JPS6127586B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention converts engine vibrations into electrical signals,
The present invention relates to an ignition timing control method for detecting a knock condition in an engine and controlling ignition timing, and particularly relates to an improvement of a device for detecting a knock condition. There are two main methods for detecting a knock state of an engine. One method is to monitor cylinder internal pressure, and the other is to monitor engine vibration. This method of supervising the vibration of an engine is disclosed in, for example, Japanese Patent Laid-Open No. 54-25334 and Japanese Patent Laid-Open No. 52-87537. The former method has the advantage of being relatively easy to detect, but requires an expensive sensor for each cylinder, which is disadvantageous for the commercialization of ignition control equipment. The latter is
The disadvantage is that when the engine rotates at high speed, the vibration of the engine itself becomes large and it becomes difficult to detect the knock signal.
Since only one sensor is required, this method is advantageous compared to the former method in terms of cost. Conventionally, in systems that detect knock signals by replacing engine vibration with electrical signals, this signal is integrated after passing through a bandpass filter (hereinafter abbreviated as BPF) that has a frequency band specific to the knock signal. A method was used to detect a knock state by shaping the waveform and comparing it with a certain reference signal, but since the frequency band of the knock signal is 8kHz, which is a relatively high frequency, small knock signals are distorted. There was a flaw that was difficult to detect. In the present invention, in order to eliminate the above-mentioned drawbacks,
The BPF signal is input directly into a comparator, compared with the level of a noise signal other than the knock signal, and the resulting output is converted into a voltage. This makes it possible to detect knock signals of a relatively small level, and also makes it possible to control ignition according to the strength of the knock. The purpose of the present invention is to detect that the engine is knocking using the knock signal described above, and to control the combustion start timing of the engine, that is, the ignition timing of the ignition plug, to match it to the most efficient point of operation. It is said that In the embodiment of the present invention, the following method is used to detect the knock signal described above and control the ignition timing of the ignition plug. That is, as shown in Fig. 8A, the reference ignition timing is adjusted to the engine's intake manifold negative pressure and
The period shall be determined by the person concerned, so that the timing is always within the organization's knowledge area. A circuit is added to delay the ignition timing according to the appearance of the knock with respect to the reference ignition timing. Optimal control cannot be performed if the ignition timing remains delayed, so a circuit is added to gradually reduce the amount of delay with a certain time constant, thereby returning the ignition timing to the reference position. When analyzing the vibration waveform of an internal combustion engine, it is a known fact that when the engine is knocking, the amplitude becomes large only in a certain frequency band. When the engine vibration is replaced with an electric signal and the signal waveform is observed after passing through a bandpass filter that passes only the frequency band specific to the knock signal, the signal waveform is as shown in FIG. 5B. On the other hand, the waveform in Figure 5A is
Indicates the primary current of the coil. When the primary current is cut off, a high voltage is generated on the secondary side of the coil, causing sparks to fly to the ignition plug. Below, we will look at the signal waveform over time using the timing of primary current cutoff as a reference.
The vibration wave that appears from the cutoff to point a is the spark noise of the ignition plug picked up by the bandpass filter. From point a to point b is a oscillating waveform in a frequency band that is included in engine vibration and can pass through a bandpass filter. The wave from point b to point c is a knock signal that appears when the engine knocks. The waves after point c are similar to the waves from point a to point b. The present invention is based on this observed waveform. FIG. 1 is a block diagram of one embodiment of the present invention, and FIG. 2 is a circuit diagram showing one embodiment of the present invention. The explanation will be given below with reference to FIGS. 1 and 2. FIG. 1 schematically shows the circuit of FIG. 2,
A knock sensor that replaces engine vibration with an electrical signal, a band pass filter that passes only the frequency band specific to the knock signal from the electrical signal from the knock sensor, (hereinafter band pass filter is abbreviated as BPF) Vibration after passing through the band pass filter It is superimposed on the vibration waveform after passing through a bandpass filter and an amplifier that amplifies the waveform. Vibration waveform other than knock signal =
A sample-and-hold circuit is used to grasp the level of the wave from time a to point b in Fig. 5B (hereinafter, this vibration waveform is abbreviated as background noise = BGU), and the sample-and-hold BGN is compared with the engine vibration. A comparator that detects the knock signal, when comparing the sampled and held voltage with engine vibration.
An attenuator to provide a noise margin, a knock amount that converts the knock signal detected by the comparator into a voltage, and a voltage converter, the time during which spark noise is generated, which is an obstacle to detecting the knock signal, and the sample hold. A noise mask time generator to mask the vibration waveform at the time when the noise occurs, and a sampling time generator to provide sample and hold timing. An engine speed detection circuit, a voltage control circuit during high-speed operation, and a maximum voltage limiting circuit and minimum voltage limiting circuit to limit the maximum and minimum values of the delay angle to keep the delay angle constant when a knock detection unit A, a reference ignition timing signal generator that generates a reference ignition timing signal in synchronization with engine rotation, and a knock detection unit A that is delayed by a certain angle from the reference ignition timing in response to the voltage of the knock amount → voltage converter. a delay signal generator that generates an ignition timing signal; an ignition control circuit that controls the primary current of the ignition coil using the reference ignition timing signal and the delay signal; amplifying the signal from the ignition control circuit;
FIG. 3 is a block diagram consisting of an ignition control section B consisting of an amplifier that switches a power transistor and cuts off and on the primary current. Next, a circuit embodying the block diagram of FIG. 1 will be explained with reference to FIG. 2. The vibration waveform of the engine, which has been converted into an electrical signal by the knock sensor, is input to point a. The vibration waveform input to point a passes through a bandpass filter composed of resistors 1 to 9, resistor 11, capacitors 90 to 92, and operational amplifiers 191 and 192. This BPF is a generally known two-stage active filter, but in order to enable use with single grounding, the reference terminal (terminal in this circuit diagram) of the operational amplifier is connected to resistors 3 and 6 and resistors 5 and 7. It is fixed near the midpoint of the operating voltage (around 3V). The center frequency is f ₀ (Hz), the gain at the midpoint frequency is H ₀ , and the cutoff frequencies above and below the center frequency α ₁ , α
Let Q be the _ratio of _{the center frequency to}
R ₂ , R ₄ , R ₈ , R ₉ , R ₁₁ ) and capacitor 1,
2, C ₁ and C ₂ are each given by the following equations.

[Table] In the circuit example of the present invention, H ₀ ≒ 10,
Since it was found that Q≒20, δ ₀ ≒7kHz was the easiest to distinguish the knock signal, the resistance value was selected using the above formula. The signal waveform after passing through the BPF is the operational amplifier 1
93, transistor 131, diode 101,
102, resistance 10, 12, 13, 14, 16, 1
7 (=R ₁₀ , R ₁₂ , R ₁₃ , R ₁₄ , R ₁₆ , R ₁₇ ), the signal is amplified approximately 15 times. Fourth
The details and operating waveforms of the amplifier and sample-and-hold circuit are shown in FIGS. The circuit consisting of the resistors R ₁₀ and R ₁₂ diodes 101 and the transistor 151 is a circuit for determining the DC operation levels V _B and V _B ' of the operational amplifier 193, and the circuit consisting of the diode 102 and the resistors R ₁₆ and R ₁₇ is , is an attenuator that divides the output value of the operational amplifier 193 by resistance division. In FIG. 4, the voltage at the output point d of the operational amplifier 193 is V _O1 , and the voltage at the point e obtained by dividing the resistor is V _O ′ ₁ , c
Letting the voltage at point V _C be the voltage V _C at point c', and the values of V _O1 and V _O1 ' are expressed by the following equations. V _O1 =R ₁ +R ₂ +R ₃ /R ₃・V _i −R ₁ +R ₂ /
R ₃・V _c +V _f … (6) V _O1 ′=R ₂ +R ₃ /R ₃・V _i −R ₂ /R ₃・V _c …
...(7) Also, if the voltage at the reference terminal g point of the operational amplifier 192 is V _g , then the DC operation level V of V _O1 and V _O1 '
_B and V _B ' are as follows. V _B =R ₁ +R ₂ +R ₃ /R ₃・V _g −R ₁ +R ₂ /R
₃・V _c +V _f … …(8) V _B ′=R ₂ +R ₃ /R ₃・V _g −R ₂ /R ₃・V _c …
...(9) The amplified vibration waveform V _O1 is the transistor 152
Diode 103 and resistors 19 and 20 (=
R ₁₉ , R ₂₀ ). The base of PNP transistor 152 is pulled by transistor 162 via resistor _R19 according to timing from a sampling time generator, which will be described later, and transistor 152 is turned on. Meanwhile, the vibration waveform V _O1 is the transistor 152, diode 10
However, due to the rectifying action of the diode 103, the maximum voltage of V _O1 during the sampling time is held in the capacitor 93. Here, the terminal voltage V _S of the capacitor 93 is the value obtained by subtracting the voltage drop V _f of the diode 103 from V _O1 , and the term V _f in equation (6) is eliminated, and V _s
is a value that is not affected by the temperature characteristics of the voltage drop of the diode. The terminal voltage V _s of the capacitor 93 is held until the next spark occurs, but at the moment when the spark occurs, a transistor 161, which is a part of the mask time generator described later, is turned on, so that it is discharged through the resistor 20. At the same time as the transistor 161 is turned off, the transistor 162 is turned on and a new maximum voltage V _s of V _O1 is sampled and held in the capacitor 93. The operation from the amplifier output to sampling is shown in FIG. As a result of the experiment, the maximum time to point a where the spark noise occurs is about 1.8 msec, and this period is considered as the discharge time of the sample voltage V _s of the capacitor 93.
It is appropriate to set the period up to a′ as the sampling time. The sampling time is approximately
0.7msec, background noise of V _O1 =
The highest value of BGN is sampled. Subsequently, the operation of the comparator 194 will be explained with reference to FIGS. 4 and 5. The sample voltage V _s is connected to the terminal of the comparator ₁₉₄ , and the attenuated voltage V _O1 ' of V _O1 is connected _to the terminal of the comparator 194 _.
If the knock signal is present at ₁ ', V _O1 ' becomes a higher voltage than the sample voltage V _s , and the output voltage V _O2 of the comparator 194 becomes High. This operation is detailed in C and D in FIG. In D, V _O2 is also High during 0-a' on the time axis.
This process will be described later. The output voltage V _O1 of the amplifier is reduced to V _O1 by the attenuator.
The reason for lowering it to ₁ ' is to provide a noise margin to avoid misidentification as a knock signal when noise is added to the signal waveform for some reason, and this is calculated using equations (6) to (9).
As is clear from the above, it can be set appropriately by changing the value of the resistor 17. The knock signal detected by comparator 194 is applied to transistor 155 and diode 104,1.
05, 106, resistance 23, 26 (=R ₂₃ , R ₂₆ ),
It is connected to a knock amount to voltage conversion circuit constituted by a capacitor 94 (=C ₉₄ ). Diode 106 is a Zener diode of around 3V,
The voltage between the base of the PNP transistor 155 and the power supply _Vcc is determined. On the other hand, PNP transistor 15
A resistor is connected between the emitter of No. 5 and _Vcc . Now the zener voltage of diode 106 is V _Z10
6. The voltage between B and E of the transistor 155 is V _B-E1
55 , the collector current I _C155 is I _c155 ≒ V _Z106 −V _R−E155 /R ₂₃ ...(1
0), which is a constant current. This constant current is determined by the output of the comparator 194.
When the output is high, it is charged to the capacitor C ₉₄ through the diode 105, and when the output is low, it is pulled through the diode 104 to the output transistor of the comparator. In this method, transistor 155
The negative influence of natural charging on the capacitor 94 due to leakage current can be prevented. Therefore, the terminal voltage V _C194 of the capacitor 94 is the sixth
As shown in the figure, if the rectangular wave width of the comparator 194 output is th, then V _C194 =I _C155 ·th/C ₉₄ (11), and the battery is linearly charged only when the output is High. Here, Fig. 5 C, D and Fig. 3 j, k, l
As shown in , when the knock amount is large, the signal is output for a long time.
As the number of square waves increases, the amount of charge per knot increases. Here, the features of the present invention are as follows:
The amount of charge per knot corresponds to the size of the knot. Since the knock signal itself is understood as a rectangular pulse, this method is different from the method of integrating the knock signal and converting it into a gentle curve for detection.
Relatively small knock signals can be detected. 2
It is a point. Therefore, if a method is adopted in which the delay angle from the reference ignition timing signal is controlled in proportion to the amount of charge in the capacitor 94, the optimum delay angle can be set depending on the magnitude of the knock at that time. It is possible. Here, in order to facilitate understanding of this embodiment, the delay signal generator will be explained. terminal voltage,
The signal is received by a voltage follower circuit composed of transistor 107 and resistors 80 and 51, and generates a delayed signal approximately in direct proportion to the voltage of the converter. The delay signal is an angle signal based on the reference ignition timing signal from the ignition control circuit (specifically, the ON signal of transistor 173), and if the voltage of the converter is constant, the reference ignition timing is always at a constant angle. Generates a more delayed ignition timing signal. In the capacitor 97 in FIG. 2, charging and discharging are repeated by the phase-stable multivibrator in the delay signal generator to generate a triangular wave. The switching point from charging to discharging coincides with the reference ignition timing, and the discharging time is a delay time from the reference ignition timing. The switching point from discharging to charging occurs when the terminal voltage of the capacitor 97 reaches the reference voltage (in this example, approximately
1.5V) is reached. Now, assuming that the charging current is i ₁ and the discharge current is i ₂ , the discharging time Td is expressed as Td=T·i ₂ /i ₁ (12). T in equation (12) represents the time from the reference ignition timing signal to the next reference ignition timing signal. Therefore, converting equation (12) into angle Ad results in the following equation. Ad=Td/T×120゜=i ₂ /i ₁ ×120゜ ...(13) 120゜... Crank angle between reference signals in case of 6 cylinders As is clear from equations (12) and (13) , the delay angle is i ₂
is directly proportional to Since this charging current i ₂ is determined by the resistor 51 connected to the emitter of the voltage follower transistor 107, the terminal voltage of the capacitor 94 and the delay angle are directly proportional. Below, we will return to the explanation of the operation of the circuit. diode 10
7, 108 and resistor 173 is a maximum voltage limiting circuit for capacitor 94. The leakage current of the Zener diode 108 causes the capacitor 94 to pass through the diode 107.
In order to prevent the charging voltage from spontaneously discharging, the Zener diode 10 is connected by the resistor 173.
8 is always biased to maintain a constant voltage. As a result, the maximum voltage of the capacitor 94 is clamped, so in a method of setting the delay angle based on the charging voltage, it is possible to set the maximum delay angle on the circuit. The circuit configuration in the middle part of Figure 2 is the noise mask time generator, sampling time generator, and
These are an engine rotational speed detection circuit and a high-speed operation control circuit, and these are all related to the control of the knock amount→voltage conversion circuit. The noise mask time generator and the sampling time generator are constructed by one-shot circuits. The operation will be explained with reference to the timing chart in FIG. 3 and FIG. 2. By interrupting the primary current of the ignition coil 201, an induced voltage as shown in FIG. 3b is generated at the collector of the power transistor 172. The pulse-like voltage that occurs immediately after shutoff is due to the leakage inductance of the ignition coil and has a width of 10 to 20μ.
sec, the height is around 300V. In the embodiment, in order to use this as a trigger pulse for the one-shot circuit, resistors 74 and 75 are inserted in series between the collector of the power transistor 172 and the ground, and from the midpoint, the voltage is divided by the resistors to approximately 1/5 to 1/5 of the collector voltage. 1/8
A voltage dropped to 1 is applied to the base of the transistor 158 via the Zener diode 131.
The Zener diode 131 is for extracting only the peak wave of the induced voltage, and in the embodiment,
I am using a 27V one. transistor 158
The diode 113 connected to the base of the transistor 158 is for compensating for the reverse breakdown voltage between E and B of the transistor 158.
This is to ensure the switching operation. Triggered by the peak voltage of the power transistor 172, the transistor 158 turns on as shown in FIG.
Turn off. In this OFF state, the electric charge stored in the capacitor 95 is reversely charged, and the transistor 1
59,160 base points, (Fig. 2 b, c
point) until it reaches the V _BE threshold.
Here, in this embodiment, one capacitor allows
In order to obtain two types of time, the resistors 34 and 35 are connected in series to form a reverse charging path for the capacitor 95, and from the dividing point to the base of the transistor 160 via the resistor 97, and the connection between the resistor 34 and the capacitor 95. The points are connected to the base of transistor 159, respectively. Therefore, the time for the base potential of transistor 160 to reach the threshold value of V _BE is shorter than that of transistor 159. The OFF time of transistor 159 is determined by resistors 34 and 35.
The OFF time of the transistor 160 is determined by the OFF time of the transistor 159 and the division ratio between the resistors 34 and 35. The operation of transistor 158 is opposite to that of transistor 159 because its base is connected to the collector of transistor 159 through a resistor. On the other hand, transistor 1 that creates the shorter time
The base of a transistor 161 is connected to the collector of the transistor 60, and the transistor 161 outputs a phase-inverted waveform of the transistor 160. The collector voltage of the transistor 161 and the transistor 15 are connected to the base of the transistor 162.
9 collector voltage is resistor 39, diode 1
14 and 115 in an AND relationship, and the transistor 162 is turned on only from the time when the transistor 101 is turned off until the transistor 159 is turned on. The operations of the transistors 158 to 162 have been described in sequence from the time of sparking, and the waveforms in FIGS. 3a to 3g illustrate these operations. In the sample hold circuit described above, the collector of the transistor 162 is connected to the base of the transistor 152 through the resistor 19 and the diode 116, and the collector of the transistor 161 is connected to the capacitor 93 through the resistor 20, and when the transistor 161 is turned on, Therefore, while noise is being output from the amplifier, the previous noise level data held in the capacitor 93 is released, and the operation of sampling new noise level data by turning on the transistor 162 is as described above. This part functions as a sampling time generator. In addition, between points 0 and a' on the time axis in Fig.
In order to process the output voltage V _O2 of the comparator 194, which has reached the current level, the collector of the transistor 158 is connected via the diode 120 to the collector of the transistor 155 which constitutes the knock amount to voltage conversion circuit in FIG. It is carried out by As a result, timing chart c, h in Figure 3
As is clear from the waveform diagram, during the time when spark noise appears in the amplifier output, the collector current of the transistor 155 is drawn into the transistor 158 via the diode 120, so that charge is stored in the capacitor 94. There isn't. That is, the transistor 158 has the function of a noise mask time generator. Here, the collector of transistor 158 and resistor 3
The diode 112 inserted between 3 and 3 is
This is for temperature compensation of the threshold value of the base-emitter voltage _VBF of the transistors 159 and 160, thereby reducing temperature changes in the sampling time and noise mask time. Transistors 156, 157, diode 10
9, 110, 111 and resistors 28, 29, 32 is a voltage reduction circuit for returning the ignition timing, which has been delayed due to the generation of the knock signal, to the reference position. The transistor 156 and the diode 109 are a constant current circuit using two adjacent IC transistors on one chip, and the current value is determined by the resistor 2 connected to the power supply.
Determined by 8. The collector of the transistor 157 is connected to the base of the transistor 156, and the base is connected to the collector of the transistor 161 constituting the above-mentioned sampling time generator via diodes 110 and 111, and the transistor 161 is turned on. During this period, the transistor 157 is turned off, and the transistor 156 uses a constant current to draw out the charge from the capacitor 94 to which its collector is connected. As mentioned above, the transistor 161 is turned on for a certain period of time for each ignition, so the capacitor 9
The terminal voltage of No. 4 decreases by a constant voltage each time. Therefore, according to the present system, the ignition timing delayed by the knock signal approaches the reference ignition timing by a fixed angle every time the ignition occurs. This method has the advantage of being able to control the ignition timing in accordance with the operating speed of the engine, compared to naturally discharging the capacitor 94 using a resistor. In other words, in the case of natural discharge, the time required to return to the reference ignition timing is constant, whereas in this method, the time required to return to the reference ignition timing varies in proportion to the engine speed, so it is possible to obtain good control characteristics. Transistors 153, 165, diode 13
2. The portion composed of resistors 24 and 25 is a minimum voltage limiting circuit and a voltage control circuit during high-speed operation. Resistors 21, 24, and 25 (=R ₂₁ , R ₂₄ , R ₂₅ ) are connected in series from the power source to the ground, and a diode 132 is inserted between the resistors 21 and 24. Since the base of the transistor 153 is connected to the connection point between the diode 132 and the resistor 21, the emitter is connected to the capacitor 94 with the voltage drop V _f of the diode 132 and the base-emitter voltage V _BE of the transistor 153 eliminated. It is connected. Therefore, since the transistor 153 forms a voltage follower circuit, the terminal voltage of the capacitor 93 is connected to the diode 132 and the resistor 24.
If the voltage is higher than the voltage at point f, which is the connection point between the Because of this, the lowest value of the terminal voltage of the capacitor 93 is always clamped to the potential at point f. During normal operation of the engine, transistor 165 is always
Since it is ON, the potential at point f is the Zener voltage V _Z112 of the Zener diode 112.
In this example, the potential is divided by 4.
It is set around 1.1V. The voltage V-f at point f is expressed by the following equation. V−f=(V _Z112 −V _f )・R ₂₄ /R ₂₄ +R
₂₄ ...(14) The reason why the minimum voltage limiting circuit is necessary is to prevent the above-mentioned charging current of the delay signal generator from becoming too small and deteriorating the accuracy of the delay angle.In this embodiment, the minimum voltage limiting circuit is required. It is set to 5 μA or more. On the other hand, in high-speed operation, the movement of the transistor 165 is as shown by the broken line and hatching in FIG. V−f=(V _Z112 −V _f )・(R ₂₄ +R ₂₅ )/
R ₂₁ +R ₂₄ +R ₂₅ ...(15) The voltage V-f shown by equation (15) is higher than that shown by equation (14) by R ₂₅ , and is set to about 3V in the example. are doing. At this time, if the terminal voltage of the capacitor 94 is operating at 3V or less, the transistor 153
The terminal voltage is suddenly raised to 3V, which is the same as V-f. When the capacitor 94 is operating at a terminal voltage of 3V or more, the transistor 164 is always turned on at high speed, as shown by the broken line in FIG. The voltage is constantly drawn into the transistor 164 and stops charging, and is gradually discharged by the transistor 156 of the voltage reduction circuit described above and settles to 3V. Here, the movement of transistor 165 at high speed is
ON and OFF occur once each in one cycle, and the voltage at point f alternates between approximately 1V and approximately 3V as expressed by equations (14) and (15). Since the voltage of the capacitor 94, which is lowered by the transistor 156 during one cycle, is converted into an angle of about 0.1 degree, its terminal voltage is clamped at the higher potential of 3V. Therefore, at high speeds, ignition is always delayed by a certain angle (10 degrees in the example) from the basic ignition timing, so ignition is always delayed slightly from the knock zone. Next, resistors 40-50 and 60, 78, comparator 195, transistors 163, 164, diodes 117-122 and 122, 124,
133, and the operation of the rotation detector constituted by the capacitor 96 will be explained with reference to FIGS. 2 and 7. The capacitor 96 includes a resistor 40 and a diode 11.
7 to the collector of the transistor 162, and when the transistor is ON as shown in FIG. 7f and g, the charge of the capacitor is discharged,
When OFF, resistor 41 (R = ₄₁ ) and capacitor 9
6 (C ₉₆ ). When this charging voltage exceeds the reference voltage V _X obtained by dividing the constant power source V _Z125 of the Zener diode 125 by the resistors 42 and 43 (=R ₄₂ , R ₄₃ ), the output of the comparator 195 becomes High. If this reference voltage _V appears, and at higher rotations, only the Low signal is always available. Transistors 163, 164, resistors 45-4
9. The part formed by the diode 121 is a well-known phase stable circuit, but if the signal that turns on the transistor 164 is used as a set signal, and the signal that turns it off is used as a reset signal, the set signal is generated by the diode 119. of transistor 161 connected directly to the base of transistor 163.
The ON signal and the reset signal are an AND signal of the output of the comparator 195 and the OFF signal of the transistor 159 connected by the diode 134. To explain this using the time chart in FIG. 7, regardless of other conditions, the transistor 161
When ON, the base current of transistor 163 is drawn into transistor 161 via diode 119, and transistor 163 is turned on.
OFF, transistor 164 is turned ON. here,
When the voltage drop V _f119 across diode 119 becomes higher than the base voltage threshold of transistor 163,
The effect that cannot be turned off is transistor 16.
It is removed by a diode 121 inserted between the emitter connection point of No. 3,164 and ground. This diode 121 also has a Schmitt effect when both transistors are switched on and off. On the other hand, the reset signal is an AND of the high signals of the comparator 195 and the transistor 159, so it is output at the same time as the set signal is output.
Since the set signal takes priority, there is no effect on circuit operation. When the set signal disappears, that is, when the transistor 161 switches from ON to OFF, the transistor 161 switches from OFF to ON, the capacitor 96 is discharged as described above, and the comparator output also switches to Low. Therefore, almost at the same time as the set signal disappears, the set signal also disappears. The set time at this point is guaranteed by the delay in the operation of transistor 162 relative to 161 and the time constant provided by resistor 117, which drains the charge from capacitor 96. In other words, while transistor 161 is on, transistor 163
The base current of transistor 161 is drawn through diode 119 to transistor 161.
3 is ON, and transistor 164 is OFF.
Next, when the transistor 160 is turned off, the output of the comparator is still high and at the same time, the transistor 159 is turned off, so the base current flows into the transistor 163, and the transistor 163 is turned on and the transistor 164 is turned off. 164 is
When turned OFF, transistor 1 is transferred from its collector.
Since the base current continues to be supplied by the resistor 47 connected to the base of 63, this state is maintained until the next set signal arrives. When the phase stabilizer circuit is reset, transistor 164 is in the OFF state. At this time, a resistor 44 connected to the collector of the transistor,
If only the resistor 49 connected to the power supply among the resistors 47, 49, and 50 is made smaller than the other resistors, the collector voltage can be made sufficiently high. If this voltage is set higher than the reference voltage (negative terminal) of the comparator 195 for rotation detection, which is set by the resistance division between the resistors 42 and 43, the influence on the reference voltage side will be increased by the diode 118. will disappear. On the other hand, when the phase stabilization circuit is set, the transistor 164 is on, so current flows from the reference voltage side through the diode 118 and the resistor 44, and the resistors 43 and 44 are connected in parallel, lowering the reference voltage value. . Therefore, at low speed rotation, the reference voltage takes on a waveform as shown by the solid line in FIG. 7e, giving the comparator a Schmitt effect. Next, when the rotation becomes high speed and the comparator output is always at a low level as described above, the reset signal is no longer output and the transistor 163 is always OFF and the transistor 164 is always ON. For the above reasons, the transistor 164 operates in the same manner as the transistor 114 during low speed rotation, as shown by the solid line in FIG. 7i, and is always turned on during high speed rotation, as shown by the broken line in the same figure. The collector of this transistor 164 is connected via a diode 122 to the collector of a transistor 155 that supplies a charging current for the knock signal to voltage conversion section, and the transistor 164 is turned on.
When the voltage is on, a charging current is drawn into the transistor, and the knock signal
Even if it appears, the knock amount is not converted to voltage. Therefore, during high speed rotation, transistor 164 is always ON.
Due to the normal state, the charging current to the capacitor 94 is stopped, and the terminal voltage is clamped at around 3V as described above. A transistor 16 that switches the minimum voltage of the capacitor 94 in response to high-speed operation and low-speed operation.
The base current of 5 is connected to a resistor 77 connected to the collector of the transistor 160, a resistor 124 connected to the collector of the transistor 164 via a diode 124, a resistor 78 connected to the output of the comparator 195 of the rotation detector, and a resistor 78 connected to the output of the comparator 195 of the rotation detector. It is supplied by four resistors, resistor 62 connected to the collector of transistor 167 in the control circuit. As shown in FIG. 7, the operation of the transistor 160 is opposite in phase to that of the transistor 161, and therefore, the phase is also opposite to that of the transistor 164 during low speed rotation. Therefore, when the transistor 160 is ON, the transistor 164 supplies the base current of the transistor 165 through the resistor 56, and when the transistor 164 is ON, the transistor 160 supplies the base current of the transistor 165 through the resistor 77, so the transistor is always ON during low speed rotation. are doing. On the other hand, during high-speed rotation, the transistor 164 is always
Since the transistor 165 is in the ON state, the transistor 165 is ON only when the transistor 160 is OFF. The moment the ignition key 204 is turned on, the transistor 160 turns on, and the capacitor 96 begins to be charged through the resistor 41. During the time td until the terminal voltage of the capacitor 96 reaches the reference voltage of the comparator 195, the output of the comparator is at a low level, so the transistor 164 is in the ON state during this period. Therefore, the base current of transistor 165 is the same as that of transistor 160,
164 is not supplied from either collector, and the transistor 165 is turned off during time td, the voltage of the capacitor rises to the voltage required for delaying at high speed, and at the time of starting the ignition timing is delayed from the normal ignition timing. The phenomenon of ignition occurs. Engine startability may or may not be better if the ignition timing is delayed from the normal timing.For the former, it is better to leave it as it is, but for the latter, Therefore, it is necessary to take measures against this. To counter this, resistors 60 and 7 were added.
It is 8. That is, the resistor 60 causes the ignition key 204 to
A high signal is output the moment it is turned on, and at high speed, it cannot prevent the transistor 165 from turning on and off. In this embodiment, the transistor 165 is connected to the collector of the transistor 167 in the ignition timing control circuit.
Transistor 165 is turned on at the same time as the ignition key is turned on. Also, the transistor 164 is activated even during low speed rotation.
There is a time when it is ON, but this time is within the noise mask time, so there is no adverse effect on the operation of converting the knock amount to voltage. However, when I turned the starter to start, voltage was generated in the pick-up coil 207, and until now it was ON.
Since the transistor 166 that had been in operation is turned off, the transistor 167 in the next stage is turned on, and there is a time when the base current of the transistor 165 is not supplied and the transistor 165 is turned off. What makes up for this time is
Resistor 78 connected to the output of comparator 195
It is. In other words, from the moment the ignition key 204 is turned on until the capacitor 96 is charged, the current flows from the transistor 167, and after the capacitor 96 is charged and the output of the comparator 195 becomes High, the current from that output flows into the transistor 165. It is always turned on when starting. This can prevent the ignition timing from being delayed from the normal timing at the time of starting. Next, we will briefly touch on the reference ignition position signal generator and the ignition control circuit. Positive and negative alternating current waveforms as shown in FIG. 3m appear in the pickup coil 203 in synchronization with the rotation of the engine. The transistor 166 is normally biased and turned on by the resistor 82, but due to the voltage generated by the pickup coil 203, the
When the point becomes negative, it turns OFF, so the transistor 167, which had been OFF, turns ON, and in the same way, transistor 173 turns OFF, 168 turns ON, 169 and 170 turn OFF, and the power transistor 1 turns OFF.
72 is turned on and the primary current of the second ignition coil 201 flows. Next, when the y point changes relatively steeply from negative to positive, transistor 166 turns on, and 167 follows.
OFF, 173 turns ON. of this transistor
The ON signal is sent to the delay signal generator as a reference ignition timing signal. When transistor 173 turns on, transistor 168 turns off, but since the delay signal from the delay signal generator is sent to the base of transistor 169 at low level,
ON is delayed from the reference position signal by the amount of the delay signal. Hereinafter, this delay is due to the delay of the transistor 170.
ON, followed by OFF of the power transistor 172,
The interruption of the primary current 201 of the ignition coil 201 is delayed by the delay signal. In other words, the ignition timing is delayed.
Transistor 171 is for limiting the maximum value of the primary current. As detailed above, according to the present invention, the delay angle from the reference ignition timing is determined by the strength of the notch.
It can be set optimally each time. Furthermore, since the pulse of the knock signal is detected as it is, relatively small knock signals can be detected.

[Brief explanation of the drawing]

FIG. 1 is a block diagram establishing an embodiment of the present invention, FIG. 2 is a circuit diagram of an embodiment of the present invention, and FIG. 3 a
~m is a timing chart, Figures 4 and 5A
~D are detailed explanatory diagrams of the sample and hold circuit and comparator, Figure 6 is D', E is an operation diagram of knock amount → voltage conversion, Figures 7 a to h are timing charts of the rotation detection section, Figure 8 a , b are principle diagrams showing a system according to an embodiment of the present invention.

Claims (1)

[Claims] 1. A first means for converting the vibrations of the internal combustion engine into an electrical signal, and a second means for extracting and outputting an electrical signal corresponding to a frequency band of vibrations related to the engine's knots determined in advance from the output signal of the first means.
a third means for maintaining the level of a signal voltage that is not an electrical signal related to engine knock among the output signals of the second means; and a signal voltage level maintained by the third means. , a fourth means for comparing the output signal of the second means and extracting only the electrical signal related to the engine knock;
fifth means for converting the electrical signal regarding the knock taken out by the means to a voltage level; and the output signal of the second means is amplified and inputted to the comparator, and in one ignition cycle period, An ignition system with a knock signal detection device, characterized in that the number of output pulses of the comparator is configured to vary depending on the intensity of the knock, and the ignition timing is delayed in accordance with the number of output pulses of the comparator.