JPS6338547B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a control device for an automobile, and more particularly to a point control device. BACKGROUND ART In recent years, automobile exhaust gas regulations and safety regulations have become particularly strict, and increasingly large-scale electronic control devices have begun to be installed in automobiles, and these control devices are required to have complex and highly accurate control characteristics. The control characteristics include, for example, optimal fuel injection characteristics and ignition advance characteristics in response to input information such as engine speed and intake air flow rate in an electronically controlled fuel injection device. It is necessary for the control device itself to store control characteristics for each of the various input information, and in order to improve control accuracy, it is necessary to store the control characteristics in digital values using a digital memory. However, since the input/output characteristics stored as digital values are quantized and represented stepwise, a memory with a huge storage capacity is required to store the output (control) characteristics with high precision, making it difficult to effectively Memory usage is a problem. Looking at the ignition advance characteristics in particular, firstly, if a characteristic table is constructed using the engine's air flow rate and rotational speed as parameters, a situation where the rotational speed is relatively low and the air flow rate is large cannot occur under actual operating conditions. If a characteristic table is constructed using the engine's air flow rate and rotational speed as parameters, there will be approximately 40% of such situations that cannot occur in reality. Therefore, the storage capacity for this area is wasted, and since the characteristics are concentrated in other areas, the capacity in other areas is further increased in order to ensure accuracy. Second, when a characteristic table is constructed using the throttle valve opening and the engine rotational speed as parameters, the throttle valve opening has an apparently close relationship with changes in air flow rate, and there are problems similar to those of air flow rate. There is a problem in that the change characteristic is not as linear as the air flow rate and is not actually suitable as a parameter for the ignition advance characteristic. In order to perform control that adapts to the vehicle's driving conditions, it is necessary to control the fuel injection time and ignition timing in synchronization with the engine rotation, but the engine rotation speed varies from 600 to 6000 RPM. At high speeds, it is 10ms per revolution, and in the case of a 6-cylinder engine, it is necessary to ignite every 3.3ms, and ignition advance control (setting how many times before a certain point in engine rotation the ignition timing is) is required. control) etc., high-speed processing is required. Therefore, it is necessary to shorten the search time for the ignition advance characteristic table, and it is also necessary to reduce the storage capacity of the memory that constitutes the ignition advance characteristic table. In addition, when a characteristic table is configured using the engine's suction negative pressure and engine rotational speed as parameters, there are fluctuations in the suction negative pressure due to EGR operation and air cleaner clogging, resulting in accurate ignition advance values. There is a problem with the parameters that cannot be searched. The present invention has been made to solve the above-mentioned drawbacks, and its purpose is to provide an ignition timing control device that can effectively utilize memory and reduce storage capacity. The features of the present invention include: (a) air flow rate detection means for detecting information on the amount of air taken into the engine; (b) rotation speed detection means for detecting information on engine rotation speed; (c) from the air amount detection means. Load information determining means that determines load information based on the obtained air amount information and the rotation speed information obtained from the rotation speed detection means; (d) The rotation speed information and the load information are each determined at a predetermined number of dividing points. Storage means that divides the ignition data and stores the ignition data in advance at addresses determined according to the order of division points; (e) Based on the load information from the load information determining means and the rotation speed information from the rotation speed detection means; The ignition timing control device comprises ignition control means for reading ignition data determined by corresponding load information and rotation speed information from the storage means and controlling ignition timing based on the read ignition data. Next, embodiments of the present invention will be described using the drawings. FIG. 1 is a system diagram showing the main configuration of an electronic engine control device. The flow rate of the air taken in through the air cleaner 12 is measured by an air flow meter, and an output QA representing the air flow rate is input from the air flow meter 14 to the control circuit 10. The air flow meter 14 is provided with an intake temperature sensor 16 for detecting the temperature of intake air, and an output TA representing the temperature of the intake air is input to the control circuit 10. The air that has passed through the air flow meter 14 passes through the throttle chamber 18 and is drawn into the combustion chamber 34 of the engine 30 from the intake manifold 26 via the intake valve 32. The amount of air taken into the combustion chamber 34 is controlled by changing the opening degree of a throttle valve 20 provided in the throttle chamber in mechanical interlock with the accelerator pedal 22. The opening degree of the throttle valve 20 is determined by the throttle position detector 24.
0 position is detected, and a signal QTH representing the position of this throttle valve 20 is obtained.
is input from the throttle position detector 24 to the control circuit 10. The throttle chamber 18 is provided with an idle bypass passage 42 and an idle adjustment screw 44 for adjusting the amount of air passing through the bypass passage 42. When the engine is operating in an idling state, the throttle valve 20 is in a fully closed state. Intake air from the air flow meter 14 flows through the bypass passage 42;
It is sucked into the combustion chamber 34. Therefore, the amount of intake air during idling operation can be changed by adjusting the idle adjustment screw. Since the energy generated in the combustion chamber is almost determined by the amount of air from the bypass passage 42, the idle adjust screw 44
By adjusting the amount of air taken into the engine and changing the amount of air taken into the engine, it is possible to adjust the engine speed during idling to an appropriate value. The throttle chamber 18 is further provided with another bypass passage 46 and an air regulator 48. The air regulator 48 controls the amount of air passing through the passage 46 in accordance with the output signal NIDL of the control circuit 10, and controls the engine speed during warm-up operation and supplies an appropriate amount of air to the engine when the throttle valve 20 suddenly changes. I do. Additionally, the air flow rate during idling operation can be changed as needed. Next, to explain the fuel supply system, the fuel stored in the fuel tank 50 is sucked into the fuel pump 52, and the fuel damper 54
is pumped to. The fuel damper 54 absorbs pressure pulsations in the fuel from the fuel pump 52 and sends fuel at a predetermined pressure to the fuel pressure regulator 62 via the fuel filter 56. Fuel from the fuel pressure regulator is fed under pressure to a fuel injector 66 via a fuel pipe 60, and the fuel injector 66 opens in response to the output INJ from the control circuit 10 to inject fuel. The amount of fuel injected from the fuel injector 66 is determined by the valve opening time of the injector 66 and the pressure difference between the pressure of the fuel fed to the injector and the pressure of the intake manifold 26 into which the fuel is injected. However, it is desirable that the fuel injected from the fuel injector 66 depend only on the valve opening time determined by the signal from the control circuit 10. Therefore, the pressure of the fuel fed to the fuel injector 66 is controlled by the fuel pressure regulator 62 so that the difference between the fuel pressure to the fuel injector 66 and the manifold pressure of the intake manifold 26 is always constant. Intake manifold pressure is applied to the fuel pressure regulator 62 via a pressure guide pipe 64, and when the fuel pressure in the fuel pipe 60 exceeds a certain level with respect to this pressure, the fuel pipe 60 and the fuel return pipe 58 are brought into communication with each other, and the excess pressure is removed. The fuel corresponding to the amount is returned to the fuel tank 50 via the fuel return pipe 58. In this way, the difference between the fuel pressure in the fuel pipe 60 and the manifold pressure in the intake manifold is always kept constant. The fuel tank 50 is further provided with a pipe 68 and a canister 70 for absorbing vaporized fuel gas, and an atmosphere opening 74 is provided during engine operation.
Air is sucked in from the intake manifold, and the absorbed vaporized fuel gas is guided to the intake manifold through a pipe 72 and then to the engine 30. As explained above, fuel is injected from the fuel injector, the intake valve 32 opens in synchronization with the movement of the piston 74, and a mixture of air and fuel is introduced into the combustion chamber 34. This air-fuel mixture is compressed and combusted by the spark energy from the spark plug 36, thereby converting the combustion energy of the air-fuel mixture into kinetic energy that moves the piston. The combusted air-fuel mixture is exhausted as exhaust gas from an exhaust valve (not shown) to the atmosphere via an exhaust pipe 76, a catalytic converter 82, and a muffler 86. exhaust pipe 76
is the exhaust gas recirculation pipe 78 (hereinafter referred to as EGR pipe).
A portion of the exhaust gas is guided to the intake manifold 26 through this pipe. That is, part of the exhaust gas is recirculated to the intake side of the engine. This recirculation amount is determined by the valve opening amount of the exhaust gas recirculation device 28. This valve opening amount is the output of the control circuit 10.
Controlled by EGR and further exhaust gas recirculation device 28
The valve position is converted into an electrical signal and input to the control circuit 10 as a signal QE. A λ sensor 80 is provided in the exhaust pipe 76 and detects the mixture ratio of the air-fuel mixture sucked into the combustion chamber 34. Specifically O ₂ sensor (oxygen sensor)
is generally used to detect the oxygen concentration in the exhaust gas and generate a voltage Vλ according to the oxygen concentration. The output Vλ of the λ sensor 80 is input to the control circuit 10.
The catalytic converter 82 is provided with an exhaust temperature sensor 84, and an output TE corresponding to the exhaust temperature is input to the control circuit 10. The control circuit 10 is provided with a negative power terminal 88 and a positive power terminal 90. Further, the control circuit 10 applies the above-described signal IGN to the primary coil of the ignition coil 40 to control the spark generation of the spark plug 36, and the high voltage generated in the secondary coil is transferred to the power distributor 38.
is applied to the spark plug 36 via the combustion chamber 34.
generates a spark for combustion within. More specifically, the ignition coil 40 has a positive power terminal 92.
The control circuit 10 is further provided with a power transistor for controlling the primary coil current of the ignition coil 40. Positive power terminal 92 of ignition coil 40 and negative power terminal 88 of control circuit 10
A series circuit is formed between the primary coil of the ignition coil 40 and the power transistor, and when the power transistor becomes conductive, electromagnetic energy is accumulated in the ignition coil 40, and when the power transistor is cut off, electromagnetic energy is accumulated in the ignition coil 40. The above-mentioned electromagnetic energy is energy with high voltage that is generated by the spark plug 3.
6. The engine 30 is provided with a water temperature sensor 96,
The temperature of engine cooling water 94 is detected, and a signal TW corresponding to this temperature is input to control circuit 10. Further, the engine 30 is provided with an angle sensor 98 that detects the rotational position of the engine, and this sensor 98 generates a reference signal PR every 120 degrees, for example, in synchronization with the rotation of the engine. degree) Angle signal every time it rotates
Generate PC. These signals are input to the control circuit 10. 100 indicated by a dotted line in the figure is a negative pressure sensor,
Voltage according to negative pressure of intake manifold 26
Input VD to the control circuit 10. Specifically, a semiconductor negative pressure sensor can be considered as the negative pressure sensor 10. Apply the boost pressure of the intake manifold to one side of the silicon chip,
Atmospheric pressure or constant pressure is applied to the other side. In some cases, a vacuum may be used. With this structure, a voltage VD corresponding to the manifold pressure is generated due to the piezoresistance effect, etc., and the control circuit 1
Applied to 0. FIG. 2 is an operational diagram illustrating the ignition timing and fuel injection timing with respect to the crank angle of a six-cylinder engine. A represents the crank angle, and a reference signal PR is output from the angle sensor 98 every 120 degrees of crank angle. In other words, 0° of crank angle,
The reality signal PR is input to the control circuit 10 at every 120°, 240°, 360°, 480°, 600°, and 720°. In the figure, the first cylinders are
5th cylinder, 3rd cylinder, 6th cylinder, 2nd cylinder, 4th cylinder
Represents the operation of the cylinder. Further, J1 to J6 represent the opening positions of the intake valves of each cylinder. As shown in Fig. 2, the valve opening positions of each cylinder are shifted by 120° in terms of crank angle. Although the valve opening position and valve opening width differ somewhat depending on the structure of each engine, they are approximately as shown in the figure. In the diagram, A1 to A5 are fuel injectors 66
This represents the valve opening timing, that is, the fuel injection timing. The time width JD of each injection timing A1 to A5 represents the valve opening time of the fuel injector 66. This time range
JD can be considered to represent the fuel injection amount of the fuel injector 66. A fuel injector 66 is provided corresponding to each cylinder, and these injectors are each connected in parallel to a drive circuit within the control circuit 10. Therefore, in response to the signal INJ from the control circuit 10, the fuel injectors corresponding to each cylinder open simultaneously and inject fuel. The first cylinder shown in FIG. 2B will be explained. In synchronization with the reference signal INTIS generated at a crank angle of 360°, an output signal INJ is sent from the control circuit 10 to the fuel injector 66 provided in the manifold or intake port of each cylinder.
is applied to As a result, fuel is injected for the time JD calculated by the control circuit 10 as shown by A2. However, since the intake valve of the first cylinder is closed, the injected fuel is held near the intake port of the first cylinder and is not sucked into the cylinder. Next, a signal is again sent from the control circuit to each fuel injector 66 in response to the reference signal INTIS generated at a crank angle of 720 degrees, and fuel injection indicated by A3 is performed.
Almost simultaneously with this injection, the intake valve of the first cylinder opens, and with this opening, both the fuel injected at A2 and the fuel injected at A3 are sucked into the combustion chamber. The same can be said for other cylinders. In other words, in the fifth cylinder shown in C, A2 and A are at the intake valve opening position J5.
The fuel injected in step 3 is inhaled. 3 shown in d.
In the cylinder, at the intake valve opening position J3, part of the fuel injected at A2, the fuel injected at A3, and further A
A portion of the fuel injected at step 4 is inhaled. The sum of the part of the fuel injected at A2 and the part of the fuel injected at A4 becomes the injection amount for one injection. Therefore, each intake stroke of the third cylinder also receives two injection amounts. Similarly, in the sixth cylinder, second cylinder, and fourth cylinder shown in E, H, and G, two injections from the fuel injector are inhaled in one intake stroke. As can be seen from the above explanation, the fuel injection amount specified by the fuel injection signal INJ from the control circuit 10 is half of the fuel required to generate fuel, and two injections from the fuel injector 66 cause the combustion chamber to
4, the required amount of fuel corresponding to the air taken in is obtained. In FIG. 2, G1 to G6 indicate ignition timings corresponding to the first to sixth cylinders. By cutting off the power transistor provided in the control circuit 10, the primary coil current of the ignition coil 40 is cut off.
Next, generate a high voltage in the coil. This high voltage is generated at the ignition timings G1, G5, G6, G2, and G4, and is distributed by the power distributor 38 to the spark plugs provided in each cylinder. As a result, each spark plug is ignited in the order of the first cylinder, the fifth cylinder, the third cylinder, the sixth cylinder, the second cylinder, and the fourth cylinder, and the mixture of fuel and air is combusted. The detailed circuit configuration of the control circuit 10 in FIG.
As shown in the figure. A positive power supply terminal 90 of the control circuit 10 is connected to a positive terminal 110 of the battery, and a voltage VB is supplied to the control circuit 10. The power supply voltage VB is set to a constant voltage PVCC, for example, 5 [V] by the constant voltage circuit 112.
is held constant. This constant voltage PVCC is supplied to a central processor (hereinafter referred to as CPU), random access memory (hereinafter referred to as RAM), and read-only memory (hereinafter referred to as ROM).
Furthermore, the output PVCC of the constant voltage circuit 112 is also input to the output circuit 120. The input/output circuit 120 includes a multiplexer 122, an analog/digital converter 124, a pulse output circuit 126, a pulse input circuit 128, a discrete input/output circuit 130, and the like. Analog signals are input to the multiplexer 122, and one of the input signals is input based on a command from the CPU.
is selected and input to the analog-to-digital converter 124. As analog input signals, each sensor shown in FIG. , an analog signal TW representing the engine cooling water temperature, an analog signal TA representing the intake air temperature, an analog signal TE representing the exhaust gas temperature, an analog signal QTH representing the throttle opening, an analog signal QE representing the valve opening state of the exhaust gas recirculation device. , an analog signal Vλ representing the excess air ratio of the intake mixture, and an analog signal QA representing the intake air amount are sent to the filters 132 to 14.
4 to multiplexer 122.
However, the output Vλ of the λ sensor 80 is input to the multiplexer via an amplifier 142 having a filter circuit. In addition, an analog signal VPA representing atmospheric pressure is input from the atmospheric pressure sensor 146 to the multiplexer. Also, from the positive power supply terminal 90° to the resistor 150,1
The voltage VB is connected to the resistor 160 in a series circuit of 52,154
Further, the terminal voltage of the series circuit of the resistors is kept constant by a zener 148. The values of voltages VH and VL at junctions 156 and 158 between resistors 150 and 152 and resistors 152 and 154 are input to multiplexer 122. The CPU 114, RAM 116, and ROM mentioned above
118 and the input/output circuit 120 are connected by a data bus 162, an address bus 164, and a control bus 166, respectively. Furthermore, from the CPU
A clock signal E is applied to each of the PAM, ROM, and input/output circuit 120, and data is transmitted via the data bus 162 in synchronization with this clock signal E. The multiplexer 122 of the input/output circuit 120 has water temperature TW, intake air amount TA, exhaust gas temperature TE, throttle opening QTH, exhaust recirculation amount QE, λ sensor output Vλ, atmospheric pressure VPA, intake air amount QA, and reference voltage.
Negative pressure VD is input instead of VH, VL, and intake air amount QA. These inputs are in ROM11
Based on the instruction program stored in 8.
The address of the CPU 114 is specified via the address bus, and the analog input of the specified address is taken in. This analog input is routed from multiplexer 122 to analog-to-digital converter 124.
The digitally converted values are held in registers corresponding to the respective inputs, and are taken into the CPU 114 or RAM 116 as required based on instructions from the CPU 114 sent via the control bus 166. A reference pulse PR and an angle signal PC are input from the angle sensor 98 to the pulse input circuit 128 in the form of a pulse train via a filter 168.
Furthermore, a pulse PS having a frequency corresponding to the vehicle speed is input from the vehicle speed sensor 170 to the pulse input circuit 128 via the filter 172 in the form of a pulse train. The signal processed by CPU 114 is held in pulse output circuit 126. Pulse output circuit 12
The output from 6 is applied to a power amplifier circuit 188, and the fuel injector is controlled based on this signal. Reference numerals 188, 194, and 198 are power amplification circuits that control the primary coil current of the ignition coil 40, the opening degree of the exhaust gas recirculation device 28, and the air regulator 48, respectively.
The opening degree of the pulse output circuit 126 is controlled according to the output pulse from the pulse output circuit 126. Discrete input/output circuit 1
30 receives signals from a switch 174 that detects that the throttle valve 20 is in a fully closed state, a starter switch 176, and a gear switch 178 that indicates that the tranmission gear is a top gear through filters 180, 182, and 184, respectively. received and retained. Furthermore, it holds processing signals from the central processor CPU 114. The signals related to the discrete input/output circuit 130 are signals whose contents can be displayed with one bit. Next, the power amplifier circuits 196, 200, 20
A signal is sent from the discrete input/output circuit to 2 and 204, respectively, to close the exhaust gas recirculation device 28 and stop the recirculation of exhaust gas, control the fuel pump, display abnormal temperature of the catalyst, Indicates engine overheating. FIG. 4 shows a specific circuit of the pulse output circuit 126. A register group 470 is the reference register group mentioned above, and is used to hold data processed by the CPU 114 or to store data at a predetermined constant level. Holds data that indicates a value. This data is
It is sent from the CPU 114 via the data bus 162. The register to hold is specified using address bus 1.
64, and the above data is input to the designated register and held. The register group 472 is a momentary register group and holds the instantaneous state of the engine and the like. Instantaneous register group 472, latch circuit 476, and incrementer 4
78 exhibits a so-called counter function. The output register group 474 includes, for example, a register 430 that holds the engine rotation speed and a register 432 that holds the vehicle speed. These values are obtained by reading the values of instantaneous registers when certain conditions are met. Output register group 47
The data held in CPU 162 is sent to the relevant register by a signal sent from the CPU via the address bus, and sent from this register to the CPU 114 via the data bus 162. Comparator 480 receives reference data from a selected register of the group of reference registers and instantaneous data from a selected register of the group of instantaneous registers at inputs 482 and 484, respectively, and performs a comparison operation. The comparison result is output from the output terminal 486. The output terminal is set in a predetermined register in the first comparison output register group 502 which functions as a comparison result holding circuit. Furthermore, it is then set in a predetermined register of the second comparison output register group 504. Reference register group 470, instantaneous register group 47
2. Read and write operations of the output register group 474, incrementer 478 and comparator 480
The operations of setting the output to the first comparison output register 502 and the second comparison output register 504 are processed within a certain predetermined time. Further, various processes are performed in a time-sharing manner according to the stage order of the stage counter 572. For each stage, a predetermined register in each of the reference register group 470, instantaneous register group 472, first and second comparison result register groups, and, if necessary, a predetermined register in the output register group 474 is selected. . Also, incrementer 478 and comparator 4
80 is commonly used. FIG. 5 is a diagram for explaining the timing of FIG. 4. A clock signal E is supplied from the CPU 114 to the input/output circuit 120. This signal is shown in A. From this clock signal E, a circuit 574 generates two non-overlapping clock signals φ1 and φ2. This signal is shown in (b) and (c). The circuit shown in FIG. 4 operates according to these clock signals φ1 and φ2. FIG. 5D is a stage signal, which is switched at the rising edge of clock signal φ2, and processing of each stage is performed in synchronization with φ2. In Figure 5
THROUGH indicates that the latch circuit or register circuit is enabled, and indicates that the output of these circuits is dependent on the input. Also
LATCH indicates that these circuits hold certain data and the output of this circuit does not depend on the input. The stage signal shown in (d) becomes a readout signal for the reference register 470 and instantaneous register 472, and the contents are read out from a certain selected predetermined register. E and F show the operation of reference register 470 and instantaneous register 472, respectively. This operation is performed in synchronization with the clock φ. The operation of latch circuit 476 is shown in FIG. This circuit enters the THROUGH state when φ2 is at a high level, writes data in a specific register read from the instantaneous register group 472, and clocks φ2.
When 2 becomes low level, it becomes LATCH state. In this way, the data of a predetermined register in the instantaneous register group corresponding to that stage is held. The data held in the latch circuit 476 is
Incrementer 478 not synchronized to clock signal
modified based on external conditions. Here, the incrementer 478 has the following functions based on the signal from the incrementer controller 490. The first function is an increment function that increases the value indicated by the input data by one. Second
The function is a non-increment function, which allows the input to pass through without increasing it. The third function is a reset function, which changes all inputs to data indicating a value of 0. Looking at the data flow of the instantaneous registers, one register in the instantaneous register group 472 is selected by the stage counter 572, and its held data is input to the comparator 480 via the latch circuit 476 and the incrementer 478. Additionally, a closed loop is created from the output of incrementer 478 back to the originally selected register. Therefore, when the incrementer functions to increase data by one, this closed loop functions as a counter. However, in this closed loop, if a situation occurs in which the data of the instantaneous register group is output from a specific selected register while the data is looped around and input, malfunction occurs. Therefore, a latch circuit 476 is provided to cut off the data. The latch circuit 476 enters the THROUGH state in synchronization with clock φ2, while the THROUGH state in which the input is written to the instantaneous register is synchronized with clock φ1. Therefore, data is cut between clocks φ2 and φ1. In other words, even if the value of a specific register in the register 472 is changed, the latch circuit 47
The output of 6 remains unchanged. Comparator 480, like incrementer 476, also operates out of synchronization with the clock signal. The input of the comparator 480 is transmitted through a latch circuit and an incrementer for the data held in one reference register selected from the reference register group 470 and the data held in one register selected from the instantaneous register group. received data. The result of this data comparison is synchronized with clock signal φ1.
It is set to the first comparison result register group 502 which enters the THROUGH state. Furthermore, this data is set in the second comparison result register group 504 which enters the THROUGH state at clock φ2. The output of this register 504 becomes a signal for controlling each function of the incrementer and a drive signal for the fuel injector, ignition coil, exhaust gas recirculation device, etc. Also, based on this signal, the measurement results of the engine rotational speed and vehicle speed at each stage are written from the instantaneous register group to the output register group 474.
Now, for example, when writing the engine rotation speed, the second signal indicating that a certain period of time has elapsed is
It is held in the comparison result register RPMWBF552,
At the RPM stage in Table 1, which will be described later, data held in the instantaneous register 462 is input to the register 430 of the output register group based on the output of this register 552. At this time, the second comparison result register
If the signal indicating that a certain period of time has elapsed is not held in the RPMWBF 552, the operation of inputting the data held in the register 462 to the register 430 is not performed even in the RPM stage. On the other hand, based on the signal held in the second comparison result register VSPWBF 556, data in the instantaneous register 468 is inputted to the output register 432 as data representing the vehicle speed at the timing of the stage VSP. Writing of data representing the engine rotational speed RPM and vehicle speed VSP to the output register group 474 is performed as follows. In Figure 5,
The stage signal STG is RPM or VSP, and the data in the instantaneous register 462 or 468 is at the high level of the clock φ2, and the latch circuit 476 is activated.
It becomes THROUGH state and is written, and the clock φ
2 becomes low level, the above data becomes
LATCHed. The data held in this way is stored in the register RPMWBF552 or
Based on the signal from the VSPWBF556, the output register group 474 enters the THROUGH state as shown in FIG.
be done. When the CPU 114 reads data held in the output register group 474, the CPU 114 specifies the register via the address bus 164, and the data is taken in in synchronization with the clock signal E shown in FIG. 5A. FIG. 6 shows a generation circuit for the stage signal STG.
Stage counter with signal φ1 from circuit 574
SC570 is counted up, and the outputs C0 to C6 of the stage counter SC570 and the output of the T register in FIG. 4 are input to the stage decoder.
Added to SDC. The stage decoder SDC outputs signals 01 to 017 to the stage latch circuit.
Write to STGL in synchronization with clock φ2. The stage latch STGL reset input has a fourth
When the ^20- bit signal GO of the MODE register shown in the figure is input and the ^20- bit GO signal of the MODE register becomes low level, all outputs of STGL become low level and all processing operations are stopped. On the other hand, when the GO signal becomes high level, the stage signal STG is outputted again in a fixed order, and processing is performed based on it. The above stage decoder SDC is READ, ONLY,
This can be easily achieved by using MEMORY etc. The detailed contents of stages 00 to 6F of the stage signal STG, which is the output of the stage latch STGL, are shown in Table 1.

【table】

[Table] First, the general reset signal GR is input to the reset terminal of the stage counter SC570 shown in FIG. 6, so that the counter outputs C0 to C6 all become 0. This general reset signal is sent from the CPU when this control circuit is activated. When clock signal φ2 is input in this state, at the rising edge of φ2,
EGRP stage signal STG is output. EGRP processing is performed based on this stage signal. Next, the stage counter SC570 counts up by one at clock φ1, and INTL of the next stage signal STG is output at clock φ2. INTL processing is performed based on this stage signal INTLSTG. Furthermore, next, a stage signal CYLSTG is output and CYL processing is performed, and next, a stage signal ADV is output and ADV processing is performed. In this way, the stage counter SC570
continues to count up in synchronization with φ1, φ
A stage signal STG is output in synchronization with 2, and processing is performed in accordance with this signal. When C0 to C6 of stage counter SC570 all become 1, stage signal INJSTG is output.
The INJ process is performed and all the processes in Table 1 are completed. With the next clock signal φ1, all C0 to C6 of the stage counter SC570 become 0, and with the clock signal φ2, the stage signal EGRPSTG is output, and STG processing is performed. In this way, the processing in Table 1 is repeated. Table 2 shows the processing contents of each stage shown in Table 1.

【table】

[Table] Output from the stage latch circuit STGL in Figure 6
The STG0 and STG7 signals are circuits for synchronizing the input input from the outside with the internal clock signal of the input/output circuit 120.
Output STG7 is the stage counter
It is output when C0 to C2 of SC570 are all 1. External signals include, for example, a reference signal PR generated in synchronization with engine rotation, an angle signal PC, and a vehicle speed pulse PS generated in synchronization with wheel rotation. The periods of these pulses change greatly, and as they are, they are not synchronized with the clock signals φ1 and φ2. Therefore, it is not possible to determine whether to increment at the ADVSTG stage, VSPSTG stage, or RPMSTG stage in Table 1. Therefore, it is necessary to synchronize external pulses, such as pulses from a sensor, with the stages of the input/output circuit. Moreover, in order to improve the detection accuracy, the angle signal PC and the vehicle speed signal PS need to be synchronized with the stage with respect to the rising and rising edges of their input pulses. Reference signal PR
For this, it is sufficient to synchronize it with the rise. Output of stage latch circuit STGL in Figure 6
Using STG0 and STG7, the above synchronized signal is generated at φ2 timing. The circuit is shown in FIG. Further, the operation timing is shown in FIG. External input pulses such as sensor output, such as reference pulse PR, angle signal PC, vehicle speed signal
PS is latched by the STG0 output shown in FIG. 6 in latch circuits 600, 602, and 604 shown in FIG. 7, respectively. In Figure 8, A is the clock signal φ2, B is the clock signal φ1, and C and D are the stage signals STG7 and STG.
It is 0. This stage signal is generated in synchronization with φ2, as explained in FIG. The signal shown in E is the output pulse from the angle sensor or vehicle speed sensor and is the reference pulse PR or angle pulse PC.
Alternatively, it indicates the vehicle speed pulse PS, and the generation timing, pulse duty, and period of this signal are irregular, and are inputted regardless of the stage signal. Now, a signal as shown in FIG.
Assuming that the pulses are input to 00, 602, and 604, they are latched by the stage signal STG0 (pulse shown in the figure). Therefore, as shown in FIG. 8, the level becomes high at the time point. Furthermore, since the input signals RP, PC, and PS are at high level in the stage signal STG0 indicated by wo, the latch circuits 600, 602, and 6
The high level is latched at 04 respectively. However, in the stage signal STG0 shown by w, the input signal
Since PR, PC, and PS are at low level, the low level is latched. Therefore, the latch circuit 60
The outputs A1, A2, and A3 of 0, 602, and 604 are as shown in F. Latch circuits 606, 608,
610 latches the outputs A1, A2, and A3 respectively with the power of the stage signal STG7, so that the outputs rise from the point indicated by y. Furthermore, the stage signal STG7 also latches at a high level, so it continues to be at a high level. Therefore, the latch circuits 606, 608, 6
The ten output signals B1, B2, and B3 are as shown in FIG. The signal A1 and the signal B1 sent through the inverter 608 are input to the NOR circuit 612, and a synchronized reference signal PRS is generated as shown in FIG. This synchronized reference signal PRS
captures the rising edge of the reference signal PR and becomes the pulse width of the stage signals STG0 to STG7. EXCLUSIVELY OR circuits 614 and 616 receive signals A2 and B2 and signals A3 and B3, respectively, and the rising edge of the signals PC and PV generates the signal shown in (i), and the rising edge of the signals PC and PV generates the signal (g). do. The duty of signals R and G is the same as the duty shown in C, and the duty of the stage signal STG
Determined by 0 and STG7. In the above explanation, it was determined that the signals PR, PC, and PS were input at the same time with the same duty, but in reality, these signals are not input at the same time and their duties are different. Furthermore, even when looking at the same signal itself, its period and duty differ each time. However, due to FIG. 7 and the synchronization circuit, the pulse has a constant width. This pulse width is the stage signal STG
It is determined by the time difference between STG 0 and STG7. Therefore, latch circuits 600, 602, 604 and 606, 608, 6
By changing the stage signal applied to 10, the pulse width can be adjusted and changed. This pulse width is determined in relation to the timing of the stages in Table 1. In other words, as shown in Table 1, the INTL stage has a stage counter (C0~
C2, C3 to C6) are allocated in the state of (1,0), and further (1,1), (1,2), (1,3)
...and are assigned to each 8th stage. Since each stage is set to one microsec, an INTL stage is assigned every eight microsecs. In the INTL stage, the angle signal
Since it is necessary to detect the PC and control the incrementer, when the output PC of the angle sensor 98 is applied to the synchronization circuit shown in FIG. Make a pulse and based on this synchronized pulse PCS
Control the incrementer controller in the INTL stage. This synchronized angle signal PCS is also detected in stages ADV and RPM. This stage ADV
The RPM is assigned each time the values of C3 to C6 count up by one when the stage counters C0 to C2 are at 3 and 6, respectively. The assigned stage rotates in a cycle of 8 microsecs. The STG0 signal in Figure 7 is the C of the stage counter.
Output when the value of 0 to C2 is 0, while STG7
is output when C0 to C2 have a value of 7. This output is produced independently of C3-C6. Therefore, the eighth
As can be seen from the figure, the synchronized angle signal PCS always has a varying pulse width from the value of 0 to the value of 6 for the stage counter outputs C0 to C2, and this pulse is detected by the stages INTL, ADV, and RPM, and the increment controller control. As above, the CYL stage that detects the synchronization reference PRS uses the stage counter output C0~
When the reference pulse PR is input from the angle sensor 98, which is always assigned when the value of C2 is 2, the synchronized reference is always assigned to this input when the stage counter C0 to C2 is 2.
It is necessary for PRS to be issued. The circuit in Figure 7 is
Since the pulse width is between STG0 and STG7, this information is fully satisfied. Next, the VSP stage for detecting the wheel speed is assigned when the stage counter outputs C0 to C2 always have a value of 5. Therefore, it is sufficient that the synchronized PSS signal is output when the value of C0 to C2 is 5. In the circuit shown in FIG. 7, the values of C0 to C2 range from 0 to 6, so this value is satisfied. In Figure 7, instead of the STG0 signal, create a signal STG4 that always appears when the value of C0 to C2 is 4, use this signal, and use this signal instead of the signal of STG7.
A signal STG6 that is always output when the value of is 6 may be used. In this case, when the signal PS is input, the synchronization signal PSS is the output C0 of the stage counter.
It will always be output when the value of ~C2 is 4 and 5. The stage cycle will now be explained.
In Table 1, stage counter outputs C0 to C6
128 types of stage signals with values from 0 to 127 are created, and when all of these signals have been generated, the large cycle is completed and a new large cycle begins again. This large cycle is further composed of 16 small cycles, and this small cycle is composed of 8 types of stage signals. This small cycle corresponds to the values of stage counter outputs C0 to C2 from 0 to 7, respectively, and is completed in 8 microsecs. The pulse outputs PR, PC, and PS from the sensor are reliably synchronized, and the synchronized pulses PRS, PCS,
In order to reliably generate PSS, it is necessary that the output from the sensor has a pulse width longer than this short cycle. For example, the duty of the angular pulse becomes narrower as the engine rotates faster. For example, at 9000 rpm, it will be about 9 microsecs. Therefore, in order to achieve sufficient synchronization with respect to 9000 revolutions per minute, it is necessary to make this small cycle shorter than this, and in this embodiment it is 8 microsecs. Next, the operation of incrementer 478 shown in FIG. 4 will be explained. A detailed circuit of incrementer 478 is shown in FIG. As mentioned above, this incrementer has three functions: the first function is to increase the input data by a value of 1, the second function is to reset the input data, and the third function is to increase the input data by a value of 1. The function is a function that outputs input data as is. The increment function is performed by the ICNT signal, and the reset function is performed by the IRST signal.
When ICNT signal is high level, increment function, when low level, non-increment function,
When the IRST signal is high level, it becomes a reset function and the IRST signal has priority over the ICNT signal. The conditions may be selected based on the stage signals commanded by each process. The conditions include synchronized external input and the register group 504 of the second comparison result.
This is the output of Further, the conditions for transferring and writing data to the output register 474 are also similar to the conditions for the incrementer. FIG. 10 is a diagram explaining the processing of the fuel injection signal INJ. Since the start of injection differs depending on the number of cylinders, the register 442 acting as a CYL COUNTER counts the initial angular pulse INTLD generated from the reference signal PRS,
The result is held as a value related to the number of cylinders.
When compared with the CYL register 404 and when they are greater than or equal, the first register group 502 is
CYLFF 506 is set to 1, and CYLBF 508 of the second register group 504 is set to 1. With this CYL BF=1, CYL COUNTER44
2 is reset. Also, when CYL BF=1, the INJ TIMER 450 that measures the injection time is reset. Always unconditionally incremented by time, the injection time is set.
Compare with INJD register 412, and if greater or equal, INJ FF522 of the first register group
is set to 1. Also, the second register group
INJ BF524 is set to 1. This INJBF
When =1, increment by time is prohibited. This reversed output of INJ BF becomes the fuel injection time width, which becomes the valve opening time of the fuel injector. FIG. 11 is a diagram illustrating processing of signals that control ignition. By the initial angular pulse INTLD,
Register 45 acts as ADV COUNTER
2 and is incremented by the synchronized angle pulse PC being high. Then, compare it with the ADV register 414 that holds the ignition angle from INTLD, and if it is greater or equal, the first register 502
ADV FF 526 is set to 1, and ADV BF 528 of the second register 504 is set to 1. ADVD indicating the rise of ADV BF resets the DWL COUNTER 454, which starts energization, and is incremented when the synchronized angle pulse PC is at a high level. Then, it is compared with the DWL register 416 that holds the angle at which energization starts from the previous ignition position, and if the angle is greater or equal, the first register 502 is set.
The DWL FF 530 is set to 1, and the DWL BF 532 of the second register 504 is set to 1. The output of this DWL BF532 becomes the ignition control signal ING1. FIG. 12 is a diagram illustrating the processing of EGR (NIDL). Since these are all proportional solenoids, they perform duty control. maintain the cycle
Holds EGRP register 418 and on time
There are two EGRD registers 420, and
The TIMER is measured by EGR TIMER456. In processing, when processing the EGRP STG, the data held in the EGRP register 418 and the EGR TIMER 456 are compared, and if they are greater or equal, 1 is set in the EGRP FF 534 of the first register group 502. Set. Furthermore, the second register group 504
EGRP BF 536 is set to 1. When processing EGRD STG, unconditional non-increment, or EGR with EGRP BF=1.
TIMER 456 is reset. EGRDFF5
38 is the EGRD register 420 and EGR TIMER
456 and the results are greater than or equal, EGRD BF540 is set to 1.
is set to The inverted output of this EGRD BF540 is the EGR control signal. The operation is similar to NIDL. Figure 13 shows the engine speed RPM (and vehicle speed).
FIG. The measurement method uses a certain measurement time width as RPMW.
It is determined by the TIMER 460 and obtained by counting the synchronized angle pulses PC within that time width. RPMW TIMER460, which measures time width,
is unconditionally incremented and also RPMW
It is reset when BF552=1. RPMW
FF550 is set to 1 because the RPMW register 426 that holds the time width and the RPMW
This is when the TIMER 460 is compared and the result is greater than or equal. RPM COUNTER462 that counted the PC by RPMWD indicating the rise of RPMW BF552
The contents of are transferred to the RPM register 430 of the output register 474 and written. Also, RPMW BF
When 552=1, RPM COUNTER462
will be reset. VSP STG processing is also similar to RPM. Table 3 shows the functions of each register.

【table】

【table】

【table】

【table】

[Table] Next, a method for setting reference data in the reference register 470 will be explained. Register 402,4
04, 406, and 410 are set when the device of this embodiment is started. Once set, these values do not change. Next, data set in register 408 is performed by program processing. The register 412 contains the fuel injector 6.
Data INJD representing the valve opening time of No. 6 is input. This data INJD is determined, for example, as follows. Output signal QA of air flow meter 14
is input to an analog-to-digital converter 124 via a multiplexer 122. Here, it is converted into digital data and held in a register (not shown). The load data TP is obtained by calculation processing from the data representing the intake air amount and the engine rotational speed data N held in the register 430 in FIG. 4, or by searching the information stored in the map state. Furthermore, the outputs of the intake air temperature sensor 16, water temperature sensor 96, and atmospheric pressure sensor 146 are converted into digital data, and correction is performed based on this data and the operating state of the engine. This correction factor is K1
shall be. Furthermore, the battery voltage is also digitized, and correction is performed according to this data. Let this correction coefficient be TS. Correction is then performed by the λ sensor 80. Let this correction coefficient be α.
In other words, data INJD is expressed as follows. The valve opening time of the fuel injector is determined as follows: INJD=α(K1.TP+TS). However, the method shown here is just one example, and it is of course possible to define it using other methods. The register 414 contains data representing the ignition timing.
ADV is set. This data ADV is created as follows. The ignition data θIG in the map state with the above load data TP and rotation speed as factors.
It is stored in the ROM 118 and determined from this map. Furthermore, start correction, water temperature correction, acceleration correction, etc. are added to this θIG. In this way, the ignition timing data ADV is created. The register 416 contains data DWL as data for controlling the primary current charging time of the ignition coil.
is set. This data DWL is the above data
It is calculated from the ADV value and the digital value of battery voltage. Next, an interpolation method for determining the ignition timing, which is a control amount corresponding to the input information, based on the input information will be described. Figure 14 shows the ignition advance angle in the Z-axis direction (represented as contour lines in the figure) when the x-axis is the engine speed data N and the y-axis is the load data TP representing the basic fuel injection time.
It was plotted on. As shown in this figure, the ignition advance angle, which is a control characteristic, is determined by the engine rotational speed N
The contour lines are steep because the load TP is low and changes rapidly around medium load TP. On the other hand, when the engine speed N is medium or higher and the load TP is large, the change is small and the contour lines are gentle. Uniform accuracy can be maintained by dividing the x-axis and y-axis according to the slope of the contour lines. Here, a case is shown in which the x-axis and y-axis are divided into 11 parts with the minimum interval being 2 to the 4th power. Now, a case will be shown in which the ignition advance angle Z is determined when the engine rotational speed N and the load TP are obtained as input information. The digital value of the value of each dividing point shown in FIG. 14 and the value of the ignition advance angle at the intersection of the dividing points on the x-axis and y-axis is expressed as data as shown in FIG. 15.
It is stored in ROM118. Figure 15a shows the value of the dividing point on the x-axis, Figure 15b shows the value of the dividing point on the y-axis, and Figure 15e shows the control output at the intersection of the dividing points shown in a and b above. It means the value of the characteristic (ignition advance angle). Of the data a, b, and e, only the upper 5 bits are stored in the ROM 118, excluding the lower ⁴ bits (the shaded area in the figure), which means the minimum interval of 24 bits between division points. 1st
Figure 5c and d are the input information for x shown in Figure 14,
Shows the values of x and y when y is input.
Here, all input information and output control characteristics are expressed in 9 bits. A method for determining control characteristic values corresponding to input values will be explained according to the processing flow shown in FIG. first,
Set the engine speed N to the RPM register at 4 in Figure 4.
30, and the load state TP is determined from the air flow rate QA given by the air flow sensor 14 from the value obtained via the analog-to-digital converter and the engine rotational speed N. The engine rotational speed N obtained in this manner is used as an input value x, and the load state TP determined by the calculation is temporarily stored as y in a work area in the RAM. (Step 800 in FIG. 16) Next, using the digital values x and y obtained as described above (FIG. 15 c and d), find a dividing point that sandwiches x and y. Figure 17 shows how to find the target.To explain specifically using the values shown in Figure 15, first divide the input value x [001111010] by the fourth power of the minimum dividing point interval 2. In other words, The digits are shifted 4 bits to the right by means of digit shifting. This makes x become [0111]. At step 850 in Fig. 17, this input value x [0111] is converted to the minimum dividing point x0 [00000] and the maximum dividing point [0111]. 10011].This ROM1
As explained in FIG. 3, access to the data xi in
164 and 166. Hereinafter, the data of the division points will be handled in the same way. In the above check, since x0≦x≦x11 is satisfied, first, in step 854, i=(11+1)/2
Determine the initial value of i which becomes =6. Based on this, the initial value of h is also determined. Then, in step 860,
Compare with the value of x6 [00110]. As a result, x＞x6
Since the following relationship holds true, proceed to step 862 i
=6+3=9 (h is set to h=3 by steps 856 and 858), step 8 from h=3
The determination of 64 is NO. Next, as above, x and
Compare with x9 [01011]. In this case, the relationship x<x9 holds, so proceed to step 870 and i=
9-1=8 (h is set to h=1 in step 858), and x and x are determined in step 872.
Compare with x8 [01001] in step 860. In this case as well, since x<x8, proceed to step 870 and i=
8-0=8 (h is set to h=0 in step 858). Since the determination h=0 in the next step 872 holds true, the process advances to step 874.
In this case, since x is between xi-1 and xi, i is set to the value of i-1 in step 874 for later processing. Therefore, in step 866, the input value x is x7.
It can be seen that it is between x8. In addition, in the judgment at step 850, it is determined whether the input value x is smaller than the minimum value x0 of the dividing point or the input value
If larger than xm, the input value is x0 respectively
and xm. The above has been described for the input value x, but the same applies to the input value y. The division point values xi, xi+1, yj, yj+1 obtained by the above method are temporarily stored in a work area in the RAM 116. Next, from i and j obtained by the above method, the control amount of the intersection of the dividing points
Find Zij. In other words, if CONT is the first address of the table that stores the control amount, then the data of the control amount Zij at the division point (xi, yj) is MAij=CONT+2(i×jm+j), where MAij is the storage address. (1) Stored at the address. Note that the coefficient 2 in this equation (1) means that the ROM 118 is 1 byte/1
This indicates that it is assumed to be a word. Therefore, as explained in Figure 3, CPU11
4, the data storage address MAij is specified via the address bus 164, and the control bus 16
By controlling the ROM through 6,
Data Zij in ROM 118 is stored in a register in CPU 114 via data bus 162.
This data Zij is temporarily stored in a work area in the RAM 116 using the above three types of buses. In a similar manner, the control amounts Zij+1, Zi+1j, Zi+1j+1 of the division points sandwiching the input values x and y are stored in the work area in the RAM 116 (step 830 in FIG. 16). Next, the control amount Z for the input values x and y is determined by interpolation based on the control amount for each division point determined in step 830 of FIG. 16 above. Here, the control amount Z is obtained by using linear interpolation three times. First, interpolation in the x-axis direction is performed twice, and interpolation in the y-axis direction is performed based on the control amounts Zj and Zj+1 obtained as a result. The interpolation formula is as follows. Zj=Zij+(Zi+1j-Zij)x-xi/xi+1-xi...(2) Zi+1=Zij+1(Zi+1j-Zij+1) x-xi/xi+1-xi...(3) Z=Zi+(Zj+1-Zj)y- yj/yj+1-yi...(4) RAM1 at steps 820 and 830 in Figure 16
The value temporarily stored in the work area in 16,
xi, xi+1, yi, yi+1, Zij, Zij+1, Zi+1j,
Calculate equations (2), (3), and (4) based on Zi+1j and j+1. Taking equation (2) as an example, first,
The data of Zi+1j and Zij in the RAM is read into the register in the CPU via the data bus 162, and the subtraction means in the CPU is used to calculate (Zi+1j−
Zij) is subtracted, and the result W1 is temporarily stored in the work area in the RAM. Next, in order to match the number of digits of x and xi before subtracting (x - xi), xi
Multiply the value of 2 to the 4th power (minimum interval between dividing points) and obtain (x
−xi) is subtracted. As a result, W2 and the above RAM
The multiplication of the operation result W1 in is performed by adding W2 once W1 using the addition means in the CPU,
This calculation result W3 is similarly stored in the RAM. This multiplication can be done by
It's easier to do. Next, the denominator of equation (2) is (xi
+1−xi) is determined by the subtraction means. Here, the above xi and xi+1 are as shown in Figure 15a,
Since it is the value of the upper 5 bits excluding the lower 4 bits, the subtraction result W4 of (xi + 1 - xi) is a part of the number of digits L of digit shift corresponding to division by (xj + 1 - xi) in equation (2). It means. Here, L is expressed as L=(xi+1-xi)/2+k (5), assuming that the minimum interval between dividing points is 2 to the k power. xi and xi+1 in equation (5) are the 5-bit data shown in FIG. 15a. Therefore, Zij is obtained by shifting the calculation result W3 to the right by L digits using a digit shifting means and adding Zij to the result using an adding means. Similarly to what has been explained above, the control amount Z corresponding to the input values x and y is finally obtained using equations (3) and (4). This control amount Z represents the ignition advance angle, so
The initial angular pulse INTLD shown in Figure 11 (this pulse is related to the number of cylinders in the engine; for 4 cylinders it is 180
If the interval is W, and the predetermined angle of the signal PC generated every time the engine rotates is 1 degree, as explained in Figure 1, then the ignition timing is The value ADVV set in the ADV register for control is obtained by ADVV=W-Z (6). By setting this value in the ADV register in the pulse output circuit 126 described in FIG. 4, the ignition timing can be controlled as described in FIG. 11. From the above, the effect of this embodiment is that no matter what the relationship between the control characteristics corresponding to the input information, the accuracy of the control characteristics corresponding to the input information can be maintained to the maximum necessary extent for the entire input information. Since the required storage capacity can be kept to the necessary minimum, the storage capacity can be significantly reduced compared to the conventional method. For example, in the case of the embodiment shown in FIG. 14, input information is equally divided at intervals of 2 to the n power while maintaining the accuracy of control characteristics to the necessary minimum (in this embodiment, n=4, so 19 Storage capacity required for (to be divided) 20 x 20 x 9 = 3600
According to the invention, 12×12×9+
Since it only requires 12 x 2 x 9 = 1512 bits, the storage capacity can be significantly reduced. In addition, in the present invention, when finding division points that sandwich arbitrary input values, by using the two-part search method, the processing is faster than a linear search, so the processing speed is faster than that of a linear search. This is highly effective when processing is performed in accordance with the engine speed. Furthermore, the interval between the dividing points of the input information is set to 2 to the n1 power,
By raising 2 to the power of n2...2 to the power of nm (n1, n2...nm is a positive integer) and a power of 2, the division required in interpolation calculations can be performed by moving digits, so there is no need for a division function. This reduces costs because a simple CPU is required. Furthermore, as an effect of the embodiment of the present invention, the storage capacity can be reduced by storing the data of the division points of input information as information only of the upper bits excluding the k bits of the minimum interval 2 of the division points to the k power. . According to the present invention, since the load condition indicating the basic fuel supply amount of the engine and the engine speed are used as the parameters of the ignition timing characteristic table, the memory can be used effectively and the storage capacity can be reduced. This allows accurate ignition timing to be found even if the engine is clogged.

[Brief explanation of drawings]

FIG. 1 is a layout diagram showing the positions of sensors and actuators in one embodiment of the present invention, FIG. 2 is an operation explanatory diagram for explaining the operation of FIG. 1, and FIG. 3 is an illustration of the control circuit of FIG. 1. Detailed diagram, Figure 4 is a partial detailed diagram of the input/output circuit in Figure 3, Figure 5 is an explanatory diagram of the operation of Figure 4,
Fig. 6 is a detailed diagram of the stage counter in Fig. 4, Fig. 7 is a detailed diagram of the synchronization circuit, Fig. 8 is an explanatory diagram of the operation of Fig. 7, Fig. 9 is a detailed diagram of the incrementer controller, and Fig. 10 is a detailed diagram of the stage counter in Fig. 4. Figure 11 is a diagram explaining the operation of fuel injection signal processing, Figure 11 is a diagram explaining the operation of ignition timing control, Figure 12 is a diagram explaining the operation of EGR or NIDL processing, and Figure 13 is a diagram explaining the operation of engine rotation speed RPM or vehicle speed VSP detection. It is an operation explanatory diagram. Fig. 14 is a characteristic diagram of the ignition advance angle, Fig. 15 is a characteristic diagram showing the characteristics of the intersection of x and y in Fig. 4, Fig. 16 is an operation explanatory diagram,
FIG. 17 is an explanatory diagram illustrating the search for division points. 10...control circuit, 12...air cleaner, 14...
Air flow meter, 36... Spark plug, 38... Power distributor, 40... Ignition coil, 98... Angle sensor, 1
14...CPU, 116...RAM, 118...ROM,
120...Input/output circuit.

Claims (1)

[Scope of Claims] 1 (a) Air flow rate detection means for detecting information on the amount of air taken into the engine; (b) Rotation speed detection means for detecting engine rotation speed information; (c) Said air amount detection means load information determining means for determining load information based on the air amount information obtained from the rotation speed information and the rotation speed information obtained from the rotation speed detection means; (d) dividing the rotation speed information and the load information into a predetermined number of division points, respectively; (e) storage means that divides the ignition data into segments and stores the ignition data in advance at addresses determined according to the order of the division points; (e) based on the load information from the load information determining means and the rotational speed information from the rotational speed detection means; An ignition timing control device for an engine comprising ignition control means for reading ignition data determined by corresponding load information and rotation speed information from the storage means and controlling ignition timing based on the read ignition data.