Method and Apparatus for Ignition System Spark Timing Control During Engine Cranking

References are made to the following related co-pending applications, each filed in the name of Kenji Ikeura on Apr. 3, 1980:

(1) U.S. patent application Ser. No. 136,959;

(2) U.S. patent application Ser. No. 137,000;

(3) U.S. patent application Ser. No. 136,996; and

(4) U.S. patent application Ser. No. 136,994, now U.S. Pat. No. 4,314,540, issued Feb. 9, 1982.

The present invention relates to a method and apparatus for the control of an ignition system spark timing for a spark ignition internal combustion engine, and more particularly to a method and apparatus for the control of an ignition system spark timing during cranking of a spark ignition internal combustion engine.

Description is now provided concerning the conventional art and the problems associated therewith. The spark timing control device of an internal combustion engine adjusts the timing of spark to a certain degree before the top dead center (for example, 12.degree. C. BTDC/600 rpm) so as to cause the ignition to take place at the optimum instance to provide the maximum output because there is an ignition delay from the passage of a spark to the actual ignition of the fuel within the cylinder of the engine. The ignition delay in terms of time is substantially unchanged even if the engine revolution speed increases, but if it is considered in terms of crank angle, the ignition delay increases in accordance with an increase in the engine revolution speed because the piston speed increases in accordance with the increase in engine revolution speed. Thus, a centrifugal spark advance unit of the mechanical type, for example, is operatively connected to a distributor, thereby to advance the spark timing in accordance with an increase in the revolution speed because if the spark timing remains at the same level as that for the low revolution speed, an adequate output characteristic can not be provided when the revolution speed increases.

FIGS. 1 and 2 are graphs showing spark advance characteristics provided by the conventional devices wherein the axis of abscissa designates engine revolution speed and the axis of ordinate designates spark advance value. The reference letter N.sub.1 designates a revolution speed beyond or above which the spark advance value starts to increase, and the letter N.sub.2 designates a revolution speed beyond or above which the spark advance value becomes maximum. In the case of FIG. 1, the characteristic features in that the spark advance value linearly increases as the revolution speed increases when the revolution speed is between N.sub.1 and N.sub.2. In the case of FIG. 2, the characteristic features in that the gradient of a slope between N.sub.1 and N.sub.2 changes at N.sub.3. In both of the cases, spark advance values within a low revolution speed range below N.sub.1 and within a high revolution speed range above N.sub.2 are constant, thereby to maintain the spark advance constant irrespective of the variation in the revolution speed. The letter N.sub.4 designates the idle revolution speed. As will be understood from an observation of the characteristics shown in FIGS. 1 and 2, according to the conventional devices, the spark advance value remains constant when the engine is under cranking, so that there is a problem that if the cranking speed is low, the ignition resulting from the passage of a spark causes a torque tending not to rotate the engine crankshaft in the forward direction, but to rotate in the reverse direction, thereby to deteriorate the engine start-up performance when the atmospheric temperature is low and the battery voltage is low.

An object of the present invention is to improve start-up of a spark ignition internal combustion engine when the engine is under cranking even if the cranking speed is very low. The method and apparatus provide a control wherein, when the cranking speed is low, the value of spark advance is decreased in accordance with a drop in engine revolution speed.

The method and apparatus of the invention concerns the control wherein it is detected whether a spark ignition internal combustion engine is under cranking or not, and the revolution or cranking speed of the engine is sensed, and, when the engine is under cranking and at the same time the cranking speed is lower than a predetermined engine revolution value, the value of spark advance is decreased in accordance with a drop in the cranking speed.

The invention may be better understood by reference to the detailed description which follows and to the drawings.

FIGS. 1 and 2 are graphs of engine-speed spark-advance versus engine speed according to the conventional art;

FIG. 3 is a block diagram illustrating one embodiment of the present invention;

FIG. 4 is a flow chart illustrative of the operation of the digital microcomputer used to determine the value of spark advance;

FIG. 5 is a flow chart illustrative of the operation of the digital microcomputer used to determine the value of spark advance; and

FIG. 6 is a graph of engine-speed spark-advance versus engine speed N of the present invention.

FIG. 3 is a block diagram showing a preferred embodiment of the present invention. In FIG. 3, the reference numeral 1 designates a start switch which provides an ON or OFF signal S.sub.1 in response to whether an engine is under cranking or not. The numeral 2 designates a basic angle sensor which provides basic angle pulses S.sub.2 each upon an expiration of a basic angle (for example, 120.degree.) as the crank shaft rotates, and numeral 3 designates a unit angle sensor which provides unit angle pulses S.sub.3 each upon an expiration of a unit angle (for example, 1.degree.) as the crank shaft rotates. The revolution speed of the engine is found based upon the signal S.sub.3 from the unit angle sensor 3. These signal S.sub.1 to S.sub.3 (if necessary, other signals including a signal corresponding to engine induction vacuum and pulse signals for fuel injection may be used) are read in via an input-output control unit 4 that is constructed of semiconductors by a central processor 5 including a ROM (read only memory), a RAM (random access memory), and a CPU (central processor unit). From the read signals S.sub.1 to S.sub.3, the central processor 5 detects whether the engine is under cranking or not, finds by calculation or table look-up a spark advance value, and instructs this value to that section of the input-output control unit 4 which performs spark advance control. The input-output control unit 4 provides, at a spark output terminal 6, an output that carries information that current be passed or interrupted for a spark based upon the signals S.sub.2 and S.sub.3 of the basic angle sensor 2 and the unit angle sensor 3 and the instructed spark advance value, thereby controlling the operation of a transistor 7 which turns on or off electric current passing through an ignition coil 9 of an ignition system as enclosed by the rectangular one-dot chain line shown in FIG. 3. The numeral 8 designates a distributor, the numeral 10 designates a spark plug, and the numeral 11 designates a battery for a source of electricity. Owing to the benefit of recent progress in semiconductor techniques, it has been possible to easily manufacture and use a unit integrally combining a central processor 5 with an input-output control unit 4 and which also has a built-in transistor 7, as shown in the two-dot chain line box of FIG. 3.

FIGS. 4 and 5 show flow charts illustrative of the operation of the central processor 5 from a determination of spark advance value to the output thereof. FIG. 4 is a flow chart wherein the same spark advance characteristic is employed for engine cranking and for normal or ordinary-in-use engine operation, while FIG. 5 is a flow chart wherein distinct spark advance characteristics are used for engine cranking and for normal or ordinary-in-use engine operation, respectively. Referring to FIG. 4, when the program starts, the basic spark advance value (abbreviated hereinafter as ADV) is determined based upon engine revolution speed, load, and other factors in a step P.sub.1. In a step P.sub.2 it is decided the ON or OFF state of the start switch 1 whether the engine is under cranking or not. If the start switch 1 is in an OFF state, the engine is not under cranking so that the program goes to a step P.sub.3 to provide the ADV value found in the step P.sub.1. If the start switch 1 is in an ON state, the engine is under cranking so that the program goes the step from P.sub.2 to a step P.sub.4 wherein it is decided whether the engine revolution speed N is higher than a predetermined value N.sub.L or lower than the value N.sub.L. If N>N.sub.L, the program goes from P.sub.4 to P.sub.3 wherein the spark advance value found in the previous step P.sub.1 is issued as an output. If N.ltoreq.N.sub.L as a result of the decision in the step P.sub.4, the program goes to the step P.sub.5 wherein the ADV value is multiplied by a coefficient that is proportional to the engine revolution speed N and is sent to step P.sub.3 for issuance. Therefore, the spark advance value is rendered small when N/N.sub.L .ltoreq.1.

According to the operation shown in the flow chart shown in FIG. 5, it is decided in step P.sub.2 whether the start switch 1 is in an ON state or in an OFF state and if the start switch 1 is in an ON state thereby to indicate that the engine is under cranking, the program goes to the step P.sub.12, while, if the start switch 1 is in an OFF state, the program goes to the step P.sub.11. In the step P.sub.11, a spark advance value for ordinary-in-use engine operation is found based upon the engine revolution speed and load and the result is issued via step P.sub.3. In the step P.sub.12, a spark advance value for the engine cranking is found and the program goes to the step P.sub.4. The program after the step P.sub.4 inclusive is the same as that of the case shown in FIG. 4; that is, if the comparison of N with N.sub.L shows that N.ltoreq.N.sub.L, the step goes to the step P.sub.5 wherein the ADV value found in the step P.sub.12 is multiplied with N/N.sub.L and this result is issued via step P.sub.3.

Hereinafter the meaning of the operation described in connection with FIGS. 4 and 5 is explained. Properly speaking, the spark advance value should be adjusted, taking as a correction factor into account the time delay from the passage of a spark to the actual ignition such that at the optimum instance after the top dead center, the combustion pressure becomes maximum, thereby to give rise to the efficient conversion from thermal energy to torque. In this respect, ADV value should be given by spark advance time rather than by crank angle.

Since, in ordinary-in-use operating ranges, the engine revolution speed is relatively stable, it can be considered that the advance angle is equal to the advance time. Considering that parts of engine main portion are designed to move in a timed relation with the crank angle, it is most convenient and practical to give an advance value by the crank angle.

However, to give a spark advance by the crank angle is inconvenient upon engine cranking, particularly under cold start-up when the engine revolution speeds are excessively low. If, for example, a spark advance of 18.degree. is optimum for an engine revolution speed of 120 rpm, this spark advance value corresponds to 25 milliseconds before top dead center in terms of time. This spark advance value of 18.degree. corresponds to 50 milliseconds before top dead center when the engine revolution speed is only 60 rpm, thus giving a spark too early to allow the combustion pressure to become maximum before the top dead center, causing reverse rotation or suspension of rotation, with the result that the engine fails to start. If, for the purpose of avoiding it, the spark advance value is set excessively small, enough torque fails to be produced, causing the engine to fail to start.

To solve this problem, when the cranking or engine speed N is lower than the predetermined value N.sub.L, a spark advance time before top dead center becomes constant by, according to the present invention, multiplying the optimum ADV value for the speed N.sub.L with a coefficient N/N.sub.L, thereby attaining the most effective conversion from combustion pressure to torque to enable optimum engine start-up.

The reason why such correction of the spark advance value is unnecessary when N>N.sub.L is that when the cranking speed N is relatively high, the other environmental conditions such as spark energy and fuel atomization are favourable and in a direction tending to decrease the combustion delay, thus rendering the correction of spark advance angle unnecessary. Another reason is that startup of the engine is not affected even if the spark advance is not minutely set when the cranking speed is sufficiently high, and if, actually, the spark advance is minutely set, a reaction takes place such as an occurrence of knocking after the completion of combustion. It is appropriate that the predetermined value N.sub.L be a value corresponding to a value within a range from 60 rpm to 200 rpm. FIG. 6 shows engine-speed spark-advance versus engine revolution speed N, wherein N.sub.L designates the predetermined value, while N.sub.1, N.sub.2, N.sub.4 designate revolution speed values having similar meanings to the counterparts shown in FIG. 1.

In the description regarding the embodiment the case was explained wherein the microcomputer constructed of semiconductors is used to calculate ADV.times.N/N.sub.L, but the present invention is not limited to this application and the spark advance characteristic as shown in FIG. 6 can be attained with a centrifugal spark advance device used in the conventional distributor.

However, if a microcomputer system is already employed, the present invention can be embodied with a slight addition of a program, thus practicing the invention without any substantial cost increase. If a centrifugal spark advance mechanism of the conventional distributor is used to practice the invention, the invention can be embodied and adaptable to almost all kinds of conventional internal combustion engines with substantially acceptable cost increase.

It will now be understood that since, according to the present invention, when the cranking speed N is lower than a predetermined value N.sub.L upon cranking of the engine, the spark advance is rendered small, and reverse revolution of the engine is prevented even when the cranking speed is low so that effective conversion from the combustion pressure to torque takes place, thus providing good start-up performance.

In the practice of the invention, the inventor has used a Hitachi Model No. HD 46802 central processor. A Hitachi Model No. HD 46506 input-output control unit compatible with the processor has been used.

The start switch 1 is entirely conventional and may be a pair of contacts which are closed while the engine starter motor is in operation.

The basic angle sensor and unit angle sensor are also conventional and may be in the form of a detecting unit 24 described in U.S. Pat. No. 4,015,565, filed Apr. 5, 1977 in the name of Aono et al. and entitled "SPARK-ADVANCE CONTROL APPARATUS FOR INTERNAL COMBUSTION ENGINE" (which is herein incorporated by reference) or may be in the form of a speed-electrical transducer 10 described in U.S. Pat. No. 3,853,103, filed Dec. 10, 1974 in the name of Wahl et al. (assignee: Robert Bosch GmbH) and entitled "IGNITION TIMING CONTROL SYSTEM FOR INTERNAL COMBUSTION ENGINE IGNITION SYSTEMS," (which is also incorporated by reference).

A way to find or sense the revolution speed of the engine, i.e., engine rpm, based upon a signal from the unit angle sensor is conventional and described in the above-mentioned U.S. Pat. No. 3,853,103. In this respect, reference is made to U.S. Pat. No. 3,969,614, filed July 13, 1976 in the name of Moyer et al. and entitled "METHOD AND APPARATUS FOR ENGINE CONTROL" and reference is also made to U.S. Pat. No. 4,009,699, filed Mar. 1, 1977 in the name of Hetzler et al. and entitled "DIGITAL IGNITION SPARK TIMING ANGLE CONTROL WITH READ ONLY MEMORY."

The inventor has used a breakerless ignition system employing an ignition coil and a power transistor.

A way to determine the instance of ignition based upon binary signals of the basic angle sensor and the unit angle sensor is disclosed in U.S. Pat. No. 3,853,103 in the name of Wahl et al. and U.S. Pat. No. 4,015,565 in the name of Aono et al.

In a four-cylinder, four-cycle internal combustion engine, each one ignition pulse must be obtained after each revolution of the crank shaft by i.e. at 180.degree.; 180.degree. after the first ignition pulse is derived, a second basic angle pulse must then be provided by the basic angle sensor. In this case, the basic angle sensor must be designed to provide a basic angle pulse after each revolution of the crank shaft by 180.degree..

In a six-cylinder, four-cycle internal combustion engine, each one ignition pulse must be obtained after each revolution of the crank shaft by i.e. at 120.degree.; 120.degree. after the first ignition pulse is derived, a second basic angle pulse must then be provided by the basic angle sensor. In this case, the basic angle sensor must be designed to provide a basic angle pulse after each revolution of the crank shaft by 120.degree..