JPS611868A - Ignition timing controller - Google Patents

Abstract

PURPOSE:To avoid knocking and to improve the operationability by providing an ignition timing control means for correcting the basic ignition timing on the basis of the learning value of lagging angle correction corresponding with the operating condition fed from a memory and setting the final ignition timing. CONSTITUTION:Basic value setting means (c) will set the basic firing timing for high octane number fuel. Memory (e) will learn and store the output from correction operating means (d) when the knocking is not occurring or being suppressed. While ignition timing control means (f) will read the learning value of lagging angle correction from the memory (e) to correct the basic ignition timing on the basis of said learning value thus to set the final ignition timing. Consequently, the ignition timing can be optimized for wide operating range to avoid the ignition timing thus to improve the operating performance of engine.

Description

[Detailed description of the invention] (Technical field) The present invention relates to an ignition timing control device for an engine.

(Prior art) In recent years, from the viewpoint of air pollution, the use of high octane fuel (hereinafter referred to as "high octane") containing lead compounds such as tetraethyl lead has been restricted, and the use of unleaded normal octane fuel (hereinafter referred to as "high octane") has been restricted.
The engine's compression ratio is designed to be relatively small to suppress knocking, assuming that it will be used as a regular engine.

On the other hand, considering the efficiency of the engine, a higher compression ratio is preferable, and high-octane engines that do not contain lead compounds have recently been developed, so engines intended for use with high-octane engines are once again being provided. However, since the ignition timing at which knocking occurs is different between high-octane and regular engines, it is necessary to appropriately adjust the ignition timing in engines that use both.

A conventional ignition timing control device of this type is described in, for example, Japanese Patent Laid-Open No. 143169/1983.

This device performs ignition timing control using a preset required advance value for high-octane engines, and when knocking is detected under predetermined conditions (for example, within a predetermined rotation range (1500 to 4000 rpm)), performs processing to correct the ignition timing to the retard side by a predetermined amount, and continues this ignition timing correction processing until the engine stops after the correction or until a certain period of time has elapsed to suppress knocking and prevent engine damage. etc. are prevented.

Note that the reason why knocking detection is limited to a predetermined rotation range is as follows. That is, in a rotation range lower than 1500 rpm, knocking is likely to occur, especially when the load is high, even when using a high-octane engine. Therefore, in such a low rotation range, it is difficult to determine whether knocking is occurring due to a difference in octane number, and even if knocking is detected, the ignition timing is not switched to regular ignition timing.

On the other hand, in a high rotation range of 4000 rpm or more, vibration noise is large and it is difficult to accurately determine knock vibration.

Therefore, the ignition timing is not changed because it is expected that there will be a misjudgment of knocking. The ignition timing for high-octane engines is also set in an advanced range where knocking vibrations hardly occur.

However, in such conventional ignition timing control devices, operation is started based on the optimum ignition timing for high-octane engines, and if knocking occurs within a predetermined rotation range, the ignition timing is subsequently retarded by a predetermined amount. Since the ignition timing was configured to be corrected and controlled for regular use, when the occurrence of knocking is detected, the ignition timing is corrected to the retarded side even if the engine goes out of the above-mentioned predetermined rotation range. Ru. Therefore, even though knocking is likely to occur in a rotation range lower than 1500 rpm during regular use, the amount of correction to the retard side is limited to the above-mentioned predetermined amount. Therefore, in such a low rotation range, knocking cannot be sufficiently suppressed, which may lead to engine overheating, reduced performance, and even damage. On the other hand, 4000rp
In the high rotation range of m or more, it is preferable to advance the ignition timing as long as knock vibration does not occur as it will improve output, but the ignition timing is set to such an extent that almost no knock vibration occurs, and the ignition timing is not necessarily sufficiently advanced. Not yet. This resulted in insufficient output in the high rotation range and poor fuel efficiency.

(Purpose of the Invention) Therefore, the present invention sets the basic ignition timing for high-octane engines according to the operating condition, and when knocking occurs, the basic ignition timing is corrected to the retarded side so as to suppress knocking, and the knocking is suppressed. The retardation correction amount when the retardation correction is performed is learned as the one corresponding to the driving state at that time, and on the other hand, when knocking is not occurring, the retardation correction amount for performing the retardation correction is assumed to be zero. The basic ignition timing is corrected based on these learning values, and the final ignition timing is set, so that it can be used for a wide range of operating conditions, regardless of the use of fuel with a different octane rating. to always keep the ignition timing optimal,
The purpose is to avoid knocking and improve engine drivability.

(Structure of the invention) FIG. 1 is an overall configuration diagram for clearly explaining the present invention.

The operating state detecting means a detects the operating state of the engine, and the knock detecting means detects knock vibrations generated in the engine. The basic value setting means C sets the basic ignition timing for high octane fuel based on the operating condition,
The correction amount calculation means d calculates a retardation correction amount to correct the basic ignition timing to the retard side so as to suppress knocking when knocking occurs, and calculates the retardation correction amount when knocking does not occur. stop. On the other hand, the storage means e learns the output of the correction amount calculation means d when knocking is not occurring and when knocking is suppressed as corresponding to the driving state at that time, and stores this learned value, The ignition timing control means f reads the learned value of the retardation correction amount corresponding to the operating state from the storage means e, corrects the basic ignition timing based on this learned value, and sets the final ignition timing. The ignition means g ignites the air-fuel mixture based on the ignition signal from the ignition timing control means f, so that the ignition timing is always optimally controlled over a wide range of operating conditions to avoid knocking.

(Example).

Hereinafter, the present invention will be explained based on the drawings.

2 to 12 are diagrams showing one embodiment of the present invention.

First, the configuration will be explained. In FIG. 2, reference numeral 1 denotes a six-cylinder engine. Intake air is supplied from an air cleaner 2 to each cylinder through an intake pipe 3, and fuel is injected by an injector 4 based on an injection signal St. Each cylinder is equipped with an ignition plug 5, and the ignition plug 5 receives a high-voltage pulse P from an ignition coil 7 via a distributor 6.
i is supplied. The ignition plug 5, distributor 6, and ignition coil 7 constitute an ignition means 8 that ignites the air-fuel mixture, and the ignition means 8 generates a high-pressure pulse pi based on the ignition signal Sp. Then, the air-fuel mixture in the cylinder is ignited and exploded by the high pressure pulse Pf, and is discharged through the exhaust pipe 9 as exhaust gas. The flow rate Qa of intake air is detected by an air flow meter 1o and controlled by a throttle valve 11 in the intake pipe 3. Vibration Ve of engine 1 body
is detected by a knock sensor (knock detection means) 12, and the temperature Tw of the cooling water flowing through the water jacket is detected by a water temperature sensor 13. Further, the crank angle Ca of the engine 1 is detected by the crank angle sensor 14, and the crank angle sensor 14 detects a reference crank angle CaI and a 2° crank angle C based on 70° before compression top dead center of each cylinder.
(The air flow meter IO1 water temperature sensor 13 and crank angle sensor 14 constitute the operating state detecting means 15, and the operating state detecting means 15 and the knock sensor] 2
The signals from the control unit 16 are input to the control unit 16.

The control unit 16 has functions as a basic value setting means, a correction amount calculation means, a storage means, and an ignition timing control means, and as shown in detail in FIG.
ROM22, RAM23, interface circuit u, 2
5. Knock vibration discrimination circuit, multiplexer 27, A/
The D conversion conversion circuit is composed of a power counter 29, a clock circuit 3o, and a flip-flop circuit 31. The knock vibration determination circuit 26 compares the frequency and amplitude of the vibration Ve of the engine 1 body with respective predetermined reference values to detect a predetermined knock vibration. For example, a vibration ■e having a frequency of about 7 KHz and higher than a predetermined vibration level is detected as a knock vibration, and the knock signal Sn becomes (H) when knock vibration is occurring and (L) when knock vibration is not occurring. is output to the flip-flop circuit 31. Flip-flop circuit 3]
is set when knock vibration is occurring, and the first reset signal Sr! from the interface circuit U, which will be described later, is set.
The CPU 2 outputs the discrimination signal sh, which becomes ()I) when set and becomes (L) when cleared.
Output to 1.

Signals Qa and Tw from the air flow meter IO and the water temperature sensor 13 are input to the multiplexer 27, and the multiplexer 27 switches these signals Qa and Tw at predetermined timing to power the A/D conversion circuit. The A/D conversion circuit converts the signals Qa and TW inputted as analog signals into digital signals for the CPU 21 .

Further, the crank angle 2° signal Cυ from the crank angle sensor 14 is input to the counter 29, and the counter 4 counts this crank angle 2° signal CaZ to calculate the rotation speed N of the engine 1 and outputs it to the CPU 21. do. CPU2
The crank angle reference signal Cat from the crank angle sensor 14 is further input to the CPU 21.
According to the program written in the RAM 2, necessary external data is taken in, data is transferred to and from the RAM 23, and arithmetic processing is performed, and the processed data is sent to the interface circuits 24 and 25 as necessary. Output to. Note that the CPU 21 executes arithmetic processing every 10 m5 based on the clock signal Sφ from the clock circuit 30. Crank angle signals Cat and Cd2 from the crank angle sensor 14 are further input to the interface circuit 24, and the ink-face circuit U generates an ignition signal at a predetermined ignition timing based on data and signals Cat and C6Lz from the CPU 21. sp is output to the ignition means 8.

Further, the crank angle reference signal Cat is similarly input to the interface circuit δ, and the interface circuit 5 outputs the injection signal 31 to the injector 4 based on the data from the CPU 21 and the signal Cai.

As shown in detail in FIG. 4, the interface circuit 24 is broadly divided into an advance angle cent circuit 41, a standby circuit 42, and an ignition signal generation circuit 43. The lead angle set circuit 41 includes a clip-flop circuit 44 and an AND circuit 45.
, counter 46, register 42 and comparator 4
The flip-flop circuit 44 is set by the crank angle reference signal Cat. When the clip-flop circuit 44 is turned on, the AND circuit 45 opens its gate and outputs the 2° crank angle signal C612 to the counter 46, and the counter 46 counts the 2° signal C612.

On the other hand, the final advance angle value F from the CPU 21 is stored in the register 47.
ADV is cented. And comparator 48
When the output of the counter 46 (the count value of the 2'' signal Cα2) becomes equal to the basic advance value FADV, it outputs the first reset signal Sr, resets the clip-flop circuit 44 and the counter 46, and resets the clip-flop circuit 44 and the counter 46. The reset signal Sr□ is output to the standby circuit 42, the ignition signal generation circuit, and the above-mentioned clip-flop circuit 31.The standby circuit 42 includes a flip-flop circuit 49, an AND circuit,
It is composed of a counter 51, a register 52, and a comparator, and the flip-flop circuit 49 is set by the first reset signal Sr1 from the comparator 48. When the flip-flop circuit is set, the AND circuit 50 opens the gate and outputs the crank angle 2° signal CtL.
z is output to the counter 51, and the counter 51 counts this 2° signal C0. On the other hand, the CPU
The de-energization angle value TDWE from 21 is set. Then, when the count value of the counter 51 becomes equal to the de-energization angle value TDWE, the comparator 53 outputs a second reset signal S.
r2 to reset the clip-flop circuit 49 and counter 51, and also output the second reset signal S.
r2 is output to the ignition signal generation circuit 43. The ignition signal generating circuit is composed of a flip-flop circuit 54 and a transistor Q1, and the flip-flop circuit is set by a second reset signal Sr2 and reset by a first reset signal Sr. Then, at the reset timing, the transistor Q1 is turned off and the ignition signal Sp corresponding to the final ignition timing GK (details will be described later) is output. This ignition signal S
The primary current from the pantry 5 that is being supplied to the ignition coil 7 is cut off by p, and a high voltage pulse p is applied to the spark plug 5.
t occurs (note that the distribution value is not shown). .

Next, the action will be explained.

In general, if a regulator is mistakenly used in an engine designed for high-octane use, or if a regulator is unavoidably used because high-octane is not available, knocking will occur frequently and the engine will not be able to fully demonstrate its performance. In the worst case scenario, the engine may be damaged. This is because the octane number, which indicates anti-knock properties, is different between the two. For this reason, when knocking occurs under predetermined conditions, the ignition timing is automatically switched to regular ignition timing, but according to this method, knocking can be appropriately avoided under various operating conditions other than the above predetermined conditions. Can not. In addition, emphasis is placed on avoiding knocking, and it is desirable to advance the ignition timing as much as possible within the range where knocking does not occur, in terms of improving output and fuel efficiency, but this kind of control No emphasis is placed on it.

Therefore, in this embodiment, the concept of learning control is adopted, and the retardation correction amount when knocking is suppressed is sequentially learned as corresponding to the driving state at that time, and the same learning is performed for the next same driving state. By determining the retardation correction amount from the value, engine performance is improved by advancing the ignition timing as much as possible while avoiding knocking.

5 to 9 and FIG. 11 are flowcharts showing the engine control program written in the ROM 22, and P and -P in the figures indicate each step of the flowchart.

The engine control by this program is performed by the CPU 21, and the processing by the CPU 21 can be roughly divided into the following three types of processing.

(I) Back Ground Job (hereinafter referred to as BG
(referred to as J processing). When an interrupt occurs during this BGJ processing, the BGJ processing is temporarily interrupted and the corresponding interrupt processing is executed with priority. The interrupt processing is of the following two types depending on the corresponding interrupt signal.

(II) Interrupt processing A Interrupt processing by clock signal Sφ every 10m5 (Ill
) Interrupt processing B Interrupt processing by crank angle reference signal Cai Figure 5 shows (1
) This is a flowchart of a program that performs BGJ processing.
When the N position is reached, power is supplied to the control unit 16 and the steps proceed. First, initial processing is performed at P. Initial processing involves initializing the memory contents of the RAM 23, setting initial values for the registers 47 and 52, and finally setting interrupt permission. Next, at P, it is determined whether or not the crank angle reference signal CaL has been input, and if it has been input, it is determined that the engine l has started rotating, and P3
If no input is received, it is determined that rotation has not yet started and the process waits at step P2.

When the engine 1 starts rotating, the mixture ratio correction coefficient KMP is calculated using P. This is, for example, the engine speed N
The optimum value is determined by looking up the value corresponding to the operating state at that time from a table map with the basic injection amount Tp and the basic injection amount Tp as parameters. This improves drivability by optimizing the mixture ratio for the operating conditions, mainly at high speeds and high loads. Next, use P- to set the basic ignition timing GT for high-octane engine (details will be described in subroutine 5UB-1 below).
. Then, the processing by this program is repeatedly executed until the power supply is stopped.

FIG. 6 is a flowchart of a program that performs (■) interrupt processing A, and this program is executed once every 10 m3. First, read the rotation speed N with P, and use A/D with P.
Read the converted intake air amount Qa and cooling water temperature Tw0
Next, in P7, the basic injection amount 'rp is calculated according to the following equation (2).

Tp-'K・(Qa/N-)−・−・■However, K: Constant This basic injection amount Tp indicates the injection amount corresponding to the intake air amount Qa per engine rotation, and this is the injection amount corresponding to the intake air amount Qa per engine rotation. This is because fuel is injected in synchronization with the rotation, for example, every rotation. Then, at P, the final injection amount T1 is determined according to the following formula (■)
Calculate.

'T'1=Tpx (1+KMR)xcoEF+Ts
-・・・・・■ However, COF, F: Various increase coefficients Ta: Injector response delay correction coefficient ■In the formula, C0EF is for various increase corrections to the basic injection amount Tp based on, for example, cooling water temperature or acceleration increase. , this C
The values of 0EF and Ts are calculated in interrupt processing B, which will be described later.

Then, P is the retardation correction amount BETA of the basic ignition timing GT.
(Details will be described in subroutine 5UB-2 below).

FIG. 7 is a flowchart of a program that performs interrupt processing B, and this program is based on the crank angle reference signal CaL"
It is executed once for each input (that is, every 120 degrees since there are 6 cylinders).

First, the knock signal Sn indicating whether or not knock vibration is occurring is read in Plo, and the final ignition timing GK is set depending on the presence or absence of knock vibration in pH (details will be described in subroutine 5UB-3 below). ). Then Pl2
The injection signal Si is output. Therefore, the final injection amount Tl of fuel is injected from the injector 4 into the intake pipe 3, and the mixture ratio of the intake air-fuel mixture is controlled so as to always correspond to the operating state. Then, Plm calculates the values of COEF and TS according to the operating state at that time.

Figure 8 shows subroutine 5U for setting the basic ignition timing GT.
It is a flowchart which shows B-1. P2. + reads the rotational speed N and basic injection amount 'rp, and Pμ looks up the optimum value of the basic ignition timing GT corresponding to the current N and Tp from the MA knob MA shown in the following first table.

Table 1 Here, the basic ignition timing GT is the optimum ignition timing corresponding to the driving condition at that time when using high-octane gasoline (for example, gasoline with an octane rating). It is set to a value that is as advanced as possible. Next, the basic ignition timing GT obtained by lookup is set as the basic advance value TADV using Pl, and the de-energized angle value TDWE is calculated from the rotational speed N and battery voltage Vb using PA+, and the value is stored. .

Figure 9 shows the final advance value FAD corresponding to the final ignition timing GK.
It is a flowchart which shows subroutine 5UB-2 which calculates V. The rotational speed N and the basic injection amount 'rp are read in the Pit, and the retardation correction amount BETA in the area corresponding to the current N and Tp is looked up from the map MB shown in FIG. 10 in the Piz. Here, the retard angle correction amount BETA is a value for retarding the final ignition timing GK by correcting the basic advance value TADV to the retard side to avoid knocking.
As shown in FIG. 10, optimum values are stored corresponding to a plurality of operating ranges A-L. The value of this retardation correction amount BETA is always maintained at the latest data by appropriately learning according to the driving conditions at that time in learning flow 0F, which will be described in the next section.

In the learning flow CF, first, Phi stores a value for the number of knocking occurrences (hereinafter referred to as the knock door value; this value is calculated in subroutine 5UB-3 described later) KNCT.
It is determined whether or not KNCT=0, and when KNCT=O, it is determined that knocking has not occurred and the process proceeds to Pl4. On the other hand, when KNCT≠0, the value of the retard correction amount BETA is corrected (increases the BETAO value) using P% according to the following equation (2).

BETA=BETA+KNCTxDNT-・-・-〇However, in the DNT Ni-no-Soku correction coefficient 0 formula, the knock correction coefficient DNT is multiplied by the knock door value KNCT, for example, DNT=1 (d
eg). Therefore, the value of the retard angle correction amount BETA increases depending on the number of occurrences of knocking. After performing the above correction, the non-quest store value KN is calculated using Pl et al.
Clear the CT (KNCT-0) and prepare for the next correction. Next, in p37, the limit value KLM of the retard correction amount BETA is looked up from the map MC according to N and 'rp. The limit value KLM regulates the maximum value of the retard correction amount B ETA. For example, when the knock sensor 12 is abnormal, the knock door value KNCT increases abnormally and the final ignition timing G
In order to prevent K from being delayed more than necessary, the map MC (not shown, for example, the 10th
It is stored in a map similar to the one shown in the figure).

Next, at P3+1, the retard angle correction amount BETA is set to the limit value KLM.
Compared with, when BETA>KLM, BETA with Piq
Restrict P to the value KLM and proceed to P. On the other hand, BETA
If ≦KLM, jump P5Q and proceed to PI, o. Therefore, it is possible to prevent the ignition timing from being delayed more than necessary, and to avoid deterioration in drivability such as poor acceleration and engine stalling. In P2O, the above steps P3q to P5
The value of the retard correction amount BETA corrected through q is used as a learned value corresponding to the current driving condition, and the map value of the corresponding area in the map MB is rewritten to this learned value. Further, when KNCT=0 in steps p and s, it is assumed that the value of the retard angle correction amount BETA is appropriate, and the map value is not rewritten.

To summarize the above, if knocking occurs in the current driving range, the retardation correction amount B is adjusted according to the degree of knocking.
Increase the value of ETA, rewrite the map value in the corresponding area, and if knocking does not occur, change the current retardation correction amount B.
Rewriting is not performed since ETA is the optimal value. By repeatedly executing such learning flow CF, it is possible to learn the optimum value of the retard angle correction amount BETA, and the learned value is stored in the map MB. Therefore, the value of the retardation correction amount BETA is always stored as the latest and optimal value for each applicable operating region regardless of changes in operating conditions. It can also be regarded as an optimal value.

After passing through the above-described learning flow 〇F, the final advance angle value FADV is calculated according to the following equation (②) in P4.

F ADV = 70°-(TADV-BETA)
-・-■ This final advance angle value F: ADV is advance angle data that indicates the degree at which the ignition is to be performed counting from 70 degrees before the top dead center of each cylinder. It is determined by correcting TADV and BETA corresponding to the current operating state. In this case, since the retard angle correction amount BETA is the optimum value corresponding to the driving range at that time, it is possible to improve the output (torque) and fuel efficiency while avoiding knocking by setting the final advance angle value FADV to the optimum value. can.

FIG. 11 is a flowchart showing a subroutine 5UB-3 for outputting the ignition signal sp corresponding to the final ignition timing 6K. P1 sets the final advance angle value FADV in the register 47, and P4z sets the de-energized angle value TDWE in the register 52.
Set to . Therefore, from 70° before top dead center, FAD
The timing advanced by the value of V becomes the final ignition timing GK, and at this timing, the ignition signal Sp is output and the air-fuel mixture is ignited. Next, read the discrimination signal sh with P@, and
@ is used to determine the level of the determination signal sh (ie, the presence or absence of knocking).

When the discrimination signal sh is (H) (when knocking occurs), the knock door value KNCT is set to (1) using P@.
), and when the discrimination signal sh is CL) (when knocking is not occurring), the P disease process is not executed. Therefore, when knocking occurs in the relevant operating range, the number of knocking occurrences is equal to the knock door value K.
Stored as NCT.

FIG. 12 (81 to +d) is a timing chart of ignition timing control.

- As shown in Figure 12 (@), when the operating region is B, the value of the retard correction amount BETA calculated for each Ioms is zero before knocking occurs, as shown in Figure 12 (a). (see figure (b)), the final ignition timing GK is equal to the basic ignition timing GT for high-octane engines (see figure (dl)). When knocking n1 occurs due to use, a retardation correction amount BETA is calculated according to this knocking n1, and the final ignition timing GK is delayed compared to the basic ignition timing GT by this BETA value.Therefore, the next time Knocking can be avoided at the ignition timing of .In this region B, knocking n2,
Each time n3 occurs, the value of the retard angle correction amount BETA increases, and the value of the map MB is rewritten with the assumption that the retard angle correction amount BETA corresponds to this B area. Next, timing 1. When the operating region shifts from B to C,
Initially, BETA=O, and the final ignition timing GK is equal to the basic ignition timing GT.

When knocking n4 occurs, the retardation correction amount BETA is calculated accordingly, and the final ignition timing GK is changed to the basic ignition timing G.
It lags behind T, and knocking is avoided. At this time, the value of the retard angle correction amount BETA is also rewritten as the value of the map MB as the retardation learning value corresponding to the C area. and,
When transitioning from area C to area B again at timing t2,
At this time, the retard angle correction amount BETA is not BETA=0, but the learning value learned in the previous B area is used. Therefore, the final ignition timing GK is appropriately corrected to the retard side before ).7 king occurs. Therefore, damage to the engine due to knocking can be prevented and output (torque) can be improved. This means that even if it is difficult to determine whether knocking has occurred, knocking can be avoided by appropriately delaying the ignition timing based on previous learning results. Furthermore, when new knocking n5 occurs in this B region, the retardation correction amount BETA becomes even larger.

Therefore, even in the same region, it is possible to appropriately deal with changes in operating conditions (for example, changes in outside temperature, changes over time in the device, etc.), and always store the optimum retardation correction amount BETA as a learned value. In addition, the above-mentioned action is performed at 1500 r.p.m.
(e.g. A, E, Z shown in Figure 10)
The same thing is done in . Therefore, unlike in the past, knocking can be sufficiently suppressed even in such a low rotation range, thereby preventing engine overheating, performance deterioration, and even damage. On the other hand, the final ignition timing GK is sufficiently advanced to be equal to the basic ignition timing GT as long as knocking does not occur even in a high rotation range (for example, H, L) of 4000 rpm or more. Therefore, unlike the conventional engine, it is possible to improve output and fuel efficiency even in such a high rotation range. In this way, the ignition timing can be optimally controlled as described above for the use of fuels with different octane numbers, and knocking can be avoided and drivability can be improved.

(Effects) According to the present invention, the ignition timing can always be optimized for a wide range of operating conditions regardless of the use of fuels with different octane numbers, thereby improving engine drivability by avoiding knocking. be able to.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram of the present invention, FIGS. 2 to 12 are diagrams showing an embodiment of the present invention, FIG. 2 is a schematic configuration diagram thereof, and FIG. 3 is a circuit configuration diagram of the control unit. Figure 4 is a circuit configuration diagram of the interface circuit, Figure 5 is a flowchart of the program that performs the BGJ processing, Figure 6 is a flowchart of the program that performs the interrupt processing A, and Figure 7 is the flowchart of the program that performs the interrupt processing B. 8 is a flow chart showing a subroutine for setting the basic ignition timing, FIG. 9 is a flow chart showing a subroutine for calculating the final advance angle value, and FIG. 10 is a flow chart showing a subroutine for calculating the final advance angle value.
11 is a flowchart showing a subroutine for outputting the ignition signal, and FIG. 12 is a timing chart for explaining the action. i-11-engine, 8--ignition means, 12--knock detection means, 15--operating state detection means, 16--control unit (basic value setting means, correction amount calculation means, storage means, ignition timing control means).

Claims (1)

[Scope of Claims] a) Operating state detection means for detecting the operating state of the engine; b) Knock detection means for detecting knock vibration generated in the engine; and c) Basics for high octane fuel based on the operating state. basic value setting means for setting ignition timing; and d) calculating a retardation correction amount for correcting the basic ignition timing to the retard side so as to suppress knocking when knocking occurs, and when knocking does not occur. a correction amount calculation means for stopping the calculation of the retardation correction amount; e) learning the output of the correction amount calculation means when knocking is not occurring and when knocking is suppressed as corresponding to the driving state at that time; , a storage means for storing the learned value; f) reading out the learned value of the retard correction amount corresponding to the operating condition from the storage means, correcting the basic ignition timing based on the learned value and setting the final ignition timing; An ignition timing control device comprising: ignition timing control means; and g) ignition means for igniting an air-fuel mixture based on an ignition signal from the ignition timing control means.