JP3859789B2 - Engine misfire diagnostic device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an engine misfire diagnosis apparatus for diagnosing misfire based on a change in an engine rotation state amount for each predetermined crank angle and an output value of an air-fuel ratio sensor.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, as an engine misfire diagnosis apparatus, an engine misfire diagnosis apparatus that diagnoses engine misfire based on a change in an engine rotation state amount for each predetermined crank angle has been employed.
[0003]
That is, there is a strong correlation between the in-cylinder pressure (combustion pressure) of the cylinder during the combustion stroke and the engine rotation between the combustion stroke and the next combustion stroke. A misfire can be determined for each cylinder.
[0004]
However, because there are manufacturing tolerances, mounting tolerances, etc., in the crank rotor and crank angle sensor for detecting the crank angle that is the basis for obtaining the engine rotational state quantity, the detected engine rotational state quantity is not included. Variations occur, particularly in the high engine speed range, and this effect becomes large, making it difficult to accurately detect misfire.
[0005]
In order to cope with this, the present applicant has proposed a technique according to Japanese Patent Laid-Open No. 6-42397.
[0006]
In this prior example, the engine rotation speed (engine rotation speed) is used as the engine rotation state quantity, the difference in engine rotation speed for each predetermined crank angle between cylinders, that is, the differential rotation DELNE is calculated, and the current combustion stroke cylinder # By subtracting the differential rotation correction value AVEDNX for the cylinder #n calculated one cycle (720 ° CA) by statistical processing using a weighted average from the differential rotation DELNEn of n, and calculating the corrected differential rotation DELNAn for the cylinder. The effects of manufacturing tolerances, mounting tolerances, etc. of the crank rotor and crank angle sensor for detecting the crank angle are eliminated. Further, a correction differential rotation change DDNAnn for the cylinder #n is calculated by subtracting the correction differential rotation DELNAn-1 of the misfire diagnosis target cylinder # n-1 before the one combustion stroke from the correction differential rotation DELNAn, and this correction differential rotation is calculated. The change DDNEAn is compared with the misfire determination level LVLMIS set based on the engine operating state, and the correction differential rotation change DDNEAn-1 for the misfire diagnosis target cylinder # n-1 calculated at the time of the previous routine execution is negative. Compare with misfire determination level-LVLMIS. When the current corrected differential rotation change DDNEAn of the combustion stroke cylinder #n is equal to or higher than the misfire determination level LVLMIS and the corrected differential rotation change DDNEAn-1 of the misfire diagnosis target cylinder # n-1 is equal to or lower than the negative misfire determination level, Judgment is made on misfiring of cylinder # n-1 to be diagnosed.
[0007]
However, in addition to variations in engine rotation state (engine rotation speed) due to manufacturing tolerances, mounting tolerances, etc. of the crank rotor and crank angle sensor, engine rotation state amounts due to variations in combustion pressure for each cylinder There are variations.
[0008]
That is, the difference in intake air distribution ratio for each cylinder due to the complexity of the intake pipe shape, the intake air interference between the cylinders, the difference in the combustion temperature for each cylinder caused by the cooling route by the cooling water, the combustion chamber volume and the piston of each cylinder Variation in shape, slight variation in air-fuel ratio of each cylinder resulting from differences in fuel injection amount due to injector manufacturing error, etc., and these causes cause variation in combustion pressure for each cylinder, Variations in engine rotation occur. Since the inertia energy is large in the high engine speed range, the amount of change in the engine rotation at the time of misfire is small, and misfire determination becomes difficult.
[0009]
Therefore, if the misfire determination level as the determination value is set strictly, the misfire may be erroneously determined due to the above factors even though the misfire has not actually occurred. In order to cope with this, if the misfire determination level as a determination value is set loosely, misfire cannot be detected at the time of misfire.
[0010]
For this reason, in JP-A-4-318256, when the engine is in a low engine speed range, a misfire diagnosis is performed based on the engine speed, while the oxygen concentration in the exhaust gas increases when a misfire occurs in the engine. At least in the high engine speed range, a technique for performing misfire diagnosis based on the oxygen concentration in the exhaust gas is disclosed.
[0011]
[Problems to be solved by the invention]
However, in the preceding example, since the misfire diagnosis is performed only based on the oxygen concentration in the exhaust gas at least in the high engine speed range, it can be detected that misfire has occurred, but the misfire in which misfire has occurred. There is a disadvantage that the cylinder cannot be identified.
[0012]
In other words, the misfire determination itself can be detected without any significant influence on the engine speed if the oxygen sensor (O2 sensor) is active and a certain amount of engine load is applied. The time from the occurrence of misfire to the change in oxygen concentration varies depending on the oxygen sensor mounting position, engine rotational speed (exhaust gas flow rate), and the like. Therefore, the time from the occurrence of misfire until the change in output of the oxygen sensor occurs is uneven, and the misfire cylinder cannot be determined.
[0013]
In view of the above circumstances, an object of the present invention is to provide an engine misfire diagnosis apparatus that can perform a strict misfire determination and reliably determine a misfire cylinder in the entire engine rotation range.
[0015]
[Means for Solving the Problems]
  To achieve the above objective,Claim1The engine misfire diagnostic device according to the invention described1'sAs shown in the basic configuration diagram, PlaceEngine rotation state amount detection means for detecting the engine rotation state amount for each constant crank angle, cylinder determination means for determining the current combustion stroke cylinder, and subtracting the currently detected engine rotation state amount from the previously detected engine rotation state amount Differential rotation state quantity calculating means for calculating a difference in rotational state quantity with respect to the combustion stroke cylinder, and averaging the differential rotation state quantity for each cylinder in a predetermined period to obtain an average differential rotation state quantity value for each cylinder. The difference rotation state quantity average value calculating means to calculate, and the difference between the maximum value of the respective difference rotation state quantity average values and the difference rotation state quantity average value for each cylinder, respectively, A first misfire determination means for comparing an average value difference for each cylinder with a misfire determination level and determining a cylinder having a larger average value difference than the misfire determination level as a misfire cylinder, and misfire based on an output value of an air-fuel ratio sensor. Discriminate And a second misfire determination means, and a misfire diagnosis means for determining that the misfire of the misfire cylinder is determined every time when the misfire is determined based on an output value of the air-fuel ratio sensor and diagnosing the misfire of the cylinder every time. And
[0016]
  Claim2An engine misfire diagnosis apparatus according to the invention described in claim1In the described invention, each of the misfire determination means performs misfire determination when the engine operating state is in the fuel increase region.
[0017]
  Claim3An engine misfire diagnosis apparatus according to the invention described in claim1In the invention described above, the second misfire determination unit includes an integration unit that integrates the output voltage of the air-fuel ratio sensor and the absolute value of the change amount of the output voltage every predetermined period, and the second misfire determination unit includes a fuel increase amount when the engine operating state is increased. When the air-fuel ratio sensor output voltage average value for a predetermined period is calculated and the air-fuel ratio sensor output voltage average value does not reach the first determination value, When the absolute value integrated value of the change amount is larger than the second determination value, it is determined that misfire has occurred.
[0018]
  Claim4An engine misfire diagnosis apparatus according to the invention described in claim1 toClaim3In the described invention, each of the average values is calculated by a simple average over a predetermined period.
[0019]
  Claim5An engine misfire diagnosis apparatus according to the invention described in claim1Or claims4In the described invention, the predetermined period is a period every time the number of ignition reaches a predetermined number.
[0020]
  Claim6An engine misfire diagnosis apparatus according to the invention described in claims 1 to5In the described invention, the engine rotation state quantity is an engine rotation speed.
[0023]
  Claim1In the described invention, the engine rotation state amount for each predetermined crank angle between the cylinders is detected, the current combustion stroke cylinder is determined, and the currently detected engine rotation state amount is subtracted from the previously detected engine rotation state amount. A difference in rotational state amount with respect to the combustion stroke cylinder is calculated. Then, the differential rotation state quantity for each cylinder in the predetermined period is averaged to calculate a differential rotation state quantity average value for each cylinder, and the maximum value of these differential rotation state quantity average values and each cylinder are calculated. The difference from the average value of the difference in rotation state is calculated. Then, an average value difference for each cylinder with respect to the maximum value is compared with a misfire determination level, and a cylinder having a larger average value difference than the misfire determination level is determined as a misfire cylinder. Further, misfire is determined based on the output value of the air-fuel ratio sensor, and when misfire is determined, the misfire is determined every time for the misfire cylinder, and the misfire is diagnosed every time for the cylinder.
[0024]
  At this time, the claim2In the described invention, when the engine operation state is in the fuel increase region for catalyst protection, misfire determination is performed based on the engine rotation state amount and the air-fuel ratio sensor output value.
[0025]
  Claims3In the described invention, the output voltage of the air-fuel ratio sensor and the absolute value of the change amount of the output voltage are integrated for each predetermined period. When the misfire diagnosis is performed based on the output value of the air-fuel ratio sensor, when the engine operating state is in the fuel increase region for catalyst protection, the air-fuel ratio sensor output voltage average is calculated based on the integrated value of the air-fuel ratio sensor output voltage for a predetermined period. When the average value of the air-fuel ratio sensor output voltage does not reach the first determination value, or the absolute value integrated value of the air-fuel ratio sensor output voltage change amount in a predetermined period is greater than the second determination value. Is also determined to be misfire.
[0026]
  Claims4In the described invention, each average value is calculated by a simple average over a predetermined period.
[0027]
  And claims5In the described invention, the predetermined period is a period each time the number of ignitions reaches a predetermined number.6In the description, the engine rotational speed is adopted as the rotational state quantity.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to FIGS.
[0029]
First, the overall configuration of the engine will be described with reference to FIG. In the figure, reference numeral 1 denotes an engine, which is a horizontally opposed four-cylinder gasoline engine in this embodiment. Cylinder heads 2 are provided on both the left and right banks of the cylinder block 1a of the engine 1, and an intake port 2a and an exhaust port 2b are formed in each cylinder head 2.
[0030]
In the intake system of the engine 1, an intake manifold 3 is communicated with each intake port 2a, and a throttle chamber 5 is communicated with the intake manifold 3 via an air chamber 4 in which intake passages of the respective cylinders are gathered. An air cleaner 7 is attached to the upstream side of the throttle chamber 5 via an intake pipe 6, and the air cleaner 7 is communicated with an air intake chamber 8.
[0031]
The throttle chamber 5 is provided with a throttle valve 5a that is linked to an accelerator pedal. A bypass passage 9 that bypasses the throttle valve 5a is connected to the intake pipe 6. The idle speed is adjusted by adjusting the amount of bypass air flowing through the bypass passage 9 according to the valve opening degree during idle. An idle speed control valve (ISC valve) 10 for controlling the engine is interposed.
[0032]
Further, an injector 11 is disposed immediately upstream of the intake port 2a of each cylinder of the intake manifold 3. The injector 11 communicates with a fuel tank 13 through a fuel supply path 12, and an in-tank type fuel pump 14 is provided in the fuel tank 13. The fuel from the fuel pump 14 is pumped to the injector 11 and the pressure regulator 16 through the fuel filter 15 interposed in the fuel supply passage 12, and is returned from the pressure regulator 16 to the fuel tank 13 so as to The fuel pressure to the injector 11 is adjusted to a predetermined pressure.
[0033]
On the other hand, a spark plug 17 that exposes the discharge electrode at the tip to the combustion chamber is attached to each cylinder of the cylinder head 2, and an igniter is connected to the spark plug 17 via an ignition coil 18 provided for each cylinder. 19 is connected.
[0034]
Further, as an exhaust system of the engine 1, an exhaust pipe 21 is communicated with a collecting portion of the exhaust manifold 20 communicating with each exhaust port 2 b of the cylinder head 2, and a catalytic converter 22 is interposed in the exhaust pipe 21 so that the muffler 23 is communicated.
[0035]
Next, sensors for detecting the engine operating state will be described. A thermal intake air amount sensor 24 using a hot wire or a hot film is interposed immediately downstream of the air cleaner 7 of the intake pipe 6. Further, a throttle valve 5 a provided in the throttle chamber 5 is connected to a throttle valve 5 a. A throttle sensor 25 having a built-in opening sensor 25a and an idle switch 25b that is turned on when the throttle valve 5a is fully closed is connected.
[0036]
A knock sensor 26 is attached to the cylinder block 1 a of the engine 1, a cooling water temperature sensor 28 is exposed to a cooling water passage 27 communicating with the left and right banks of the cylinder block 1 a, and further upstream of the catalytic converter 22. An O2 sensor 29 is provided as an example of the air-fuel ratio sensor.
[0037]
In addition, a crank angle sensor 32 is provided on the outer periphery of a crank rotor 31 that is axially attached to the crankshaft 30 of the engine 1, and further, a cam rotor 34 that is provided continuously with a camshaft 33 that rotates 1/2 with respect to the crankshaft 30. In addition, a cam angle sensor 35 for cylinder discrimination is provided.
[0038]
As shown in FIG. 17, the crank rotor 31 has protrusions 31a, 31b, 31c formed on the outer periphery thereof, and these protrusions 31a, 31b, 31c are connected to the cylinders (# 1, # 2 cylinders and # 3, # 3). (# 4 cylinder) before compression top dead center (BTDC) θ1, θ2, θ3. In the present embodiment, θ1 = 97 ° CA, θ2 = 65 ° CA, and θ3 = 10 ° CA.
[0039]
Further, as shown in FIG. 18, cylinder discrimination protrusions 34a, 34b and 34c are formed on the outer periphery of the cam rotor 34, and the protrusion 34a is located after compression top dead center (ATDC) θ4 of the # 3 and # 4 cylinders. The projection 34b is composed of three projections, and the first projection is formed at the position of ATDCθ5 of the # 1 cylinder. Further, the projection 34c is composed of two projections, and the first projection is formed at the position of ATDCθ6 of the # 2 cylinder. In the present embodiment, θ4 = 20 ° CA, θ5 = 5 ° CA, and θ6 = 20 ° CA.
[0040]
Then, as shown in the time chart of FIG. 8, the crank rotor 31 and the cam rotor 34 are rotated by the rotation of the crankshaft 30 and the camshaft 33 as the engine is operated, and each protrusion of the crank rotor 31 is connected to the crank angle sensor 32. And crank pulses of θ1, θ2, and θ3 (BTDC 97 °, 65 °, and 10 °) are output every 1/2 engine rotation (180 ° CA). On the other hand, each projection of the cam rotor 34 is detected by the cam angle sensor 35 between the θ3 crank pulse and the θ1 crank pulse, and a predetermined number of cam pulses are output from the cam angle sensor 35.
[0041]
Then, in an electronic control unit 40 (see FIG. 19) to be described later, an engine speed (engine speed) NE is calculated based on the input interval time Tθ of the crank pulse output from the crank angle sensor 32, and each Based on the pattern of the combustion stroke order of the cylinders (for example, # 1 cylinder → # 3 cylinder → # 2 cylinder → # 4 cylinder) and the value obtained by counting the cam pulses from the cam angle sensor 35 by the counter, Cylinder discrimination such as stroke cylinder, fuel injection target cylinder and ignition target cylinder is performed.
[0042]
Calculation of control amounts for actuators such as the injector 11, spark plug 17, and ISC valve 10, and output of control signals, that is, engine control such as fuel injection control, ignition timing control, idle speed control, etc., are performed as shown in FIG. This is performed by a control unit (ECU) 40.
[0043]
The ECU 40 is composed mainly of a microcomputer in which a CPU 41, a ROM 42, a RAM 43, a backup RAM 44, a counter / timer group 45, and an I / O interface 46 are connected to each other via a bus line, and supplies stabilized power to each part. Peripheral circuits such as a constant voltage circuit 47, a drive circuit 48 connected to the I / O interface 46, and an A / D converter 49 are incorporated.
[0044]
The counter / timer group 45 is a free-run counter, various counters such as a cam angle sensor signal (cam pulse) input counter, a fuel injection timer, an ignition timer, and a periodic interrupt for generating a periodic interrupt. Various timers such as timers, timers for measuring the crank angle sensor signal (crank pulse) interval, and watchdog timers for monitoring system anomalies are collectively referred to for convenience, and various other software counters and timers are used. .
[0045]
The constant voltage circuit 47 is connected to the battery 51 via a first relay contact of a power relay 50 having two relay contacts. The relay coil of the power relay 50 is connected to the battery 51 via an ignition switch 52. It is connected. The constant voltage circuit 47 is directly connected to the battery 51. When the ignition switch 52 is turned on and the contact of the power relay 50 is closed, power is supplied to each part in the ECU 40. On the other hand, the ignition switch Regardless of whether 52 is ON or OFF, the backup RAM 44 is always supplied with backup power. Further, the fuel pump 14 is connected to the battery 51 through a relay contact of the fuel pump relay 53. A power line for supplying power from the battery 51 to each actuator is connected to the second relay contact of the power relay 50.
[0046]
The input port of the I / O interface 46 includes an idle switch 25b, a knock sensor 26, a crank angle sensor 32, a cam angle sensor 35, a vehicle speed sensor 36 for detecting a vehicle speed, and a starter switch for detecting a start state. 37, and the intake air amount sensor 24, throttle opening sensor 25a, cooling water temperature sensor 28, and O2 sensor 29 are connected via the A / D converter 49, and the battery voltage VB. Is input and monitored.
[0047]
On the other hand, at the output port of the I / O interface 46, an alarm lamp 38 is provided on the relay coil of the fuel pump relay 53, the ISC valve 10, the injector 11, and an instrument panel (not shown) to centrally display various alarms. Are connected via the drive circuit 48 and the igniter 19 is connected.
[0048]
Further, an external connection connector 55 is connected to the I / O interface 46. By connecting a serial monitor (portable failure diagnosis device) 60 to the external connection connector 55, the ECU 40 is connected to the ECU 40 by the serial monitor 60. By making use of the I / O data and the self-diagnosis function of the ECU 40, it is possible to make a diagnosis by reading out the failure part including the cylinder misfire determination NG flag FNG # i, which will be described later, indicating the misfire cylinder stored in the backup RAM 44, and the trouble data indicating the failure content Yes. Further, the trouble monitor data can be initially set (cleared) by the serial monitor 60.
[0049]
The trouble data diagnosis and initial set by the serial monitor 60 are described in detail in Japanese Patent Publication No. 7-76730 by the present applicant.
[0050]
In the CPU 41, in accordance with a control program stored in the ROM 42, the detection signals from the sensors and switches inputted through the I / O interface 46, the battery voltage, etc. are processed, and various data stored in the RAM 43, Based on various learning value data stored in the backup RAM 44, fixed data stored in the ROM 42, etc., the fuel injection amount, ignition timing, duty ratio of the drive signal for the ISC valve 10, etc. are calculated, fuel injection control, Engine control such as ignition timing control and idle speed control is performed.
[0051]
In such an engine control system, the ECU 40 performs a misfire diagnosis based on the change in the engine rotation state quantity and the output value (output voltage) of the O2 sensor 29.
[0052]
That is, a change in the engine rotation state amount for each predetermined crank angle is detected, and the cylinder is determined. Then, the change in the engine rotation state amount for each cylinder is compared with the misfire determination level, and misfire is determined for each cylinder. Further, misfire is determined based on the output voltage O2S of the O2 sensor 29 as the output value of the air-fuel ratio sensor. When the misfire of a predetermined cylinder is determined based on the change in the engine rotation state amount and the misfire is determined based on the O2 sensor output voltage O2S, the misfire of the cylinder is diagnosed.
[0053]
That is, when the change in the engine rotation state amount for each cylinder is compared with the misfire determination level to determine misfire for each cylinder, and when misfire is determined for a predetermined cylinder, O2 indicating the oxygen concentration in the exhaust gas is further displayed. It is determined whether misfire has occurred based on the sensor output voltage O2S. Then, because the misfire of the corresponding cylinder is determined and diagnosed based on the consistency of these discrimination results, the misfire judgment level is set strictly (small) in the high engine speed range, for example, misfire due to variations in engine rotation of each cylinder. Even if it is determined that the misfire of the predetermined cylinder is not performed, this is verified by the misfire diagnosis based on the O2 sensor output voltage O2S, so that the erroneous diagnosis can be surely prevented.
[0054]
In addition, when there is a true misfire, misfire diagnosis is also performed by the misfire diagnosis by the O2 sensor output voltage O2S. Therefore, the misfire diagnosis accuracy can be improved by synergy with the strict setting of the misfire determination level, and Even in the high engine speed range, it becomes possible to determine the misfire cylinder.
[0055]
More specifically, in the present embodiment, the current combustion stroke cylinder #n is determined based on the crank pulse and the cam pulse output from the crank angle sensor 32 and the cam angle sensor 35, respectively. Then, the engine rotation speed MNXn calculated this time is adopted as the engine rotation state quantity for each predetermined crank angle between BTDC θ2 and θ3 between the cylinders, that is, between the θ2 and θ3 crank pulse inputs. Is subtracted from the previous engine rotational speed MNXn-1 in the same section to calculate the rotational speed difference as the misfire parameter of the combustion stroke cylinder #n, that is, the differential rotational speed DELNE # n with respect to the cylinder #n. Then, the differential rotation DELNE # i (i = 1 to 4) for each cylinder in a predetermined period until the number of ignitions reaches the set number is averaged to calculate a differential rotation average value DNAVE # i for each cylinder. Next, the maximum value DNAVEMAX among these differential rotation average values DNAVE # i is discriminated, and the difference SDNAVE # i between the maximum value DNAVEMAX and the differential rotation average value DNAVE # i for each cylinder is calculated. Then, the average value difference SDNAVE # i for each cylinder with respect to the maximum value is compared with the misfire determination level LVLMIS, and a cylinder having an average value difference SDNAVE # i larger than the misfire determination level LVLMIS is determined as a misfire cylinder.
[0056]
Further, at this time, misfire is further determined based on the output voltage O2S of the O2 sensor 29 as the output value of the air-fuel ratio sensor, and when misfire is determined, it is determined that the misfire of the misfire cylinder every time, and every time the cylinder is in question. Diagnose a misfire.
[0057]
In the misfire diagnosis using the O2 sensor output voltage O2S, the absolute values of the O2 sensor output voltage O2S and the change amount of the output voltage O2S are integrated for each A / D conversion input of the O2 sensor output voltage O2S as a predetermined period. Then, the O2 sensor output voltage average value O2AVE is calculated based on the integrated value of the O2 sensor output voltage O2S in a predetermined period in which the number of ignitions reaches the set number, and the O2 sensor output voltage average value O2AVE reaches the first determination value. If the absolute value integrated value DO2S of the O2 sensor output voltage change amount in the predetermined period is larger than the second determination value, it is determined that the misfire has occurred.
[0058]
The misfire diagnosis based on the engine speed change for each cylinder and the O2 sensor output voltage O2S is performed when the engine operation state is in the fuel increase region for catalyst protection.
[0059]
That is, the misfire determination level LVLMIS for determining misfire based on a change in engine rotation speed is strictly set even in the engine high rotation range, and the misfire cylinder determined by the change in rotation speed for each cylinder has truly misfired. Whether the engine is high or not is verified based on the O2 sensor output voltage O2S.
[0060]
In this full increase region, when there is no misfire, where no misfire has occurred, the exhaust gas air-fuel ratio after combustion becomes rich due to the increase in fuel and the output voltage O2S of the O2 sensor 29 becomes a substantially constant rich output. Sometimes the oxygen concentration in the exhaust gas increases, resulting in a substantially lean output. Therefore, in the full increase region for protecting the catalyst, there is a large difference in the O2 sensor output voltage O2S between the time of no misfiring and the time of misfiring, so that misfire can be reliably verified based on the O2 sensor output voltage O2S. It becomes possible.
[0061]
That is, the ECU 40 has functions of a rotational state quantity change detecting means, a cylinder judging means, a first misfire judging means, a second misfire judging means, and a misfire diagnostic means according to the present invention, and further calculates a differential rotational state quantity. Also, functions as means, differential rotation state quantity average value calculation means, and integration means are realized.
[0062]
Hereinafter, the misfire diagnosis process according to the present invention executed by the ECU 40 will be described with reference to the flowcharts shown in FIGS.
[0063]
First, when the ignition switch 52 is turned on and the ECU 40 is powered on, the system is initialized, and the flags and counters except for trouble data and various learning values stored in the backup RAM 44 are initialized. It becomes. When the starter switch 37 is turned on and the engine is started, a cylinder discrimination / engine rotation speed calculation routine shown in FIG. 2 is executed for each crank pulse input from the crank angle sensor 32.
[0064]
In this cylinder discrimination / engine rotation speed calculation routine, when the crank rotor 31 rotates and a crank pulse is input from the crank angle sensor 32 in accordance with engine operation, first, in step S1, the crank pulse input this time is θ1. , Θ2, and θ3 are identified based on the cam pulse input pattern from the cam angle sensor 35, and in step S2, the current combustion stroke cylinder is determined from the crank pulse and cam pulse input pattern. Cylinder discrimination such as ignition target cylinder and fuel injection target cylinder is performed.
[0065]
That is, as shown in the time chart of FIG. 8, for example, if there is a cam pulse input between the previous crank pulse input and the current crank pulse input, the current crank pulse is a θ1 crank pulse. Further, the crank pulse to be input next time can be identified as the θ2 crank pulse.
[0066]
In addition, when there is no cam pulse input between the previous and current crank pulse inputs, and there is a cam pulse input between the previous and previous crank pulse inputs, the current crank pulse can be identified as the θ2 crank pulse and is input next time. The crank pulse can be distinguished from the θ3 crank pulse. Also, if there is no cam pulse input between the previous time and this time, and between the previous and the previous crank pulse input, the crank pulse input this time can be identified as the θ3 crank pulse, and the crank pulse input next time is Can be distinguished from θ1 crank pulse.
[0067]
Further, when three cam pulses are input between the previous and current crank pulse inputs (θ5 cam pulse corresponding to the protrusion 34b), the next compression top dead center is the # 3 cylinder, and the current combustion stroke cylinder is # 1. It can be determined that the cylinder and ignition target cylinder are the # 3 cylinder, and the fuel injection target cylinder is the second # 4 cylinder. When two cam pulses are input between the previous and current crank pulse inputs (θ6 cam pulse corresponding to the projection 34c), the next compression top dead center is the # 4 cylinder, and the current combustion stroke cylinder is the # 2 cylinder. It can be determined that the ignition target cylinder is the # 4 cylinder and the fuel injection target cylinder is the # 3 cylinder.
[0068]
In addition, when one cam pulse is input between the previous and current crank pulse inputs (θ4 cam pulse corresponding to the protrusion 34a) and the previous compression top dead center determination is # 4 cylinder, the next compression top dead center Is the # 1 cylinder, the current combustion stroke cylinder can be identified as the # 4 cylinder, the ignition target cylinder can be identified as the # 1 cylinder, and the fuel injection cylinder can be identified as the # 2 cylinder. Similarly, when one cam pulse is input between the previous and current crank pulse inputs and the previous compression top dead center determination is # 3 cylinder, the next compression top dead center is # 2 cylinder, It can be determined that the combustion stroke cylinder is the # 3 cylinder, the ignition target cylinder is the # 2 cylinder, and the fuel injection target cylinder is the # 1 cylinder.
[0069]
In the four-cycle four-cylinder engine 1 of this embodiment, the combustion stroke is in the order of cylinders # 1 → # 3 → # 2 → # 4, and as shown in the time chart of FIG. If the combustion stroke cylinder #n is # 1, the ignition target cylinder reaches compression top dead center # n + 1 = # 3 cylinder, and the fuel injection target cylinder at this time is # n + 3 = #, which is the next two cylinders. It becomes 4 cylinders. Then, the current cylinder discrimination result of the combustion stroke cylinder #n is referred to in the differential rotation calculation routine of FIG. 3 executed for each θ3 crank pulse input, and the differential rotation DELNE # n for the corresponding fuel stroke cylinder #n is calculated. Is accumulated.
[0070]
Although not described in detail here, the ignition timing for the corresponding cylinder # n + 1 is set according to the discrimination result of the ignition target cylinder # n + 1, and ignition timing control is performed for each cylinder. Further, the discrimination result of the fuel injection target cylinder # n + 3 cylinder is referred to in a fuel injection amount setting routine (not shown) executed every predetermined cycle, and the fuel injection amount is set for each cylinder.
[0071]
Thereafter, the process proceeds from step S2 to step S3, and the time from the previous crank pulse input to the current crank pulse input measured by the crank pulse input interval timing timer, that is, the crank pulse input interval time (θ1 crank pulse and θ2 crank The pulse input interval time Tθ12, θ2 crank pulse and θ3 crank pulse input interval time Tθ23, or θ3 crank pulse and θ1 crank pulse input interval time Tθ31) are read out, and the crank pulse input interval time Tθ is detected.
[0072]
Next, the process proceeds to step S4, in which the crank pulse angle corresponding to the crank pulse identified this time is read out, and the current engine speed (engine speed) NE is determined based on the crank pulse angle and the crank pulse input interval time Tθ. Is stored at a predetermined address in the RAM 43 and the routine is exited. The angle between the crank pulses is known and is stored in advance as fixed data in the ROM 42. In the present embodiment, the angle between the θ1 crank pulse and the θ2 crank pulse is 32 ° CA, The angle between the θ2 crank pulse and the θ3 crank pulse is 55 ° CA, and the angle between the θ3 crank pulse and the θ1 crank pulse is 93 ° CA. In consideration of the engine start time, the engine speed NE is calculated at, for example, 150 rpm or more.
[0073]
Then, in the differential rotation calculation routine shown in FIG. 3 executed for each θ3 crank pulse input, each data of the combustion stroke cylinder #n and the engine rotational speed NE is read, and the differential rotation DELNE # for the cylinder #n is read. n is calculated, and this differential rotation DELNE # n is integrated.
[0074]
Here, the differential rotation calculation routine is executed in response to the θ3 crank pulse input. At this time, BTDCθ2 based on the input interval time Tθ23 between the θ2 crank pulse and the θ3 crank pulse by the cylinder discrimination / engine speed calculation routine. , Θ3 is calculated, and the differential rotation DELNE # n representing the misfire parameter for the cylinder #n is calculated based on the engine speed NE between BTDC θ2 and θ3.
[0075]
In this differential rotation calculation routine, first, in step S11, the current combustion stroke cylinder data #n that has been subjected to cylinder determination in the cylinder determination / engine rotation speed calculation routine is read to identify the current combustion stroke cylinder #n. .
[0076]
In step S12, the current engine speed NE, that is, the latest engine speed NE between BTDC θ2 and θ3, is read out and set as the engine speed MNXn corresponding to the current combustion stroke cylinder #n (MNXn ← NE).
[0077]
Next, the process proceeds to step S13, and from the engine speed MNXn corresponding to the current cylinder #n set in step S12, the cylinder # n-1 before the one combustion stroke set in the previous routine execution and stored in the work area. The engine rotational speed MNXn−1 corresponding to is subtracted, and the change in engine rotational speed at each predetermined crank angle between the cylinders, that is, the differential rotational speed DELNE # n with respect to the current combustion stroke cylinder #n is calculated (DELNE # n ← MNXn). -MNXn-1).
[0078]
In step S14, the differential rotation DELNE # n is integrated to obtain a differential rotation integrated value ΣDELNE # n for the cylinder #n (ΣDELNE # n ← ΣDELNE # n + DELNE # n).
[0079]
Next, in step S15, the number of ignition times CIGN giving the misfire diagnosis execution period is counted up (CIGN ← CIGN + 1). Here, this routine is executed every θ3 crank pulse input, that is, every 180 ° CA, and the engine 1 in this embodiment is a four-cylinder engine. Therefore, by counting the number of executions of this routine, the number of ignitions is increased. Is substantially the same as counting.
[0080]
In step S16, in preparation for the calculation of the differential rotation for the next combustion stroke cylinder, the current rotational speed MNXn is stored in the work area as the previous engine rotational speed MNXn-1 (MNXn-1 ← MNXn), and the routine is exited. .
[0081]
With the above differential rotation calculation routine, as shown in FIG. 9, the differential rotation DELNE is calculated for each cylinder, and the differential rotation integrated value ΣDELNE # i corresponding to each cylinder #i (i = 1 to 4 cylinders). Is calculated.
[0082]
On the other hand, FIG. 4 is an O2 sensor output voltage integration routine, which is executed for each A / D conversion input of the output voltage O2S of the O2 sensor 29, and calculates the absolute value of the O2 sensor output voltage O2S and the change amount of the output voltage O2S. Accumulate.
[0083]
Next, the O2 sensor output voltage integration routine will be described. First, in step S21, the latest A2D-converted O2 sensor output voltage O2Sn is read, and the O2 sensor output voltage O2S is read as the O2 sensor output voltage integration value. Add to ΣO2S and integrate (ΣO2S ← ΣO2S + O2S).
[0084]
Next, in step S22, the O2 sensor output voltage O2Sn-1 at the time of the previous A / D conversion stored in the work area at the time of the previous routine execution is subtracted from the O2 sensor output voltage O2Sn, and the absolute value of this subtraction value is obtained. Is added to the absolute value integrated value DO2S of the O2 sensor output voltage change amount and integrated (DO2S ← DO2S + | O2Sn−O2Sn−1 |).
[0085]
Thereafter, in step S23, the O2 sensor output voltage sampling count CO2S is counted up (CO2S ← CO2S + 1). In subsequent step S24, the O2 sensor output voltage O2Sn is used in preparation for the calculation of the absolute value integrated value DO2S at the next routine execution. The O2 sensor output voltage O2Sn-1 at the time of previous A / D conversion is stored in the work area (O2Sn-1 ← O2Sn), and the routine is exited.
[0086]
Then, in the misfire diagnosis routine shown in FIG. 5, when the number of ignition times reaches the set number of times that determines the execution period of the misfire diagnosis and the engine operating state is in the fuel increase region for catalyst protection, The calculated differential rotation integration value ΣDELNE # i for each cylinder, the O2 sensor output voltage integration value ΣO2S integrated by the O2 sensor output voltage integration routine, and the absolute value integration value DO2S of the O2 sensor output voltage change amount are read out. . Then, a differential rotation DELNE # i (i = 1 to 4) for each cylinder in a predetermined period until the number of ignition reaches the set number is averaged to calculate a differential rotation average value DNAVE # i for each cylinder, The maximum value DNAVEMAX among these differential rotation average values DNAVE # i is discriminated, and the difference SDNAVE # i between the maximum value DNAVEMAX and the differential rotation average value DNAVE # i for each cylinder is calculated. Then, the average value difference SDNAVE # i for each cylinder with respect to the maximum value is compared with the misfire determination level LVLMIS, and a cylinder having an average value difference SDNAVE # i larger than the misfire determination level LVLMIS is determined as a misfire cylinder.
[0087]
When a misfiring cylinder is determined, the O2 sensor output voltage average value O2AVE is calculated based on the O2 sensor output voltage integrated value ΣO2S in a predetermined period until the number of ignition reaches the set number, and the O2 sensor output is calculated. When the voltage average value O2AVE does not reach the first determination value LVLO2AVE, or when the absolute value integrated value DO2S of the O2 sensor output voltage change amount in the predetermined period is larger than the second determination value LVLDO2S, it is determined that misfire has occurred. Then, the misfire is determined every time in the misfire cylinder, and the misfire is diagnosed every time in the cylinder.
[0088]
Next, the misfire diagnosis routine shown in FIG. 5 will be described.
[0089]
This misfire diagnosis routine is executed every predetermined crank angle (for example, every 180 ° CA) or every predetermined time (for example, 5 msec), and in step S31, the ignition number count value CIGN is set to the set number of times IGN (for example, 400). Times).
[0090]
The set number of times IGN determines the execution cycle of the misfire diagnosis. The average process for the differential rotation DELNE # i for each cylinder, which will be described later, and the calculation of the O2 sensor output voltage average value O2AVE based on the O2 sensor output voltage integrated value ΣO2S. By processing, and the absolute value integrated value DO2S of the O2 sensor sensor output voltage change amount, the influence of engine rotation fluctuation due to snatch or disturbance is eliminated, and the influence of temporary misfire due to acceleration or the like is eliminated. An ignition frequency value suitable for giving a period to be obtained is obtained in advance by simulation or experiment, and this ignition frequency value is set as the set frequency IGN and stored in the ROM 42 as fixed data.
[0091]
That is, the number of ignitions CIGN that directly depends on the number of misfires is adopted, and the ignition number CIGN is compared with the set number of times IGN to determine the execution period of the misfire diagnosis. It becomes possible to make a diagnosis reliably.
[0092]
When CIGN <IGN and the number of ignition times CIGN has not reached the set number of times IGN, the routine is directly exited. On the other hand, when CIGN ≧ IGN and the number of times of ignition CIGN has reached the set number of times IGN, the routine proceeds to step S32. That is, every time the number of ignition times CIGN reaches the set number of times IGN, a misfire diagnosis is performed based on data for a period corresponding to the set number of times IGN.
[0093]
In step S32, it is determined by the full increase coefficient KFULL whether the engine operating state is in a fuel increase area so-called full increase area for catalyst protection.
[0094]
As is well known, the full increase coefficient KFULL is for protecting the catalyst by preventing an abnormal increase in the catalyst temperature by correcting the fuel increase when the engine operating state is at least one of high speed and high load. And is given as a correction term for various increase coefficients COEF in the calculation formula of the fuel injection pulse width Ti that determines the fuel injection amount.
[0095]
As is well known, the fuel injection pulse width Ti is air-fuel ratio corrected by multiplying the basic fuel injection pulse width Tp by the various increase correction coefficients COEF and air-fuel ratio feedback correction coefficient α, and an air-fuel ratio learning correction coefficient. The learning correction is performed by multiplying KBLRC, and the voltage correction pulse width Ts for compensating the invalid injection time of the injector is added, and is given by the following equation.
[0096]
Ti ← Tp × COEF × α × KBLRC + Ts
The full increase coefficient is given as a correction term in various increase coefficients COEF. An example of an arithmetic expression of the above various increase coefficients COEF is shown in the following expression.
[0097]
COEF ← KST × (1 + KFULL + KTW + KAS + KAI + KACC)
Here, KST is a start increase coefficient, and is used to increase the fuel only during cranking during starter motor operation to ensure startability. KTW is a water temperature increase coefficient that determines the fuel increase rate for ensuring drivability when the engine is cold, and KAS is an increase coefficient after start for ensuring the stability of the engine rotation immediately after the engine is started. The initial value is set based on the coolant temperature TW by the coolant temperature sensor 28, and gradually decreases until KAS = 0 after the engine is started by turning off the starter switch 37. KAI is a post-idle increase coefficient for preventing slack at idle release, and is set to an initial value when the idle switch 25b shifts from ON (throttle valve fully closed) to OFF (throttle valve open). It is gradually decreased until KAI = 0. Further, KACC is an acceleration increase coefficient for preventing lean spikes in the air-fuel ratio due to the intake air amount detection delay of the intake air amount sensor 24 during acceleration when the throttle valve 5a is suddenly opened. When acceleration of the throttle valve sudden opening is detected based on the speed, the initial value is set and then gradually decreased until KACC = 0.
[0098]
On the other hand, the full increase coefficient KFULL for protecting the catalyst is set by referring to the full increase coefficient table stored in the ROM 42 based on the engine speed NE and the basic fuel injection pulse width Tp representing the engine load. The
[0099]
The full increase coefficient table obtains an appropriate fuel increase value that can prevent an abnormal increase in the catalyst temperature for each region based on the basic fuel injection pulse width Tp representing the engine load and the engine speed NE by simulation or experiment in advance. The fuel increase appropriate value is set as a full increase coefficient KFULL as a table using the basic fuel injection pulse width Tp and the engine rotational speed NE as parameters, and is stored in a series of addresses in the ROM 42.
[0100]
An example of the full increase coefficient table is shown in FIG. In this full increase coefficient table, as shown by the oblique lines in FIG. 10, when the basic fuel injection pulse width Tp is in at least one of the engine high load area and engine high speed area, the full increase coefficient KFULL is KFULL. > 0 is set. This region is a so-called full increase region where fuel increase correction is performed using the full increase coefficient KFULL. Outside this full increase area, the full increase coefficient KFULL is set to KFULL = 0, and fuel increase correction by the full increase coefficient KFULL is not performed. The full increase coefficient table stores a larger value of the full increase coefficient KFULL as the basic fuel injection pulse width Tp is larger and the engine rotational speed NE is higher, that is, as the engine speed is higher.
[0101]
That is, in the present embodiment, the differential rotation DELNE # i (i = 1 to 4) for each cylinder in a predetermined period until the number of ignition times CIGN reaches the set number of times IGN is averaged by the processing described later. The differential rotation average value DNAVE # i for each cylinder is calculated, the maximum value DNAVEMAX among these differential rotation average values DNAVE # i is determined, and the maximum value DNAVEMAX and the differential rotation average value DNAVE # i for each cylinder are determined. Difference SDNAVE # i is calculated. Then, the misfire cylinder is determined by comparing the average value difference SDNAVE # i for each cylinder with respect to the maximum value with the misfire determination level LVLMIS, and the misfire determination level LVLMIS for determining misfire is determined at the high engine speed. It is also set strictly in the region, and it is verified based on the O2 sensor output voltage O2S whether or not the determined misfire cylinder is truly misfiring.
[0102]
Here, the high engine speed range is adapted to a full increase region in which fuel increase correction is performed using the full increase coefficient KFULL for catalyst protection.
[0103]
In this full increase region, when there is no misfire without misfire, the exhaust gas air-fuel ratio after combustion becomes rich by the fuel increase correction by the full increase coefficient KFULL, and the output voltage O2S of the O2 sensor 29 is almost constant rich. Output. On the other hand, at the time of misfire where combustion is not performed, the oxygen concentration in the exhaust gas increases, so the O2 sensor output voltage O2S becomes a substantially lean output.
[0104]
Therefore, in the full increase range for catalyst protection by this full increase coefficient KFULL, the difference in O2 sensor output voltage O2S between no misfire and no misfire is large, and misfire is reliably verified based on the O2 sensor output voltage O2S. It becomes possible to do.
[0105]
In step S32, it is determined whether or not the engine operation region is in the fuel increase region depending on whether or not the full increase coefficient KFULL is greater than 0. Instead, for example, the engine speed NE and the engine load are represented. A region table based on the basic fuel injection pulse width Tp may be adopted, and the fuel increase region determination may be performed by referring to this region table, and appropriate means can be employed.
[0106]
When the condition is not satisfied outside the fuel increase range of KFULL = 0, the routine jumps to step S40, and in steps S40 to S44, the differential rotation integrated value ΣDELNE # i, the number of ignition times CIGN, the O2 sensor output voltage of each cylinder. Clear accumulated value ΣO2S, absolute value accumulated value DO2S of O2 sensor output voltage change amount, and O2 sensor output voltage sampling count CO2S (ΣDELNE # i ← 0, CIGN ← 0, ΣO2S ← 0, DO2S ← 0, CO2S ←) 0) Exit the routine.
[0107]
On the other hand, when the condition in the fuel increase region where KFULL> 0 is satisfied in step S32, the process proceeds to step S33, where the misfire cylinder detection subroutine shown in FIG. 6 is executed, and the engine rotation fluctuation is based on the differential rotation DELNE # i for each cylinder. Determine the misfire cylinder.
[0108]
The misfire cylinder detection subroutine of FIG. 6 will be described. For each cylinder #i (i = 1 to 4 cylinders) in a predetermined period in which the number of ignition times CIGN integrated in the differential rotation calculation routine reaches the set number of times IGN in step S51. The differential rotation integrated value ΣDELNE # i is read out, and based on the differential rotation integrated value ΣDELNE # i for each cylinder, the differential rotation average value DNAVE # i for each cylinder in a predetermined period is calculated by the following equation.
[0109]
DNAVE # i ← ΣDELNE # i / (IGN / N)
Here, N is the number of cylinders, and the differential rotation DELNE is accumulated for each cylinder by (IGN / N) times until the number of ignition times CIGN reaches the set number of times IGN. Accordingly, the differential rotation average value ΣDELNE # i for each cylinder in the predetermined period when the ignition number CIGN reaches the set number IGN is divided by (IGN / N), respectively, so that the differential rotation average for each cylinder in this predetermined period is obtained. The value DNAVE # i can be calculated. Further, the above (IGN / N) is known and is previously stored in the ROM 42 as fixed data.
[0110]
In step S52, the magnitude relations of the differential rotation average value DNAVE # i for each cylinder are compared, and the maximum of these differential rotation average values DNAVE # i is set as the maximum value DNAVEMAX.
[0111]
That is, as shown in FIG. 11, for example, of the differential rotation average values DNAVE # 1, DNAVE # 2, DNAVE # 3, and DNAVE # 4 for each cylinder in a predetermined period (400 ignition), the differential rotation of # 1 cylinder The average value DNAVE # 1 is the maximum, and the differential rotation average value DNAVE # 1 of the # 1 cylinder is set as the maximum value DNAVEMAX.
[0112]
Next, the process proceeds to step S53, where the difference rotational average value DNAVE # i for each cylinder is subtracted from the maximum value DNAVEMAX to calculate the average value difference SDNAVE # i for each cylinder with respect to the maximum value DNAVEMAX (SDNAVE # i). <-DNAVEMAX-DNAVE # i).
[0113]
In step S54, the average value difference SDNAVE # i for each cylinder is compared with the misfire determination level LVLMIS, and a cylinder having an average value difference SDNAVE # i larger than the misfire determination level LVLMIS is satisfied with SDNAVE # i> LVVLMIS. It is determined that the misfire cylinder is continuously misfired every time, and in step S55, a cylinder misfire flag FMIS # i representing the continuous misfire of the corresponding cylinder is set (FMIS # i ← 1).
[0114]
Further, it is determined that at least a constant continuous misfire has not occurred at least for each cylinder in which SDNAVE # i ≦ LVLMIS and the average value difference SDNAVE # i is equal to or lower than the misfire determination level LVLMIS, and in step S56, the corresponding cylinder is assigned. The corresponding cylinder misfire flag FMIS # i is cleared (FMIS # i ← 0), and the process proceeds to step S34 of the misfire diagnosis routine (FIG. 5).
[0115]
That is, instead of directly using the differential rotation DELNE # i for each cylinder, by adopting the differential rotation average value DNAVE # i for each cylinder until the number of ignitions CIGN reaches the set number of times IGN, the temporary rotation due to snatch, disturbance, etc. This eliminates the effects of typical engine rotation fluctuations, and eliminates the effects of temporary misfires due to acceleration and the like. Then, the maximum value DNAVEMAX among the differential rotation average values DNAVE # i is discriminated, and the difference SDNAVE # i between the maximum value DNAVEMAX and the differential rotation average value DNAVE # i for each cylinder is respectively determined. This is accurately reflected as a misfire parameter indicating continuous misfire of the cylinder every time. Further, since the average value difference SDNAVE # i for each cylinder with respect to the maximum value DNAVEMAX is compared with the misfire determination level LVLMIS, continuous misfire is determined for each cylinder. Therefore, the differential rotation average value DNAVE # i for each cylinder is determined. Relative to the value (misfire determination level), the above-described average value difference SDNAVE # i for each cylinder as a misfire parameter representing continuous misfire changes greatly depending on the presence or absence of continuous misfire each time and is dynamic. The range is large, and therefore it is possible to more accurately determine the continuous misfire for each cylinder.
[0116]
In addition, since the simple average of the differential rotations in the predetermined period is adopted, it is possible to significantly reduce the calculation burden on the CPU 41 related to the misfire diagnosis.
[0117]
For example, in the example of FIG. 11 described above, the differential rotation average value DNAVE # 1 of the # 1 cylinder is the maximum maximum value DNAVEMAX, and therefore the average value difference SDNAVE # 1 of the # 1 cylinder is SDNAVE # 1 = 0 rpm. The average value difference SDNAVE # 2 of the # 2 cylinder with respect to the maximum value DNAVEMAX is about 18 rpm, the average value difference SDNAVE # 3 of the # 3 cylinder is about 25 rpm, and the average value difference SDNAVE # 4 of the # 4 cylinder is about 65 rpm. In this case, it can be seen that the # 4 cylinder causes continuous misfire every time.
[0118]
However, in the high engine speed range (usually 5000 rpm or more), as described above, the engine rotation caused by manufacturing tolerances and mounting tolerances of the crank rotor and crank angle sensor, and variations in combustion pressure for each cylinder, etc. The influence of speed variation becomes large, and since the inertial energy is large in the high rotation range, the engine rotation change amount at the time of misfire is small, and misfire determination becomes difficult.
[0119]
Therefore, in the high engine speed range, if the misfire determination level LVLMIS as the determination value is set strictly (small), the misfire is erroneously determined even though the misfire does not actually occur due to the rotation variation of each cylinder. End up. In order to cope with this, if the misfire determination level LVLMIS is set to be large, the actual misfire cannot be detected. In any case, in the misfire determination based on the engine rotation change, it is not possible to appropriately determine continuous misfire every time for each cylinder in the high engine speed range.
[0120]
Therefore, at this time, the misfire determination based on the O2 sensor output voltage O2S indicating the oxygen concentration in the exhaust gas is used in combination, and it is verified whether or not a true misfire has occurred based on the O2 sensor output voltage O2S by the processing described later. . In other words, the misfire determination level LVLMIS can be set strictly by combining the misfire determination based on the engine rotation change and the misfire determination based on the O2 sensor output voltage O2S, and the synergy with the strict setting of the misfire determination level LVLMIS. It is possible to greatly improve the misfire diagnosis accuracy in the high engine speed range, and to determine the misfire cylinder in the high engine speed range.
[0121]
When the misfire cylinder detection subroutine is completed and the process proceeds to step S34 of the misfire diagnosis routine (see FIG. 5), the cylinder misfire flag FMIS # i for each cylinder is referred to, and all FMIS # i = 0. When the cylinder-specific misfire flags FMIS # 1, FMIS # 2, FMIS # 3, and FMIS # 4 are cleared for all of the cylinders, and it is determined that no continuous misfire has occurred for all the cylinders # 1 to # 4 7 jumps to step S65 of the misfire verification subroutine of FIG. 7 and stores all the misfire determination NG flags FNG # 1, FNG # 2, FNG # 3, FNG for each cylinder that are stored in the backup RAM 44 and indicate continuous misfire determination for each cylinder. Clear # 4.
[0122]
On the other hand, in step S34, FMIS # i = 1, and any one of the cylinder misfire flags FMIS # 1 to FMIS # 4 is set, and at least one cylinder is continuously misfired every time. If it is determined that the O2 sensor 29 is active, the flow advances to step S35 to determine whether the O2 sensor 29 is active.
[0123]
That is, when the O2 sensor 29 is inactive, the oxygen concentration in the exhaust gas cannot be determined from the output voltage O2S of the O2 sensor 29, and a misfire diagnosis based on the O2 sensor output voltage O2S cannot be performed.
[0124]
Therefore, when the O2 sensor 29 is inactive, the routine jumps to step S40, and in steps S40 to S44, the differential rotation integrated value ΣDELNE # i, the number of ignition times CIGN, the O2 sensor output voltage integrated value ΣO2S, O2 sensor for each cylinder. The absolute value integrated value DO2S of the output voltage change amount and the O2 sensor output voltage sampling count CO2S are cleared, and the routine is exited.
[0125]
The activation determination of the O2 sensor 29 can be determined as the activation of the O2 sensor 29 when, for example, the output voltage O2S of the O2 sensor 29 is equal to or higher than a set value or a predetermined range is maintained for a set value or more. .
[0126]
On the other hand, when the O2 sensor 29 is active in step S35, the process proceeds to step S36, and a misfire verification subroutine shown in FIG. 7 is executed to verify the misfire based on the O2 sensor output voltage O2S indicating the oxygen concentration in the exhaust gas. .
[0127]
The misfire verification subroutine of FIG. 7 will be described. In step S61, the O2 sensor output voltage integration value ΣO2S integrated in the O2 sensor output voltage integration routine and the O2 sensor output voltage sampling count CO2S are read, and the O2 sensor output voltage integration is performed. The value ΣO2S is divided by the O2 sensor output voltage sampling count CO2S to calculate the O2 sensor output voltage average value O2AVE during a predetermined period when the ignition count CIGN reaches the set count IGN (O2AVE ← ΣO2S / CO2S).
[0128]
In step S62, the O2 sensor output voltage average value O2AVE is compared with a first determination value LVLO2AVE for determining misfire based on the O2 sensor output voltage average value O2AVE.
[0129]
When O2AVE <LVLO2AVE and the O2 sensor output voltage average value O2AVE does not reach the first determination value LVLO2AVE, the O2 sensor 29 is in spite of the fuel increase region for catalyst protection, that is, the full increase region. Output voltage O2S continuously indicates a lean output, and it is determined that a continuous misfire has occurred in the predetermined cylinder every time, the process proceeds to step S64, and the corresponding cylinder specified by the cylinder misfire flag FMIS # i The cylinder misfire determination NG flag FNG # i indicating the final misfire diagnosis result and indicating the final misfire diagnosis result and indicating the continuous misfire determination for each cylinder stored in the backup RAM 44 is set (FNG # i ← 0). ), The process proceeds to step S37 of the misfire diagnosis routine (FIG. 5).
[0130]
If O2AVE ≧ LVLO2AVE in step S62, the process proceeds to step S63, where the absolute value integrated value DO2S of the O2 sensor output voltage change amount integrated in the O2 sensor output voltage integration routine is read, and the O2 sensor in this predetermined period is read. The absolute value integrated value DO2S of the output voltage change amount is compared with a second determination value LVLDO2S for determining misfire based on the absolute value integrated value DO2S.
[0131]
When DO2S> LVLDO2S and the absolute value integrated value DO2S of the O2 sensor output voltage change amount is larger than the second determination value LVLDO2S, the O2 sensor output voltage O2S fluctuates due to air-fuel ratio fluctuation due to continuous misfire of the predetermined cylinder. Accordingly, it is determined that the continuous misfire has occurred in the predetermined cylinder every time, and similarly, the process proceeds to step S64, and the continuous misfire of the corresponding cylinder specified by the cylinder misfire flag FMIS # i is determined. Then, the cylinder misfire determination NG flag FNG # i indicating the continuous misfire determination of the corresponding cylinder stored in the backup RAM 44 is set, and the process proceeds to step S37 of the misfire diagnosis routine.
[0132]
That is, when continuous misfire is determined each time for a specific cylinder by misfire determination due to engine rotation change, misfire determination is further performed based on the O2 sensor output voltage O2S to verify misfire, and by misfire determination based on the O2 sensor output voltage O2S. When misfire is determined, it is determined that the specific cylinder is continuously misfired every time.
[0133]
Here, even when the engine operating state is in the fuel increase region for protecting the catalyst, that is, the full increase region, as described above, the full increase coefficient KFULL is set to a larger value as the engine high load and higher rotation. The fuel increase by the full increase coefficient KFULL is large, and the full increase coefficient KFULL is set to a smaller value and the fuel increase is reduced as the engine shifts to the low engine speed or engine load side.
[0134]
The relationship between the O2 sensor output voltage O2S at the time of no misfire and the O2 sensor output voltage O2S at the time of continuous misfire in a specific cylinder when the value of the full increase coefficient KFULL is small and the fuel increase is small. The time chart of FIG. 12 with CIGN as the time axis is shown.
[0135]
As shown in the figure, when there is no misfire, the O2 sensor output voltage O2S becomes a substantially constant rich output due to the fuel increase by the above-mentioned full increase coefficient KFULL, and during misfire, the oxygen concentration in the exhaust gas increases and the O2 sensor output The voltage O2S continues the lean output.
[0136]
Therefore, the O2 sensor output voltage average value O2AVE obtained by averaging the O2 sensor output voltage O2S is large when there is no misfire, and is small when the specific cylinder is continuously misfired every time.
[0137]
On the other hand, the absolute value integrated value DO2S of the O2 sensor output voltage change amount has a small fluctuation of the O2 sensor output voltage O2S regardless of the presence or absence of misfire when the value of the full increase coefficient KFULL is small and the fuel increase amount is small. As shown in FIG. 13, it is a small value for both no misfire and misfire. Therefore, when the fuel increase is small, it is difficult to determine the misfire based on the absolute value integrated value DO2S of the O2 sensor output voltage change amount.
[0138]
That is, when the value of the full increase coefficient KFULL is small and the fuel increase is small, it is possible to determine misfire based on the O2 sensor output voltage average value O2AVE. At this time, the O2 sensor output voltage average value O2AVE is the first value. It can be determined that a continuous misfire has occurred in the predetermined cylinder every time by being below the determination value LVLO2AVE.
[0139]
In the misfire determination based only on the O2 sensor output voltage O2S, only the misfire and the continuous misfire of the predetermined cylinder can be determined each time, and it is not possible to specify the cylinder that causes the continuous misfire every time. Since the misfire cylinder is specified by the above, by using the misfire determination based on the engine rotation change and the misfire determination based on the O2 sensor output voltage O2S in combination, the misfire cylinder causing the continuous misfire every time even in the high engine speed range. Certainly can be identified.
[0140]
On the other hand, when the value of the full increase coefficient KFULL is large and the fuel increase is large, the relationship between the O2 sensor output voltage O2S when no misfire occurs and the O2 sensor output voltage O2S when a continuous misfire occurs in a specific cylinder each time, This is shown in the time chart of FIG.
[0141]
As shown in FIG. 14, at the time of no misfire, the O2 sensor output voltage O2S becomes a substantially constant rich output by the fuel increase by the full increase coefficient KFULL. In addition, at the time of continuous misfire of the predetermined cylinder, even if the oxygen concentration in the exhaust gas increases due to misfire, the richness of the base air-fuel ratio is large when the fuel increase is large, so the O2 sensor output voltage O2S is still Rich output. However, the O2 sensor output voltage O2S fluctuates due to air-fuel ratio fluctuation due to misfire.
[0142]
Therefore, when the fuel increase is large, the O2 sensor output voltage O2S becomes a rich output regardless of the presence or absence of misfire. Therefore, the O2 sensor output voltage average value O2AVE obtained by averaging the O2 sensor output voltage O2S is obtained when there is no misfire and during misfire. Both are rich and the difference is small. For this reason, when the fuel increase amount is large, it is difficult to determine misfire by the O2 sensor output voltage average value O2AVE.
[0143]
On the other hand, as shown in FIG. 15, the absolute value integrated value DO2S of the O2 sensor output voltage change amount is small when there is no misfire because the O2 sensor output voltage O2S is a substantially constant rich output with little fluctuation. In addition, at each successive misfire, the absolute value integrated value DO2S of the O2 sensor output voltage change amount increases due to the fluctuation of the O2 sensor output voltage O2S.
[0144]
Therefore, when the value of the full increase coefficient KFULL is large and the fuel increase is large, it is possible to determine misfire by the absolute value integrated value DO2S of the O2 sensor output voltage change amount. At this time, the O2 sensor output voltage change amount When the absolute value integrated value DO2S exceeds the second determination value LVLDO2S, it can be determined that continuous misfire has occurred in the predetermined cylinder every time.
[0145]
That is, by determining both the O2 sensor output voltage average value O2AVE and the absolute value integrated value of the O2 sensor output voltage change amount, it is possible to appropriately determine misfire regardless of the difference in fuel increase by the full increase coefficient KFULL. It becomes possible.
[0146]
On the other hand, in step S63, when DO2S ≦ LVLDO2S, that is, when O2AVE ≧ LVLO2AVE and DO2S ≦ LVLDO2S, the output voltage O2S of the O2 sensor 29 corresponds to the fuel increase region for catalyst protection, that is, the full increase region. It is determined that the rich output is continuously displayed, the air-fuel ratio fluctuation due to continuous misfire does not occur every time, and no continuous misfire occurs in any cylinder, and the process proceeds to step S65, and the backup RAM 44 All cylinder misfire determination NG flags FNG # 1, FNG # 2, FNG # 3, and FNG # 4 that are stored and indicate continuous misfire determination for each cylinder are cleared, and the process proceeds to step S37 of the misfire diagnosis routine.
[0147]
That is, even if it is determined that the specific cylinder is continuously misfired every time by the misfire determination due to the engine rotation change described above, if it is determined that there is no misfire based on the misfire determination based on the O2 sensor output voltage O2S, the misfire due to this engine rotation change. The determination result is determined to be due to the manufacturing tolerances and mounting tolerances of the crank rotor and crank angle sensor, and the influence of variations in engine rotation speed due to variations in combustion pressure for each cylinder. At this time, no misfire is determined by verification based on the O2 sensor output voltage O2S.
[0148]
When the misfire verification subroutine is completed and the process proceeds to step S37 of the misfire diagnosis routine (see FIG. 5), the cylinder misfire determination NG flag FNG # i for each cylinder stored in the backup RAM 44 is referred to. If any one of the cylinder misfire determination NG flags FNG # 1, FNG # 2, FNG # 3, and FNG # 4 is set for each cylinder, the process proceeds to step S38, and the flashing cycle is short or long. The warning lamp 38 is flashed with a predetermined flashing code based on the number of flashes or a combination thereof, and the driver is notified of the occurrence of misfire. Then, in preparation for the next diagnosis, after the above steps S40 to S44, the differential rotation integrated value ΣDELNE # i, the number of ignitions CIGN, the O2 sensor output voltage integrated value ΣO2S, the absolute value integrated value of the O2 sensor output voltage change amount of each cylinder The DO2S and O2 sensor output voltage sampling times CO2S are cleared, and the routine is exited.
[0149]
On the other hand, when FNG # i = 0 in step S37, the cylinder misfire determination NG flags FNG # 1, FNG # 2, FNG # 3, and FNG # 4 for all cylinders are cleared, and all cylinders # 1 to # 1 are cleared. When continuous misfire does not occur every time for # 4, the process proceeds to step S39, the flashing of the alarm lamp 38 by the predetermined blinking code is stopped due to no misfire, and the routine is exited through steps S40 to S44.
[0150]
As a result, when misfiring has occurred in a predetermined cylinder, the warning lamp 38 is notified by blinking, and the driver can easily determine that misfiring has occurred.
[0151]
Further, when troubleshooting at a service factory such as a dealer, the serial monitor 60 is connected to the external connection connector 55, so that the serial monitor 60 uses the cylinder misfire determination NG flag FNG # i for each cylinder in the ECU 40. By reading the trouble data, it is possible to easily and accurately determine which cylinder is misfiring.
[0152]
After repairing the corresponding part, the serial monitor 60 clears the cylinder specific misfire determination NG flag FNG # i. In the present embodiment, even if the cylinder specific misfire determination NG flag FNG # i is not cleared by the serial monitor 60 after the corresponding part is repaired, if the misfire does not occur due to the return to the normal state, the misfire may occur. In the diagnosis routine, the misfire determination NG flag FNG # i for each cylinder is cleared.
[0153]
In this embodiment, the engine rotation speed (engine speed) is used as the engine rotation state quantity, and misfire is detected based on the engine rotation change amount for each predetermined cycle. However, the engine rotation speed, the engine rotation angular velocity, or the engine rotation angular acceleration may be used instead of the engine rotation speed.
[0154]
In this embodiment, when the engine operating state is in the fuel increase region for protecting the catalyst, misfire is diagnosed by using the engine rotation change amount and the O2 sensor output voltage O2S together. When it is outside, the misfire diagnosis may be performed only by the engine rotation change amount. That is, when outside the fuel increase region for catalyst protection, the engine is outside the high engine speed range, the inertia energy is small with respect to the engine high engine speed range, the engine rotation change amount at the time of misfiring is large, and the combustion pressure for each cylinder The influence of variations in engine rotation for each cylinder due to variations in the engine is small. Accordingly, at this time, there is no problem in terms of diagnosis accuracy even if the misfire diagnosis is performed only by the engine rotation change amount.
[0158]
【The invention's effect】
  As described above, according to the first aspect of the invention,The engine rotation state amount for each constant crank angle is detected, the current combustion stroke cylinder is discriminated, and the engine rotation state amount detected this time is subtracted from the previously detected engine rotation state amount to determine the rotation state amount for the combustion stroke cylinder. Calculate the difference. Then, the differential rotation state quantity for each cylinder in the predetermined period is averaged to calculate a differential rotation state quantity average value for each cylinder, and the maximum value of these differential rotation state quantity average values and each cylinder are calculated. The difference from the average value of the difference in rotation state is calculated. Then, an average value difference for each cylinder with respect to the maximum value is compared with a misfire determination level, and a cylinder having a larger average value difference than the misfire determination level is determined as a misfire cylinder. Further, when misfiring is determined based on the output value of the air-fuel ratio sensor, and misfiring is determined, it is determined that the misfiring cylinder is misfired every time and the misfiring is diagnosed every time in the cylinder., DWhen determining misfire due to engine rotation change, the difference in the rotation state quantity for each cylinder is not directly used as the change in the engine rotation state quantity, but the average value of the difference rotation state quantity for each cylinder in a predetermined period is adopted. It is possible to eliminate the effects of temporary engine rotation fluctuations due to acceleration and to the effects of temporary misfires due to acceleration or the like. Further, the difference between the maximum value of the average value of the differential rotation state quantities and the average value of the differential rotation state quantity for each cylinder is calculated, and the average value difference for each cylinder with respect to the maximum value is calculated as a final misfire parameter. Therefore, it is possible to accurately reflect the continuous misfire of the specific cylinder every time.
[0159]
Furthermore, since the average value difference for each cylinder with respect to this maximum value is compared with the misfire determination level, continuous misfire is determined every time for each cylinder, so the average value of the rotational speed difference for each cylinder is determined as the determination value (misfire determination level). Compared to the misfire diagnosis, the above average value difference for each cylinder as a misfire parameter representing continuous misfire changes greatly depending on the presence or absence of continuous misfire every time, and the dynamic range is large. This has the effect of making it possible to determine consecutive misfires for each cylinder.
[0160]
  At that time, the claim2In the described invention, when the engine operation state is in the fuel increase region for catalyst protection, misfire determination is performed based on the engine rotation state amount and the air-fuel ratio sensor output voltage. That is, the fuel increase area for catalyst protection is compatible with the high engine speed range, and when there is no misfire under this fuel increase area, the exhaust gas air-fuel ratio after combustion becomes rich due to the fuel increase and the air-fuel ratio sensor output value is almost constant. In the case of misfire where combustion is not performed, the oxygen concentration in the exhaust gas increases, so the air-fuel ratio sensor output value becomes a substantially lean output. Therefore, in the fuel increase region for protecting the catalyst, the difference in the air-fuel ratio sensor output value between the no-fire and the misfire is large.2According to the described invention, the above claims1In addition to the effects of the present invention, the misfire determination result due to the engine rotation change can be reliably verified based on the air-fuel ratio sensor output value.
[0161]
  Claim3In the described invention, the output voltage of the air-fuel ratio sensor and the absolute value of the change amount of the output voltage are integrated for each predetermined period. When the misfire diagnosis is performed based on the output value of the air-fuel ratio sensor, when the engine operating state is in the fuel increase region for catalyst protection, the air-fuel ratio sensor output voltage average is calculated based on the integrated value of the air-fuel ratio sensor output voltage for a predetermined period. When the average value of the air-fuel ratio sensor output voltage does not reach the first determination value, or the absolute value integrated value of the air-fuel ratio sensor output voltage change amount in a predetermined period is greater than the second determination value. If it is too large, it is determined that there is a misfire.1In addition to the effects of the present invention, when the engine operating state is in the fuel increase region for protecting the catalyst and the fuel increase is small, the presence or absence of misfire is reliably reflected in the average value of the air-fuel ratio sensor output voltage in a predetermined period. Therefore, misfire can be accurately determined by comparing the air-fuel ratio sensor output voltage with the first determination value. When the fuel increase is large, the absolute value integration of the air-fuel ratio sensor output voltage change amount during a predetermined period is performed. The value accurately reflects the presence or absence of misfire, and the misfire can be accurately determined by comparing the absolute value integrated value of the air-fuel ratio sensor output voltage change amount with the second determination value. Regardless, it has the effect of being able to determine misfire appropriately.
[0162]
  Claims4In the described invention, since each average value is calculated by a simple average over a predetermined period,1 toClaim3In addition to the effects of the described invention, it has the effect of remarkably reducing the computation burden on the CPU related to misfire diagnosis.
[0163]
  Claim5In the described invention, the predetermined period is a period every time the number of ignition reaches the predetermined number.1Or claims4In addition to the effects of the described invention, the number of ignitions that directly depends on the number of misfires is adopted, and the execution period cycle of the misfire diagnosis is determined by this number of ignitions, so it is caused by the effect of snatch, disturbance, etc., temporary misfire due to acceleration, etc. This has the effect of eliminating the influence and reliably diagnosing the constant misfire of the predetermined cylinder, that is, the continuous misfire of the predetermined cylinder every time.
[0164]
  At that time, the claim6In the description, the engine rotational speed is used as the rotational state quantity.5The described invention can be incorporated into a conventional engine misfire diagnostic apparatus very easily, and the misfire diagnostic accuracy can be improved.
[Brief description of the drawings]
FIG. 1 is a basic configuration diagram of the present invention.
FIG. 2 is a flowchart of a cylinder discrimination / engine speed calculation routine.
FIG. 3 is a flowchart of a differential rotation calculation routine.
FIG. 4 is a flowchart of an O2 sensor output voltage integration routine.
FIG. 5 is a flowchart of a misfire diagnosis routine.
FIG. 6 is a flowchart of a misfire cylinder detection subroutine based on misfire determination based on engine rotation change.
FIG. 7 is a flowchart of a misfire verification subroutine based on an O2 sensor output voltage.
FIG. 8 is a time chart showing the relationship between crank pulse, cam pulse, combustion stroke cylinder, ignition timing, and fuel injection timing.
FIG. 9 is a timing chart showing a calculation state of differential rotation for each cylinder.
FIG. 10 is an explanatory diagram of a full increase coefficient table.
FIG. 11 is a time chart showing the relationship of differential rotation DELNE # i, differential rotation average value DNAVE # i, maximum value DNAVEMAX, and average value difference SDNAVE # i for each cylinder with respect to maximum value DNAVEMAX for each cylinder;
FIG. 12 is a time chart showing the relationship between the O2 sensor output voltage O2S at the time of no misfire and the O2 sensor output voltage O2S at the time of continuous misfire in a specific cylinder when the fuel increase is small.
FIG. 13 shows the absolute value integrated value DO2S of the O2 sensor output voltage change amount at the time of no misfire when the fuel increase is small and the absolute value integration of the O2 sensor output voltage change amount when the continuous misfire occurs in the specific cylinder every time. Time chart showing the relationship with the value DO2S
FIG. 14 is a time chart showing the relationship between the O 2 sensor output voltage O 2 S when no misfire occurs and the O 2 sensor output voltage O 2 S when a continuous misfire occurs in a specific cylinder each time the fuel increase is large;
FIG. 15 shows the absolute value integration value DO2S of the O2 sensor output voltage change amount when no misfire occurs when the fuel increase is large and the absolute value integration of the O2 sensor output voltage change amount when the continuous misfire occurs every time in a specific cylinder. Time chart showing the relationship with the value DO2S
FIG. 16 is an overall schematic view of an engine.
FIG. 17 is a front view of a crank rotor and a crank angle sensor.
FIG. 18 is a front view of a cam rotor and a cam angle sensor.
FIG. 19 is a circuit configuration diagram of an electronic control system.
[Explanation of symbols]
1 engine
29 O2 sensor (air-fuel ratio sensor)
32 Crank angle sensor
40 electronic control unit (rotation state quantity change detection means, cylinder discrimination means, first misfire discrimination means, second misfire discrimination means, misfire diagnosis means, differential rotation state quantity calculation means, differential rotation state quantity average value calculation means, Integration means)
NE engine speed (engine rotation state)
DELNE differential rotation (change in engine rotation state)
LVLMIS misfire level
O2S O2 sensor output voltage (air-fuel ratio sensor output value; air-fuel ratio sensor output voltage)
#N Current combustion stroke cylinder
MNXn Engine speed corresponding to the current combustion stroke cylinder #n (the amount of engine rotation detected this time)
MNXn-1 Engine rotation speed corresponding to cylinder # n-1 before one combustion stroke (previously detected engine rotation state amount)
DELNE # n Differential rotation relative to the current combustion stroke cylinder #n (difference in rotational state relative to the combustion stroke cylinder)
DELNE # i Differential rotation for each cylinder (differential rotation state quantity for each cylinder)
DNAVE # i Differential rotation average value for each cylinder (differential rotation state quantity average value for each cylinder)
DNAVEMAX maximum value (maximum value of average value of each differential rotation state quantity)
SDNAVE # i Average value difference for each cylinder with respect to the maximum value
O2AVE O2 sensor output voltage average value (air-fuel ratio sensor output voltage average value)
LVLO2AVE first judgment value
Absolute value integrated value of DO2SO2 sensor output voltage variation (absolute value integrated value of air-fuel ratio sensor output voltage variation)
LVLDO2S Second judgment value
CIGN number of ignition
IGN set times (predetermined number)

Claims (6)

An engine rotational state quantity detecting means for detecting an engine rotation state amount of Jo Tokoro crank angle each,
Cylinder discrimination means for discriminating a current combustion stroke cylinder;
Differential rotation state amount calculating means for subtracting the engine rotation state amount detected this time from the previously detected engine rotation state amount to calculate a difference in rotation state amount with respect to the combustion stroke cylinder;
A differential rotation state quantity average value calculating means for averaging the differential rotation state quantity for each cylinder in a predetermined period and calculating a differential rotation state quantity average value for each cylinder;
Calculate the difference between the maximum value of the differential rotation state quantity average value and the differential rotation state quantity average value for each cylinder, and compare the average value difference for each cylinder with respect to the maximum value to the misfire determination level. First misfire determination means for determining a cylinder having an average value difference larger than the misfire determination level as a misfire cylinder;
Second misfire determination means for determining misfire based on the output value of the air-fuel ratio sensor;
An engine misfire diagnostic apparatus, comprising: misfire diagnosis means for determining misfire every time the misfire cylinder when the misfire is determined based on an output value of an air-fuel ratio sensor and diagnosing misfire of the cylinder every time. Each misfire determining means, when the engine operating condition is in the fuel increase region, misfire diagnostic apparatus according to claim 1 Symbol placement engine and performing a misfire determination. And an integration unit that integrates the output voltage of the air-fuel ratio sensor and the absolute value of the change amount of the output voltage every predetermined period, and the second misfire determination unit is configured such that when the engine operating state is in the fuel increase region, An air-fuel ratio sensor output voltage average value in a predetermined period is calculated, and when the air-fuel ratio sensor output voltage average value does not reach the first determination value, or the absolute value of the air-fuel ratio sensor output voltage change amount in the predetermined period when the integrated value is larger than the second determination value, misfire diagnostic apparatus according to claim 1 Symbol placement engine and discriminates misfire. Each mean value, misfire diagnostic apparatus for an engine according to any one of claims 1 to claim 3, characterized in that calculated by simple average during a predetermined period. The predetermined period is misfire diagnostic apparatus for an engine according to any one of claims 1 to 4, wherein the number of times of ignition is a period of each time reaches a predetermined number of times. The engine rotation state amount misfire diagnostic apparatus for an engine according to any one of claims 1 to 5, characterized in that an engine rotational speed.

Images