System for Controlling the Ignition Timing of an Internal Combustion Engine

The present invention relates to a system for controlling the ignition timing of an internal combustion engine such as an automotive engine.

A learning control system for correcting the ignition timing has been proposed. The control system is adapted to advance the ignition timing so as to produce a maximum torque as long as the level of engine knocking does not exceed a tolerable level. The ignition timing stored in a RAM is corrected by a small correcting quantity (quantity of correction) and converged to a desired value little by little. The correcting quantity for the ignition timing at every updating operation is gradually reduced as the number of the learning increases, that is as the ignition timing approaches the desired value.

On the other hand, if a large disturbance occurs, such as a large change of engine load, the ignition timing must be corrected by a large quantity. However, in the state where the ignition timing approaches the desired ignition timing, the correcting quantity at each updating is very small as described above. Accordingly, it takes a long time to correct the ignition timing to a new desired timing.

The object of the present invention is to provide a control system which may quickly correct the ignition timing to a desired ignition timing when the ignition timing greatly deviates from the desired ignition timing.

According to the present invention, there is provided a system for controlling the ignition timing of an internal combustion engine having an ignition timing control device, comprising sensing means for sensing the operating conditions of the engine and for producing an engine operating condition signal, and a knock sensor for sensing engine knock and for producing a knock signal.

The system comprises first means responsive to the engine operating condition signal and knock signal for producing an ignition timing correcting signal representing an ignition timing correcting quantity which is applied to the ignition timing control device for correcting the timing, second means for detecting the change of engine operating conditions which will cause a deviation of ignition timing from a desired ignition timing and for producing a correction signal, said second means is means for detecting frequency of ignition timing correction higher than a predetermined value and third means responsive to the correction signal for increasing the ignition timing correcting quantity.

Other objects and features of this invention will become understood from the following description with reference to the accompanying drawings.

FIG. 1 is a block diagram showing a control system according to the present invention;

FIG. 2 is a block diagram showing a main part of the control system;

FIGS. 3a and 3b show tables storing a plurality of ignition timings;

FIG. 4 shows a range of a coefficient K;

FIGS. 5, 6, 7a and 7b are flow charts showing the operation of the system; and

FIGS. 8a and 8b show a retard coefficient table and an advance determining period table, respectively.

Referring to FIG. 1, an intake air pressure (or quantity) sensor 1, engine speed sensor 4 such as a crankangle sensor, and knock sensor 7 are provided to detect engine operating conditions. The output of the sensor 1 is applied to an A/D converter 3 through a buffer 2, and the output of the sensor 4 is applied to an interrupt processing circuit 6 through a buffer 5. The output of the knock sensor 7 is applied to a comparator 12 through a filter 8 and an amplifier 9, and, on the other hand, to the comparator 12 through a rectifier 10 and an amplifier 11. The comparator 12 compares both inputs and produces an output signal when an engine knock having a higher level than a predetermined value is generated. The outputs of the A/D converter 3, circuit 6 and comparator 12 are applied to a microprocessor 18 through an input port 13.

The microprocessor 18 comprises a CPU 15, RAM 16, ROM 17 and output port 14. The output of the microprocessor 18 is applied to an ignition timing control device 21 through a driver 19 so as to control the ignition timing in accordance with the engine operating conditions sensed by the sensors 1, 4 and 7.

FIG. 5 summarizes the operation of the control system. The operation is divided into a rough correction and a fine correction. At a step 30, it is decided whether a rough correction has been executed (if a rough correction completion flag RCMP is set). In accordance with the decision, the rough correction or fine correction is executed at a step 31 or 32. At a step 33, a real ignition timing SPK.sub.real is calculated.

The rough correction is an operation for obtaining a basic ignition timing SPK.sub.bs which is calculated in a basic ignition timing setting circuit 71 shown in FIG. 2. FIG. 6 shows the operation of the rough correction. At a step 37, engine speed and intake air pressure are calculated based on output signals of sensors 1 and 4. Thereafter, at a step 38, a first maximum ignition timing MAPSTD and a second maximum ignition timing MBT are read from tables 38a and 38b (FIGS. 3a, 3b) in the ROM 17, in accordance with the engine speed and intake air pressure. The first maximum ignition timing is maximum timing for producing maximum torque with low-octane gasoline without the occurrence of knocking and the second maximum ignition timing is maximum timing for producing maximum torque with high-octane gasoline without the occurrence of the knocking.

In the system, a coefficient K for correcting the ignition timing is provided. The value of the coefficient K is preliminarily set to a value between zero and 1 as shown in FIG. 4.

The coefficient K is stored in the RAM 16 and updated in accordance with engine operating conditions so as to roughly converge the ignition timing to a desired ignition timing. The updating is performed under a predetermined condition and the condition is determined at a step 39. When the difference between the first and second maximum ignition timings read from the tables 38a and 38b (FIG. 3a and FIG. 3b) is larger than a predetermined degree, for example 5.degree., the updating is performed. Namely, the program proceeds to a step 40, where it is determined whether a knock has occurred during the program. When the occurrence of knocking is determined, the program proceeds to a step 41, and if not, proceeds to a step 42. At step 41, the coefficient K is decremented by a correcting quantity .DELTA.K(.DELTA.K=K/2), and the remainder K-.DELTA.K is stored in the RAM 16 as a new coefficient for the next updating. Accordingly, the correcting quantity .DELTA.K at the next updating is (K-.DELTA.K)/2. Namely, the correcting quantity is one-half of the coefficient K at updating. More particularly, if the initial coefficient is 1/2, the correcting quantity is 1/4, and if it is 0 or 1, the correcting quantity is 1/2 as seen from FIG. 4.

At the step 42, it is determined whether the engine has operated without knock occurring for a predetermined period. When knocking does not occur for the period, the coefficient K is incremented by the correcting quantity .DELTA.K at a step 43.

After the updating of the coefficient K at step 41 or 43, it is determined whether the rough correction is completed at a step 44. As will be understood from the above description, the correcting quantity .DELTA.K decreases as the number of the correction increases. In the system, when the correcting quantity reaches a predetermined small value, the rough correction is completed. Accordingly, if quantity .DELTA.K reaches the predetermined value, a rough correction completion flag RCMP is set at a step 45, or if not, the flag is reset at a step 46. On the other hand, the total correcting quantity SPK.sub.prt and the number of correction NUM of the ignition timing are stored in an ignition timing correcting quantity table 73 and a table 74 (FIG. 2) for the number of the correction. At a step 47, a basic ignition timing SPK.sub.bs is calculated from the following formula

where .DELTA.MAPMBT=MBT-MAPSTD

The basic ignition timing is applied to an engine 72 (FIG. 2) to operate the engine at the ignition timing. The coefficient K is stored in the RAM 16. If the rough correction is not completed, the coefficient K is updated at the next program so as to roughly converge the ignition timing to a desired ignition timing as described above. It will be understood that if the initial coefficient K is 0, the basic ignition timing SPK.sub.bs calculated by the formula (1) is the maximum ignition timing MAPSTD at the first program. The basic ignition timing SPK.sub.bs obtained by the rough correction is further corrected by the fine correcting operation as described hereinafter.

Referring to FIGS. 7a and 7b, at a step 52, it is decided whether the engine operation is in a range which is proper to correct the basic ignition timing SPK.sub.bs. If it is in the range, the correcting quantity SPK.sub.prt and the number of correction NUM are read from tables 73 and 74 at a step 53. Then, at a step 54, a retard coefficient LN for retarding quantity RET is looked up from a retard coefficient table 75 (FIG. 2) of FIG. 8a in accordance with the number of correction NUM, and an advance determining period ADJ is looked up from an advance determining period table 76 (FIG. 2) of FIG. 8b in accordance with the number of correction NUM. Thereafter, the program proceeds to a step 55, where it is decided whether a knock has occurred during the program. When the occurrence of knocking is determined, the program proceeds to a step 56, and if not, it proceeds to a step 59. At step 56, the intensity of the knock and the interval of knocks are calculated at a calculating circuit 78 (FIG. 2), and then, retarding quantity KNK is looked up from a retarding quantity table 79 in accordance with the intensity and the interval of the knocking. At a step 57, a real retarding quantity RET.sub.real is calculated by multiplying the retarding quantity KNK and retard coefficient LN together (RET.sub.real =KNK.times.LN). Thereafter, the program proceeds to a step 58, where the correcting quantity SPK.sub.prt stored in the table 73 is subtracted with the real retarding quantity RET.sub.real to obtain a new correcting quantity SPK.sub.prtr which is stored in the table 73.

On the other hand, at the step 59, it is decided whether a knock occurred in the advance determining period ADJ, which is performed at a comparator 80 in FIG. 2. When knocking does not occur in the period, the program proceeds to a step 60, where an advancing quantity ADV of a constant small value is added to the correcting quantity SPK.sub.prt to obtain a new correcting quantity SPK.sub.prta which is performed in an advancing quantity setting circuit 81 in FIG. 2 and stored in the table 73. Thereafter, a step 61, it is determined whether the new correcting quantity SPK.sub.prta is larger than a limit value which is obtained by subtracting the basic ignition timing SPK.sub.bs from the maximum ignition timing MBT (MBT-SPK.sub.bs). When the new correcting quantity SPK.sub.prta is smaller than the limit value, the new correcting quantity is stored in the table 73 at a step 63. If it is larger than the limit value, value of MBT-SPK.sub.bs is used as a new correcting quantity (at a step 62) and stored in the table 73.

Thereafter the program proceeds to a step 64, where it is decided whether knock larger than a predetermined intensity has occurred or the frequency of correction in the same direction (advance or retard) is higher than a predetermined value, which is caused by large disturbance. When such a phenomenon occurs, the program proceeds to a step 65 (circuit 82), where the number of correction NUM which is applied to the tables 75 and 76 (FIGS. 8a and 8b) is reduced. Accordingly, the retard coefficient LN to be obtained in the next program is increased, and the advance determining period ADJ is reduced as seen from FIGS. 8a and 8b, which means that the new correcting quantity SPK.sub.prtr or SPK.sub.prta increases. Thus, the ignition timing is largely corrected at succeeding programs, so that the timing can quickly converge to a desired value.

While the presently preferred embodiment of the present invention has been shown and described, it is to be understood that this disclosure is for the purpose of illustration and that various changes and modifications may be made without departing from the scope of the invention as set forth in the appended claims.