JPH0438904B2 - - Google Patents

Description

TECHNICAL FIELD The present invention relates to a fuel injection method for a direct injection internal combustion engine such as a diesel engine.

Problems to be Solved by the Prior Art and the Invention In a cylinder injection engine (for example, a diesel engine), fuel injected into the cylinder is ignited sequentially at the same time as the injection, and combustion spreads. It is required from the viewpoint of combustion efficiency that the distance is short.
Ease of ignition and combustion is strongly influenced by temperature and pressure since this is a chemical reaction, and the higher these are, the easier it is to ignite and burn. Therefore, high temperature
Although it is desirable to inject a large amount of fuel in the later stages of combustion when the pressure is high, it has been difficult to implement such a control method.

On the other hand, in a direct injection engine, unlike a normal intake pipe injection engine (for example, a general gasoline internal combustion engine), the ignition delay period of the fuel injected into the compressed air in the engine cylinder varies greatly depending on the engine condition. Change. In particular, the lower the temperature of the intake air and the combustion chamber wall, the greater the ignition delay.As a result, when the ignition delay period is large, the amount of fuel injected during the ignition delay period increases, and the fuel is When it ignited, it instantly combusted, creating a loud explosion and noise.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a fuel injection control method for a direct injection internal combustion engine that increases combustion efficiency and achieves both increased combustion efficiency and noise prevention at low temperatures.

Means for Solving the Problems In order to solve the above-mentioned problems, the present invention provides a plurality of injection times for one combustion, and sets the injection stop period of each injection to become longer as the temperature of the engine becomes lower. At the same time, the injection amount at the early injection point is made smaller and the injection amount at the later injection point is increased.

Effect According to the above-mentioned means, by making the number of injections plural for one combustion, the amount of fuel increases as the later stage of the combustion progresses.

Furthermore, by increasing the injection stop period of each injection as the engine temperature decreases, the amount of fuel during the ignition delay period, which becomes longer at low temperatures, can be reduced.

Embodiment FIGS. 1 and 2 are a hardware configuration diagram of a diesel engine, a control device, and a side sectional view of a fuel injection valve for realizing the fuel injection method for an internal combustion engine according to the present invention.

In FIG. 1, 1 is an electrostrictive actuator 2
This is an injection valve that is driven to open and close by the An injection valve 1 is attached to a diesel engine 8 so as to inject fuel from a nozzle into a cylinder of the diesel engine 8, and high-pressure fuel is supplied to the injection valve 1 from an accumulator 3. A fuel pressure of, for example, 200 kg/cm ² is supplied to the accumulator 3 from a high-pressure pump 9 driven by an internal combustion engine and is stored therein.

FIG. 2 is a side sectional view of the injection valve 1. FIG. The voltage applied to the electrostrictive actuator 2 is controlled by the control device 4, and when the applied voltage is -500V, the injection valve 1 opens and injects fuel from the nozzle 120 into the cylinder of the internal combustion engine. , and the applied voltage is +500V
At this time, the injection valve 1 closes and stops fuel injection. This injection valve is described in a patent application filed on April 15, 1982 by the present applicant entitled "Fuel Injection Device for Internal Combustion Engine."

In the device shown in FIG. 2, the injection valve 1 is operated by the expansion and contraction of the electrostrictive actuator 2. The electrostrictive actuator 2 is a stack of thin disc-shaped piezoelectric elements having an electrostrictive effect. For example, if a ceramic made by sintering lead zirconate titanate called PZT is used as a piezoelectric element, and the diameter is 15 mm and the thickness is 0.5 mm, when 500 V is applied in the thickness direction, it will elongate by about 0.5 μm, and vice versa. When −500V is applied to
Shrinks by 0.5μm. Therefore, 100 such elements
If you stack them, you can get 100 times more expansion and contraction.

Silver electrodes are formed on both sides of each piezoelectric element of the electrostrictive actuator 2, one of which is connected to the lead wire 201 and the other is grounded.
The lead wire 201 passes through the casing upper 104 via the grommet 103, is taken out to the outside, and is connected to the control device 4. The expansion and contraction motion of the electrostrictive actuator 2 is directly transmitted to the piston 105, and as a result, the piston 105 performs vertical reciprocating motion.

The piston 105 slides within the cylinder 106 within the casing upper 104, expands and contracts the volume of the pump chamber 107, and performs a pump action. A disc spring 108 is provided within the pump chamber 107 to bias the piston 105 in the direction of contraction of the electrostrictive actuator. This compensates for the contraction force, which is weaker than the expansion force of the electrostrictive actuator 2.

The pump chamber 107 communicates with the inside of the needle cylinder 112 of the nozzle body 111 via a disc-shaped distance piece 110. A nozzle needle 113 is slidably housed in the needle cylinder 112, and the pressure of the pump chamber 107 acts on the upper end surface of the nozzle needle 113. A ring-shaped recess 114 is formed on the outer periphery of the upper end of the nozzle needle 113, and a disc spring 115 is fitted into this recess 114 for urging the nozzle needle 113 downward using the lower end surface of the distance piece 110 as a fulcrum. It is.

Further, a plurality of conduction holes 116 are provided in the axial direction of the distance piece 110, whereby the pump chamber 107 and the needle cylinder 112 are electrically connected. Note that the conduction hole 116 opens into the needle cylinder 112, facing the recess 114 of the nozzle needle 113.

A ring-shaped recess 117 is provided on the outer periphery of the lower end of the nozzle needle 113, and a protrusion 118 is provided in the center. And 11
9 indicates a flat surface at the lower end of the nozzle needle 113. A nozzle 120 is provided in the center of the needle cylinder 112 to penetrate the protrusion 118 of the nozzle needle 113 . Note that the flat surface 119 of the nozzle needle 113 and the lower end of the needle cylinder 112 can be in close contact with each other.
As a result, if they are in close contact, the nozzle 120 will be closed, and if they are not in close contact, the nozzle 120 will be opened.

A ring-shaped enlarged fuel reservoir 122 is provided at the lower end of the needle cylinder 112.
This fuel reservoir 122 is connected to a fuel passage 123. The fuel passage 123 is connected to the nozzle body 11
1. It passes through the distance piece 110 and the casing upper 104. Further, a passage 124 communicating with a fuel reservoir 122 is provided in the nozzle needle 113, and an opening 125 is provided at the upper end of the passage 124.
is formed. This opening 125 has an upper end surface of the nozzle needle 113 connected to the distance piece 110.
It is closed when it is in close contact with the lower end surface of.

The above-described casing upper 104, distance piece 110, and nozzle body 111 have the same diameter, are stacked in this order, and are pushed out and fixed in the axial direction by a bag-shaped casing lower 126. At this time, the casing lower 12
The female thread 6 is screwed into the male thread of the casing upper 104. Further, at the opening of the lower end of the casing lower 126, the protrusion 118 and the nozzle 12
0 is exposed. Furthermore, the casing lower 12
A male thread 128 is formed on the outer periphery of the engine 6 so that it can be fixed to the diesel engine 8.

In addition, 129 is an O-ring, 130 is a knock pin, and 131 is a fuel inlet provided in the casing upper 104.

Next, the operation of the fuel injection valve shown in FIG. 2 will be explained.
First, when +500V is applied to the electrostrictive actuator 2, the electrostrictive actuator 2 expands, the volume of the pump chamber 106 contracts, and the fuel pressure therein increases. This high pressure is applied to nozzle needle 1
13 to press the nozzle needle 113 against the lower end surface of the needle cylinder 112.
As a result, the injection port 120 is closed and no fuel injection is performed. At this time, the fuel reservoir 122 is filled with 200
Kg/cm ² of fuel is supplied, and this fuel pressure acts on the entire upper end surface of the nozzle needle 113 through the passage 124 of the nozzle needle 113, so that the flat surface 119 under the nozzle needle 113 is Continue to adhere to the lower end surface of.

In the above state, the electrostrictive actuator 2
When a voltage of -500V is applied to , the electrostrictive actuator 2 contracts, and therefore the volume of the pump chamber 106 expands. As a result, the nozzle needle 113
is sucked upward. At this time, the fuel reservoir 12 is passed through the passage 124 of the nozzle needle 113.
2 fuel is also sucked in, but the nozzle needle 113
When the nozzle needle 113 is in close contact with the lower end of the needle cylinder 112, the distance between the upper end of the nozzle needle 113 and the lower end of the distance piece 110 is approximately 0.2 mm, and the diameter of the passage 24 is approximately 0.5 mm.
The amount of fuel drawn through passage 124 is negligible. Further, at this time, if the upper end surface of the nozzle needle 113 comes into close contact with the lower end surface of the distance piece 110, the opening 125 of the passage 124 is closed. Therefore, in this state, the nozzle needle 11
3, the upper end surface of the nozzle needle ¹¹³ continues to be in close contact with the lower end surface of the distance piece 110, and during this time,
Fuel injection is performed from the nozzle 120. Note that during fuel injection, the protrusion 118 adjusts the fuel injection angle and prevents the nozzle port 120 from clogging.

Returning to FIG. 1, the control device 4 includes four sensors 5.
Signals S ₅₁ to _{S 54} are input from nodes 1 to 54, respectively.

The sensor 51 is a rotational speed sensor using, for example, a magnetic pickup, and emits an angle signal _S51 corresponding to the rotational speed N _E (rpm) of the internal combustion engine. This rotational speed sensor 51 is placed near a signal plate 56 attached to a shaft 55 that rotates once in synchronization with 1/2 rotation of the engine crankshaft, and is engraved on the outer periphery of the signal plate 56. It detects the protrusion 57 and generates an angle signal _S51 consisting of 180 pulses per revolution of the signal plate.
One pulse of S ₅₁ is for the engine/crankshaft.
Compatible with rotation of 1°CA.

The sensor 52 is a reference position sensor using, for example, a magnetic pickup or the like, and detects one protrusion 58 provided on the signal plate 56 to generate a reference signal _S52 . This reference signal S ₅₂
is generated at a reference position, for example, in this embodiment, 30° CA before compression top dead center of the internal combustion engine,
A protrusion 58 is provided at an appropriate position for this purpose.

The sensor 53 is, for example, a load sensor using a potentiometer that is interlocked with the accelerator pedal 59, and generates a voltage signal _S53 corresponding to the opening degree θ (deg) of the accelerator pedal.

The sensor 54 is a temperature sensor using a thermistor or the like, and is installed in the water jacket of the internal combustion engine, and its resistance value changes in response to the temperature T (〓) of the cooling water of the internal combustion engine, and the signal _S54 is output. Occur. Note that the temperature sensor 54 may be configured to detect the temperature of intake air instead of the cooling water.

The control device 4 receives output signals S ₅₁ , S 53 , and output signals from the sensors 51, 53, and ₅₄ corresponding to the rotational speed N _E of the internal combustion engine, the accelerator opening θ, and the cooling water temperature T, respectively.
An appropriate injection fuel amount q (g/st) is calculated based on _S54 , and an injection time τ (μsec) required to inject the fuel amount q is calculated. At the same time, the appropriate fuel injection start timing is calculated, the arrival of that timing is determined by the rotational speed sensor 51 and the reference position sensor 52, and a voltage of -500V is applied to the electrostrictive actuator 2 at the injection start timing to perform injection. Open valve 1.

4 and 3 are a block diagram and an explanatory diagram of the opening time of the injection valve for explaining the first embodiment of the fuel injection method for an internal combustion engine according to the present invention, and the fuel injection time τ is divided into a first injection time τ ₁ and a second injection time τ ₂ , with an injection stop time t placed between them. That is, τ=τ ₁ +τ ₂ ...(1a) holds true.

Here, the first injection time τ ₁ and the second injection time τ ₂ are set so that τ ₁ <τ ₂ ...(1b) is satisfied.
Establish. The injection stop time t is determined by the following equation (2) based on the rotational speed N _E , the fuel amount q, and the cooling water temperature T, and is the ignition delay time estimated under each of the engine operating conditions described above.

t=m/N _E ×q×T ……(2) Here, m is a constant. For example, if m=2.5×10 ⁶ (msec.rpm.mg/st.〓), N _E =600rpm, q=7mg/st, T=353
When 〓, t=1.69 msec. Also, N _E
= 600 rpm, q = 7 mg/st, T = 273〓, t = 2.18 msec. The appropriate injection time τ when q=7 mg/st is 200 μsec, and if the first injection time τ ₁ is set to 30 μsec, for example, the second injection time τ ₂ is 170 μsec.

The injection stop time t (=2.18 msec) is set between the first injection time τ ₁ and the second injection time τ ₂ .

In this way, by setting the injection stop time t, the main fuel injection is stopped until the minute amount of fuel injected as a pilot into the cylinder during the first injection time τ ₁ is ignited, and after the ignition, fuel injection is resumed. is the second
Since the injection is carried out over the injection time τ ₂ , there is no problem in the conventional case where the fuel injected during the ignition delay time burns instantaneously at the same time as ignition and produces a loud explosion sound.

Note that, as is clear from equation (2), the injection stop time t becomes shorter as the rotational speed N _E , fuel amount q, and cooling water temperature T increase; for example, N _E =
At 2000rpm, q=15mg/st, T=353〓, t=
It becomes 0.24msec. Although not adopted in this embodiment, the injection stop time t is
When the time is about 0.2 msec or less, control that eliminates the injection stop time is also effective. The reason for this is that under conditions where the value of N _E ×q × T is greater than a certain level, the ignition delay time becomes extremely small, so that noise caused by premixed combustion does not become a problem.

Next, the control device 4 will be explained.

In FIG. 4, the output signals S ₅₁ and S 52 of the rotational speed sensor 51 and the reference position sensor ₅₂ are waveform-shaped by a first shaping circuit 411 and a second shaping circuit 412, respectively, and are shaped into "O" (ov) or “1”
(5v) level pulses are converted into an angle signal _S11 and a reference signal _S12 . A signal S ₅₃ from the load sensor 53 is converted into a 16-bit digital signal D ₁₃ by a first A/D (analog/digital) conversion circuit 413 and guided to a bus line 494. As described above, the resistance value of the temperature sensor 54 changes according to the cooling water temperature T, and this low resistance value change is converted into a 16-bit digital signal _D14 by the second A/D conversion circuit 414, and is sent to the bus line as described above. 494.

The engine rotation counter 415 is a 16-bit counter, and the second shaping circuit 4 is connected to its reset input.
A reference signal S ₁₂ from 12 is introduced, and a 100 KHz clock signal CLK ₁ generated by a clock signal generation circuit 416 is also introduced to the clock input. This engine revolution counter 415 has a function that automatically stops counting at the maximum value of the counter so that the counter does not overflow before a reset signal is received. It is assumed that each counter described below is also provided with this function.

The count contents of the engine revolution counter 415 are latched when the reference signal _S12 is input to its reset input, and are led to the bus line 494 in the form of a digital output signal _D15 . Therefore, the output signal D ₁₅ of this engine rotation counter 415 corresponds to the engine rotation period _TN .

The angle counter 417 is a 16-bit counter,
A reference signal _S12 is introduced to its reset input terminal,
The angle signal S ₁₁ from the first shaping circuit 411 is led to its clock input terminal, and its output signal D ₁₇ is led to one input terminal of the angle comparator 419. Therefore, the count contents of the angle counter 411 represent the engine rotation angle φ at every moment after the reference signal _S12 is generated.

The angle latch circuit 418 is a 16-bit latch circuit that latches the engine rotation angle φi corresponding to the appropriate injection start time calculated by the CPU (central processing unit) 491, which will be described later, and outputs the angle in the form of a digital signal _D18 . It is output to the other input terminal of the comparator 419.

The angle comparator 419 is a 16-bit comparator, which compares the output D17 of the angle counter 417 and the output _D18 of the angle latch circuit ₄₁₈ , and outputs _D17.
=D ₁₈ , that is, when the engine rotation angle φ from the reference position reaches the rotation angle φi corresponding to the appropriate injection start timing, a coincidence signal S ₁₉ of “1” level is output.

The one-shot multi 420 has a trigger input terminal fed with a coincidence signal _S19 from the angle comparator 419, and outputs a "1" level pulse signal having a constant time width of 30 μsec in synchronization with this signal _S19 . This constant time width corresponds to the first injection time τ ₁ .

The first counter 421 is a 16-bit counter, and its reset input terminal receives the output signal _S20 of the one-shot multi 420, and its clock input terminal receives the output signal S20 from the clock signal generation circuit 416.
A 1MHz clock signal CLK2 is introduced. Therefore, the output signal D ₂₁ of the first counter 421 is
It represents the momentary elapsed time _T1 after the one-shot multi 420 finished generating a 30 μsec pulse.

The first latch circuit 422 is a 16-bit latch circuit, and latches the injection stop time t calculated by the CPU 491 and outputs it as a signal _D22 .

The first comparator 423 is a 16-bit comparator, and outputs an output signal D ₂₁ of the first counter 421 corresponding to the elapsed time T ₁ and an output signal of the first latch circuit 422 corresponding to the injection stop time t.
D ₂₂ is compared, and when D ₂₁ = D ₂₂ , that is, the elapsed time T _{1 after the end of the first injection time τ 1 becomes} equal to the injection stop time t, a coincidence signal S ₂₃ of “1” level is generated. is output to the set input terminal of flip-flop 427.

The second counter 424 is a 16-bit counter, and its reset input terminal receives the coincidence signal S23 of the first comparator 423, and its clock input terminal receives the match signal _S23 from the clock signal generation circuit 416.
A 1 MHz clock signal CLK2 is introduced and its output signal D ₂₄ is introduced to one input terminal of a second comparator 426. Therefore, the second counter 424
The output D ₂₄ represents the momentary elapsed time T ₂ since T ₁ =t.

The second latch circuit 425 is a 16-bit latch circuit, and the second injection time τ ₂ calculated by the CPU 491
is latched and output to the second comparator 426 in the form of a digital output signal _D25 .

The second comparator 426 is a 16-bit comparator, and outputs an output signal D ₂₄ of the second counter 424 corresponding to the elapsed time T ₂ and an output signal D ₂₅ of the second latch circuit 425 corresponding to the second injection time τ ₂ . When D ₂₄ =D ₂₅ , that is, when the elapsed time T ₂ since the end of the injection stop time t becomes equal to the second injection time τ ₂ , the coincidence signal S ₂₆ at the “1” level is sent to the flip-flop 427. Output to the reset input terminal.

The reset flip-flop 427 is
As mentioned above, the coincidence signal S 23 of T ₁ =t, which is the output of the first comparator 423, is introduced to the set input terminal, and the second signal S ₂₃ is introduced to the reset input terminal.
A coincidence signal S ₂₆ with T ₂ =τ ₂ which is the output of the comparator 426 is derived. Therefore, from the output terminal Q of the flip-flop 427, a signal _S27 is outputted to the OR circuit 428, which becomes the "1" level when T ₁ =t and becomes the "0" level when T ₂ =τ ₂ .

The OR circuit 428 logically ORs the signal S ₂₀ with a time width τ ₁ of 30 μsec, which is the output of the one-shot multi 420, and the output signal S ₂₇ of the flip-flop 427, and outputs the resulting signal S ₂₈ to the drive circuit 495. do.

The drive circuit 495 is a circuit that drives the electrostrictive actuator, and its output voltage is applied to the electrostrictive actuator 2. This drive circuit 495 outputs a voltage of -500V to open the injection valve 1 when the output signal _S28 of the OR circuit 428 is at the "1" level, and the output signal S28 of the OR circuit 428 is at the "1" level.
When the level is reached, a voltage of +500V is output and the injection valve 1 is closed.

The power supply circuit 496 stabilizes the voltage supplied from the battery 6 via the key switch 61 and supplies it as a power supply voltage to each part of the device, and also generates a high voltage of ±500V for driving the electrostrictive actuator and supplies it to the drive circuit 495. This is a supply circuit.

CPU491 is a 16-bit central processing unit,
The reference signal _S12 is connected to the interrupt input terminal INT1.
is guided. The ROM 492 is a read-only memory that stores programs and various data for the CPU 491, and the RAM 493 is a random access memory for the CPU 491 to work with. these
ROM492, RAM493 are bus line 494
It is connected to the CPU 491 via.

Next, the operation of the embodiment device including the control device 4 having the above circuit configuration will be explained.

FIG. 5 is a program flowchart of the CPU 491, and FIG. 6 is a signal waveform diagram of each part in the device shown in FIG.

First, processing in the CPU 491 will be explained. The processing routine of the CPU 491 consists of a main routine and an INT1 routine that is activated each time an interrupt INT1 is input. In the main routine, after initializing each part, interrupts are enabled and the program enters an idle loop.

The INT1 routine is activated every time the reference signal _S12 is generated, calculates the required amount of injected fuel and injection start timing, and calculates the engine rotation angle φi, injection stop time t, and second injection time at the appropriate injection start time mentioned above. Calculate τ ₂ and apply the calculation results to the latch circuit 41.
8, 422, and 425 to latch, thereby controlling the fuel injection.

At the beginning of the INT1 routine, the accelerator opening degree θ is read from the first A/D conversion circuit 413, and
2 Read the water temperature T from the A/D conversion circuit 414,
Next, the rotation period is determined from the engine rotation counter 415.
Load T _N. Then, the engine speed N _E is calculated from the rotation period T _N , and the basic injection amount is calculated by interpolation from the map of the basic injection amount q _BASE for the accelerator opening θ and the engine speed N _E , which have been determined in advance by bench tests etc. q Find _BASE .

Next, by finding the correction amount P for the water temperature T from the map and multiplying it by the basic injection amount q _BASE ,
The temperature-corrected injected fuel amount q is calculated. Then, the injection time τ during which the injection valve is opened is determined from the map of the injection time τ with respect to the injected internal combustion amount q.
Furthermore, apart from these calculations, the rotation angle φi of the injection start time under the above conditions is determined by interpolation from a map of the rotation angle φi of the injection start time with respect to the injected fuel amount q and the engine speed _NE .

Next, the injection stop time t is calculated based on the formula: t=m/N _E ×q×T, and the second injection time τ ₂ is determined using the following formula: τ ₂ =τ−30 μsec. Finally, the injection start time engine rotation angle φi, injection stop time t, and second injection time τ ₂ determined by the above calculations are output to each latch circuit 418, 422, and 425, and then the process returns to the main routine. Thereafter, signals are generated at predetermined timings by each circuit within the control device 4.

First, the reference position sensor 52 generates a reference signal at a reference position of 30° CA before compression top dead center of the diesel engine.
A _{signal S52} is generated, which is waveform-shaped by the second shaping circuit 412 to generate a reference signal _S12 shown in FIG. 6. This reference signal S ₁₂ is applied to the angle counter 417
The angle counter 417 starts counting the number of pulses of the angle signal S ₁₁ corresponding to the engine rotation angle from this point on, and outputs the count output signal D ₁₇ to the angle comparator 419.

Angle comparator 419 is an angle counter 417
When a predetermined number of pulses are input to the output D ₁₇ and the output D 17 becomes equal to the output D ₁₈ of the angle latch circuit 418, the momentary engine rotation angle φ becomes the rotation angle φi corresponding to the proper injection start timing. At that point, the pulse output signal _S19 of "1" level shown in FIG. 6 is output to the one-shot multi 420. As a result, the one-shot multi 420 is triggered and outputs a pulse output signal _S20 having a time width of 30 μsec shown in FIG. 6 to the OR circuit 428 and the first counter 421.

The first counter 421 starts in synchronization with the fall of this signal _S20 , and the elapsed time _T1 counted by this first counter 421 is the first latch circuit 4.
At the time when the injection stop time t becomes equal to the injection stop time t set as 22, a pulse signal _S23 of the "1" level shown in FIG. Sent out. Therefore, flip-flop 427 is set as shown in FIG. 6.
At the same time, the second counter 424 starts and T ₂ =
When τ ₂ , that is, the elapsed time T ₂ measured by the second counter 424, becomes equal to the second injection time τ ₂ set in the second latch circuit 425, the second
A pulse signal S26 shown in FIG. 6 is generated at the output of the comparator 426, and this signal _S26 causes the pulse signal _S26 shown in FIG.
The output of flip-flop 427 is reset as shown in FIG.

The output signal S ₂₆ of the one-shot multi 420 and the output signal S ₇ of the flip-flop 427 are added in an OR circuit 428, and the OR circuit 428 sends an output signal S ₂₈ as shown in FIG. 6 to the drive circuit 495. As shown in FIGS. 9 and 10, the drive circuit 495 performs fuel injection by outputting a voltage of -500V to the electrostrictive actuator 2 when the signal _S28 is at the "1" level, and when the signal S28 is at the "0" level. Sometimes +
Outputs a voltage of 500V and stops fuel injection. 7 and 8 are a block diagram and an explanatory diagram of the opening time of the injection valve for explaining a second embodiment of the fuel injection method for an internal combustion engine according to the present invention. The injection method using this second embodiment device is to divide the fuel injection into three times, lengthen the total fuel injection period to improve air utilization, and sequentially This improves combustion efficiency by increasing the valve opening time.

The injection method of this second embodiment is the same as the first embodiment up to the point where the injection time τ is determined by the control device 4'. In the second embodiment, as shown in FIG. 8, the injection time τ is the first injection time τ ₁ , the second injection time τ ₂ , and the third injection time τ ₃
It is divided into three parts. τ ₁ , τ ₂ and τ ₃ have a 1:2:3 relationship, τ ₁ =τ/6 τ ₂ =τ/3 τ ₃ =τ/2 τ ₁ +τ ₂ +τ ₃ =τ. The interval between τ ₁ and τ ₂ and between τ ₂ and τ ₃ , that is, the injection stop time t, is a function of the rotational speed N _E and is determined by the following equation.

t=n/N _EHere , n is a value determined by the engine cooling water temperature, and the lower the temperature, the larger the value. This n can be obtained by calculation as in the first embodiment.
For simplicity, let the water temperature be constant and treat n as a constant. For example, assuming the water temperature is 80℃, n=
If the rotation speed N _E is 600 rpm, the injection stop time t will be 1.5 msec. now,
When the injection fuel amount q is 42 mg/st at full load, the injection time τ is 1.2 msec, the first injection time τ 1 is 0.2 msec, the second injection time τ ₂ is 0.4 msec, and the third injection time τ ₂ is 0.4 msec.
The injection time τ ₃ is 0.6 msec, and an injection stop time t of 1.5 msec is placed between τ ₁ and τ ₂ and between τ ₂ and τ ₃ , respectively. Rotation speed like this
When N _E is low speed, the period from the start of the first injection time τ ₁ to the end of the third injection time τ ₃ is
4.2msec, improving air utilization at low speeds.

On the other hand, when the rotational speed N _E is high, for example 4000 rpm, the injection stop time t is
If the injection fuel amount q is 42 mg/st, the period from the start of the first injection time τ ₁ to the end of the third injection time τ ₃ is only 1.63 msec. In other words, the total fuel injection period at low speed (600 rpm) is 2.6 times longer than at high speed (4000 rpm).

Further, in this second embodiment, the relationship τ ₁ <τ ₂ <τ ₃ holds. In this way, if the injection valve is opened multiple times and the later valve opening time is made longer than the earlier valve opening time, combustion will become more intense in the later stages of combustion, which will reduce noise generation in the diesel engine. Prevention and improvement of combustion efficiency are both achieved.

To explain the control device 4' in FIG. 7, sensors 51 to 54, circuits 412 to 419, 491 to
The configuration of 495, etc. is almost the same as that of the first embodiment. The different parts are a circuit part for obtaining the first to third injection times τ ₁ to _{τ 3} and a circuit part for obtaining the injection stop time t between each injection time, and these will be described below.

The third to seventh counters 430 to 434 are 16-bit counters, each of which has its reset input terminal inputted with the coincidence signal of the previous stage comparator, and whose clock input receives a 1MHz clock signal CLK2. There is. The third to seventh latch circuits 435 to 439 are 16-bit latch circuits, respectively, and the τ ₁ , t, τ ₂ ,
t and τ ₃ are each latched and output. The third to seventh comparators 440 to 444 are 16-bit comparators, and output a "1" level match signal when the outputs of the respective input counters and the latch circuits match. The third to fifth flip-flops 445 to 447 have their set input terminals and reset input terminals led to the match signal of the previous stage comparator and the match signal of the next stage comparator, respectively. The OR circuit 448 adds the outputs of the third to fifth flip-flops 445 to 447 and outputs the sum to the drive circuit 495.

Next, the operation of this control device 4' will be explained. Note that explanations of parts similar to those in the first embodiment will be omitted.

FIG. 9 is a program flowchart of the CPU 491, and FIG. 10 is a signal waveform diagram of each part of the device shown in FIG. The INT routine is activated by the reference signal _S12 , reads the accelerator opening degree θ, water temperature T, and rotation period T _N , and calculates the injection time τ _i and the rotation angle φi at the injection start time, which is the same as the first embodiment. The same is true. After that, calculate the first to third injection times τ ₁ to τ ₃ , τ ₁ = τ/6 τ ₂ = τ/3 τ ₃ = τ/2, and further calculate the injection stop time t corresponding to the cooling water temperature T. Calculate t=n/N _E using the calculated value n and the rotational speed N _E , and set τ ₁ , t, τ ₂ , t, τ ₃ as the third to third values, respectively.
It is output to the seventh latch circuits 435 to 439 and returns. Thereafter, signals are generated at predetermined timings by hardware.

First, after the reference signal S ₁₂ (see FIG. 10 1) is generated, the angle signal S ₁₁ (see FIG. 10 2) of a predetermined number of pulses is generated.
When φ=φi is input (see FIG. 10), a "1" level pulse signal _S19 (see FIG. 10) is generated at the output of the angle comparator 419, and the third flip-flop 445 is set (see FIG. 10). 9). At the same time, the third counter 430 starts,
When the count time reaches τ ₁ , a “1” level pulse signal is sent to the output of the third comparator 440.
_S40 (see FIG. 10) occurs and resets flip-flop 445 (see FIG. 10). At the same time, the fourth counter 431 starts, and when the count time reaches t, the fourth comparator 431 starts.
A “1” level pulse signal S ₄₁ (first
0) occurs, and the fourth flip-flop 4
46 (see FIG. 10). Thereafter, the counters operate one after another in the same manner, and the 3rd to 5th counters operate one after another in the same manner.
The outputs of the flip-flops 445 to 447 include the first
0 As shown in Figures 9 to 11, τ ₁ , τ ₂ , τ ₃ respectively
Signals S ₄₅ to _{S 47} with pulse widths as follows are obtained.

These signals S ₄₅ to _{S 47} are added in an OR circuit 448 and inputted from the OR circuit 448 to the drive circuit 495 as a signal S ₄₈ (see FIG. 10).
As shown in FIGS. 13 and 14, the drive circuit 495 outputs - when the signal _S48 is at the "1" level.
Fuel injection is performed by applying 500V to the electrostrictive actuator 2, and when the level is "0", +500V is output and fuel injection is stopped.

Third fuel injection method for an internal combustion engine according to the present invention
In the embodiment, fuel injection is performed as follows.

The fuel injection time τ is divided into three parts as in the second embodiment.

The injection stop time during the confirmation fuel injection is determined based on equation (2) as in the first embodiment.

Various modifications are possible in implementing the invention.

For example, the first and second embodiments show examples in which the present invention is applied to a diesel engine, but the present invention is not limited to this and can be applied to any direct injection engine, including an electrically ignited internal combustion engine. In the case of direct injection engines, their full load performance is determined by the availability of air in combustion. It goes without saying that the higher the degree of air utilization, the better the performance, but in order to improve this degree of air utilization, fuel should be injected into as wide a part of the air flow (generally the intake swirl) as possible. There is a need. The speed of this air flow is
Since it is normally proportional to the rotational speed of the engine,
If the injection time is constant, the lower the speed, the lower the degree of air utilization, and the lower the full load performance will be. In the present invention, as the engine rotational speed becomes lower, the injection stop time becomes longer and the total fuel injection period becomes longer, so that air utilization is always maintained at a good level regardless of the engine speed.

Further, in the first and second embodiments, only the circuit configuration for one cylinder was shown to simplify the explanation, but it is of course possible to apply the present invention to a circuit with multiple cylinders. It is easy to change the configuration.

Furthermore, the engine rotation sensor 51 and the reference position sensor 52 may be optical sensors using a signal rotor having a slit on the outer periphery and a photointerrupter that combines a light emitting diode and a phototransistor, or may be an optical sensor that uses an accelerator opening detection sensor. The potentiometer used may be a sensor whose inductance changes depending on the rotation angle, or a pulse output sensor such as a rotary encoder. In addition, in the first and second embodiments, the angle signal is directly used for the injection start timing, so the resolution is 1° CA, but the angle signal is
It can also be multiplied to 0.1°CA.
Alternatively, in order to obtain a resolution of 1° CA or less, it is also possible to control the fraction by time calculation from the engine rotation speed.

Effects of the Invention As explained above, according to the present invention, the number of injections for one combustion is made plural, and the fuel is increased in the latter stage, for example, after injecting a small amount of fuel as a pilot, a large amount of fuel is The primary fuel injection is performed, thereby increasing the combustion efficiency. In addition, by increasing the injection stop period of each injection as the engine temperature decreases, it is possible to reduce the amount of fuel during the ignition delay period, which becomes longer as the temperature decreases, thereby reducing noise caused by ignition delay at low temperatures.
Therefore, at low temperatures, it is possible to both increase combustion efficiency and prevent noise.

[Brief explanation of drawings]

Fig. 1 is a block diagram of the entire internal combustion engine to which the method of the present invention is applied, Fig. 2 is a side sectional view of the injection valve in Fig. 1, Fig. 3 is a diagram illustrating the opening time of the injection valve, Fig. 4 is a block diagram of the control device in Fig. 1, Fig. 5 is a flowchart of the CPU program in the Fig. 4 device, Fig. 6 is a signal waveform diagram of each part of the Fig. 4 device, and Fig. 7 is a diagram of the second control device. A block configuration diagram showing the control device of the embodiment, FIG. 8 is a diagram explaining the opening time of the injection valve by the device shown in FIG. 7, and FIG.
FIG. 10 is a program flowchart of the CPU, and a signal waveform diagram of each part of the device shown in FIG. 7. 1... Injection valve, 2... Electrostrictive actuator,
4, 4'...control device, 51...rotation speed sensor,
52...Reference position sensor, 53...Load sensor,
54...Temperature sensor, 56...Signal plate, 57, 58...Protrusion, 8...Diesel engine.

Claims (1)

[Claims] 1. In an internal combustion engine that injects fuel into each cylinder,
A plurality of injection amounts (τ ₁ , τ ₂ , . . . ) are injected at intervals of a predetermined injection stop period (t) for one combustion in each cylinder of the engine, and the injection stop period is the engine The internal combustion engine is characterized in that the lower the temperature, the larger the injection amount, and the predetermined ratio of the division of the total injection amount is determined such that the injection amount at an earlier injection point is smaller and the injection amount at a later injection point is larger. fuel injection method.