JPS631461B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an ignition timing control method that detects knocking in an internal combustion engine and uses this detection signal to control the amount of retardation of ignition timing to an optimal state that prevents knocking.

In recent years, an ignition timing control method has been proposed in which the ignition timing is retarded when knocking of an internal combustion engine is detected, and the ignition timing is advanced when no knocking occurs, so that the amount of advance and the amount of retardation are the same.

When controlling the ignition timing using this ignition timing control method, the average ratio of the number of times knocking occurs to the number of times no knocking occurs is 1:1, and the amount of advance and the amount of retardation are balanced. Even if a small knock (trace knock) is detected, a knock occurs once every two times, so the engine noise is high and the driver feels uncomfortable.

The present invention has been made in view of the above-mentioned drawbacks, and it is necessary to determine whether or not the internal combustion engine is knocking, and to determine whether or not the internal combustion engine is knocking. to reduce the retard amount when the retard amount is not present, to store this retard amount in correspondence with the operating state of the engine including at least the rotational speed for each operating state classification, and to reduce the retard amount at that time among the stored retard amounts By controlling the ignition timing according to the amount of retardation in the classification corresponding to the engine operating state, the purpose is to reduce the occurrence rate of knocking and to maximize the engine torque.

Furthermore, in the present invention, by storing the retardation amount for each engine operating state category, the ignition timing is controlled to the optimum ignition timing that can suppress the occurrence of knocking and draw out sufficient torque even during transient times such as during engine acceleration and deceleration. It also aims to compensate for engine deterioration over time.

The present invention will be described below with reference to embodiments shown in the drawings. In FIG. 1, an internal combustion engine 1 is a well-known 4-cylinder 4-stroke spark ignition engine.
The air-fuel mixture generated by the carburetor 2 is sucked in through the intake manifold 3, and a high voltage is applied to the ignition plug (not shown) from the ignition coil 4 through the distributor 5.

6 is a rotating disk attached to the crankshaft,
This disc 6 is provided with two protrusions that are divided into two equal parts, and the protrusions correspond to the top dead center (TDC) of the internal combustion engine. 7 is a commercially available electromagnetic pick-up,
A signal is output at a location facing the protrusion of the disk 6 (top dead center of each cylinder).

Reference numeral 8 designates a first waveform shaping circuit, which shapes the waveform of the signal from the contacts of the distributor 5, and is a known circuit that includes a contact chattering waveform absorption circuit, and FIG. 3A shows the waveform. The shaping circuit 8 is an output waveform. Here, the distributor 5 is equipped with a centrifugal governor and an intake negative pressure responsive diaphragm, whereby the advance characteristic is the basic ignition timing corresponding to the engine speed and engine load.
MBT (Minimum advance for Best Torque)
It is set to . Therefore, the data distributor 5
The signal from the contact point is the same as igniting the MBT.

The vibration detector (sensor) 9 is a piezoelectric type mounted on the cylinder block of the engine 1, and outputs a voltage according to vibration acceleration using a piezo element.

10 is a second waveform shaping circuit, and 11 is an intake negative pressure sensor. Reference numeral 12 denotes a knocking discrimination circuit that detects knocking based on a signal from the vibration sensor 9. Reference numeral 13 denotes a learning control circuit which calculates and stores the ignition retard amount from the MBT based on the discrimination signal of the knocking discrimination circuit 12, and performs learning control by repeating these two processes. outputs a voltage signal representing a retard time corresponding to the amount. 1
4 is a conversion circuit, which is the MBT of the first waveform shaping circuit 8.
A signal representing the delay time and a signal representing the retard amount from the learning control circuit 13 are input, and the MBT
A signal for controlling the ignition timing is output at a timing delayed by a time corresponding to the retard amount. 16
is an AND gate, into which the signal from the conversion circuit 14 and the signal from the first waveform shaping circuit 8 are input, and generates a signal for controlling the ignition timing and the energization start timing (that is, the energization period) of the ignition coil 4. . Reference numeral 17 denotes a known igniter, which controls energization of the ignition coil 4 and ignition based on the signal from the AND gate 16.

Next, the knocking discrimination circuit 12 will be explained with reference to FIGS. 2 and 3. FIG. 2 shows a circuit diagram, and FIG. 3 shows an operating waveform diagram. In FIG. 2, the control pulse generating circuit 12a is composed of two sets of monostable multivibrators, an R-S flip-flop, and a NOR gate. The MBT signal from the first waveform shaping circuit 8 is input to the input 1B of the monostable multivibrator 120, and the input 1A is grounded. The monostable multivibrator 120
uses monostable multivibrator IC product number SN74123 manufactured by Texas Instruments Inc. together with monostable multivibrator 125 described later. Connect capacitor 1 between terminals 14 and 15 of the IC.
21. By connecting a resistor 122 between the terminal 15 and the power supply Vc, a pulse of about 10 μS determined by the capacitor 121 and the resistor 122 is generated at the output Q ₁ after the output of the first waveform shaping circuit 8 rises. .

Next, the NOR gates 123 and 124 are connected to form an R-S flip-flop as shown in the figure.
One input terminal of the NOR gate 123 is connected to the _Q1 output of the monostable multivibrator 120, and a signal representing approximately the top dead center (TDC) of each cylinder shown in FIG. 3B is input to the input of the NOR gate 124. Ru. Therefore, the output of the NOR gate 123 has the waveform shown in FIG. 3C, and the output of the NOR gate 124 has the waveform shown in FIG. The monostable multivibrator 125 is the same as the monostable multivibrator 12.
0, and when a top dead center (TDC) signal is input, a pulse signal of about 100 .mu.S shown in _FIG . NOR gate 128 is
It performs a NOR operation between the output signal of the NOR gate 124 and the output signal of the monostable multivibrator 125, and outputs the signal shown in FIG. 3E.

The buffer amplifier 12b is an impedance converter, which converts the output signal of the vibration sensor 9 into a low impedance signal, and simultaneously amplifies the signal.
The absolute value circuit 12c is a known circuit that takes the absolute value of a positive or negative signal, and operates to return the negative side portion of the vibration waveform signal of the vibration sensor 9 input via the buffer amplifier 12b to the positive side. Thus, if the output signal of the buffer amplifier 12b is as shown in FIG. 3F, the output signal of the absolute value circuit 12c is as shown in FIG. 3G.

The first integrator 12d converts the waveform shown in FIG.
It integrates the period from MBT to top dead center,
OP amplifier 130, resistors 131, 132, 133,
capacitor 134 and analog switch 135,
It consists of 136 analog switches, 1
35 is turned on and off by the signal shown in FIG. 3E from the control pulse generating circuit 12a, and the analog switch 136 is turned on and off by the signal shown in FIG. 3C.

When the C signal shown in FIG. 3C is "1", the analog switch 136 is turned on, and when the E signal shown in FIG. 3E is "1", the analog switch 135 is turned on.
is on, the output of the first integrator 12d is
When the signal becomes OV and the C and E signals are inverted to "0", the analog switches 135 and 136 are both turned off and the first integrator 12d starts integrating in the negative direction. The integration time is the time until the analog switch 135 is turned on again, but the absolute value circuit 1
Since the output signal 2c is input during the time when the analog switch 136 is turned off, the output signal is inputted until the time t ₁ , that is, the top dead center of each cylinder.

The period t in which the D signal is "1" shown in FIG. 3D is the holding time of the integrator 12d, and the output does not change. Therefore, the output signal waveform of the first integrator 12d becomes as shown _{in FIG} . −∫ ^T1 ₀ Gdt”.

The second integrator 12e has almost the same circuit configuration as the first integrator 15d, except that a constant voltage is applied by a voltage divider 137 instead of the vibration waveform as an integral input. Therefore, the output signal waveform _of the second integrator 12e becomes as shown in FIG.
Then, it becomes "-∫ ^T1 ₀ VRdt", which is proportional to time T ₁ .

The divider 12f divides the output voltage of the first integrator 12d by the output voltage of the second integrator 12e, and is a divisible multiplier (8013 manufactured by Intersil).
140, resistors 141, 142, voltage divider 143, diodes 144, 145, and analog switch 146.
46 is controlled to be turned on or off by the E signal shown in FIG. 3E, and a negative power supply voltage Vs is applied thereto.

Since the analog switch 146 is on when the E signal is "1", the negative power supply voltage Vs is input to the X and Z input terminals of the multiplier 140. When the E signal is "0", the output voltage of the first integrator 12d is input to the Z input terminal, and the output voltage of the second integrator 12e is input to the X input terminal. The multiplier 140 is adjusted by adjusting the voltage divider 143.
Since the calculation of 10Z/X is performed, the output V ₃ is becomes. Note that K ₁ is a proportionality constant (=10/VR).

The sample and hold circuit 12g includes a sample and hold IC (1H5110 manufactured by Intersil) 147, a resistor 148, a capacitor 149, and a voltage divider 150.
The output signal of the divider 12f is sampled and held according to the D signal shown in FIG. 3D, and a signal as shown in FIG. 3J is output. Here, the output voltage of the IC 147 is the same as the output voltage V ₃ of the divider 140, and this V ₃ is the output voltage of the voltage divider 150.
The voltage is divided by and output as voltage _V4 .

Therefore, the output voltage V ₄ is expressed by the following equation.

However, K ₂ is the voltage division ratio of the voltage divider 150, K=
_K1 and _K2 . As can be seen from the above equation, this output voltage V ₄ represents the average value of the vibration values of each cylinder of the engine 1 from the MBT point to the top dead center.

The comparator 12h receives the output signals of the absolute value circuit 12c and the sample hold circuit 12g, and compares the instantaneous value of vibration detected by the vibration sensor 9 with the output signal of the absolute value circuit 12c and the sample hold circuit 12g.
The average value from MBT to top dead center is compared, and if the instantaneous value is greater than the average value, a signal of "1" is output.

Therefore, the knocking of the engine 1 occurs after the top dead center of each cylinder, and when the knocking occurs, the instantaneous value of the vibration becomes larger than the average value, so when the knocking occurs, the comparator 12h changes as shown in FIG. 3K. Outputs a pulse signal.

As described above, it is possible to determine whether or not knocking is occurring in the engine 1 based on the output signal of the comparator 12h, that is, the output signal of the knocking determination circuit 12.

Next, the learning control circuit 13 will be explained using FIG. 4. Reference numeral 131 denotes a pulse width-to-voltage conversion circuit that converts the output of the second waveform shaping circuit 10 into a voltage according to the rotational speed, and the detailed circuit is well known and will therefore be omitted. 132 is an A/D conversion circuit that converts the output of the pulse width/voltage conversion circuit 131, the output of the intake negative pressure sensor 11, and the output of the knocking discrimination circuit 12 into digital quantities. 133 is a microcomputer that controls the amount of retardation from MBT; 134
is a non-volatile memory that stores retard amounts for each engine operating state category, that is, in correspondence with engine speed and suction negative pressure. 135 is a D-A that converts the output of the microcomputer 133 into an analog quantity.
It is a conversion circuit.

FIG. 5 is a flowchart for schematically explaining the functions of the microcomputer 133.
This will be explained below. When the engine is started and the processing routine starts, first in step 1301, the A-D
The rotational speed Ne and intake negative pressure Pi, which have been converted into digital quantities by the conversion circuit 132, are read. Next, in step 1302, the value of the retardation amount Q ₀ (MBT, that is, the retardation amount from the set advance angle) corresponding to the rotational speed Ne and the suction negative pressure Pi is read from the memory 134, and this retardation amount Q ₀ is rotated. It is converted into a delay time _TQ according to the number and outputted to the DA conversion circuit 135. Next, in step 1303, the retard amount Q ₀ outputted in step 1302 is
Based on the signal from the knocking determination circuit 12, it is determined whether or not knocking has occurred in the engine when the engine is ignited at the ignition timing determined by the _engine .
Add Q. In other words, the amount of retardation is increased in order to suppress the occurrence of knocking. Next, in step 1305, this new retard amount _Q0 is written to the address in the memory 134 corresponding to the current operating state, that is, the current rotational speed Ne and intake negative pressure Pi, and the data is updated. When this step 1305 is completed, the step is repeated.
Return to 1301. If no knocking has occurred, proceed from step 1303 to step 1306, and then
In 1306, the rotation speed Ne read in the previous routine
and intake negative pressure Oi and Ne read in this routine,
If they are equal or the difference is very small below a predetermined value, that is, if the engine operating range is the same as the previous time, the process proceeds to steps 1307 and 1308, and knocking occurs only for a predetermined ignition circuit X period. Performs processing to determine whether or not it has been removed. That is, in step 1307, 1 is added to the count value N to process N=N+1, and in step 1308 it is determined whether the count value N is greater than the set number of ignitions X, and if N<X, the process returns to step 1301. Then, the above-mentioned process is repeated, and when N≧X, the process advances to step 1309, where the count value N is reset to 0, and then, at step 1310, ΔQ is subtracted from the retard amount _Q0 . In other words, the engine operating state (Ne, Pi) does not change, and the number of ignitions is set to
When it does not occur during this period, the retard amount Q ₀ is reduced to bring the ignition timing closer to the MBT and work to optimally draw out the engine torque. Next step 1311
Then, it is determined whether the retardation amount Q ₀ is 0 or more. If it is 0 or more, the process proceeds to step 1305, and this Q ₀ is stored in the memory 13 corresponding to the operating state (Ne, Pi) at that time.
4, and return to step 1301. If the retard amount Q ₀ is negative (Q ₀ <0) in step 1311, the process advances to step 1312, where Q ₀ =0, and the step
Proceed to 1305. In other words, the retard amount Q ₀ is designed not to take a negative value, so that the ignition timing is not advanced beyond the MBT. Above steps
At 1306, the rotation speed Ne in the previous processing routine,
If the intake negative pressure Pi is different from the current values Ne and Pi, the process proceeds to step 1313, resets the count value N to 0, and returns to step 1301. In other words, when the engine operating state (Ne, Pi) changes, the retard amount Q ₀ is not rewritten.

The retardation amount Q ₀ is stored in the memory 134 in the form of a map using the rotational speed Ne and the intake negative pressure Pi as parameters, as shown in FIG.

Retard amount Q ₀ from microcomputer 133
The digital signal after being converted as a retard time _TQ based on the rotational speed is converted into a voltage signal by a DA conversion circuit 135 and supplied to the conversion circuit 14.

Next, the conversion circuit 14 will be explained using the electric circuit shown in FIG. An analog switch 141 is turned on when the output of the first waveform shaping circuit 8 (shown in FIG. 3A) becomes "1", that is, at MBT, and is turned off when it becomes "0". 142 is a capacitor forming an integrating circuit, which integrates the voltage signal representing the retard time from the MBT from the learning control circuit 13 while the analog switch 141 is on. 143 is an analog switch that controls the discharge of this capacitor 142, and this analog switch 143 is turned off when the output of the first waveform shaping circuit 8 becomes "1" and becomes "0".
When this happens, it turns on. 144 is a capacitor 142
This is a comparator that compares the integrated output of the Vref with a preset comparison level Vref.

To explain the operation of the conversion circuit 14, when the MBT signal is input from the first waveform shaping circuit 8 (that is, when the output becomes "1"), the analog switch 141 is turned on, and the delay from the learning control circuit 13 is turned on. A capacitor 142 integrates a voltage signal representing angular time, and this integrated output is a comparison level of a comparator 144.
When it becomes higher than Vref, the comparator 144 outputs a high level signal. In other words, this conversion circuit 14 retards the MBT signal from the first waveform shaping circuit 8 by the amount of retardation.
A signal delayed by a time corresponding to Q ₀ is generated, and this delayed signal is supplied to the igniter 17 via the AND gate 16. Therefore, the ignition timing is controlled to a timing that is retarded by the retardation amount Q ₀ calculated from the MBT to prevent knocking.

In the embodiments described above, engine speed and intake negative pressure are used as factors for determining the engine operating state classification, but other factors such as engine cooling water temperature may also be used. Further, although the above embodiment does not control the ignition retard amount for each cylinder, it is of course possible to control the ignition retard amount for each cylinder.

As described above, the present invention determines whether or not the internal combustion engine is knocking, and by this determination, if the engine is knocking, the amount of retardation of the ignition timing is increased, and if the engine is not knocking, the ignition timing is retarded. To reduce the amount of retardation, to store this retardation amount in correspondence with the operating state of the engine including the rotation speed for each operating state classification, and to correspond to the engine operating state at that time among the stored retardation amounts. It is characterized by controlling the ignition timing according to the amount of retardation in each category, and has the excellent effect of maximizing engine torque while reducing the occurrence rate of knocking. Furthermore, since the amount of retardation is memorized for each engine operating state category and control is performed using this memorized value, the ignition timing can be controlled to the optimum ignition timing to quickly suppress knocking and maximize torque even during transients such as during engine acceleration/deceleration. Furthermore, even if the engine is stopped once, by memorizing the retardation amount, the optimum ignition timing can be controlled immediately the next time the engine is operated.Furthermore, the memorized value of this retardation amount can be updated to perform learning control. This has the effect of being able to deal with changes over time in various parts of the engine. Furthermore, even if the stored retard amount is lost for some reason, the internal combustion engine will be ignited at the basic ignition timing depending on the engine operating condition, and after an appropriate period of time, the internal combustion engine will be ignited at the basic ignition timing depending on the engine operating condition. The amount of retardation of the ignition timing from the basic ignition timing is calculated and the internal combustion engine can be operated again at the optimum ignition timing, and the amount of retardation is reduced when no knocking occurs continuously for a predetermined period of time. Also, it is determined whether or not the determination result of knocking is in the same engine operating region as the previous time, and from the time when it is determined that the determination result is not in the same engine operating region as the previous time, the retardation amount is increased when no knocking is occurring. Since a new calculation is started for a predetermined period of time to reduce the engine speed, it has the excellent effect of preventing erroneous learning immediately after the engine operating range changes.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention;
Figure 2 is an electric circuit diagram of the knocking discrimination circuit shown in Figure 1, Figure 3 is a signal waveform diagram of each part in Figure 2, and Figure 4
The figure is a detailed block diagram of the learning control circuit shown in Fig. 1, and Fig. 5 is a flowchart for explaining the general functions of the microcomputer shown in Fig. 4.
Figure 6 is a schematic map of the memory shown in Figure 4;
FIG. 7 is an electrical circuit diagram of the conversion circuit shown in FIG. 1. 1... Internal combustion engine (engine), 8... First waveform shaping circuit, 9... Vibration detector (sensor), 12
...Knocking discrimination circuit, 13...Learning control circuit, 133...Microcomputer, 134...
...Nonvolatile memory, 14... Conversion circuit.

Claims (1)

[Claims] 1. Setting the basic ignition timing according to the engine operating condition, determining whether or not the internal combustion engine is knocking, and increasing the amount of retardation from the basic ignition timing if the internal combustion engine is knocking based on this determination. and to reduce the retard amount when no knocking occurs continuously for a predetermined period of time, and to have this retard amount separate from the basic ignition timing, and to adjust the retard amount to at least the operating state of the engine including the engine speed. to store information for each operating state classification in correspondence with the above, to determine whether or not the above-mentioned knocking determination result is in the same engine operating region as the previous time, and to determine whether the determination result is not in the same engine operating region as the previous time. The method includes starting a new calculation for a predetermined period of time to reduce the amount of retardation when no knocking is performed, and calculating the retardation of the category of the stored retardation amount that corresponds to the engine operating state at that time. An ignition timing control method characterized by controlling ignition timing according to the amount of ignition.