JPH06241124A - Fuel injector - Google Patents

Abstract

(57) [Abstract] [Purpose] To provide a fuel injection device capable of suppressing engine stall that occurs when the recovered fuel obtained as a result of recovering fuel evaporated in a fuel tank is purged into an engine. The purge position of the recovered fuel is set to the intake pipe branch portion 11, the air-fuel ratio sensor 22A is attached to the exhaust pipe branch portion 21 corresponding to the specific cylinder, and the ignition timing of the specific cylinder is changed to other values during purging. The cylinder is configured to be retarded. [Effects] According to the present invention, the fuel vapor is purged into the specific cylinder. Therefore, even if the rich air-fuel mixture is temporarily sucked, the combustion fluctuation remains only in the specific cylinder, so that unnecessary engine stall is prevented. It is possible to obtain stable engine rotation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuel injection device for supplying recovered fuel to an engine.

[0002]

2. Description of the Related Art It is known that fuel vapor discharged from a fuel tank of an automobile produces photochemical smog due to a reaction with ultraviolet rays and leads to air pollution. Therefore,
In each country, we are trying to prevent environmental damage by setting a regulated value for the amount of fuel vapor emission.

A vehicle fuel vapor recovery system shown in FIG. 2 is generally known as a means for complying with the above fuel vapor emission regulation.

The mechanism for collecting and supplying the fuel vapor generated from the fuel tank (“supply” will also be referred to as “purge” below) will be described with reference to FIG.

The amount of intake air is controlled by a throttle valve incorporated in the throttle body 5, passes through the surge tank 9, and is sucked into the engine 1 main body (which has four cylinders from the first cylinder to the fourth cylinder). To be done.

On the other hand, the fuel vapor generated in the fuel tank 13 passes through the pipe 46 and is temporarily collected in the canister 40. The recovered fuel is introduced into the surge tank 9 through the pipe 47, the canister purge valve 41, and the pipe 48 together with the fresh air introduced from the fresh air introduction port 45 attached to the canister 40 during the operation of the engine 1. He
Exhaust to the outside is suppressed by being sucked into the engine 1 and burning.

The canister purge valve 41 is mounted for the purpose of controlling the purge flow rate by the control unit 301. Further, the canister purge valve 41 is in a state where no drive signal is input,
That is, since the leakage air is generated even in the closed state, the leakage air is shut off by the purge cut valve 44. A pipe 50, a purge cut solenoid 43, and a pipe 47 are connected to the purge cut valve 44, and the negative pressure of the surge tank 9 is introduced into the purge cut valve 44 by turning the purge cut solenoid 43 ON and OFF to operate. Has become.

[0008]

By the way, as described above, when the pipe 48 is connected to the surge tank 9 in front of the branch of the intake pipe, it is rich in a plurality of cylinders (air-fuel ratio is wide). A (small) fuel / air mixer will be introduced, and engine stalling will occur more easily.

A first object of the present invention is to provide a fuel injection device which prevents engine stalling.

Further, there is the following problem even if the engine does not stall without largely deviating from the stoichiometric air-fuel ratio. That is, as shown in FIG. 2, when the pipe 48 is connected to the surge tank 9, the fuel vapor to be purged may be biased toward a specific cylinder depending on engine operating conditions, The air-fuel ratio of the specific cylinder becomes rich. This is shown in Figure 3
Is. FIG. 3A shows the A / F (air-fuel ratio) of each cylinder when there is no purge and feedback (F / B) control is performed by an O2 sensor (air-fuel ratio sensor). The fluctuation range of A / F between them is about 0.6. On the other hand, FIG. 3B shows the case where the purge is performed, and shows that the fuel vapor is easily sucked into the first cylinder and the second cylinder, and the fluctuation range of the A / F between the cylinders is also as follows.
It has expanded to 1.2. FIG. 3C shows the problem caused by the large fluctuation range of the A / F between the cylinders. This shows the conversion efficiency of the three-way catalyst, and is the three elements of harmful exhaust gas, HC and C.
The conversion efficiency of O and NOx depends on the air-fuel ratio to the catalyst, and also the theoretical air-fuel ratio (A / F = 1 when gasoline is burned).
In the vicinity of 4.7), the conversion efficiencies of the above three elements are both high, which shows that it is necessary to minimize the fluctuation range of A / F. Therefore, it is necessary to suppress the fluctuation range of A / F even when purging fuel vapor to the engine.

A second object of the present invention is to provide a fuel injection device which suppresses the fluctuation range of the air-fuel ratio.

[0012]

In order to solve the first problem, a fuel vapor recovery means for recovering the evaporated fuel generated in the fuel tank and a recovered fuel recovered by the fuel vapor recovery means are provided. In a fuel injection device provided with a recovered fuel supply means for supplying to an engine composed of a plurality of cylinders, the recovered fuel supply means has a pipe for supplying the recovered fuel to the engine, and a connection position between the pipe and the engine. Is an intake pipe branching portion in which an intake pipe through which intake air flows to the engine is branched for each of the cylinders.

Further, in order to solve the second problem, an exhaust pipe branching portion in which an exhaust pipe through which exhaust gas from the cylinder to which the recovered fuel is supplied by the recovered fuel supply means flows is branched for each cylinder. Is installed in the first to detect the air-fuel ratio
An air-fuel ratio sensor, a control unit for controlling the fuel injection amount of the cylinder to which the recovered fuel is supplied by the air-fuel ratio, and a fuel which is controlled by the control unit and injects fuel to the cylinder to which the recovered fuel is supplied. And an injection section.

[0014]

[Operation] By connecting the supply position of the fuel vapor to the intake pipe branch portion, the fuel vapor is sucked into only the specific cylinder,
The other cylinders are not affected by the fuel vapor supplied. For the supplied specific cylinder, the amount of fuel vapor is A /
F becomes rich, but only a few cylinders become rich,
Since the other cylinders are normal, even if a small number of cylinders stall, the other cylinders do not stall, so that the engine stall can be prevented.

Further, an air-fuel ratio sensor is attached to the exhaust pipe branch portion corresponding to the specific cylinder, and the A / F is controlled independently for only the specific cylinder, so that the control is always performed in the vicinity of the theoretical air-fuel ratio. It is possible to suppress a decrease in conversion efficiency of the catalyst.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An electronically controlled fuel injection device having a canister purge control device according to the present invention will be described below with reference to the illustrated embodiments.

FIG. 1 shows an electronically controlled fuel injection system for controlling an automobile internal combustion engine 1 to which the present invention is applied. In this embodiment, the engine has four cylinders, and the evaporated fuel is supplied to the first cylinder.

This device comprises an air cleaner 2, an air inlet 3, an intake duct 4, a throttle body 5, a throttle valve 6, an air flow meter (AFM) 7 for measuring the intake flow rate,
Surge tank 9, intake pipe branch 11a for branching intake air to the first cylinder, intake pipe branch 11b for branching intake air to the second cylinder, intake air for branching intake air to the third cylinder A pipe branch portion 11c, an intake pipe branch portion 11d for branching intake air to the fourth cylinder, a fuel injection valve (injector) 12, which is a fuel injection portion, a fuel tank 13, a fuel pump 26, and a fuel pressure regulating valve. (Pressure Regulator) 16
An exhaust manifold 21, an oxygen sensor 22A which is a second air-fuel ratio sensor (monitors the oxygen concentration of all cylinders), a first air-fuel ratio sensor 22B (which monitors the oxygen concentration of one cylinder), It has a control unit 30 which is a control unit, a fuel vapor recovery means, and a recovered fuel supply means. The fuel vapor recovery means has a pipe 46 and a canister 40. The recovered fuel supply means has a purge cut valve 44, a fresh air introduction port 45, a canister purge valve 41, pipes 47, 48 and 50, and a purge cut solenoid 43.

Further, the exhaust gas from the engine 1 is supplied to the front catalyst 2
3 and the main catalyst 24 and the muffler 25 to the atmosphere.

As shown in FIG. 1, the supply position of the fuel vapor through the pipe 49 is connected to the intake pipe branch portion 11a, the air-fuel ratio sensor 22B is attached to the exhaust pipe branch portion 21 corresponding to the specific cylinder, and While the ignition timing of the specific cylinder is being supplied, the retard angle control is performed on the other cylinders.

Here, the mechanism for collecting and supplying the fuel vapor generated from the fuel tank (“supply” is also referred to as purge below) will be described with reference to FIG.

The amount of intake air is controlled by a throttle valve built in the throttle body 5, passes through the surge tank 9, and is drawn into the engine 1 main body.

On the other hand, the fuel vapor generated in the fuel tank 13 passes through the pipe 46 and is temporarily collected in the canister 40. The recovered fuel is introduced into the surge tank 9 via the pipe 47, the canister purge valve 41, and the pipe 49 together with the fresh air introduced from the fresh air introduction port 45 attached to the canister 40 during the operation of the engine 1. He
Exhaust to the outside is suppressed by being sucked into the engine 1 and burning.

The canister purge valve 41 is mounted for the purpose of controlling the purge flow rate by the control unit 301. Further, the canister purge valve 41 is in a state where no drive signal is input,
That is, since the leakage air is generated even in the closed state, the leakage air is shut off by the purge cut valve 44. A pipe 50, a purge cut solenoid 43, and a pipe 47 are connected to the purge cut valve 44, and the negative pressure of the surge tank 9 is introduced into the purge cut valve 44 by turning the purge cut solenoid 43 ON and OFF to operate. Has become.

The fuel vapor supply position is set to the intake pipe branch portion 11a.
Since the fuel vapor is sucked into only the 77th cylinder by connecting to No. 1, the other cylinders are not affected by the fuel vapor supplied.

Further, with respect to the supplied first cylinder, the A / F becomes Rich by the amount of the fuel vapor.
An air-fuel ratio sensor 22A is provided at the exhaust pipe branch portion 21 corresponding to the cylinder.
Since the A / F is independently controlled only for the first cylinder, since the control is always performed in the vicinity of the theoretical air-fuel ratio, it is possible to suppress a decrease in catalyst conversion efficiency.

Further, since the supply is made only to the first cylinder,
Volume efficiency is higher than other cylinders, and the generated torque is increased only in the first cylinder, causing torque fluctuation between the cylinders. However, by delaying the ignition timing of the first cylinder more than the other cylinders by the amount of increase in torque, , I tried to suppress the torque fluctuation.

FIG. 4 shows another embodiment in which the embodiment of FIG. 1 is further detailed. FIG. 4 shows an electronically controlled fuel injection device for controlling an internal combustion engine 1 for an automobile to which the present invention is applied. This device includes an air cleaner 2, an air inlet portion 3, an intake duct 4, and a throttle. Body 5,
A throttle valve 6 and an air flow meter (AFM) 7 for measuring the intake air flow rate
A throttle sensor 8, a surge tank 9, an auxiliary air valve (ISC valve) 10, and an intake pipe branch portion 11.
, A fuel injection valve (injector) 12, and a fuel tank 1
3, a fuel pump 26, a fuel damper 14, a fuel filter 15, and a fuel pressure regulating valve (pressure regulator) 16
A cam angle sensor 17, an ignition coil 18, an igniter 19, a water temperature sensor 20, and an exhaust manifold 21.
It has oxygen sensors 22A and 22B, a control unit 30, a fuel vapor recovery means, and a recovered fuel supply means. The fuel vapor recovery means has a pipe 46 and a canister 40. The recovered fuel supply means has a purge cut valve 44, a fresh air introduction port 45, a canister purge valve 41, pipes 47, 48 and 50, and a purge cut solenoid 43.

Further, the exhaust gas from the engine 1 is supplied to the front catalyst 2
3 and the main catalyst 24 and the muffler 25 to the atmosphere.

The intake air is supplied to the inlet portion 3 of the air cleaner 2.
And enters the surge tank 9 through the air flow meter 7 for detecting the intake air and the throttle valve 6 for controlling the air flow rate. Here, the air is distributed by the intake pipe branch portion 11 that directly communicates with each cylinder of the internal combustion engine 1, and enters the cylinder of the internal combustion engine 1.

At this time, the air flow meter 7 outputs a signal for detecting the amount of intake air, and the control unit 30
Entered in.

On the other hand, the fuel is sucked and pressurized by the fuel pump 26 from the fuel tank 13, passes through the fuel damper 14 and the fuel filter 15, and is supplied to the fuel injection valve 12 provided in the intake pipe branch portion 11 for control. Fuel is injected according to the injection signal from the unit 30. At this time, the fuel pressure acting on the fuel injection valve 12 is regulated by the fuel pressure regulating valve 16, and therefore, the fuel pressure regulating valve 16 is disposed in the intake pipe branch portion 11.
The negative pressure is introduced to keep the fuel pressure and the pressure difference in the intake pipe branch 11 constant.

A throttle sensor 8 for detecting the opening of the throttle valve is attached to the throttle body 5, and a signal indicating the opening of the throttle valve is sent to the control unit 3.
Input to 0. Similarly, by bypassing the throttle valve 6, IS
The C valve 10 is mounted, and the throttle valve 6 is bypassed by a signal from the control unit 30 to control the air amount, thereby keeping the idle rotation speed constant.

Further, a reference signal for detecting the engine speed, controlling fuel injection timing, and ignition timing is generated from the cam angle sensor 17 and input to the control unit 30.

The temperature of the internal combustion engine 1 is detected by the water temperature sensor 20 and input to the control unit 30.

A pressure sensor 51 and a temperature sensor 52 are mounted on the fuel tank 13, which detect the pressure and temperature inside the fuel tank and send a signal to the control unit 30.

The control unit 30 calculates the optimum fuel amount according to the engine state signal (air flow meter 7, throttle sensor 8, cam angle sensor 17, water temperature sensor 20) and drives the fuel injection valve 12, Fuel is supplied to the internal combustion engine 1. Similarly, regarding the ignition timing, the igniter 1
When electricity is supplied to 9, ignition is performed through the ignition coil 18.

FIG. 5 shows the internal structure of the control unit 30 in one embodiment of the present invention.
100, a RAM 101 that allows free reading and writing, a ROM 102 dedicated to reading, and an I / O LSI 103 that controls input / output.
Are connected by buses 104, 105 and 106, respectively, and data is exchanged. MPU100 is
A signal indicating the engine state (in addition to the sensor signals described, a fuel tank pressure sensor 51, a fuel temperature sensor 52, a battery voltage 53, an ignition SW 54, a starter S
W55, neutral SW56, air conditioner SW57, and vehicle speed sensor 58) are received from the I / O LSI 103 through the bus 106, the processing contents stored in the ROM 102 are sequentially called, and predetermined calculation is performed, and then each actuator (injector) 12, igniter 19,
The drive signal to the ISC valve 10, etc.
Supply through I103.

Next, the contents of control for implementing the present invention will be described. The order of ignition is the order of the first, third, fourth, and second cylinders for smooth rotation.
FIG. 6 shows the calculation timing of each task such as fuel calculation and ignition calculation, and the method of outputting fuel injection and ignition signals. 1, 2, 3 of the cylinder identification task 61 shown in the figure
At 4, the cylinder identification task is executed. The fuel calculation task is performed in the fuel calculation tasks 63, 1, 4, 7 and 10 shown in the figure. At time 1, the fuel calculation for the fourth cylinder is performed, and at time 4, the fuel calculation for the second cylinder is performed. The ignition timing calculation task is 1, 4, 7, 10 of the illustrated ignition timing calculation task 64.
In.

First, regarding fuel injection, as a precondition of the present invention, injection is performed independently for each cylinder. The method will be briefly described below. Since fuel injection is performed independently for each cylinder, the cylinder identification signal 61 is input to the control unit 30 in the illustrated form. The cylinder identification signal 61
Indicates that the signal width is wide only for the signal generated at 110 degrees before the compression top dead center of the first cylinder, and by utilizing this, the first cylinder is identified.

When the cylinder identification signal 61 is input to the control unit 30, the cylinder identification task 62 is activated to identify the cylinder to which fuel should be injected from the cylinder identification signal 61 and the calculated fuel injection width (fuel injection width). The width is also necessary for identifying the cylinder, because the fuel injection must be completed before the intake starts. For example, in FIG. 6, the fuel is injected into the second cylinder in the cylinder identification task 62-2. When the fuel injection time is extended, it is identified in 1 of the cylinder identification task 62 that fuel injection must be started), and the above I /
The injection start position is set in the injection control register corresponding to each injector in the OLSI 103. The fuel injection width is calculated by a fuel calculation task that is started every predetermined time. At the end of the calculation result, the I / O LSI 10
The data of the injection width control register corresponding to each injector in 3 is updated.

These registers are compared with the timer, fuel injection is started when the values of the injection control register and the timer become the same, and values larger than those of the injection control register are stored. The injection width control register and the timer are compared, and the injection ends when the values of the timer and the injection width control register become the same. This is shown in FIG.
Shown in.

Next, regarding the ignition signal, I / OLSI
The output (igniter) from 103 is one, and the power is distributed by the power distribution circuit of the control unit 30. This is also similar to the fuel injection control, the ignition control of the I / OLSI 103 is performed by the cylinder identification task 62. Ignition timing is set by the ignition calculation task 64 that sets the energization start position in the register and is activated every predetermined time, and the result is I / OLS
Data is set in the energization time control register of I103 as the energization width. The usage of the ignition control register and the energization time control register is the same as the usage of the injection control register and the injection width control register described above.

The data acquisition and processing of the O2 sensor is performed by the O2F / B task 65 (second, third, fourth) shown in FIG.
The calculation for the cylinder) and the O2F / B task 66 (calculation for the first cylinder) are performed at all timings. That is, the O2F / B task A65 and the O2F / B task B66 are activated at predetermined time intervals,
Following the O2F / B task A, the O2F / B task B is performed to calculate the O2F / B coefficients α and αev. However, during the period in which the first cylinder is in the exhaust process, the O2 sensor 22A does not output an appropriate output due to the exhaust of the first cylinder. Therefore, in the O2F / B task A, as will be described later,
O2F / B coefficients α and αev are not calculated. The detailed flow will be described below with reference to the flowchart.

FIG. 7 shows the flow of the cylinder identification task. First, in step 1000, the cylinder identification counter CYLCNT is incremented, and in step 1001, it is determined whether or not it is larger than 5. If it is larger, step 1
In 002, the cylinder identification counter CYLCNT is set to 1 so that the cylinder identification counter CYLCNT operates between 1 and 4,
Moreover, by defining that the cylinder identification counter CYLCNT = 1 is 110 degrees before the compression top dead center of the first cylinder,
The cylinder position can be known from the value of the cylinder identification counter CYLCNT. According to the above-mentioned prescribed method, the control unit 30 starts the control and the cylinder identification counter is incremented at the same time when the cylinder identification signal 61 is input, but the falling edge is detected although the width of the cylinder identification signal 61 is wide. At this point, the cylinder identification counter CYLCNT is reset to 1.

Next, at step 1003, the fuel injection width T
i is read from the RAM, the injection cylinder is selected in step 1004, and the injection start timing is set in step 1005. Next, in step 1006, the ignition timing is read,
The energization angle is calculated in step 1007, and step 1008 is calculated.
Set the energization start position with and finish.

FIG. 8 shows the flow of the fuel calculation task. First, in step 1100, the intake air amount Qa
Is read, in step 1101 the engine speed Ne is read, in step 1102 various correction factors COEF due to engine water temperature and the like are read, and step 1103
The O2F / B coefficient α and the learning value αL of α are read from the RAM, and the fuel injection width Ti is calculated in step 1104 by the following calculation formula.

Ti = k × Qa / Ne × COEF × α × αL + TB where Ti: fuel injection width, K: injector injection amount adjustment coefficient (read from ROM), Qa: intake air amount, N
e: engine speed, COEF: various correction factors, α: O
2 F / B coefficient by sensor A (22A), αL: learning value α, TB: injector invalid pulse width (read from ROM).

Next, in step 1105, the second to fourth
The above calculation result is set in the injection register corresponding to the cylinder. Next, in step 1106, the O2 F / B coefficient α
ev is read, and in step 1107, the fuel injection width Ti1 for the first cylinder is calculated by the following formula, and step 11
At 08, the result is set in the injection width control register of the first cylinder.

Ti = k × Qa / Ne × COEF × α × α
L × αev + TB αev: F / B coefficient by O 2 sensor B (22B) FIG. 9 shows the O 2 F / B task A control flow. First, in step 1200, the O2 mask continuation number O2MSK is read. This is because the exhaust gas is emitted from the purge cylinder because the concentration of the gas discharged from the purge cylinder does not affect the other cylinders that have not been purged.
2 When the sensor 22A detects the exhaust gas, O2
The value of the sensor 22A is not used. That is, it is set for the purpose of temporarily interrupting the O2 F / B control, and the details will be described later.

Next, in step 1201, the value of O2MSK is checked, and if it is not 0, this flow is ended, and if it is 0, it proceeds to step 1203. In step 1203, the O2 sensor A (22A) is A / D converted, the converted value is compared with the air-fuel ratio determination slice level, and if the converted value is large,
Rich, if it is smaller, it is determined to be Lean, and step 12
Go to 04. In step 1204, it is checked whether the previous air-fuel ratio determination result is Lean. If it is not lean, the process proceeds to step 1205 to check the determination result of this time. If the result of this determination is not Rich, it means that the air-fuel ratio has changed from Rich to Lean.
In step 1207, the proportional component is subtracted, and step 1211
Go to. Further, if it is judged as Rich in step 1205, since the air-fuel ratio is still in the Rich state, the integral is subtracted in step 1206, and the process proceeds to step 1211. The processes from step 1208 to step 1210 are performed by the same method.

Next, at step 1211, the results of the proportional and integral corrections are stored in the O2F / B coefficient α. Step 1
At 212, the weighted average processing of α is performed, and step 12
At 13, the deviation between the weighted average result and the target control median is calculated,
The result is stored as the learning value αL in step 1214 and the present flow ends.

FIG. 10 is one of the two flows for calculating the O2 mask period, and the process is performed after the cylinder identification task described in FIG. In step 1300, it is checked whether or not the canister is being purged. If the canister is being purged, the process proceeds to step 1304, and the O2 mask continuation number O2MSK
Is set to 0 and αev is fixed to 1.0. If the canister is being purged in step 1300, step 13
The flow proceeds to 01 and O2MSK is checked, and if it is not 0, this flow is ended, and if it is 0, it proceeds to step 1302. In step 1302, the cylinder identification counter CYLCNT is checked, and if it is not 2, this flow is ended, and if it is 2, the first
Since this corresponds to the exhaust stroke of the cylinder, the routine proceeds to step 1303, where data is set in the O2 mask start position 02STCNT.
The data to be set corresponds to the delay time from when the cylinder identification signal is generated until the exhaust gas of the first cylinder reaches the O2 sensor A is converted into one cycle of the flow of the O2F / B task A. Value.

FIG. 11 shows the remaining flow for calculating the O2 mask period, which is executed after the flow of the O2 F / B task A described in FIG. Step 140
At 0, check if the above O2STCNT is 0,
If it is not 0, the process proceeds to step 1401 and O2STCNT is decremented. Then, in step 1402, O2STCNT is checked. If it is 0, the process proceeds to step 1403,
Set the data in the mask continuation count O2MSK. On the other hand, in step 1400, if the O2STCNT is 0, the process proceeds to step 1404, and it is checked whether 02MSK is 0. If it is 0, this flow is ended. Decrement and end this flow.

FIG. 12 shows the flow of the O2F / B task B. This flow is for the purge cylinder (first cylinder)
O2 sensor B (22B) installed in the exhaust pipe corresponding to
By this, the A / F of the purge cylinder is individually controlled. The control contents are steps 1203 to 1211 in FIG.
Since it corresponds to the above, the description is omitted here.

FIG. 13 shows the ignition calculation task control flow. Basic ignition timing BS in step 1600
Calculate ADV. In step 1601, it is checked whether it is the ignition timing of the first cylinder. As a specific method, as shown in FIG. 6, when the cylinder identification counter CYLCNT processed in the cylinder identification task is 1, it is recognized as the ignition timing of the first cylinder. If it is the ignition timing of the first cylinder, the routine proceeds to step 1602, and the increase rate of the intake air amount of the first cylinder due to the purge is calculated. This is because the air amount Qe flowing through the canister purge valve 41 is determined from the control signal output to the canister purge valve 41.
Take the method of estimating vp. Next, step 1603.
In order to correct the torque increase amount due to the increased air amount at the ignition timing, the retard amount is calculated by the following formula.

ΔADV = BSADV × Qevadv where ΔADV is the retard angle amount, BSADV is the basic ignition timing, and Qevadv is the retard amount correction coefficient (a table search value based on the air amount Qevp).

The basic ignition timing is corrected with the obtained retardation amount, the energization time is set in step 1604, and the process is ended. If it is not the ignition timing of the first cylinder in step 1601, the basic ignition timing calculated in step 1600 is set in step 1604, and the process ends.

In the present embodiment, only the first cylinder is purged in order to prevent engine stalling, but the present invention is not limited to this, and the type of engine,
Alternatively, two or more cylinders may be purged depending on the operating state.

Further, in order to prevent the fluctuation of the air-fuel ratio, the air-fuel ratio sensor is attached only to the exhaust pipe branch portion of the first cylinder, but it is assumed that a plurality of cylinders are purged and the exhaust pipe branch portion of the cylinder to be purged. An air-fuel ratio sensor may be attached to each.

[0061]

As described above, according to the present invention, the fuel vapor is purged into the specific cylinder. Therefore, even if the rich air-fuel mixture is temporarily sucked, the combustion fluctuation remains only in the specific cylinder. Can be prevented. Further, since an O2 sensor is provided in the exhaust pipe corresponding to the purge cylinder and the A / F is controlled independently of the other cylinders, unnecessary A / F fluctuation can be suppressed.

Furthermore, since the increased torque due to the purge is controlled by retarding the ignition timing, the torque increase between the cylinders can be suppressed, and stable engine rotation can be obtained.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of an electronically controlled fuel injection device according to the present invention.

FIG. 2 is an explanatory diagram of an electronically controlled fuel injection device in a conventional example.

FIG. 3 is an explanatory diagram showing an influence of unbalance of air-fuel mixture distribution between cylinders according to a conventional technique.

FIG. 4 is an explanatory diagram of an electronically controlled fuel injection device according to the present invention.

FIG. 5 is a block diagram of a control unit of the electronically controlled fuel injection device according to the present invention.

FIG. 6 is an explanatory diagram showing task calculation timing in the electronically controlled fuel injection device according to the present invention.

FIG. 7 is a flow chart of a cylinder identification task in one embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 8 is a fuel calculation task flow in one embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 9 is a flow of O 2 F / B task A in an embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 10 is a flow of O 2 mask period control 1 in one embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 11 is a flow of O 2 mask period control 2 in one embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 12 is a flow chart of O 2 F / B task B in one embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 13 is a flow chart of an ignition calculation task in an embodiment of the electronically controlled fuel injection device according to the present invention.

FIG. 14 is an explanatory diagram of operations of a fuel injection width control register and a fuel injection control register.

[Explanation of symbols]

 1 Engine 5 Throttle Body 9 Surge Tank 11 Intake Pipe Branch 13 Fuel Tank 22A O2 Sensor 22B O2 Sensor 40 Canister 41 Canister Purge Valve 43 Purge Cut Solenoid

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yutaka Takahisa 2520 Takaba, Katsuta City, Ibaraki Prefecture Hitachi Automotive Systems Division

Claims (5)

[Claims] 1. A fuel vapor recovery means for recovering vaporized fuel generated in a fuel tank, and a recovered fuel supply means for supplying the recovered fuel recovered by the fuel vapor recovery means to an engine having a plurality of cylinders. In the fuel injection device, the recovered fuel supply means has a pipe for supplying the recovered fuel to the engine, and a connecting position of the pipe and the engine is provided with an intake pipe through which intake air flows into the cylinder. A fuel injection device characterized in that an intake pipe branching portion is branched for each. 2. The fuel injection device according to claim 1, wherein an exhaust pipe through which exhaust gas from the cylinder to which the recovered fuel is supplied by the recovered fuel supply device flows is branched to each of the cylinders. A first air-fuel ratio sensor that is provided and detects an air-fuel ratio, a controller that controls the fuel injection amount of a cylinder to which the recovered fuel is supplied by the air-fuel ratio, and a controller that is controlled by the controller to perform the recovery A fuel injection device having a fuel injection unit for injecting fuel into a cylinder to which fuel is supplied. 3. The fuel injection device according to claim 2, wherein the control section controls the ignition timing of the cylinder, and the recovered fuel is supplied while the recovered fuel is being supplied by the recovered fuel supply means. The fuel injection device is characterized in that the ignition timing of the cylinder to be set is delayed from the ignition timing of the cylinder to which the recovered fuel is not supplied. 4. The fuel injection device according to claim 2 or 3, wherein an air-fuel ratio of intake air supplied to a cylinder not provided with the recovered fuel, which is provided in an exhaust pipe merging portion where exhaust gas of an engine joins is detected. The second air-fuel ratio sensor, the control unit controls the fuel injection amount of the cylinder to which the recovered fuel is not supplied by the output from the second air-fuel ratio sensor, and the recovered fuel is supplied. A fuel injection device characterized in that the output from the second air-fuel ratio sensor is not used when the existing cylinder is in the exhaust stroke. 5. The fuel injection device according to claim 1, 2, 3 or 4, wherein the number of cylinders to which the recovered fuel is supplied is one.