustrate in Fig. 14, when the heater 23d made up of the PTC pill has had its temperature raised up to its Curie point after the activation, the consumption current I begins to decrease on account c= a sudden rise in resistance. It is therefore impossible to decide the warmed-up condition of the heater wit the consumption power only.

Accordingly, the heating finish of the heater 23d is judged after the lapse of the time period TSET by avoiding the initial starting stage of the heater activation, thereby to prevent a misjudgement.

The heating finish judging power WI is the power which is consumed when the heater 23d has been warmed up to a sufficient temperature for promoting the vaporization of the fuel, after starting the fuel injection. As illustrated in Fig. 15, the heating finish judgement power map MP HW includes the coolant temperature TW and the alcohol concentration M as the parameters. Herein, as the alcohol concentration M is higher, the amount of the latent heat of vaporization is larger, and as the coolant temperature TW is lower, the heater 23d needs to be heated more sufficiently for vaporizing the injected fuel. Therefore, the values of the heating finnish judging power W1 stored in the respective addresses of the map MP HW are smaller for the higher alcohol concentration and for the lower coolant temperature.

Referring back to Fig. 2, when W ~ W1 holds at the step S121, the loop of reading the consumed current I of the heater 23d from the current sensor 63 and calculating the heater power consumption W and thereafter comparing the calculated power consumption W with the heating finish judging power W1 is iterated again. On the other hand, when W < W1 holds, the finish of the heating is decIded, and the control flow proceeds to a step S122.

At the step S122, the starter motor actuation prohibition flag FLAG1 is cleared (FLAG1 * 0) to allow the actuation of the starter motor 62. At a step S123, the fuel pump actuation allowance flag FLAG2 is set (FLAG2 * 1) to drive the fuel pump 27 again. Then, the ECS lamp 59 is changed-over from the flashed state into continuous lighting at a step S124, which is followed by a step S125.

At the step S125, a loop is iterated until the counter COUNTST reaches a preset value TC. The counter COUNTST counts a cranking time period at starter motor control steps to be described later, and the preset value TC is previously set at a value, for example, 2 3 sec. On this occasicn, since the fuel InjectIon prohibition flag FLAG3 is kept set, the engine undergoes idle cranking in which the starter motor 62 is driven without the fel Injection.

Thus, the temperature of the combustion chamber is raised by the idle crankIng. Accordingly, when the fuel mixture is fed into the combustion chamber by performing the fuel injection, the fuel has its vaporization promoted and becomes easy to ignite. Therefore, the warming-up time period of the engine 1 can be shortened.

Thereafter, when COUNTST > TC has been met at the step S125, the loop is quitted to a step S126 et seq.

At the steps S126, S127 and S128, the unstartable condition control discrimination flag F4, the counter COUNTST and the fuel injection prohibition flag FLAG3 are respectively cleared (FLAG4 + O, COUNTST ffi O, FLAG3 c O). Then, the processing in the case where the engine is not startable has been entirely done, and it is followed by a step S133.

Meanwhile, when TW > TWMET holds at the step S111 in Fig. 1, the control flow proceeds from this step to the step S129 et seq. in Fig. 3, which execute processing in the case where the engine is startable.

More specifically, at the step S129, the starter motor actuation prohibition flag FLAG1 is cleared (FLAG1 + 0) to allow the actuation of the starter motor 62, and at the step 5130, the fuel injection prohibition flag FLAG3 is cleared (FLAG3 + 0) to allow the fuel injection.

Whether or not the coolant temperature TW has reached a warming-up finishing temperature TWLA4 (for example, 50 - 60"C) is decided at the step S131.

When TW > TWLA4 holds at the step S131, the control flow jumps from this step to a step S147 in Fig. 4, at which the ECS lamp 59 is put out, and the program is quitted. On the other hand, when TW < TWLA4 holds, the step S131 is followed by the step S132, at which the heater relay 52 is turned ON so as to start the activation of the heater 23d. Thereafter, the control flow proceeds to the step S133.

Subsequently, at the step S133 which follows the step S128 of the unstartable condition processing or the step S132 of the startable condition processing, a time period for keeping an ignition timing fixed at a specified timing, namely, a fixed ignition time TADV is set in such a way that a ixed ignition time map MP IGST is derived from an interpolative calculation by using the coolant temperature TW as a parameter.

As illustrated in Fig. 16, values obtained by an experiment etc. beforehand are stored in the respective addresses of the fixed ignition time map MP IGST, and the stored values of the fled Ignition time TADV are greater as the coolant temperature TW is lower. Herein, until the fixed Ignition time TADV elapses, the ignition timing is fixed to the specified timing retarded with respect to a usual IgnitIon timing, for example, to the timing of the input of a e3 crank pulse from the crank angle sensor 37.

In this way, the ignition timing is retarded relative to the usual one in accordance with the engine temperature, and the temperature of the combustion chamber is raised, so that the mixture can be reliably ignited, and the starting characteristics of the engine can be enhanced.

Further, the control flow proceeds from the step S133 to a step S134, at which the counting operation of the timer is started. At a step S135, it is decided whether or not the engine speed Ne has reached a speed NKAN which represents that the engine is completely started. When Ne < NKAN holds, that is, when the engine is not completely started yet, the control flow branches from the step S135 to a step S141, at which a count value COUNT is incremented (COUNT + COUNT + 1). Whether or not the count value COUNT has exceeded a preset value COUNTSET is decided at z step S142.

When COUNT < COUNTSET holds at the step S142, this step is followed by a step S143, at which the timer is cleared (TIMER * 0). Further, the control flow returns from the step S143 to the foregoing step S133, at which the fixed ignition time TADV is reset, whereupon the steps stated above are iterate. On the other hand, when COUNT > COUNTSET holds, It is decided that the engine stalls, and the control flow branches from the step S142 to a step S144. Herein, the count value COUNT is cleared at the step Slut4 (COUNT + 0), and the timer is cleared at a step S145 (TIMER c 0). Subsequently, whether or not the coolant temperature TW has exceeded the startability judging coolant temperature TWMET is decided at a step S146.

Besides, when TW > TWMET holds at the step S146, the control flow returns to the step S133, and when TW < TWMET holds, it returns to the step S113 et seq. of the unstartable condition control. On the other hand, when Ne > NKAN holds at the step S135, that is, when the engine has been completely started, the step S135 is followed by a step S136, at which whether or not the coolant temperature TW has exceeded the warming-up finishing temperature TWLA4 is decided again. Subject to TW < TWLA4, the control flow branches from the step S136 to the step 5143, at which the timer is cleared (TIMER + 0) and from which the control flow returns to the step S133, whereas subject to TW > TWLA4, the control flow proceeds from the step S136 to a step S137.

At the step Si37, it is decided whether or not the time period TIMER counted by the timer has reached a preset time period L, in other words, whether or not the warming-up of the engine has been finished owing to the fact that the state in which the engine speed Ne is not lower than the completely-started engine speed NKAN and in which the coolant temperature TW is higher than the warming-up finishing temperature TWLA4 has continued for the preset time period TL. Subject to TIMER < TL, the control flow returns to the step 5135 so as to iterate the judgement of the complete starting of the engine, whereas subject to TIMER > TL, it is decided that the engine warming-up has been finished after the starting, and the step S137 is followed by a step S138.

Herein, the count value COUNT is cleared (COUNT c O) at the step S138, the timer is cleared (TIMER + 0) at a step S139, the heater relay 52 is turned OFF to deactivate the heater 23d at a step S140, and the ECS lamp 59 is put out at a step S147, whereupon the program is ended. In this way, the fuel injected from the injector 24 is vaporized by the heater 23d till the finish of the engine warming-up, and the injected fuel is vaporized by the heat of the engine itself after the finish of the engine warming-up, whereby the fuel is favorably vaporized at all times.

Owing to the precise starting-mode control as stated above, the fuel quantity after starting the engine can be sharply reduced by quickly raising the temperature of the combustion chamber of the engine, so that a shortened warming-up time period and an enhanced fuel cost can be accomplished.

(Starter Motor Control Steps) Meanwhile, the program of the starter motor control steps shown in Fig. 17 is executed interrupting the program of the initial control every predetermined time period.

In the temporally-interrupting program, at the first step S201, the value of the starter motor actuation prohibition flag FLAG1 is checked to decide whether or not the actuation of the starter motor 62 is allowed.

When FLAG1 = 0 holds at the step S201, that is, when the actuation cf the starter motor 62 is allowed, the control flow proceeds from the step S201 to a step 5202, which decides whether or not the starter switch 60 is turned ON. Subject to the ON state of the starter switch 60, the control flow proceeds to a step S203, at which the value of the unstartable condition control discrimination flag FLAG4 is checked.

When FLAG4 = 0 holds at the step S203, the control flow jumps to a step S205 When FLAG4 = 1 holds, the coolant temperature TW is lower than the startability judging coolant temperature TW > ET, and the engine is in the state of the unstartable condition contra.

Therefore, the control flow proceeds rom the step S203 to a step S204, at which the count value COUNTST for measuring the idle cranking time period cranking without injecting the fuel as the before-mentioned starting-mode control steps is incremented (COUNTST + COUNTST + 1).

Then, at the step S205, the starter motor relay 61 is turned ON to drive the starter motor 62, whereupon the program is quitted. Thus, the engine 1 is cranked.

On the other hand, when FLAG1 = 1 holds at the step S201, so the actuation of the starter motor 62 is prohibited, or when it is decided at the step S202 that the starter switch 60 is turned OFF, the control flow branches from the pertinent step to a step S206, at which the starter motor relay 61 is turned OFF to hold the starter motor 62 stopped, whereupon the program is quitted.

(Steps for Discriminating Cylinder Nos. and 'or Calculating Engine R.P.M.) Fig. 18 shows a routine for discriminating cylinder Nos. and aor calculating engine R.P.M. which is interruptively started on the basis of the input of a crank pulse from the crank angle sensor 37. At a step S301, cylinder No. vi for ignition is discrIminated on the basis of the output signals of the crank angle sensor 37 and the cam angle sensor 39. subsequently, at a step S302, cylinder No. Xi -2 for fuel injectIon is discriminated.

More specifically, as illustrated in a time chart of Fig. 29, in a case where the cam pulses of the position OS (projection group 38b) have been output from the cam angle sensor 39 by way of example, the next top dead center of compression is of the cylinder No. a3.

It is accordingly possible to discriminate that the cylinder No. a3 becomes the cylinder for ignition and that the cylinder No. 4 becomes the cylinder for fuel Injection.

Further, in a case where the cam pulse of the position e4 (projection 38a) has been output after the cam pulses cf the position e5, the next top dead center of compression is of the cylinder No. &num;2. It is accordingly possible to discrir,inate that the cylinder No. =2 becomes the cylinder for ignition and that the cylinder No. &num;1 becomes the cylinder or fuel Injection.

Likewise, the top dead center of compression after the output of the cam pulses of the position #6 (projection group 38c) is of the cylinder No. r4. Thus, the cylinder No. &num;4 becomes the cylinder for ignition, and the cylinder No. &num;3 becomes the cylinder for fuel injection. Besides, In a case where the cam pulse of the position ea (projection 38a) has been output after the cam pulses of the position 56, the subsequent top dead center of compression is of the cylinder No. &num;1.

It is accordingly possible to discriminate that the cylinder No. #1 becomes the cylinder for ignition and that the cylinder No. 2 becomes the cylinder for fuel Injection.

Further, it is seen from Fig. 29 that the crank pulse which is output from the crank angle sensor 37 after the output of the cam pulse(s) from the cam angle sensor 39 indicates the reference crank angle (81) in the case of setting the ignition timing and fuel Injection start timing of the corresponding cylinder.

More specifically, in the four-cycle four-cylinder engine 1 in this embodiment, the combustion strokes proceed in the sequence of the cylinders Nos. &num;1 # &num;3 # 2 + X4. Assuming the cylinder No. &num;i for ignition is the cylinder No. 1, the cylinder No. vi(+2) for fuel injection at this time is the cylinder No. X2, and the next cylinder No. "num;i(+2) for fuel injection becomes the cylinder No. 4.Thus, the ignition operations are performed in the sequence of the cylinders Nos. = 3 #2 + #&num;4, and the fuel InjectIon operations are performed in such a sequential manner that the fuel is injected into the corresponding cylinder once every 7200CA (every two engine revolutions).

Subsequently, the engine R.P.M. Ne is calculated at a step S303 in Fig. 18. By way of example, the time interval between the output pulses of the crank angle sensor 37 for detecting the positions BTDC el and 62 is measured to obtain a period f, and the engine R.P.M. Ne is calculated from the period = (Ne * 60/f). The calculated engine speed is stored as R.P.M. data in the predetermIned address of the RAM 44, whereupon the routine is quitted.

(Steps for Setting Fuel InjectIon Quantity and Ignition Timing) Meanwhile, a fuel Injection quantity and an ignition timing are set by an interrupt routine of every predetermined time as shown in Figs. 19 - 21. First, at a step S401, the value of the fuel injection prohibition flag FLAG3 is checked. When FLAG3 = 1 holds, that is representing that the fuel InjectIon Is prohibited in the starting mode controi, the control flow proceeds to a step S402. At the step 5402 a fuel Injection pulse width Ti is set to "O" (Ti * 0). Besides, the ignition is prohibited at a step S403, after whIch the routine is quitted.

Thus, the initial knocking of comparatively great magnitude is prevented from occurring =mediately after the cranking under the Influence of the low-boiling components of the fuel remaInIng in te cylinder of the engine, in the intake port 2a or in the intake manifold 3 or the residual fuel heated by the heater 23d. This avoids an unpleasant feeling to the driver of the vehicle or a situation in which the driver turns OFF the starter switch 60 at the first knocking, and makes it impossible to start the engine.

~ On the other hand, when FLAG3 = 0 holds at the step S401, in other words, when the fuel injection is allowed, the control flow proceeds from the step S401 to a step S404, which decides whether or not the engine R.P.M. Ne is "0", that is, whether or not the engine is at a stop. Subject to Ne = 0, the routine is quitted via the steps S402 and S403 stated above, whereas subject to Ne + 0, the step 5404 is followed by a step S405.

At the step S405, the engine R.P.M. Ne stored in the predetermined address of the RAM 44 is read out, and a time period TIIE1/2 per 1/2 rotation of the crankshaft of the engine is calculated on the basis of the engine R.P.M. Ne from the following: TI.ME1/2 = 30/Ne ... (1) The above equation (1) gives the time period per stroke in the four-cylinder engine. As regards an engine of ecual-interval combustion operations having n cylinders, the corresponding time period can be calculated from the following: TIMEl/n/2 = (60/n/2)/Ne ... (1)' Thereafter, at a step 5406, a weighting coefficient (the weight of a weighted average) per stroke TNnew is calculated from: TNnew = TIME1/2 x COF ... (2) where COF: fixed value.

Subsequently, at a step S407, an induced air quantity Q (gr./sec.) based on the output of the intake air quantity sensor 31 is read, and a weighting coefficient TNold and a corrected induced air quantity Qaold having been set in the last routine are read out.

Incidentally, TNoid = 0 and Qaold = 0 are set in the first routine.

Thereafter, at a step S408, a compensated induced air quantity Qanew with the first-order lag compensated is calculated from: Qanew = (Qaoid.TNold + Q)/(1 t TNnew) ... (3) Also, at a step 5409, an air quantIty Qp which is induced In one cylinder at the Intake stroke is calculated from: Qp - Qanew x TIME1/2 ... (4) Thus, the first-orer lag is compensated, whereby an overshoot n a transient state can be corrected.

By the way, the theoretical formula of the compensated induced air quantity Qanew is elucidated in detail in the official gazette of Japanese Patent Application Laid-open No. 5745/1990, the application of which was filed by the same Applicant before.

Subsequently, at a step 5410, a correction coefficient COEF which corresponds to the enrichment components of e.g. enrichment corrections at the starting of the engine, in a cold engine state and in the fully open state of the throttle valve is set on the basis of the output values of the throttle opening sensor 32a, idle switch 32b and coolant temperature sensor 34. However, an acceleration enrichment correction is not made.

Thereafter, at a step S411, an air-fuel ratio feedback correction coefficient a is a set on the basis of the output signal of the 02sensor 35. Besides, at a step S412, a desired air-fuel ratio A/F is set by referring to a desired air-fuel ratio map MP A/F with an interpolative calculation, on the basis of the alcohol concentration M, the air quantity Qp which is induced in one cylinder at the intake stroke, and the engine R.P.M.

Ne.

Regarding the desired air-fuel ratio A/F, the air fuel ratio varies depending upon the alcohol concentration M. As illustrated in Fig. 22, therefore, the optimum air-fuel ratios (in general, theoretical air-fuel ratios) are stored in the map of the ROM 43 beforehand in accordance with the alcohol concentration M, the induced air amount Qp per cylinder at the intake stroke, and the engine R.P.M. as parameters.

Subsequently, the step S412 is followed by a step S413, which decides whether or not the heater is under activation. Subject to the activated state of the heater, the control flow proceeds to a step S414 et seq., whereas subject to the deactivated state, it proceeds to a step S425 et seq. First, the steps for the activated heater will be explained. At the step S414 of these steps, a rate at which fuel having adhered to the wall of the intake port 2a evaporates during two revolutions (one cycle) of the engine is fixed to "1" (S 1). The rate is the evaporating ratio ss of the fuel.

Further, at the step S415, an adhesive rate at which the fuel injected from the injector 2t adheres to the wall of the intake port 2a is fixed to "0" (X 0). The adhesive rate is the adhesive ratio X of the fuel adhering on the wail.

More specifically, the fuel injected from the injector 24 is entIrely struck against the heating element 23a of the intake-port heater unit 23. DurIng the heater activation, accordingly, the fuel is instantly vaporized by the heater 23d without adhering to the wall, and hence, the evaporation of the adhering fuel does not occur. Therefore, the fuel evaporation ratio ss is fixed to "1", and the wall adhesion ratio X is fixed to "0". Thus, the air fuel ratio can be set appropriate, and an overrich mixture can be prevented, thereby the fuel consumption is improved together with the enhancements of the starting characteristics.

When the control flow proceeds from the step S415 to a step S416, the engine R.P.M. Ne is compared with the explosion R.P.M. NKAN representing the completelystarted state of the engine, thereby to judge the explosion of the engine. Subject to Ne < NKAN, at a step S417, the engine starting judgement flag FLAGS indicating the started state of engine before the explosion is set (FLAG5 ; 1). At a step S424, the ignition timing BIG is set at a fixed ignition timing (angle) ADVCS which is synchronized with, for example, the crank pulse of the position BTDC 53 (10 CA) delivered from the crank angle sensor 37. This step S424 is followed by a step S430.

On the other hand, when Ne > NKAN has been met at the step S416, the control flow proceeds from this step to a step S418, at which the value of the engine starting judgement flag FLAGS is checked. FLAGS = 1 represents that the engine was not completely started at the last routine, and that it has first come into the completely-started state this time. Therefore, the control flow proceeds to a step S419, at which the timer for counting the elapsed time of the fixed ignition is cleared (TIMER2 + O). The counting operation of this timer is started at a step S420, and the engine starting judgement flag FLAG5 is cleared (FLAG5 + 0) at a step S421, which is followed by the step S424 stated above.

In addition, when FLAGS = 0 holds at the step S418, the control flow branches from this step to a step S422, which decides whether or not the time period TIMER2 of the elapsed time of the fixed ignition as counted by the timer has reached a fixed-ignition time TADV. Herein, subject to TIMER2 < TADV, the step S422 is followed by the foregoing step S424. Subject to TTMER2 > TADV, the step 5422 is followed by a step S423, at which the timer is cleared (TiMER2 e 0) and which is followed by a step S427.

Next, there will be explained steps in the case of heater deactivation where the routine proceeds from the step 5413 to the step S425 et seq.

At the step S425 of these steps, the fuel evaporation rate 6 corresponding to two revolutions of the crankshaft of the engine is set in such a way that a fuel evaporation rate map MP R is referred to with an interpolative calculation in according with the engine speed Ne, the coolant temperature TW and the alcohol concentration M as parameters.

The fuel evaporation rate ss is governed by a wall temperature, the period and the alcohol concentration M.

More specifically, as the wall temperature is higher, the fuel evaporation rate B becomes larger. Besides, as the engine speed Ne heightens, the period shortens, and hence, a time period till the next intake stroke shortens, so that the fuel adhesion lower and the value of the fuel evaporation rate ss decreases accordingly.

Further, as the alcohol concentration M is higher, the latent heat of evaporation increases more, and hence, the fuel is more difficult to evaporate, so that the value of the fuel evaporation rate B decreases more.

Accordingly, the fuel evaporation rate B can be grasped as a function of the coolant temperature TW, the engine R.P.M. Ne and the alcohol concentration M. In this embodiment, as Illustrated in Fig. 24, the fuel evaporation rate map MP B is formed in accordance with the engine R.P.M. Ne, the coolant temperature TW and the alcohol concentration M as parameters, and the values of the fuel evaporation rate B obtained by an experiment etc. beforehand are stored in the individual areas of this map.

Referring back to Fig. 20, at the step S426, the adhesive ratio X of the fuel adhering to the wall is set in such a way that a wall adhesion ratio map MP X is referred to with an interpolative calculation in accordance with the alcohol concentration M, the compensated induction air quantity Qanew, and the fuel injection pulse width Ti set at the last routine, as parameters. By the way, at the first routine, X = O is set because the fuel injection pulse width Ti has not been set.

Variation in the adhesive ratio X of the fuel adhering to the wall is governed by the induced air quantity Qanew, the fuel Injection pulse width Ti (the quantity of the injected fuel) and the alcohol concentration M. More specifically, when the induced air quantity Qanew enlarges, a time period for atomizing the fuel shortens, and the wail adhesion ratIo X enlarges. Besides, assuming the induced air quantity Qanew to be constant, the fluctation cf the quantity of the fuel adhering to the wall is slight relative to the change of the fuel injection quantity, so that the wall adhesion ratio X becomes a relatively small value when the fuel injection pulse width Ti increases.Further, when the alcohol concentration M of the fuel heightens, the fuel becomes more difficult to evaporate due to a higher latent heat of evaporation, so that the wall adhesion ratio X becomes a relatively large value in relation of alcohol concentration.

As illustrated in Fig. 23, the wall adhesion ratio map MP X is formed in accordance with the alcohol concentration M, the compensated induced air quantity Qanew and the fuel injection pulse width Ti as parameters, and the values of the wall adhesion ratio X obtained by an experiment etc. beforehand are stored in the individual areas of this map.

Subsequently, when the control flow proceeds to the step S427 from the step S426 or from the step S423 of the processing during the heater activation as stated before, a basic ignition timing ABASE is set in such a way that a basic ignition timing map MP ABASE is referred to with an Interpolative calculation in accordance with engine R.P.M. Ne, the induced air amount Qp per cylinder at the Intake stroke and the alcohol concentration M as parameters.

As illustrated in Fig. 25, the optimum values of the basic ignition timing ABASE (the crank angle with reference to 61) obtained by an experiment etc.

beforehand with the parameters of the engine R.P.M. Ne, the induced air amount Qp per cylinder at the intake stroke and the alcohol concentration M are stored in the respective addresses of the basic ignition timing map MP DEBASE. Under the conditions of the same values of the parameters Qp and Ne, the stored values of the basic ignition timing (angle) 6BASE are smaller for the higher alcohol concentration M in order to attain greater advanced angles.

Thereafter, at a step S428, a knock control value (angle) ONK is set on the basis of a signal from the knock sensor 33, and at a step S429, the knock control value eNK is added to the basic ignition timing ABASE set at the above step S427, thereby to calculate the ignition timing (angle) aIG (eIG # #BASE + 6NK), whereupon this step S429 is followed by the step S430.

Subsequently, when the control flow proceeds to the step S430 from the step S429 or from the step S424 of the processing in the case of the heater activation, the amount Mf4 of fuel remaining in the intake port which was set four strokes (one cycle) before is read out.

Besides, at a step 5431, a fuel Injection amount Gf corresponding to one injection operation is set in accordance with an equation given below. By the way, Mf4 = 0 holds untIl the fuel Injection routine is iterated four times.

Gf= = {(Qp/A/F) x COEF - ss Mf4)/(1 - X) ... (5) As described before, with the engine 1 in this embodiment, one fuel injection operation is done for the corresponding cylinder every 7200CA (every two revolutions of the engine). When the fuel is injected from the injector 24 of the corresponding cylinder into the intake port 2a thereof, part of the injected fuel adheres to the intake valve, the wall of the intake port, etc. without being drawn into the cylinder (combustion chamber). The fuel having adhered evaporates properly during the two revolutions of the engine, and the fuel having evaporated is drawn into the cylinder together with the fuel injected at the next intake stroke.

Here, a fuel feed amount Ge which is actually fed into the cylinder by one injection operation becomes the sum of an evaporative amount Mf4-B and a fuel amount (1 - X).Gf which does not adhere to the wall, that is: Ge = (1 - X)-Gf + Mf4-ss ... (6) From Ec. (6), the fuel amount Gf required for one injection operation is evaluated as follows: Gf = (Ge - Mf4 B)/(1 - X) ... (7) The fuel amount Ge to be actually fed into the cylinder is the desired value of fuel feed based on the desired air-fuel ratio A/F and the air quantity Qp, and it becomes the following because the air fuel ratio subjected to the enrichment correction is denoted by (A/F)/COEF:: Ge = Qp.COEF/A/F) ... (8) The substitution of Eq. (8) into Eq. (7) results in Eq.

(5) mentioned before.

Subsequently, at a step S432 in Fig. 21, the fuel amount Mf remaining in the intake port at this time is set in accordance with the following equation: Mf = (1 - 6) x Mf4 + X-Gf ... (9) That is, the fuel amount Mf remaining in the intake port immediately after the fuel injection becomes the sum of the remaining amount (1 - ss) x Mf4 obtained by subtracting the evaporative component from the adhering fuel amount of the corresponding cylinder at the last injection, and the adhesive component X-Gf in the fuel amount injected this time. By the way, Mf = X-Gf holds until the injection is performed four times from starting this procedure.

Thereafter, at a step S433, a voltage correction pulse width Ts for compensating an invalid time period is set on the basis of the battery voltage. Besides, at a step S434, the fuel Injection pulse width Ti for actually driving the injector 24 is set in accordance with the following equation: Ti = K-Gf-: + Ts ... ( 3) where K: coefficient for compensating the characteristics of the injector.

The aforementioned fuel injection amount Gf has been calculated with the predictive correction of the adhesion of the fuel to the wall and the correction of the evaporation of the fuel adhering to the wall.

Therefore, the air fuel ratio is prevented from becoming rich in a transient state, particularly in a low engine speed state, so that any slow operation in the transient state is avoided to enhance the output response of the control system.

Further, an enrichment correction for acceleration is dispensed with, so that a controllability for the air fuel ratio is enhanced, and the wasteful consumption of the fuel is prevented.

Subsequently, at a step S435, an injection starting crank angle EINJST is set on the basis of an injection starting crank angle map MP EINJST in accordance with the engine R.P.M. Ne and the fuel injection pulse width Ti as parameters.

As illustrated in Fig. 26, the injection starting crank angle map MP eINJST is formed in accordance with the engine R.P.M. Ne and the fuel injection pulse width Ti as the parameters, and the optimum values of the injection starting crank angle BINJST obtained by a computation etc. beforehand are stored in the respective areas of this map. The Injection starting crank angle EINJST is set at a more advanced angle as the engine R.P.M. Ne and the fuel injection pulse width Ti are greater.

Thereafter, the control flow proceeds to a step S436, at which the last data TNold of the weighting coefficient is renewed with the data TNnew set at the step S406 (TNold * TNnew). Besides, at a step S437, the last data Qaold of the compensated induction air quantity is renewed with the data Qanew set at the step S408 (Qaold + Qanew), whereupon the routine is quitted.

(Steps for Controlling Ignition and Fuel Injection) When the ignition timing BIG and the fuel injectIon pulse width Ti have been set by the above steps, an ignition signal and a fuel injection signal are output in accordance with the charts shown in Fig. 27 and Fig.

28, respectively.

Ignition control steps shown in Fig. 27 are executed every 1800CA Interruptingly when the current crank angle calculated on the basis of the crank pulse input becomes the ignition timing (angle) EIG set by the foregoing routine (steps S424, 5429).

More specifically, at the step 5501, the ignition signal is output to the corresponding cylinder ai for ignition as discriminated by the foregoing steps for discriminating cylinder No. and calculating engine R.P.M., whereupon the routine is quitted. Meanwhile, fuel Injection control steps shown in Fig. 28 are similarly executed every 1800CA interruptingly when the current crank angle calculated on the basis of the crank pulse input becomes the Injection starting crank angle EINJST set by the foregoing routine (step S435).

Herein, at the first step S601, the driving pulse signal having the fuel injection pulse width Ti is output to the injector 24 of the corresponding cylinder #i(+2) for fuel injection as discriminated by the foregoing steps for discriminating cylinder No. and caiculating engine R.P.M.

At the next step S602, the fuel quantity Mfl remaining in the intake port at the last time is renewed with the fuel quantity Mf remaining in the intake port at this time as has been set by the foregoing steps for setting the fuel InjectIon quantity and the ignition timing (Mfl * Mf). Likewise, the data items of the remaining fuel quantity are successively renewed (Mf2 t Mfl, Mf3 + Mf2, Mf4 i Mf3).

As a result, the fuel amount Mf4 remaining in the intake port which is read out at the step S430 of the foregoing steps for setting the fuel Injection quantity and the ignition timing becomes the fuel amount of the corresponding cylinder having existed one cycle before, at all times.

By the way, in case of an engine having n cylInders, the fuel quantity Mfn remaining in the intake port one cycle before is renewed with the fuel quantity Mfn-l remaining in the intake port at the preceding cycle.

As described above, according to the present invention, a precise starting-mode control dependent upon the temperature of an engine makes it possible to attain an appropriate air-fuel ratio and to sharply reduce the amount of fuel after the starting of the engine. Moreover, at starting the engine, the temperature of a combustion chamber can be quickly raised without incurring unpleasant initial knocking immediately after cranking due to the residual components of fuel, and a warming-up tIme period can be shortened.

Accordingly, the engine starting can be shifted into an ordinary running state smoothly and quickly, and the enhancements of starting characteristics and the reduction of a fuel cost can be simultaneously accomplishea.

Furthermore, a heater for vaporIzIng the fuel can be effectively arranged, and the fuel injected from an injector can be reliably vaporized to enhance the starting characteristics. The invention brings forth such excellent effects.

Claims

1. A control method for a flexible fuel vehicle having an engine mounted on said flexible fuel vehicle, a fuel injector provided in an intake manifold for injecting fuel into a cylinder via a cylinder head, a throttle valve in a throttle chamber connected with said intake manifold, a starter motor mounted on said engine for starting said engine with electric power, a fuel pump for supplying said fuel from a fuel tank to said fuel injector, a temperature sensor mounted on said engine for detecting a coolant temperature and for producing a temperature signal, and a concentration sensor interposed between said fuel tank and said injector for sensing a concentration of said fuel and for generating a concentration signal, the control method comprising the steps of: setting a fixed ignition time in dependency on said coolant temperature; and fixing an ignition timing at a predetermined ignition timing while said fixed ignition time elapses when starting said engine so as to effectively start said engine as soon as possible.

2. A control method for a flexible fuel vehicle, substantially as herein described, with reference to, and as illustrated in, the accompanying drawings.

3. A control system for a flexible fuel vehicle, comprising means for carrying out the control method as claimed in claim 1 or claim 2.

4. A flexible fuel vehicle comprising a control system as claimed in claim 3.