JPS60147631A - Detector for combustion pressure of internal-combustion engine - Google Patents

Abstract

PURPOSE:To separate and detect exactly only the combustion pressure in all operating states by determining compression pressure at each of respective crank angles by calculation and subtracting the value thereof from the internal pressure of a cylinder. CONSTITUTION:The internal pressure of a cylinder is detected by a pressure sensor 2 and the crank angle of an engine is detected by a crank angle sensor 1. A means 3 for calculating compression pressure calculates the compression pressure for each of prescribed crank angles. A means 5 for calculating combustion pressure calculates the combustion pressure for each of the prescribed crank angle after the top dead point by subtracting the compression pressure value determined by the means 3 from the internal pressure value of the cylinder calculated by the sensor 2. A means 6 for calculating the timing for ignition calculates the timing for ignition and a means 7 for controlling the timing for ignition operates an ignition device 8, thereby performing ignition.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Application of the Invention) The present invention relates to a device for detecting combustion pressure in a cylinder, which is closely related to the combustion state of an internal combustion engine, separately from the pressure in the cylinder.

This device can be used to control ignition timing, air-fuel ratio, exhaust gas recirculation, etc.

(Prior Art) Conventionally, technology has been developed to detect the cylinder pressure as a parameter to know the combustion state in the cylinder and control the ignition timing etc. based on the result (for example, the
No. 9-29209, JP-A-52-15143, etc.).
1 In the above technique, for example, a piezoelectric element is provided in place of the washer of the spark plaque, and the cylinder pressure is detected based on the output of the piezoelectric element.

However, in the conventional technology, the cylinder pressure itself is detected, and the combustion pressure alone is not detected separately from the cylinder pressure.

It was not possible to accurately determine the combustion state within the cylinder.

That is, the cylinder pressure is a combination of the compression pressure (also referred to as motoring pressure) generated when the air-fuel mixture in the cylinder is compressed by the vertical movement of the piston and the combustion pressure generated due to combustion of the air-fuel mixture.

The compression pressure is determined by the operating conditions (mainly the suction chamber air volume)
If it is constant, it will be constant regardless of the combustion state, and only the combustion pressure will change depending on the combustion state in the cylinder.

Therefore, in order to accurately know the combustion state, it is necessary to separate and detect only the combustion pressure from the cylinder pressure.

As a method of separating only the combustion pressure from the cylinder pressure, the following 9-order method can be considered.

The method will be explained based on FIG.

FIG. 1(A) is the cylinder pressure waveform, and the dashed line P1 is the cylinder pressure waveform at low load such as idling,
The solid line P2 is the cylinder pressure waveform at medium load and high load, and the broken line po is the compression pressure waveform, and + Pl or P2 and p
The difference between this and o becomes the combustion pressure.

As shown by Pl in Fig. 1 (A), at low load such as when idling, the cylinder pressure at BTDC (,, L, before dead center) is almost only the compression pressure, and this waveform changes under the operating conditions ( The value mainly depends on the amount of intake air.

Further, the waveform at ATDC (after top dead center) is a combination of combustion pressure and compression pressure, and changes depending on combustion conditions.

On the other hand, the compression pressure is symmetrical with respect to TDC (top dead center), so the value of po at BTDC is memorized at every predetermined crank angle (for example, every 2"), and from those values, P
Target waveform for TDC of the waveform of o.

, the compression pressure waveform Po' at ATDC can be obtained.

By subtracting Po' from PI, the combustion pressure at ATDC can be detected separately from the cylinder pressure.

However, even during medium load, high load, or low load, the ignition timing is MBT (minimum 5park a
dvancefor best torque), it is shown in the P2 waveform.

The combustion pressure increases even before TDC. Normally, the combustion flame front grows sufficiently in the cylinder, and the pressure increases.

starts to rise rapidly at around 20° BTDC when ignited at 60'' BTDC, and around 2° BTDC when ignited at 20° BTDC.

Therefore, in the case of a pressure waveform like P2.

Even in BTDC, the waveform is a combination of compression pressure and combustion pressure, making it impossible to separate only the compression pressure.

If the BTD of the P2 waveform is
If we take the value at C as the compression pressure and find the target waveform, we get
The result is as shown by the broken line P2', and the actual compression pressure Pd is much higher than the actual compression pressure Pd.
The waveform will be different for 1 J, and the combustion pressure calculated from it will also contain a large error.

Further, in the above method, since it is necessary to store the value of Po at BTDC for each predetermined crank angle, there is a problem that the storage device becomes large. ! That is, when implementing the above method, a microcomputer is usually used to perform calculations.

Although it seems obvious, in order to store and read out the AD-converted value of the cylinder pressure over a predetermined range (for example, BTDC50° to TDC), a large amount of RAM of 25 to 50 bytes is required.

This results in an increase in cost, an increase in power consumption, an increase in the amount of heat generated by the element, and a decrease in reliability.

Furthermore, in general, internal combustion engine control devices using microcomputers comprehensively control not only the five ignition timings but also the fuel injection amount, exhaust gas recirculation amount, etc. Since interrupt processing is required over a wide range of crank angles, other calculations are limited, and in some cases, other control programs may become unexecutable.

(Objective of the Invention) The present invention was made to solve the above problem, and is capable of accurately separating only the combustion pressure from the cylinder pressure in all operating conditions from low load to high load. It is an object of the present invention to provide a combustion pressure detection device that can detect the combustion pressure and simplify the RAM capacity and calculation procedure.

(Summary of the invention) In order to achieve the above object, the present invention has the following features.

The compression pressure for each predetermined crank angle under predetermined conditions is stored in advance in a storage means (for example, ROM), and is set to a position (for example, 30° BTDC) that is earlier than the position where the combustion pressure suddenly rises before top dead center. By calculating the compression pressure at each predetermined crank angle from the actual value of the cylinder pressure at the reference crank angle and the above memorized value, and subtracting that value from the actual cylinder pressure value, the compression pressure at each predetermined crank angle is calculated. It is configured to check the combustion pressure.

FIG. 2 is a diagram showing the overall configuration of the present invention.

In FIG. 2, 1 is a crank angle sensor that detects the crank angle of the engine, and 2 is a pressure sensor that detects the cylinder pressure.

Further, the compression pressure calculation means 3 calculates the compression pressure at a position (for example, 30° BTDC) before the position where the combustion pressure suddenly rises before the top dead center.
The compression pressure at every predetermined crank angle (for example, every 2 degrees) in ATDO is calculated from the cylinder pressure value at the reference crank angle set in , and the value stored in advance in the storage means 4.

The above operation will be explained in detail below.

As shown in Figure 1 (B), the waveform of the compression pressure is.

POl +P depending on operating conditions (mainly intake air amount)
o2. It changes like Po3. However, since the ratio at each crank angle is constant, the POI
The waveform of is stored in the storage means, and the ratio k to Pot' at a predetermined reference crank angle, for example, 30° BTDC.
Or, set the value for each crank angle of POI, L.

By multiplying, it is possible to calculate the value of PO2+"03 at each crank angle.

In addition, the value of each crank angle of POI is a coefficient in the form of blood.

L】 , and then multiply that value by L2 or L3.

However, the value of P02+poa can be calculated in the same way as above.

As can be seen from Fig. 1 (A), the temperature is below 30° BTDC.

In the above, it is safe to assume that the cylinder pressure is only the compression pressure in any state from low load to high load. If the compression pressure is calculated based on the in-cylinder pressure at the position before the combustion pressure rises in this way, the compression pressure can be calculated accurately under all operating conditions.

Note that the above reference POI value is measured by the pressure sensor 2 when there is no combustion (for example, when the ignition device is stopped)
It may be an actual measured value or a calculated □ value.

The POI value (or uniform value L) obtained as above is stored in the storage means 4 in advance.

Next, in the compression pressure calculation means 3, a predetermined reference crank angle (!3TDc30 on the upper side) detected by the pressure sensor 2 is used.
The compression pressure for each crank angle is calculated by multiplying the above-mentioned stored value for each crank angle by the ratio of the pressure value at Ll (or the pressure value itself).

With the above configuration, only one piece of data (in this example, the value at BTDC 30°) needs to be measured and stored in BTDc-TDC, so the required RAM is reduced.
The capacity and number of interrupts' are reduced to 11J.

Next, the combustion pressure calculation means 5 subtracts the compression pressure value calculated by the compression pressure calculation means 3 from the cylinder pressure value calculated by the pressure sensor 2 at each predetermined crank angle at ATDC. Calculate the combustion pressure for each.

Note that when calculating the ignition timing, the combustion pressure at 'ATDC' usually becomes a problem.
Although it is written to calculate the combustion pressure at only B, if necessary, if it is after the above reference crank angle,
Of course, the combustion pressure can be calculated even within the TDC range.

It is possible to separate and detect the combustion pressure by means 1 to 5 listed in "-, but when using the results to control, for example, the ignition timing, the portions surrounded by broken lines (6 to 8) in FIG.
). That part will be explained below.

The ignition timing calculation means 6 calculates the ignition timing based on the output of the crank angle sensor I and the calculation result of the combustion pressure calculation means 5.

The above calculation is performed as follows.

First, the output of the crank angle sensor 1 and the combustion pressure calculation means, 5
Based on the determined combustion pressure, determine the crank angle θm at which the combustion pressure is maximum.7 Next, calculate the ignition timing so that the above θ□ becomes the predetermined value, that is, the ignition timing is set to MBT. .

Specifically, the basic ignition timing value (characteristics shown in Figure 11 below) is calculated according to the operating condition at that time (for example, rotational speed □ degree and load), and the feedback coefficient for setting the ignition timing to MBT is calculated. Then, the actual ignition timing value is calculated by adding that value to the basic ignition timing value.

Next, the ignition timing control means 7 performs 9. Then, an ignition signal is sent out at a time corresponding to the ignition timing value calculated by the ignition timing calculation means 6 described above, thereby operating the ignition device 8 to ignite.

The ignition methods include a method that performs one ignition only once every - ignition timing, a method that performs multiple ignitions every - ignition timing, and a method that re-ignites only when a misfire is detected and a misfire occurs. (For example, Japanese Patent Application No. 57-77536).

Further, in calculating the ignition timing, it is possible to determine whether or not ignition occurred in the first ignition, and to delay the ignition timing when it is determined that ignition did not occur in the first ignition.

(Embodiment) FIG. 3 is a diagram showing an embodiment of the present invention, and shows a case where the present invention is applied to ignition timing control.

In FIG. 3, the crank angle sensor 1 outputs a unit angle signal S+ (for example, St in FIG. 6) every time the crankshaft rotates by a predetermined unit angle (1° or 2°), and also outputs a unit angle signal S+ (for example, St in FIG. 6) at a predetermined reference angle ( 180° for a 4-cylinder engine, 120° for a 6-cylinder engine
82 °) each time the reference angle signal 82 (for example, Fig. 6 82
) is output.

The pressure sensor 2 outputs a pressure signal S3 corresponding to the cylinder pressure.For example, as shown in FIG. Element 20 can be used.

The microcomputer 9 has MPU10. RAMII.

Consists of ROM 12, input/output device 13, etc.
.

Then, each of the above signals 81 to S3 is inputted, and the predetermined performance $C is
) and outputs the ignition signal S4.

The ignition device "8 includes a transistor 5, a power distributor 169, spark plugs 17A to 17F, a buffer 9182, and an ignition coil 19.
It is composed of etc.

When the ignition signal S4 is applied (becomes low level), the transistor 5 is turned off, so that a high voltage is generated in the secondary winding of the ignition coil 19.

The high voltage is applied via the power distributor 16 to the ignition plug (any one of 17A to 17F) in the order of ignition, and a spark discharge occurs in the gap between the ignition plugs, causing ignition. .

Next, the operation of the microcomputer 9 will be explained. Before explaining the overall operation, first, the free run counter provided in the input/output device 13 will be explained.

FIG. 5 is a block diagram of an example of a free run counter.

In FIG. 5, edge detector 50 detects the rising edge of input pulse signal SIO and outputs detection signal 811.

A 16-bit free-running counter 51.

It is incremented by 1 every time a predetermined clock pulse (for example, a 16 μs clock) is counted.

And if it overflows, overflow.

Detector 52 detects timer overflow interrupt signal S
12, and at the same time the initial value (that is, $0000), the fimp capture register 53 retains the value of the free running counter 51 at the time when the detection signal 811 of the edge detector 50 is input.

hold

Further, the above detection signal 811 is transmitted to the MASK register 54,
is given to the ICF bit of
”, the input capture signal 813 is notified to the MPU (10 in FIG. 3) via the internal bus 55, and an input capture interrupt is generated.

Here, the input pulse signal SIO is the pulse signal SIO.

The unit angle signal S shown in Fig. 3 is used.

The unit angle signal S1 is 9, for example, as shown at 81 in FIG.

In particular, 9 cycles have a crank angle of 2° and a duty of 505%.
n.

1, that is, the pulse signal is high level for 1° and low level for 1°, so the input capture interrupt described above is generated every 2° of crank angle, and the value of the input capture register 53 at this time is By reading this, the MPU10 can read the value of the free running counter 51 when the unit angle signal S is input.

Therefore, by calculating the difference between the previous value of the free running counter 51 and the current value, the period of the unit angle signal S1 can be detected.

In the present invention, the process described below is performed using the unit angle signal S1, and the combustion pressure reaches the maximum value P11. The crank angle θ□ is determined.

Next, the relationship between cylinder pressure and ignition timing will be explained.

Figure 7 is a general cylinder pressure waveform diagram. 1. The crank angle θm at which the cylinder pressure (strictly speaking, combustion pressure) reaches the maximum value Pm is closely related to the engine's maximum torque and fuel efficiency. Yes, θ□ is a predetermined value determined by the characteristics of the engine (ATDC
The most efficient operation can be achieved when the value is approximately IO' to 20'. The ignition timing at this time is MBT.

Further, there is a nearly proportional relationship between θ□ and the ignition timing, as shown in FIG. 8, and θ□ can be changed by changing the ignition timing.

Therefore, optimal operation can be achieved by controlling the ignition timing and changing the ignition timing so that θ□ is always at the desired value.

Note that the characteristics shown in FIG. 8 indicate that the air-fuel ratio is the stoichiometric air-fuel ratio (A/F'';
14.8), but it is known that the tendency is almost the same over a wide range of A/F of about 12=20.

Furthermore, from the characteristics shown in FIG. 8, two ignition timings can be calculated by subtracting θ and n.

By comparing the first ignition timing at that time with the above-mentioned ignition timing, it can be determined whether or not the first ignition ignited.

Next, detection of compression pressure and combustion pressure will be explained.

By writing data in the ankle register in the input/output device 13 shown in Fig. 3, when the number of unit angle signals S1 input after the reference angle signal S2 is input matches that data, that is, the crank angle When the value specified by the data is reached, the angle register generates an angle match interrupt.

Therefore, the period of the reference angle signal S2 (120
In order to generate angle matching interrupts multiple times within a range of 180° for a four-cylinder engine, a value corresponding to the second interrupt angle may be written into the angle register for each angle matching interrupt.

Note that by reading the value of the angle register, it is possible to know the number of unit angle signals S input after the reference angle signal S2 was input, that is, the crank angle at that time.

The reference angle signal S2 is 9, for example, as shown in FIG.
Occurs at 70° BTDC.

Then, first write 20 (corresponding to 40°) to the angle register value, generate the first angle match interrupt at BTDC30°C70-40=30>, and write 341 to the angle register with the program executed by this interrupt. (6
8 degrees) is written, a second angle match interrupt is generated at BTDC 2 degrees, and the value of the angle register is returned to 20 at that point. By doing this, −
Two angle match interrupts occur within one ignition cycle.

FIG. 9 is a flowchart of a program executed by the above-mentioned angle coincidence interrupt.

In FIG. 9, first, at Pl, the value of the ankle register at the time of the angle coincidence interrupt is read into register A.

Next, in P2, whether the value of A is 34 or more, that is, BTD
Interrupt at C3Q° or interrupt at BTDC2°.

Determine if it is included.

If No in P2, that is, in the case of an interrupt of BTDC 30°, the process goes to P3 and starts AD conversion of the cylinder pressure (written output of the pressure sensor 2).

Next, in P4, 34 is written to the value of the angle register.This is to generate the angle match interrupt again at <, BTDC2°, as described above.

Next, in P5, it is determined whether the AD conversion is completed or not.
When the process is completed, the AD value (the value obtained by converting the analog value into a digital value) is stored in the PCYL30.

This value, that is, the value at 30° BTDC is the reference for compression pressure calculation, and corresponds to L2 or L3 in FIG. 1(B).

Note that PCYL30 is a memory area that is allocated to a predetermined location in the RAM shown in FIG. 3 and stores the AD-converted cylinder pressure value.

Also, the end of AD conversion at P5 is specifically: AD
This can be determined by checking the AD BUSY flag in the C register and repeating the loop until it becomes 0.

On the other hand, if P2 is YES, that is, if it is a BTDC2° interrupt, go to P7 and write 2 to the angle register.
Write 0. This will cause B T D C3 again next time.
An angle coincidence interrupt will occur at 0°.

Next, at P8, EXTIRQZ, that is, the free run
Unmasks the counter/timer measurement interrupt (hereinafter referred to as input/capture interrupt).

Specifically, EX of the MASK register 54 in FIG.
TIRQZ Hira I・(EICI hit) wo”O”
Nittle.

Next, in P9, AGLCN is used for pressure signal processing as described below.
Clear T (memory $CA address).

With the above (, this angle match interrupt is BTDC2°
Therefore, by canceling (ENABLE) the input capture interrupt at P8, an input capture interrupt will occur when the secondary unit angle signal S1 is input, and from then on, every time the unit angle signal S1 is input, That is, an input capture interrupt occurs every 2 degrees of crank angle. Therefore, the input capture interrupt first occurs at n+i, when the value of the angle register becomes 35, that is, TDC, and thereafter occurs every 2 degrees of crank angle.

In this input capture interrupt, the compression pressure and combustion pressure calculation, θ111 and feedback coefficient calculation program shown in FIG. 10 are executed.

The following explanation will be given based on FIG. 1O, but first the data names in FIG. 10 will be explained.

AGLCNT is a memory used as a count allocated to the $CA address of RAM, and is
It is cleared by the C2° angle match interrupt, and is incremented by each input capture interrupt.

PIMAX is assigned to address $CB in RAM and stores the maximum value of combustion pressure.

THPMX is allocated to the $CC address of the RAM, and stores the value of AGLCNT at the time of PIMAX, that is, the angle in units of 2 degrees calculated from TDC.

THPMX+1 to THPMX+6 are 6 consecutive T
HPMX value, that is, all cylinders (6 cylinders in this case).

THPMx is stored in RAM (7)
$CD.

$CE, $CF, FQ, $F1. It is allocated to address $F2.

ADVFBK is assigned address $F3F3 in RAM and changes the ignition advance angle by ±1.0. -] This is a data area for feedback control to change only one of MBTi.

M is a 6-byte memory area designated at address $F4F4 in RAM for storing intermediate results of calculations.

C0EFP is the data assigned to a specified location in ROM.
- vertical table, and stores the values of the cylinder pressure during non-combustion, that is, the compression pressure, at each crank angle (2° interval).

Figures 12 and 13 are examples of the contents of the above data table, that is, the compression pressure at ATDC.
Figure 2 shows a numerical table, and Figure 13 shows a graph.

In Figure 12, PKpt is the air 1 during non-combustion.
This shows the ratio between the cylinder pressure pi and the cylinder pressure Pt at TDC, and FIG. 13 is a graphical representation of this value.

Further, r x 32768 J is a coefficient obtained by multiplying the above PL//Pl by 32768, and HEX is the coefficient expressed in hexadecimal digits.

As an actual compression pressure coefficient table, the above HEX will be stored at the ROM address starting from C0EFP.

The numerical values shown in Figs. 12 and 13 are obtained by calculating the values for the respective engines using the following formula +11.

However, in equation (1), VO9φ, r, and t are each as follows.

vo = 43 cm3 (combustion chamber volume, i.e. two-part space when the piston is at top dead center) φ-8,305Cm (cylinder inner diameter, hoa) r = 4.18
3 cm (Crankshaft rotation radius.

2r is the stroke) 1-= 13.3 cm (Connecting rod length)
Next, FIG. 10 will be explained.

First, in Figure 10 (A), AGLCN is
Determine whether T is 0 or not.

When AGLCNT=O, this indicates the case where the crank angle reaches TDC and an input capture interrupt is entered for the first time. In this case, go to P22 immediately.

When AGLCN'r lol 0, it's already the previous imputs;
AD conversion of the pressure signal S3 has been started by the capture interrupt (see P24 below), so go to Pll,
The AD value that has already been converted is read into register A, and the value is stored in M+5.

Next, PI2 reads the address of C0EFP into index register X, and then PI3 reads X=X+AGLCNT-1.
Calculate. The value of this X indicates the address of the compression pressure data table to be referred to in this interrupt.

Next, the compression pressure is calculated in PI4 to PI8.

The compression pressure can be determined by calculating the following equation (2).

, =, xI°30 (2) 30 In the formula (2), Px is the value of the compression pressure at the crank angle X of the current interruption, KX is the value of the data table at the above crank angle X, 30 is the value of the data table at ATDC 30° (the same value as BTDC 30° since it is a target), and in the example of Fig. 12, 4 ABF' = the value of cylinder pressure (P, the value stored in CYL30). .

Specifically, the above calculation is performed as follows.

First, load the value of address X (K,) into registers A and B, and first, assign the value of PCYL3Q (Pao) to M.

After clearing M+1, call subroutine MUDLT (DMULT is a subroutine that performs 16-bit multiplication, and the product value is stored in RAM as 32-bit data.
).

The result is placed in 4 bytes of M-M+3, so.

Next, set the value of #C0EFP30 (K3o) to register A,
B and calls subroutine DIV42 (DIV
42 is.

This is a routine that divides a 32-bit dividend by a 16-bit divisor; the 16-bit quotient is returned as 1 in registers A and B, and the 16-bit remainder is placed in RAM).

The results are placed in registers A and H, and the values are correct.

This is the compression pressure PX at the crank angle X.

In equation (2), if the x value stored in the data table is stored in the form of a stone, P can be calculated by simply multiplying the value by the value of Pao.

Therefore, there is no need for division (in general microcomputers that do not have a division instruction).

It is advantageous in terms of execution speed and memory capacity.

Next, at PI9, read the current cylinder pressure previously stored in M+5 using PH, and subtract the above compression pressure from that value to calculate the combustion pressure, and place it in register B.

The above is the calculation process for compression pressure and combustion pressure.

Next, detect the crank angle at which the combustion pressure reaches its maximum.

and calculation of ignition timing will be explained.

First, in P2O, compare the size of B and PIMAX 6P
If 20 and B are smaller than PIMAX, immediately go to P22.

When B is greater than or equal to PIMAX in P2O, that is, the new B
If the value of is greater than or equal to the past maximum value, store B as a new PIMAX at ``21'' and save the AGLCNT at that time.
(crank angle value) is stored in THPMX.

By repeating the above process, PIMAX.

Always the maximum AD value of combustion pressure at ATDC.

That is, the maximum value obtained by subtracting the compression pressure from the cylinder pressure is stored, and the crank angle at that time is stored in THPMX.

Next, in P22, AGLCNT is incremented by 1, and then the process goes to P23 in FIG. 10(B).

In P23, it is determined whether AGLCNT is 25 or not.

If P23 is No, go to P24 and press A of pressure signal S3.
Immediately after starting the D conversion, go to the “End” line “(.

Next, the case of YES in P23 will be explained.

AGLCNT of 25 means ATDC at crank angle.
This corresponds to 50°, but at this point the combustion of the air-fuel mixture in the cylinder has almost finished. That is, by this point, the combustion pressure has reached the maximum value of 1.

Therefore, the value of PIMAX and THP at this point
The value of MX indicates the maximum value Pm of combustion pressure in one combustion cycle and the crank angle θm at that time.

If P23 is YES, that is, AGLCNT=25.

Go to P25 (.

P25 P2 after masking Teha EX71'IRQZ
Go to 6 (.

In P26, after shifting THPMX+1 to THPMX+6 byte by byte, THPMx+1 to THPMX
Store at +6°.

By doing this, i' HP M X +
l~T HP M X+6 always has the latest 6 θ
This means that the value of m is held.

In other words, θm of all cylinders of a 6-cylinder engine is THPMX+1 ~
It will be stored in THPMX+6.

Next, in P27 to P31, the current ignition feedback coefficient ADVFBK is calculated.

In the case of this example engine, as shown in the characteristics of Figure 8.

9 When the ignition timing is BTDC 20°, θ□ is approximately AT
When D018°, BTDC30°, approximately ATDC12°
It is. Therefore, the values of THPMX correspond to 9 and 6, respectively.

As above, THPMX+1 to THPMX+6.

Since the values of θI11 for all cylinders are stored in temporal order, 01n of the ninth cylinder to be fired is stored in THPMX+6.

Therefore, first, in P27, the value of THPMX+6 is read and this value is set as A.

Next, in P28, the magnitude of A and 6 is determined.

In general, there is a close correlation between θ+n and maximum torque (best fuel efficiency), and it varies depending on the engine, but generally θ1n is ATD
It is known that I will produce maximum torque (and therefore best fuel economy) at certain values within the range of CIO° to 20°.

In the case of this example, it is assumed that θm, which is the maximum torque, is, for example, 12° ATDC.

ATDC 12° is equivalent to 6 in THPMx.

In this example, the ignition timing may be controlled so that THPMX is always set to 6.

Therefore, in P28, if A-6, it indicates that the previous ignition timing was appropriate, so go to P29 and set the feedback coefficient ADVFBK to 0.

If A<6 at P2B, this indicates that the 2nd ignition timing is too advanced than the appropriate value, so the process goes to P2O, where the feedback coefficient ADVFBK is set to -1, and the 9th ignition timing is controlled to be delayed.

If A>6 in P28, this indicates that nine ignition timings are delayed from the appropriate value, so the process goes to P31 and controls to advance the ignition timing by one by setting the feedback coefficient ADVFBK to +1.

Next, in P32, set the value of 6 ADVFBK for 6 cylinders to 1.
After shifting byte by byte, ADVFBK+1~゛AD
Store in VFBK+6.

By doing this, ADVFBK+1 to ADVFBK
In +6, the values of six feedback coefficients are held while being circulated in order according to the firing order of each cylinder (for example, 1st, 5th, 3rd, 6th, 2nd, 4th cylinder). That will happen.

Therefore, ADVI? The value held in BK+6.

It always indicates the feedback coefficient of the cylinder that will be fired next.

Therefore, the value of ADVFBK+6 is read out as the feedback coefficient ADVFBK.

Next, two ignition timings are calculated in P33 to P35.

First, in P33, the basic ignition timing value 1. Calculate.

This procedure is similar to conventional ignition timing control.

For example, the ignition timing values (advance values) corresponding to the rotational speed and load amount as shown in Fig. 11 are stored in memory as a data table in advance, and the A method of reading the value is used.

Note that the load amount is intake air amount, intake negative pressure, or fuel injection pulse [[JTp (calculated from intake air amount and rotation speed)
etc. can be used.

In addition to ``1,'' the calculation method for the basic ignition timing value IO is 2.
There are also methods that involve only rotational speed or suction negative pressure. There is also a method in which control is performed using different characteristics when the throttle valve is fully closed (idling) and at other times.

Next, in P34, the above basic ignition timing value 1. The above-mentioned feedback coefficient ADV FB K is added to the actual ignition timing value 1. Calculate.

Next, in P35, the above-mentioned ignition timing value IIL is output, and ignition timing control is performed according to that value.

The ignition timing control method is the same as before.

For example, when the reference angle signal S2 is BTDC70', the value of 7O-IR is stored in a register.

Unit angle signal S input after reference angle signal S2 is input
The configuration may be such that the ignition signal S4 is output when the integrated value of 1 matches the value of 1 (l in the case of a 2° signal).

In addition, in the above explanation, the case where feedback control is simply performed to set the ignition timing to MBT was illustrated, but in the case of a method in which ignition is performed multiple times at each ignition timing, ignition occurs at the first ignition. There is also a method in which it is determined whether or not it is, and the fee back coefficient is changed accordingly.

The above method can be applied to the first
This is done by inserting P'36 and P37 next to P27 in Figure 0 (B).

In other words, if the first ignition timing is set to, for example, 25° BTDC, and if ignition occurs on the first ignition, as seen from the characteristics in Figure 8, θ□1 should be 14° ATDC. .

ATDC 14° corresponds to 7 in THPMX.

Therefore, if A is 8 or more in P36, it can be determined that the ignition did not occur during the first ignition.

In addition, the value used as a comparison standard in P36 (8 in the example) is 9.
Varies depending on ignition timing value. That is, assuming that the ignition occurred in the first ignition, from the characteristics shown in Fig. 8, θ, and b
The deviation from the standard value is good.

In this example, if A is 8 or more in P36, the ignition did not ignite on the first ignition.In such a case, it is necessary to delay the ignition timing overall, so P37''-
Then, the feedback coefficient ADVFBK is set to -1.

If A is less than 8 on P36, it means that the ignition was successful on the first ignition, in which case go to P28 and check A.
Distinguish the size of and 6.

The following processing is the same as that in FIG. 10(B).

As described above, it is possible to precisely control the 9 ignition timing by determining whether or not ignition occurred during the first ignition.

In the above embodiments, the present invention is applied to ignition timing control, but it can also be applied to exhaust gas recirculation control, air-fuel ratio control, etc. 'Furthermore, in the embodiments shown in FIGS. 9 and 10, the sampling of the cylinder pressure is performed in synchronization with rotation (crank angle) t, but sampling may be performed at regular intervals (time synchronization).

Note that in the case of a constant cycle, it is necessary to set the cycle to approximately 50 μs or less based on the relationship 1' with the engine rotation speed, and to convert the crank angle at which the combustion pressure is maximum from the rotation speed detected separately.

(Effects of the Invention) As explained above, in the present invention, the value of the compression pressure under the predetermined condition I is stored in the ROM in advance, and the value and the position before the position where the combustion pressure rises before the top dead center are stored in the ROM in advance. The combustion pressure is calculated by calculating the compression pressure for each crank angle from the actual measured value of the cylinder pressure at the reference crank angle set to , and subtracting that value from the cylinder pressure. Therefore, in all operating conditions from low load such as idling to high load.

Only the combustion pressure can be accurately separated and detected, and the combustion state within the cylinder can be accurately known.

Using the results, it is possible to precisely control ignition timing, exhaust gas recirculation amount, air-fuel ratio, etc.

Also, the in-cylinder pressure at BTDC is memorized and the compression pressure at ATDC is calculated from that value! In comparison, R
Since the AM capacity and the number of interrupt processing can be reduced to 11, it is possible to reduce costs and increase the margin for other controls.

[Brief explanation of drawings]

Figure 1 shows the characteristics of cylinder pressure and compression pressure. Fig. 2 is a diagram showing the overall configuration of the present invention, Fig. 3 is a diagram showing an embodiment of the invention, Fig. 4 is a side view of a pressure sensor, and Fig. 5 is a block diagram of an example of a free run counter. , Fig. 6 shows an example of the output waveform of the crank angle sensor, Fig. 7 shows the characteristics of the cylinder pressure, Fig. 8 shows the relationship between the ignition timing and 01.1, and Fig. 9 shows the angle-dispersion. A flowchart of the program including the following. FIG. 10 is a flowchart showing one embodiment of the calculation of the present invention, FIG. 11 is a characteristic side view of the basic ignition timing value, and "12.
FIG. 13 is an example of the contents of the compression pressure coefficient table. Explanation of symbols 1... Crank angle sensor 2... Pressure sensor 3...
- Compression pressure calculation means 4... Storage means 5... Combustion pressure calculation means 6... Ignition timing calculation means 7... Ignition timing control means 8... Patent attorney representing the ignition system Junnosuke Nakamura 11'3 Figure J = -4 Figure 1'IO Figure 1P11 Figure;

Claims (1)

[Claims] a crank angle detection means for detecting the crank angle of the internal combustion engine; a pressure detection means for detecting the cylinder pressure; a storage means for pre-memorizing the compression pressure under predetermined conditions for each predetermined crank angle; The compression pressure is calculated at each predetermined crank angle from the actual measurement value of the cylinder pressure at a reference crank angle set at a position before the position where the combustion pressure suddenly rises and the value stored in the storage means. and combustion pressure calculating means for calculating the combustion pressure per force for each predetermined crank angle by subtracting the calculated compression pressure from the cylinder pressure detected by the pressure detecting means for each predetermined crank angle. Combustion pressure detection device for internal combustion engines.