JP4821808B2 - Valve system abnormality detection method - Google Patents

Description

  The present invention is applied to a valve operating system that changes the valve characteristic of an engine valve by reciprocating a drive member between two mechanical limit positions, and the valve operating system is based on the fact that the valve characteristic cannot be changed. The present invention relates to an abnormality detection method for a valve train system that detects drive adhesion of the system.

  In recent years, in order to improve the fuel efficiency and output of an internal combustion engine, a valve operating system that changes the maximum lift amount of the engine valve based on the operating state of the internal combustion engine has been widely adopted (see, for example, Patent Document 1). The following configuration is generally adopted as such a valve train.

  That is, this valve train system includes an input member that abuts on a cam rotated by an engine output shaft and swings based on the rotation, and an output member that reciprocates the engine valve by swinging together with the input member. ing. The input member and the output member are connected to a drive member that can reciprocate between two mechanical limit positions, and an actuator that drives the drive member is connected to the drive member. Yes. When the actuator is driven to reciprocate within the mechanically regulated operating range, the relative phase difference between the input member and the output member is changed, and the maximum lift amount of the engine valve is changed.

  In such a valve system control device, the maximum lift amount of the engine valve is controlled in the following manner. In other words, this control device is provided with a sensor for detecting the operation amount of the actuator, in other words, the maximum lift amount of the engine valve. The microcomputer feedback-controls the duty ratio of energization to the actuator so that the deviation between the detected value of the maximum lift amount detected by the sensor and the control target value set based on the engine operating state is small. The above-described drive member is driven to control the maximum lift amount of the engine valve.

By the way, during the above-described control of the maximum lift amount, drive sticking may occur in the valve operating system due to foreign object biting or the like, and the maximum lift amount may not be changed. Therefore, in order to detect such drive fixation at an early stage, for example, an abnormality detection method described in Patent Document 2 can be employed. Specifically, in the abnormality detection method described in Patent Document 2, the detected value of the maximum lift amount is detected even though the duty ratio of energization to the actuator and the current flowing through the actuator are larger than a predetermined threshold. When it does not change, it is determined that the maximum lift amount cannot be changed, and based on this, it is determined that the drive sticking of the valve train has occurred.
JP 2005-201117 A JP 2007-205296 A

  In this way, it is possible to determine whether or not the valve system is locked based on the fact that the maximum lift amount cannot be changed. However, a deviation may occur between the detected value of the maximum lift amount detected by the above-described sensor and the actual value of the maximum lift amount due to noise or the like. When such a deviation occurs, the detection value of the maximum lift amount detected by the sensor does not reach the set control target value, but the drive member passes the position corresponding to the control target value and the mechanical limit position is reached. May reach. In this case, it is determined that the maximum lift amount cannot be changed in a state where the maximum lift amount has not reached the control target value. That is, although the maximum lift amount cannot be changed because the drive member has reached the mechanical limit position due to a deviation between the detected value and the actual value of the maximum lift amount. In other words, it may be erroneously determined that drive sticking has occurred due to biting or the like in the valve train.

  Although the valve system abnormality detection method for changing the maximum lift amount of the engine valve has been described, this inconvenience is not limited to the same configuration, and the valve system for changing other valve characteristics such as the opening / closing timing of the engine valve. Can occur almost in common.

  The present invention has been made in view of such a conventional situation, and an object of the present invention is to provide a drive member that changes valve characteristics due to a deviation between a detected value and an actual value. Provides a valve system abnormality detection method that can accurately detect drive valve sticking and avoid misjudgment that valve drive sticking has occurred when the critical limit position is reached. There is to do.

Hereinafter, means for solving the above-described problems and the effects thereof will be described.
The invention according to claim 1 is a drive member that changes the valve characteristic of the engine valve by reciprocating between two mechanical limit positions, and an actuator that is connected to the drive member and drives the drive member. The valve characteristic control target value is set on the basis of the engine operating state, and the difference between the valve characteristic detection value detected by the detection means and the control target value is reduced. And a drive control means for drivingly controlling the drive member through the actuator, and the change of the valve characteristics becomes impossible during execution of the drive control by the drive control means. In the valve system abnormality detection method for detecting drive adhesion of the valve system, when the valve characteristic cannot be changed for the first time in a predetermined detection cycle, the valve system While deferring detection of dynamic fixation, the drive member is driven toward one of the mechanical limit positions, and when the drive member stops, it is determined that the drive member has reached the mechanical limit position. Initial value learning is performed to set the detected value of the valve characteristic at the time to the initial value of the valve characteristic corresponding to the mechanical limit position, and the valve characteristic cannot be changed again after the initial value learning is completed. The gist of this is to determine that the driving sticking of the valve operating system has occurred.

  In the same configuration, when the change of the valve characteristic becomes impossible for the first time in a predetermined detection cycle, the detection of the valve drive system sticking is suspended and the initial value learning of the valve characteristic is performed, and after the initial value learning is completed When the change of the valve characteristic becomes impossible again, it is determined that the drive sticking of the valve system has occurred. As a result, for example, there is no drive sticking of the valve system, and the drive member reaches the mechanical limit position due to a deviation between the detected value and the actual value of the valve characteristic. When the valve characteristics cannot be changed, the initial value learning of the valve characteristics eliminates the state where there is a deviation between the detected value of the valve characteristics and the actual value, and the control of the valve characteristics is resumed accurately. Can. As a result, it is possible to prevent the drive member from reaching the mechanical limit position again after the initial value learning is completed, resulting in a deviation between the detected value of the valve characteristic and the actual value. Thus, when the drive member reaches the mechanical limit position, it is possible to avoid erroneously determining that the drive sticking of the valve system has occurred. On the other hand, when drive sticking occurs due to biting or the like in the valve train, initial value learning is performed and drive control of the drive member is resumed, but the actual movable range of the drive member is determined by the drive control means. Since it is smaller than the recognized movable range, the drive member may reach the position where the drive sticking occurs or the mechanical limit position even after the initial value learning is completed. When the drive member reaches the position where the drive sticking occurs or the mechanical limit position, the valve characteristic cannot be changed again, and it is determined that the drive sticking of the valve system has occurred. Therefore, according to the above-described configuration, it is erroneously determined that the valve system drive is stuck when the drive member reaches the mechanical limit position due to a deviation between the detected value and the actual value. It is possible to avoid the determination and accurately detect the drive sticking of the valve train.

  According to a second aspect of the present invention, in the valve system abnormality detection method according to the first aspect, the detection means detects a change history of a valve characteristic from a predetermined reference value, and the change history and the reference The gist is to detect the valve characteristic based on the value.

  As a specific configuration of the detecting means for detecting the valve characteristic, for example, as described in claim 2, a change history of the valve characteristic from a predetermined reference value is detected, and the change history and the reference value are determined. A configuration in which the valve characteristic is detected based on this can be employed. When such a configuration is adopted, once a deviation occurs between the detected value of the valve characteristic and the actual value due to noise or the like when detecting the change history, even if the subsequent change history is accurately detected, the deviation is It will not be resolved. For this reason, a misjudgment that the valve system drive sticking has occurred when the drive member reaches the mechanical limit position due to a deviation between the detected value and the actual value is more likely to occur. It becomes easy to do.

  In this regard, according to the above configuration, even when the configuration for detecting the valve characteristic based on the change history from the predetermined reference value and the reference value is adopted, the detected value and the actual value of the valve characteristic In this case, it is possible to avoid erroneously determining that the valve drive system is locked when the drive member reaches the mechanical limit position due to a deviation between the It becomes possible to detect accurately.

  According to a third aspect of the present invention, in the valve system abnormality detection method according to the first or second aspect, when the valve characteristic cannot be changed for the first time in the predetermined detection period, the two mechanical The gist is to perform initial value learning at a mechanical limit position that is closer to the stop position of the drive member among the limit positions.

  According to this configuration, when the valve characteristics cannot be changed, the initial value learning is performed at the mechanical limit position of the two mechanical limit positions that is relatively short from the stop position of the drive member. As a result, the time required for initial value learning is shortened as compared with the case where initial value learning is performed at a mechanical limit position that is distant from the stop position of the drive member. The fluctuation can be suppressed, and the fluctuation of the engine operating state due to the initial value learning can be suppressed.

  According to a fourth aspect of the present invention, in the valve system abnormality detection method according to any one of the first to third aspects, when the valve characteristic cannot be changed for the first time in the predetermined detection cycle, The difference between the detected value of the valve characteristic at that time and the initial value of the valve characteristic corresponding to the mechanical limit position located in the direction in which the drive member is displaced immediately before the change of the valve characteristic becomes impossible is a predetermined amount. If it is larger than the above, the initial value learning is prohibited, and the gist is to determine that the driving sticking of the valve system has occurred.

  Although a deviation may occur between the detected value and the actual value of the valve characteristic due to noise or the like, the possibility that the deviation becomes extremely large is low. Therefore, when the valve characteristic cannot be changed, it corresponds to the detected value of the valve characteristic at that time and the mechanical limit position located in the direction in which the drive member was displaced immediately before the valve characteristic could not be changed. When the difference between the initial value of the valve characteristic to be large is large, it is unlikely that the drive member has reached the mechanical limit position due to a deviation between the detected value of the valve characteristic and the actual value, Rather, it is more likely that drive sticking actually occurs in the valve train.

  In this regard, according to the above configuration, when the valve characteristic cannot be changed for the first time in a predetermined detection cycle, the drive member is displaced immediately before the detected value of the valve characteristic and the valve characteristic cannot be changed at that time. If the difference from the initial value of the valve characteristic corresponding to the mechanical limit position located in the specified direction is larger than the predetermined value, there is a deviation between the detected value of the valve characteristic and the actual value. Thus, it is considered that the possibility that the drive member has reached the mechanical limit position is extremely low, and the initial value learning is prohibited, and it is determined that the drive sticking of the valve system has occurred. As a result, it is possible to detect the drive sticking of the valve system early and accurately.

  Hereinafter, an embodiment in which the present invention is applied to an abnormality detection method for a valve train of an internal combustion engine mounted on a vehicle will be described with reference to FIGS. Here, FIG. 1 is a sectional view showing a partial sectional structure of a valve operating system of an internal combustion engine mounted on a vehicle, and FIG. 2 is a plan view showing an arrangement of the valve operating system of the internal combustion engine. is there.

  As shown in FIGS. 1 and 2, the internal combustion engine has four cylinders (only one is shown in FIG. 1), and a cylinder head 2 includes a pair of exhaust valves 10 corresponding to these cylinders. An intake valve 20 is provided so as to be able to reciprocate. The cylinder head 2 is provided with an exhaust valve opening / closing device 90 and an intake valve opening / closing device 100 corresponding to the exhaust valve 10 and the intake valve 20, respectively.

  The exhaust valve opening / closing device 90 is provided with a lash adjuster 12 corresponding to each exhaust valve 10, and a rocker arm 13 is installed between the lash adjuster 12 and the exhaust valve 10. The base end of the rocker arm 13 is supported by the lash adjuster 12, and the tip end is in contact with the base end portion of the exhaust valve 10. Further, an exhaust camshaft 14 is rotatably supported by the cylinder head 2, and the exhaust camshaft 14 is rotated in conjunction with the rotation of the engine output shaft. A plurality of cams 15 are formed on the exhaust camshaft 14, and a roller 13 a provided at an intermediate portion of the rocker arm 13 is in contact with the outer peripheral surface of the cams 15. The exhaust valve 10 is provided with a retainer 16, and a valve spring 11 is provided between the retainer 16 and the cylinder head 2 in a compressed state. The exhaust valve 10 is urged in the valve closing direction by the urging force of the valve spring 11. Thereby, the roller 13 a of the rocker arm 13 is pressed against the outer peripheral surface of the cam 15. When the cam 15 rotates during engine operation, the rocker arm 13 swings with a portion supported by the lash adjuster 12 as a fulcrum. As a result, the exhaust valve 10 is driven to open and close by the rocker arm 13.

  On the other hand, the intake valve opening / closing device 100 is provided with a compressed valve spring 21, a retainer 26 provided on the intake valve 20, a rocker arm 23, and a lash adjuster 22 as in the exhaust side. An intake camshaft 24 in which a plurality of cams 25 are formed is rotatably supported on the cylinder head 2, and the intake camshaft 24 is also rotated in conjunction with the rotation of the engine output shaft. Here, unlike the exhaust valve opening / closing device 90, the intake valve opening / closing device 100 is provided with an intermediate drive mechanism 50 between the cam 25 and the rocker arm 23. The intermediate drive mechanism 50 has an input unit 51 and a pair of output units 52, and the input unit 51 and the output unit 52 are swingably supported by a support pipe 53 fixed to the cylinder head 2. . The rocker arm 23 is urged toward the output portion 52 by the urging force of the lash adjuster 22 and the valve spring 21, and a roller 23 a provided at an intermediate portion of the rocker arm 23 is brought into contact with the outer peripheral surface of the output portion 52. Yes. As a result, the input portion 51 is urged to swing in the counterclockwise direction W1 together with the output portion 52, and the roller 51a provided at the tip of the radially extending portion of the input portion 51 presses the outer peripheral surface of the cam 25. Is done. That is, due to the urging force of the lash adjuster 22 and the valve spring 21, a load that constantly reduces the relative phase difference between the input unit 51 and the output unit 52 is generated.

  In such an intake valve opening / closing device 100, when the cam 25 rotates during engine operation, the cam 25 presses the input portion 51 while being in sliding contact with the roller 51a, whereby the output portion 52 swings in the circumferential direction of the support pipe 53. It becomes like this. When the output unit 52 swings, the rocker arm 23 swings with the portion supported by the lash adjuster 22 as a fulcrum. As a result, the intake valve 20 is driven to open and close by the rocker arm 23.

  A control shaft 54 that can be driven along the axial direction of the support pipe 53 is inserted. The control shaft 54 is drivingly connected to the input unit 51 and the output unit 52 via a connecting member. When the control shaft 54 is driven along the axial direction, the input unit 51 and the output unit 52 are relatively swung. Next, the intermediate drive mechanism 50 that connects the control shaft 54, the input unit 51, and the output unit 52 will be described in detail with reference to FIG. FIG. 3 is a partially broken perspective view showing the internal structure of the mediation drive mechanism 50.

  As shown in FIG. 3, the input unit 51 is provided between a pair of output units 52, and a substantially cylindrical communication space is formed inside the input unit 51 and the output unit 52. Further, a helical spline 51h is formed on the inner peripheral surface of the input unit 51, and a helical spline 52h in which the helical spline 51h of the input unit 51 and its teeth are inclined in the opposite direction are formed on the inner peripheral surface of the output unit 52. Is formed.

  A substantially cylindrical slider gear 55 is provided in a space formed inside the input unit 51 and the output unit 52. A helical spline 55a that meshes with the helical spline 51h of the input portion 51 is formed at the central portion of the outer peripheral surface of the slider gear 55, and both ends of the outer peripheral surface mesh with the helical spline 52h of the output portion 52. A helical spline 55b is formed.

  Further, a groove 55c extending along the circumferential direction is formed on the inner wall of the substantially cylindrical slider gear 55, and a bush 56 is fitted in the groove 55c. The bush 56 can slide on the inner peripheral surface of the groove 55c along the direction in which the groove 55c extends, but the relative displacement in the axial direction with respect to the slider gear 55 is restricted by the groove 55c.

  The support pipe 53 is inserted into a through space formed inside the slider gear 55, and the control shaft 54 is inserted into the support pipe 53. A long hole 53 a extending in the axial direction is formed in the tube wall of the support pipe 53. A locking pin 57 is provided between the slider gear 55 and the control shaft 54 to connect the slider gear 55 and the control shaft 54 through a long hole 53a. One end of the locking pin 57 is inserted into a recess (not shown) formed in the control shaft 54, and the other end is inserted into a through hole 56 a formed in the bush 56.

  In such an intermediate drive mechanism 50, when the control shaft 54 is displaced along the axial direction, the slider gear 55 is displaced in the axial direction in conjunction with the displacement. The helical splines 55a and 55b formed on the outer peripheral surface of the slider gear 55 are meshed with the helical splines 51h and 52h formed on the inner peripheral surfaces of the input portion 51 and the output portion 52, respectively. When displaced in the axial direction, the input unit 51 and the output unit 52 rotate in opposite directions. As a result, the relative phase difference between the input unit 51 and the output unit 52 is changed, and the maximum lift amount of the intake valve 20 and the operating angle corresponding to the valve opening period of the intake valve 20 are changed. FIG. 4 is a graph showing changes in the maximum lift amount and operating angle of the intake valve 20 that accompany changes in the relative phase difference between the input unit 51 and the output unit 52. As shown in FIG. 4, the maximum lift amount and the working angle of the intake valve 20 increase as the relative phase difference between the input unit 51 and the output unit 52 increases. Note that when the relative phase difference between the input unit 51 and the output unit 52 changes, the maximum lift amount and the operating angle of the intake valve 20 change corresponding to each other. Only that will be described.

  Here, as shown in FIG. 2, an electric actuator 60 (hereinafter simply referred to as “actuator 60”) is provided at the base end portion (right end portion in the figure) of the control shaft 54. The actuator 60 is driven by a microcomputer 70. Hereinafter, the structure of the actuator 60 will be described in detail with reference to FIG. FIG. 5 is a partial cross-sectional view mainly showing the structure of the actuator 60.

  As shown in FIG. 5, the housing 60 a of the actuator 60 is attached to the cylinder head 2. Inside the housing 60a is a motor 61 having a stator 61a having a coil C and a rotor 61b having a permanent magnet, and a planetary gear mechanism for converting the rotation of the motor 61 into a linear motion and transmitting it to the control shaft 54. 62 is provided.

  This planetary gear mechanism 62 has an output shaft 63 provided with a spiral spline on the outer periphery, and a plurality of planetary gears 64 provided on the outer periphery and a roller nut 65 provided on the inner periphery. Configured.

  The output shaft 63 is supported by the housing 60a so as to be displaceable in the axial direction and not to rotate about the axis L, and the tip end portion (the left end portion in FIG. 5) is connected to the control shaft 54 by a connecting member 68. Yes. The roller nut 65 is supported by a housing 60a via a double row angular bearing 66 so as to be rotatable integrally with the rotor 61b.

  Further, the planetary gears 64 are arranged at equal angles between the output shaft 63 and the roller nut 65. The spline of the planetary gear 64 meshes with both the spline of the output shaft 63 and the spline of the roller nut 65 fitted on the output shaft 63, and the planetary gear 64 is connected between the output shaft 63 and the roller nut 65. It is formed to rotate while revolving around 63.

  When the coil 60 of the stator 61a is energized by such an actuator 60, the rotor 61b and the roller nut 65 are rotated about the axis L, and each planetary gear 64 rotates around its own axis while rotating around the axis L. Revolve. The energization of the coil C is duty-controlled by the microcomputer 70 based on the operating state of the engine. Further, as described above, the output shaft 63 cannot rotate about the axis L, but can be displaced in the axial direction, so that the output shaft 63 is displaced in the axial direction by the rotation and revolution of each planetary gear 64. To do.

  As the output shaft 63 is displaced in the axial direction, the control shaft 54 is displaced in the axial direction, so that the drive state of the intermediate drive mechanism 50 is changed, and the maximum lift amount of the intake valve 20 is continuously changed. Become. Specifically, when the control shaft 54 is driven in the direction F of FIG. 5, the maximum lift amount decreases, while when the control shaft 54 is driven in the direction R of FIG. 5, the maximum lift amount increases. The output shaft 63 is fixed with two stoppers 63a and 63b that can come into contact with the housing 60a. The output shaft 63, in other words, the control shaft 54, is driven by the stoppers 63a and 63b. Can be driven. When the output shaft 63 is displaced to a mechanical limit position where the stopper 63a contacts the housing 60a (hereinafter referred to as “Hi end”), the maximum lift amount becomes the maximum value, while the stopper 63b contacts the housing 60a. When the output shaft 63 is displaced to the mechanical limit position (hereinafter referred to as “Lo end”), the maximum lift amount becomes the minimum value.

  As shown in FIG. 5, the actuator 60 includes three electrical angle sensors D1 to D3, and an 8-pole multipolar magnet that rotates integrally with the roller nut 65 in correspondence with the electrical angle sensors D1 to D3. Is provided. These electrical angle sensors D1 to D3 are pulse-like signals as shown in FIGS. 6A to 6C, that is, a logic high level signal “H” and a logic low, according to the magnetism of an 8-pole multipole magnet. The level signal “L” is alternately output. Note that the three electrical angle sensors D1 to D3 are arranged every 120 ° in the circumferential direction of the output shaft 63 so as to obtain such a pulse signal waveform. Therefore, the edge of the pulse signal output from one of these electrical angle sensors D1 to D3 is generated every 45 ° rotation of the output shaft 63. Further, the pulse signal from one of these electrical angle sensors D1 to D3 is shifted in phase to the advance side and the delay side by 30 ° rotation of the output shaft 63 with respect to the pulse signals from the other electrical angle sensors. It is in the state.

  The actuator 60 includes two position sensors S1 and S2 that function as rotary encoders, and a 48-pole multipolar magnet (not shown) that rotates integrally with the output shaft 63 corresponding to the position sensors S1 and S2. Is provided. These position sensors S1 and S2 are pulse signals as shown in FIGS. 6D and 6E, that is, a logic high level signal “H” and a logic low level, according to the magnetism of the 48-pole multipole magnet. The signal “L” is alternately output. Note that the position sensor S1 is disposed at a distance of 176.25 ° from the position sensor S2 in the circumferential direction of the output shaft 63 so that such a pulse signal waveform can be obtained. Therefore, the edge of the pulse signal output from one of the position sensors S1, S2 is generated every 7.5 ° rotation of the output shaft 63. Further, the pulse signal from the position sensor S2 is in a state in which the phase is shifted from the pulse signal from the position sensor S1 to the advance side and the delay side by the 3.75 ° rotation of the output shaft 63.

  Here, the edge interval of the pulse signal combined with the electric angle sensors D1 to D3 is 15 °, whereas the edge interval of the pulse signal combined with the position sensors S1 and S2 is 3.75 °. Therefore, the edge of the pulse signal combined with the position sensors S1 and S2 is generated four times from the generation of the edge of the pulse signal combined with the electric angle sensors D1 to D3 to the next generation of the edge.

  The pulse signals output by the electrical angle sensors D1 to D3 and the position sensors S1 and S2 are taken into the microcomputer 70. The microcomputer 70 includes a central processing unit (CPU) 71 that performs numerical calculation and information processing by a program, a non-volatile memory (ROM) 72a that stores programs and data necessary for various controls, input data and calculation results. A volatile memory (DRAM) 72b for temporarily storing data, and a rewritable nonvolatile memory (EEPROM) 72c for storing a reference position obtained by learning control.

  Further, the microcomputer 70 is connected with sensors for detecting the operating state of the engine, such as a current sensor 73 for detecting a current flowing through the coil C and an accelerator sensor 74 for detecting an opening degree of an accelerator pedal of the vehicle. The microcomputer 70 sets a control target value for the maximum lift amount of the intake valve 20 based on the operating state of the engine, and based on the pulse signals output by the electrical angle sensors D1 to D3 and the position sensors S1 and S2 described above. The actual value of the maximum lift amount of the intake valve 20 is detected. Hereinafter, the procedure for detecting the actual value of the maximum lift amount of the intake valve 20 will be described in detail with reference to FIGS.

  Here, FIGS. 6A to 6E show the waveforms of the pulse signals output from the electrical angle sensors D1 to D3 and the position sensors S1 and S2 when the output shaft 63 of the actuator 60 rotates as described above. Yes. 6 (f) to 6 (h) show patterns in which the electrical angle count value E, the position count value P, and the stroke count value S change with respect to the change in the rotation angle when the actuator 60 rotates. . 7A shows the correspondence between the output signal patterns of the electrical angle sensors D1 to D3 and the electrical angle count value E, and FIG. 7B shows the output signals of the position sensors S1 and S2. A mode in which the position count value P increases or decreases when an edge occurs is shown.

First, each count value will be described.
[Electric angle count value E]
The electrical angle count value E is set based on the pulse signals of the electrical angle sensors D1 to D3 and represents the rotational phase of the motor 61. Specifically, as shown in FIG. 7A, depending on which of the logic high level signal “H” and the logic low level signal “L” is output from each of the electrical angle sensors D1 to D3. Thus, the electrical angle count value E is set to any one of continuous integer values in the range of “0” to “5” and stored in the DRAM 72b. The microcomputer 70 detects the rotational phase of the actuator 60 based on the electrical angle count value E stored in the DRAM 72b, and switches the energized phase of the coil C to rotate the motor forward and backward. Here, when the motor 61 rotates forward, the electrical angle count value E changes in the forward direction in the order of “0” → “1” → “2” → “3” → “4” → “5” → “0”. To do. On the other hand, when the actuator 60 rotates in the reverse direction, the electrical angle count value E changes from “5” → “4” → “3” → “2” → “1” → “0” → “5” in the reverse direction.

[Position count value P]
The position count value P represents a change history from the reference value at the time of engine start with respect to the operating angle of the motor 61, in other words, the control value of the actuator 60. Specifically, of the position sensors S1 and S2, which one of the rising edges and the falling edges of the pulse signal is generated from one sensor, and the logic high level signal “H” and the logic low level signal from the other sensor. Depending on which of “L” is output, either “+1” or “−1” is added to the position count value P (see FIG. 7B). In FIG. 7B, “↑” represents the rising edge of the pulse signal, and “↓” represents the falling edge of the pulse signal. The position count value P obtained by executing such processing is a value obtained by counting the edges of the pulse signals from the position sensors S1 and S2.

  Here, if the motor 61 is rotating forward, the position count value P is incremented by “1” for each edge of the pulse signal from the position sensors S1 and S2 shown in FIGS. 6 (d) and 6 (e). Then, the pattern changes in the direction indicated by the arrow A along the pattern shown in FIG. On the other hand, if the actuator 60 is in reverse rotation, the position count value P is subtracted by “1” for each edge of the pulse signal, and in the direction indicated by the arrow B along the pattern shown in FIG. It will change. The position count value P is reset to “0” when the operation of the internal combustion engine is stopped. Therefore, the position count value P indicates how much the control value of the actuator 60 has changed with respect to the reference value when the engine is started, in other words, how much the maximum lift amount of the intake valve 20 is relative to the reference value when the engine is started. Indicates whether it has changed. The position count value P is stored in the DRAM 72b because it needs to be quickly added and subtracted based on the drive of the intake valve opening / closing device 100.

[Stroke count value S]
The stroke count value S represents the absolute value of the control amount of the actuator 60, in other words, the actual value of the maximum lift amount. During engine operation, the microcomputer 70 adds the position count value P to the stroke count value S, and the stroke count value S is updated to the added value. Note that the final value of the stroke count value S when the engine stop is completed and the drive of the intake valve opening / closing device 100 is stopped is learned as the reference value Sg at the start of the next engine operation and stored in the EEPROM 72c.

  Therefore, the microcomputer 70 calculates the stroke count value S, in other words, the actual value of the maximum lift amount, based on the reference value Sg stored in the EEPROM 72c and the position count value P stored in the DRAM 72b. Then, the microcomputer 70 sets the duty ratio DU of energization to the actuator 60, in other words, the driving load of the actuator 60 so that the deviation between the actual value and the control target value set based on the engine operating state becomes small. Feedback control. Hereinafter, control of the duty ratio of energization to the actuator 60 will be described in detail.

  In the processing related to the feedback control of the duty ratio DU, the duty ratio DU is calculated from the following arithmetic expressions (1) to (4) based on the deviation ΔRVL between the control target value of the maximum lift amount of the intake valve 20 and its actual value. Set through.

DU ← VLP + VLI + VLD (1)

VLP ← KP ・ ΔRVL (2)
KP: Correction coefficient (proportional gain)

VLI ← KI · ΣΔRVL (i) (3)
KI: Correction coefficient (integral gain)

VLD ← KD · (ΣΔRVL (i) −ΣΔRVL (i−1)) / Δt (4)
KD: Correction coefficient (differential gain)
Δt: Control cycle

In the above equation (2), “VLP” is a proportional term in the feedback control, and when there is no divergence tendency between the actual value of the maximum lift amount of the intake valve 20 and its control target value, The proportional term VLP converges to the reference value “0”.

  On the other hand, when the actual value of the maximum lift amount of the intake valve 20 is smaller than the control target value, the proportional term VLP is set to a positive value, and the driving load that biases the control shaft 54 in the direction R in FIG. Occurs.

  On the other hand, when the actual value of the maximum lift amount of the intake valve 20 is larger than the control target value, the proportional term VLP is set to a negative value, and the driving load that biases the control shaft 54 in the direction R in FIG. Occurs.

  Thus, based on the deviation ΔRVL between the actual value of the maximum lift amount and the control target value, the proportional term VLP having a size corresponding to the deviation ΔRVL is calculated, so that it is between the actual value of the maximum lift amount and the control target value. The duty ratio DU is set to an appropriate amount so as to compensate for the deviation.

  In the above equation (3), “VLI” is an integral term in the feedback control. In the above equation (3), “ΣΔRVL” is an integrated value of the deviation ΔRVL in a predetermined period, and the subscript “i” is a value of the deviation ΔRVL calculated in each control cycle in the predetermined period. Respectively. As apparent from the equation (3), when there is a steady deviation between the control target value of the maximum lift amount and the actual value, the integral term VLI gradually increases or decreases. As a result, even if there is a steady deviation between the control target value of the maximum lift amount and the actual value that cannot be compensated for by the proportional term VLP, this deviation is caused by this integral term VLI. Will be countered.

  In the above equation (4), “VLD” is a differential term. As shown in the equation (4), when the deviation ΔRVL between the actual value of the maximum lift amount and the control target value changes abruptly, the differential term VLD changes so as to moderate the change.

  As a result, by such feedback control, as the resistance load for driving the actuator 60 increases, the duty ratio DU of energization to the actuator 60 is set larger, and the current flowing through the actuator 60, more precisely the coil C, increases.

By the feedback control described above, the maximum lift amount of the intake valve 20 can be changed to a value suitable for the engine operating state, and the fuel efficiency and output of the internal combustion engine can be improved.
Incidentally, during the above-described control of the maximum lift amount, for example, the planetary gear mechanism 62 may be stuck to drive due to foreign object biting, and the maximum lift amount may not be changed. Therefore, in order to detect such drive sticking at an early stage, for example, the maximum lift amount of the intake valve 20, in other words, the stroke count value S, although the duty ratio DU and the current flowing through the actuator 60 are larger than a predetermined threshold value. When there is no change, it can be determined that the maximum lift amount cannot be changed, and based on this, it can be determined that the drive valve is stuck.

  However, when noise is generated in the output signals of the position sensors S1 and S2, the detected value Sk of the stroke count value S is, in other words, between the detected value of the maximum lift amount of the intake valve and the actual value of the maximum lift amount. And a stroke count value S corresponding to the actual value of the maximum lift amount (hereinafter referred to as “actual value Sj of stroke count S”) may occur. When such a deviation occurs, the detected value of the maximum lift amount does not reach the set control target value, but the control shaft 54 passes through the position corresponding to the control target value and reaches the Hi end or Lo end. There is a fear. In this case, it is determined that the maximum lift amount cannot be changed while the maximum lift amount has not reached the control target value. That is, when the control shaft 54 reaches the Hi end or the Lo end due to the deviation between the detected value of the maximum lift amount and the actual value, it is erroneously determined that drive sticking has occurred due to biting or the like in the valve system. There is a risk.

Therefore, in this embodiment, such inconveniences are preferably avoided by executing the processing described below.
Hereinafter, the processing procedure of the abnormality detection method according to the present embodiment will be described with reference to the flowcharts of FIGS. The series of detection processes shown in FIGS. 8 to 10 are executed by the microcomputer 70 during a period from the start of engine operation to the next stop, that is, one engine operation trip as one detection cycle.

  In this process, first, it is determined whether or not the maximum lift amount cannot be changed (step S10). Specifically, when the change amount of the stroke count value S is larger than a predetermined change amount within a unit time, it is determined that the maximum lift amount can be changed, and the duty ratio DU and the current flowing through the actuator 60 are When the stroke count value S does not change although it is greater than the predetermined threshold, it is determined that the maximum lift amount cannot be changed. If it is determined that the maximum lift amount can be changed (step S10: NO), it is determined that drive sticking has not occurred in the valve train, and a predetermined time interval is provided. The process in step 10 is repeatedly executed.

  On the other hand, if it is determined that the maximum lift amount cannot be changed (step 10: YES), based on the change history of the stroke count value S, the control shaft 54 immediately before the maximum lift amount cannot be changed. It is determined whether or not has been displaced to the Hi end side (step S20). If it is determined that the control shaft 54 has been displaced to the Hi end side (step S20: YES), the detected value Sk of the stroke count value S when the maximum lift amount cannot be changed and the Hi end are set. It is determined whether or not the difference ΔShi (= Shi−S) between the corresponding stroke count value S and the initial value Shi is equal to or less than a preset threshold value Sth (step S30).

  Here, even if a deviation occurs between the detected value Sk of the stroke count value S and the actual value Sj due to noise or the like, the possibility that the deviation becomes extremely large is low. Therefore, when the maximum lift amount cannot be changed and there is a large difference between the detected value of the stroke count value S and the actual value at that time, there is a deviation between the detected value Sk and the actual value Sj. The possibility that the control shaft 54 has reached the Hi end due to the occurrence is low, but it is more likely that the drive sticking actually occurs in the valve train.

  Therefore, when it is determined that the difference ΔShi is larger than the threshold value Sth (step S30: NO), there is a deviation between the detected value Sk of the stroke count value S and the actual value Sj. The possibility that the control shaft 54 has reached the Hi end is considered extremely low. In this case, it is determined that drive valve sticking has occurred, and a warning lamp provided on the operation panel of the vehicle is turned on (step S32). On the other hand, when it is determined that the difference ΔShi is equal to or smaller than the threshold value Sth (step S30: YES), the deviation is generated between the detected value Sk of the stroke count value S and the actual value Sj. It is determined that there is a possibility that the control shaft 54 has reached the Hi end, and it is determined whether or not the learning history flag Fg is “ON” (step S31). The learning history flag Fg is a flag indicating whether or not the Lo end learning described later has been executed after the start of the current engine operation. The learning history flag Fg is set to “off” when the engine is started and stored in the DRAM 72b. .

  Here, when it is determined that the learning history flag Fg is “off” (step S31: NO), the Lo-end learning has not been performed after the start of the current engine operation, and the maximum lift amount is determined. It is judged that the change of the first becomes impossible. In this case, it is determined that there is a possibility that the control shaft 54 has reached the Hi end due to the deviation between the detected value Sk of the stroke count value S and the actual value Sj, and Lo end learning is executed. .

  The flowchart of FIG. 10 shows the Lo end learning processing procedure. That is, in the Lo end learning process, first, based on the detection value Sk of the stroke count value S when the maximum lift amount cannot be changed, the control target value St of the stroke count value S is obtained through the following equation (5). And the above-described feedback control of the duty ratio DU is executed (step S110).

St ← Sk−Bd (5)
Bd: Weight loss value

In equation (5), the weight loss value Bd is a positive value set in advance. Therefore, the control target value St is set to a value smaller than the detected value Sk of the stroke count value S, and the control shaft 54 is driven so as to be displaced toward the Lo end side. As a result, the maximum lift amount decreases and the position count value P decreases. It should be noted that the amount of reduction Bd is set as appropriate in order to suppress a sudden change in the maximum lift amount.

  Next, similarly to step S10 described above, it is determined whether or not the maximum lift amount cannot be changed based on the duty ratio DU, the current flowing through the actuator 60, and the change amount of the stroke count value S (step S120). . Here, when it is determined that the maximum lift amount can be changed (step S120: NO), it is determined that the control shaft 54 has not reached the Lo end, and the process returns to the previous step S110 to control the control shaft. 54 is continuously driven to the Lo end side. On the other hand, if it is determined that the maximum lift amount cannot be changed (step S120: YES), it is considered that the control shaft 54 has reached the Lo end, and the detected value Sk of the stroke count value S at that time is determined. The stroke count value corresponding to the Lo end stored in the EEPROM 72c is updated to the initial value Slo (step S130), and the position count value P is reset to “0” (step S140).

Then, after such Lo end learning is completed, the process proceeds to step S34 in FIG. 8, the learning history flag Fg is set to “ON”, and the process in step S10 is repeatedly executed.
On the other hand, when it is determined in step S31 described above that the learning history flag Fg is “ON” (step S21: YES), that is, after the Lo end learning is completed, the maximum lift amount cannot be changed again. If it is determined that the valve has been driven, it is determined that drive valve sticking has occurred, and a warning lamp provided on the operation panel of the vehicle is turned on (step S32).

  By the way, when it is determined in step S20 described above that the control shaft 54 has been displaced to the Lo end immediately before the change of the maximum lift amount becomes impossible (step S20: NO), the change of the maximum lift amount is changed. It is determined whether or not the difference ΔSlo (= Sk−Slo) between the detected value Sk of the stroke count value S when disabled and the initial value Slo of the stroke count value S corresponding to the Lo end is equal to or less than the threshold value Sth. (Step S40).

  When it is determined that the difference ΔSlo is larger than the threshold value Sth (step S40: NO), the control shaft 54 is caused by a deviation between the detected value Sk of the stroke count value S and the actual value. Is considered very unlikely to have reached the Lo end. In this case, it is determined that drive valve sticking has occurred, and a warning lamp provided on the operation panel of the vehicle is turned on (step S42). On the other hand, when it is determined that the difference ΔSlo is equal to or smaller than the threshold value Sth (step S40: YES), the control shaft is caused by a deviation between the detected value of the stroke count value S and the actual value. It is determined that 54 may have reached the Lo end, and it is determined whether or not the learning history flag Fg is “ON” (step S41).

  If it is determined that the learning history flag Fg is “OFF” (step S41: NO), the maximum lift amount cannot be changed for the first time after the start of the current engine operation. It is determined that there is a possibility that the control shaft 54 has reached the Lo end due to the deviation between the detected value Sk of S and the actual value Sj, and the above-described Lo end learning is executed (step S43). .

  Then, after such Lo end learning is completed, the learning history flag Fg is set to “ON” (step S44), and the process of step S10 is repeatedly executed. On the other hand, if it is determined in step S41 that the learning history flag Fg is “ON” (step S41: YES), that is, it is determined that the maximum lift amount cannot be changed again after the Lo end learning is completed. If it has been determined, it is determined that drive valve sticking has occurred, and a warning lamp provided on the operation panel of the vehicle is turned on (step S42).

  Next, with reference to FIGS. 11 to 17, a plurality of specific examples of an aspect in which the above-described abnormality detection method processing is executed when the maximum lift amount cannot be changed will be described. Here, FIGS. 11 to 17 illustrate the Lo end, the Hi end, the position where drive fixing occurs (hereinafter referred to as “fixing position”), and the control shaft 54 when the above-described abnormality detection method is executed. It is explanatory drawing which shows typically the relative relationship of these drive positions. As shown in FIGS. 11 to 17, in these specific examples, the initial value Slo of the stroke count value S corresponding to the Lo end is set to “100” and the stroke count value S corresponding to the Hi end is set. The initial value Shi is set to “200”, and the threshold value Sth is set to “15”.

  As shown in FIG. 11 (a), when the control shaft 54 is driven to the Hi end side and the detection value Sk becomes “140” due to the drive fixation of the valve operating system, the maximum lift amount is detected in the current detection cycle. When the change becomes impossible for the first time (step S10: YES, step S20: YES), it is determined whether or not the difference ΔShi between the detection value Sk and the initial value Shi is equal to or less than the threshold value Sth (step S30). Here, since the difference ΔShi (“200” − “140” = “60”) is larger than the threshold value Sth (“15”) (step S30: NO), the Lo end learning is not executed and the valve train is driven. It is determined that sticking has occurred (step S32).

  Further, as shown in FIG. 11B, when the control shaft 54 is driven to the Lo end side and the detection value Sk becomes “160” due to the drive fixation of the valve operating system, the maximum lift is detected in the current detection cycle. When the change of the amount becomes impossible for the first time (step S10: YES, step S20: NO), it is determined whether or not the difference ΔSlo between the detected value Sk and the initial value Slo is equal to or less than the threshold value Sth (step S40). Here, since the difference ΔSlo (“160” − “100” = “60”) is larger than the threshold value Sth (“15”) (step S40: NO), the Lo end learning is not executed and the valve train is driven. It is determined that the sticking has occurred (step S42).

  As shown in FIG. 12A, the control shaft 54 is driven to the Hi end side, and the detected value Sk becomes “190” due to the deviation between the detected value Sk of the stroke count value S and the actual value Sj. When the control shaft 54 reaches the Hi end and the maximum lift amount cannot be changed for the first time in this detection cycle (step S10: YES, step S20: YES), the difference between the detection value Sk and the initial value Shi It is determined whether or not ΔShi is equal to or less than a threshold value Sth (step S30). Here, the difference ΔShi (“200” − “190” = “10”) is equal to or less than the threshold value Sth (“15”) (step S30: YES), and the maximum lift amount cannot be changed for the first time (step S30). S31: NO), Lo end learning is executed (step S33). Then, as shown in FIG. 12B, in Lo end learning, the control shaft 54 is driven to the Lo end (steps S110 and S120), and the detected value Sk (“90”) of the stroke count value S at that time. ) Is set to the initial value Slo (“100”). Thereby, the deviation between the detected value of the stroke count value S and the actual value is eliminated, and the control of the maximum lift amount is accurately restarted.

  As shown in FIG. 13A, the control shaft 54 is driven to the Lo end side, and the stroke count value S is set to “110” due to a deviation between the detected value Sk of the stroke count value S and the actual value Sj. When the control shaft 54 reaches the Lo end at the time when the maximum lift amount cannot be changed for the first time in the current detection cycle (step S10: YES, step S20: NO), the detection value Sk and the initial value Slo It is determined whether or not the difference ΔSlo is less than or equal to the threshold value Sth (step S40). Here, the difference ΔSlo (“110” − “100” = “10”) is equal to or less than the threshold value Sth (“15”) (step S40: YES), and the maximum lift amount cannot be changed for the first time (step S40). S41: NO), Lo end learning is executed (step S43). Then, as shown in FIG. 13B, in Lo end learning, it is determined that the control shaft 54 has reached the Lo end (step S120: YES), and the detected value Sk (“ 110 ") is set to the initial value Slo (" 100 "). Thereby, the deviation between the detected value of the stroke count value S and the actual value is eliminated, and the control of the maximum lift amount is accurately restarted.

  As shown in FIG. 14A, when the control shaft 54 is driven to the Hi end side and the detection value Sk of the stroke count value S becomes “190” due to the drive fixation of the valve system, the current detection cycle When the change of the maximum lift amount is impossible for the first time (step S10: YES, step S20: YES), it is determined whether or not the difference ΔShi between the detected value Sk and the initial value Shi is equal to or less than the threshold value Sth (step S30). ). Here, the difference ΔShi (“200” − “190” = “10”) is equal to or less than the threshold value Sth (“15”) (step S30: YES), and the maximum lift amount cannot be changed for the first time (step S30). S31: NO), Lo end learning is executed (step S33). As shown in FIG. 14B, in Lo end learning, the control shaft 54 is driven to the Lo end (steps S110 and S120), and the detected value Sk (“100”) of the stroke count value S at that time is The initial value Slo (“100”) is set. Here, after the Lo end learning is completed, there is no deviation between the detected value of the stroke count value S and the actual value, but the actual movable range of the control shaft 54, that is, the Lo end (Sj = “100”). To the fixing position (Sj = “190”) is smaller than the movable range recognized by the microcomputer 70, that is, the range from the Lo (Sj = “100”) end to the Hi end (Sj = “200”). . Therefore, as shown in FIG. 14C, even after the Lo end learning is completed, the control shaft 54 may reach the fixing position, and the maximum lift amount may not be changed again (step S31: YES) In this case, it is determined that drive sticking of the valve system has occurred (step S32).

  As shown in FIG. 15A, the control shaft 54 is driven to the Lo end side, and the stroke count value S becomes “110” due to the deviation between the detected value Sk of the stroke count value S and the actual value Sj. When the control shaft 54 reaches the Lo end when it becomes, and the maximum lift amount cannot be changed for the first time in the current detection cycle (step S10: YES, step S20: NO), the detection value Sk and the initial value Slo are set. It is determined whether or not the difference ΔSlo is equal to or smaller than the threshold value Sth (step S40). Here, the difference ΔSlo (“110” − “100” = “10”) is equal to or less than the threshold value Sth (“15”) (step S40: YES), and the maximum lift amount cannot be changed for the first time (step S40). S41: NO), Lo end learning is executed (step S43). Then, as shown in FIG. 15B, in Lo end learning, it is determined that the control shaft 54 has reached the Lo end (step S120: YES), and the detected value Sk (“ 110 ") is set to the initial value Slo (" 100 "). Thereby, the deviation between the detected value Sk and the actual value Sj is eliminated. In this way, the deviation between the detected value Sk and the actual value Sj has been eliminated by the Lo end learning, but the fixing position (Sj = from the actual movable range of the control shaft 54, that is, the Lo end (Sj = “100”). The range up to “190”) is smaller than the movable range recognized by the microcomputer 70, that is, the range from the Lo (Sj = “100”) end to the Hi end (Sj = “200”). Therefore, as shown in FIG. 15C, after the Lo end learning is completed, when the control shaft 54 is driven to the Hi end side and the detection value Sk becomes “190”, the valve drive system is fixed. As a result, the maximum lift amount may not be changed again (step S31: YES). In this case, it is determined that drive sticking of the valve system has occurred (step S32).

  As shown in FIG. 16 (a), when the control shaft 54 is driven to the Lo end side and the detection value Sk becomes “110” due to the drive fixing of the valve operating system, the maximum lift amount in the current detection cycle. When it becomes impossible for the first time to be changed (step S10: YES, step S20: NO), it is determined whether or not the difference ΔSlo between the detection value Sk and the initial value Slo is equal to or less than the threshold value Sth (step S40). Here, the difference ΔShi (“110” − “100” = “10”) is equal to or less than the threshold value Sth (“15”) (step S40: YES), and the maximum lift amount cannot be changed for the first time (step S40). S41: NO), Lo end learning is executed (step S43). As shown in FIG. 16B, since the control shaft 54 cannot be driven to the Lo end side from the fixing position, it is determined that the control shaft 54 has reached the Lo end (step S120: YES), and at that time Detection value Sk (“110”) is set to the initial value Slo (“100”). Accordingly, the actual movable range of the control shaft 54, that is, the range from the fixed position (Sj = “110”) to the Hi end (Sj = “200”) is recognized by the microcomputer 70, that is, the fixed position ( Sj = “110”) to a position where the stroke count value S is “210”. Therefore, as shown in FIG. 16C, after the Lo end learning is completed, the control shaft 54 may reach the Hi end, and the maximum lift amount may not be changed again (step S31: YES). ). In this case, it is determined that drive sticking of the valve system has occurred (step S32).

  As shown in FIG. 17A, the control shaft 54 is driven to the Hi end side, and the control shaft 54 reaches the Hi end due to a deviation between the detected value Sk of the stroke count value S and the actual value Sj, If the maximum lift amount cannot be changed for the first time in the current detection cycle when the detected value Sk becomes “190” (step S10: YES, step S20: YES), the difference between the detected value Sk and the initial value Shi It is determined whether or not ΔShi is below threshold value Sth (step S30). Here, the difference ΔShi (“200” − “190” = “10”) is equal to or less than the threshold value Sth (“15”) (step S30: YES), and the maximum lift amount cannot be changed for the first time (step S30). S31: NO), Lo end learning is executed (step S33). Here, in the Lo end learning, when the control shaft 54 is driven to the Lo end side, the driving sticking of the valve system may occur. For example, as shown in FIG. 17B, when the detection value Sk becomes “100” in the Lo end learning, if the maximum lift amount cannot be changed due to drive fixation, the control shaft 54 reaches the Lo end. (Step S120: YES), and the detected value Sk (“100”) at that time is set to the initial value Slo (“100”). That is, although the Lo end learning has been executed, the deviation between the detected value Sk and the actual value Sj has not been eliminated, and the actual movable range of the control shaft 54, that is, the fixed position (Sj = “110”). The range up to the Hi end (Sj = “200”) is recognized by the microcomputer 70, that is, the range from the fixed position (Sj = “110”) to the position where the stroke count value S is “210”. Get smaller. Therefore, as shown in FIG. 17C, even after the Lo end learning is completed, the control shaft 54 may reach the Hi end, and the maximum lift amount may not be changed again (step S31: YES). In this case, it is determined that drive sticking of the valve system has occurred (step S32).

According to the embodiment described above, the following effects can be obtained.
(1) When the change of the maximum lift amount of the intake valve 20 becomes impossible for the first time in the current detection cycle, the detection of the drive sticking of the valve operating system is suspended and the Lo end learning of the maximum lift amount is performed, and the Lo end When the maximum lift amount cannot be changed again after the learning is completed, it is determined that the driving sticking of the valve system has occurred. As a result, no drive sticking of the valve system has occurred, and the control shaft 54 is at the Lo end or due to the deviation between the detected value Sk of the stroke count value S and the actual value Sj. When reaching the Hi end and the maximum lift amount cannot be changed, the state where there is a deviation between the detected value Sk of the stroke count value S and the actual value Sj is eliminated by learning the Lo end, and the maximum lift The amount control can be resumed accurately. As a result, it can be avoided that the control shaft 54 reaches the Lo end and the Hi end again after the Lo end learning is completed, and there is a deviation between the detected value Sk and the actual value Sj. As a result, when the control shaft 54 reaches the Lo end or the Hi end, it is possible to avoid erroneously determining that the drive valve is stuck. On the other hand, when drive sticking occurs due to biting or the like in the valve system, Lo end learning is performed and control of the maximum lift amount is resumed, but the actual movable range of the control shaft 54 is determined by the microcomputer 70. Therefore, even after the Lo end learning is completed, the control shaft 54 may reach the position where the drive sticking occurs or the Lo end and the Hi end. When the control shaft 54 reaches the position where the drive sticking occurs or the Lo end and the Hi end, the maximum lift amount cannot be changed again, and it is determined that the drive sticking of the valve system has occurred. Therefore, according to the above-described embodiment, when the control shaft 54 reaches the Lo end or the Hi end due to a deviation between the detected value Sk and the actual value Sj, the valve drive system is locked. Therefore, it is possible to avoid erroneously determining that the occurrence of the valve has occurred and to accurately detect the drive sticking of the valve system.

  (2) A change history of the stroke count value S from the reference value Sg, that is, the position count value P is detected, and the stroke count value S based on the reference value Sg and the position count value P, in other words, the intake valve 20 The maximum lift amount was detected. When such a configuration is adopted, noise is generated when the position count value P is detected, and once a deviation occurs between the detected value Sk of the stroke count value S and the actual value Sj, the subsequent position count value P Even if it is detected accurately, the deviation is not resolved. For this reason, there is an error that the valve drive system is stuck when the control shaft 54 reaches the Lo end or the Hi end due to a deviation between the detected value Sk and the actual value Sj. Judgment is more likely to occur.

  In this regard, by adopting the above-described abnormality detection method, even when the configuration of detecting the stroke count value S based on the position count value P and its reference value Sg as in the present embodiment is adopted, When the control shaft 54 reaches the Lo end or the Hi end due to a deviation between the detected value Sk and the actual value Sj, it is erroneously determined that the valve system drive is stuck. Thus, it is possible to accurately detect the drive sticking of the valve system.

  (3) When the maximum lift amount cannot be changed for the first time in the current detection cycle, the control shaft 54 is displaced immediately before the change of the detected value Sk of the stroke count value S and the maximum lift amount becomes impossible. If the difference from the initial value of the stroke count value S corresponding to the mechanical limit position located in the direction in which it is in a direction is larger than the threshold value Sth, the Lo end learning is prohibited and it is determined that the drive sticking of the valve system has occurred. I tried to do it. As a result, when the maximum lift amount cannot be changed for the first time in this detection cycle, the control shaft 54 is displaced immediately before the change of the detected value Sk of the stroke count value S and the maximum lift amount becomes impossible. When the difference from the initial value of the stroke count value S corresponding to the mechanical limit position that is located in the direction that has been larger than the threshold value Sth, there is a deviation between the detected value Sk of the stroke count value S and the actual value Sj. The possibility that the control shaft 54 has reached the mechanical limit position due to the occurrence is considered to be extremely low, and Lo end learning is prohibited, and it is determined that drive valve sticking has occurred. As a result, it is possible to detect the drive sticking of the valve system early and accurately.

In addition, the said embodiment can also be implemented with the following forms which changed this suitably.
In the above implementation, when the change of the maximum lift amount of the intake valve 20 becomes impossible for the first time in the current detection cycle, the Lo end learning of the maximum lift amount is executed. Not limited to this, when the change of the maximum lift amount of the intake valve 20 becomes impossible for the first time, the control shaft 54 is driven toward the Hi end, and when the control shaft 54 stops, the control shaft 54 becomes the Hi end. Hi end learning may be executed in which the detected value Sk of the stroke count value S at that time is set to the initial value Shi corresponding to the Hi end. In addition, when the change of the maximum lift amount of the intake valve 20 becomes impossible for the first time, the initial value learning is performed at the mechanical limit position closer to the stop position of the control shaft 54 out of the two mechanical limit positions. A configuration to be performed can also be adopted. By adopting such a configuration, for example, compared to the case where the initial value learning is performed at the mechanical limit position that is away from the stop position of the control shaft 54, the time required for the initial value learning is shortened and the learning is performed. Thus, fluctuations in the maximum lift amount can be suppressed, and fluctuations in the engine operating state due to initial value learning can be suppressed.

  In the above-described embodiment, one engine operation trip is set as one detection cycle. However, the present invention is not limited to this, and a configuration in which two or more operation trips are set as one detection cycle may be employed. When such a configuration is adopted, since it is necessary to store the value of the learning history flag Fg from the start to the completion of one detection cycle, the learning history flag Fg can be stored in a nonvolatile memory such as the EEPROM 72c. desirable.

  In the above embodiment, when the change of the maximum lift amount becomes impossible for the first time, the control shaft 54 is positioned in the direction in which it was displaced immediately before the change of the detected value Sk and the maximum lift amount becomes impossible. When the difference from the initial value of the stroke count value S corresponding to the mechanical limit position is larger than the threshold value Sth, there is a deviation between the detected value Sk of the stroke count value S and the actual value Sj. Therefore, it is considered that the possibility that the control shaft 54 has reached the mechanical limit position is extremely low, and the Lo end learning is prohibited and it is determined that the drive sticking of the valve system has occurred (step S20, step S30). Step S40). However, the present invention is not limited to this. For example, when there is a high possibility that a large deviation occurs between the detected value Sk of the stroke count value S and the actual value Sj due to factors other than noise, as shown in the flowchart of FIG. It is also possible to adopt a configuration in which the processing in steps S20 and 30 is omitted, and Lo end learning is executed on condition that the maximum lift amount cannot be changed for the first time (step S31: NO).

  In the above-described embodiment, the change history of the stroke count value S from the reference value Sg, that is, the position count value P is detected, and the stroke count value S based on the reference value Sg and the position count value P, in other words, the intake air The case where the present invention is applied to the valve system abnormality detection method for detecting the maximum lift amount of the valve 20 is illustrated. For example, even when a configuration in which the maximum lift amount is directly detected by a potentiometer or the like is employed, a deviation occurs between the detected value of the maximum lift amount and its actual value due to electrical variation or aging. Therefore, when the control shaft 54 reaches the Lo end or the Hi end, it may be erroneously determined that the driving sticking of the valve system has occurred. Therefore, the present invention can be basically applied in a similar manner to the valve system abnormality detection method for directly detecting the maximum lift amount.

  In the above embodiment, the case where the present invention is applied to the valve system abnormality detection method for changing the maximum lift amount and the operating angle of the engine valve is illustrated, but the present invention is not limited thereto, for example, the valve opening time of the engine valve, etc. The present invention can also be applied in basically the same manner to the valve system abnormality detection method for changing other valve characteristics.

1 is a sectional view showing a partial sectional structure of a valve train of an internal combustion engine according to an embodiment of the present invention. The top view which shows the arrangement | positioning aspect about the valve operating system of the internal combustion engine concerning the embodiment. The fracture | rupture perspective view which shows the internal structure about the mediation drive mechanism of the embodiment. The graph which shows the change of the maximum lift amount and working angle of an intake valve accompanying the change of the relative phase difference of the input part and output part of the embodiment. The fragmentary sectional view which mainly shows the structure of the actuator of the embodiment. (A)-(h) The timing chart which shows the pattern change in which the output waveform of each sensor of the embodiment and the count value of each count change. (A), (b) The figure which shows the relationship between the output signal of each sensor of the same embodiment, an electrical angle count, and a position count. The flowchart which shows the process sequence of the abnormality detection method concerning the embodiment. The flowchart which shows the process sequence of the abnormality detection method concerning the embodiment. The flowchart which shows the process sequence of the abnormality detection method concerning the embodiment. (A), (b) When the process of the abnormality detection method according to the embodiment is executed, a specific example of a relative relationship among the Lo end, the Hi end, the fixing position, and the drive position of the control shaft is schematically illustrated. Explanatory drawing shown in. (A), (b) An explanation schematically showing a specific example of the relative relationship between the Lo end, the Hi end and the drive position of the control shaft when processing of the abnormality detection method according to the embodiment is executed. Figure. (A), (b) An explanation schematically showing a specific example of the relative relationship between the Lo end, the Hi end and the drive position of the control shaft when processing of the abnormality detection method according to the embodiment is executed. Figure. (A)-(c) When the process of the abnormality detection method concerning the embodiment is performed, a specific example of the relative relationship between the Lo end, the Hi end, the fixing position, and the drive position of the control shaft is schematically illustrated. Explanatory drawing shown in. (A)-(c) When the process of the abnormality detection method concerning the embodiment is performed, a specific example of the relative relationship between the Lo end, the Hi end, the fixing position, and the drive position of the control shaft is schematically illustrated. Explanatory drawing shown in. (A)-(c) When the process of the abnormality detection method concerning the embodiment is performed, a specific example of the relative relationship between the Lo end, the Hi end, the fixing position, and the drive position of the control shaft is schematically illustrated. Explanatory drawing shown in. (A)-(c) When the process of the abnormality detection method concerning the embodiment is performed, a specific example of the relative relationship between the Lo end, the Hi end, the fixing position, and the drive position of the control shaft is schematically illustrated. Explanatory drawing shown in. The flowchart which shows the process sequence about the modification of the abnormality detection method of the said valve operating system.

Explanation of symbols

  C ... Coil, S1, S2 ... Position sensor, D1-D3 ... Electrical angle sensor, 2 ... Cylinder head, 10 ... Exhaust valve, 11 ... Valve spring, 12 ... Rush adjuster, 13 ... Rocker arm, 13a ... Roller, 14 ... Exhaust camshaft, 15 ... cam, 16 ... retainer, 20 ... intake valve, 21 ... valve spring, 22 ... lash adjuster, 23 ... rocker arm, 23a ... roller, 24 ... intake camshaft, 25 ... cam, 26 ... retainer, DESCRIPTION OF SYMBOLS 50 ... Mediation drive mechanism, 51 ... Input part, 51a ... Roller, 51h ... Helical spline, 52 ... Output part, 52h ... Helical spline, 53 ... Support pipe, 53a ... Long hole, 54 ... Control shaft, 55 ... Slider gear, 55a ... helical spline, 55b ... helical spline, 55c ... groove, 6 ... Bush, 56a ... Through hole, 57 ... Locking pin, 60 ... Actuator, 60a ... Housing, 61 ... Motor, 61a ... Stator, 61b ... Rotor, 62 ... Planetary gear mechanism, 63 ... Output shaft, 63a, 63b ... Stopper, 64 ... Planetary gear, 65 ... Roller nut, 66 ... Bearing, 68 ... Connecting member, 70 ... Microcomputer (drive control means), 71 ... Central processing unit (CPU), 72a ... Non-volatile memory (ROM) , 72b ... volatile memory (DRAM), 72c ... non-volatile memory (EEPROM), 73 ... current sensor, 74 ... accelerator sensor, 90 ... exhaust valve opening / closing mechanism, 100 ... intake valve opening / closing mechanism.

Claims (4)

A driving member that changes the valve characteristic of the engine valve by reciprocating between two mechanical limit positions; an actuator that is connected to the driving member and drives the driving member; and a detecting means that detects the valve characteristic; Then, a control target value of the valve characteristic is set based on the engine operating state, and the drive member is driven through the actuator so that a difference between the detection value of the valve characteristic detected by the detecting means and the control target value is small Applied to a valve operating system comprising a drive control means for controlling, and detecting the drive sticking of the valve operating system based on the fact that the change of the valve characteristic becomes impossible during the execution of the drive control by the drive control means. In the valve system abnormality detection method,
When the change of the valve characteristics becomes impossible for the first time in a predetermined detection cycle, the detection of the drive fixation of the valve system is suspended, and the drive member is driven toward one of the mechanical limit positions, Initially, when the drive member stops, it is determined that the drive member has reached the mechanical limit position, and the detected value of the valve characteristic at that time is set to the initial value of the valve characteristic corresponding to the mechanical limit position A valve system abnormality characterized by performing value learning and determining that the valve system has become stuck on the condition that the valve characteristic cannot be changed again after completion of the initial value learning. Detection method. In the valve system abnormality detection method according to claim 1,
The valve system abnormality detection method, wherein the detecting means detects a change history of the valve characteristic from a predetermined reference value, and detects the valve characteristic based on the change history and the reference value. In the valve system abnormality detection method according to claim 1 or 2,
When the valve characteristic cannot be changed for the first time in the predetermined detection cycle, the initial value learning is performed at the mechanical limit position that is closer to the stop position of the drive member among the two mechanical limit positions. A valve system abnormality detection method characterized by comprising: In the valve system abnormality detection method according to any one of claims 1 to 3,
When the change of the valve characteristic becomes impossible for the first time in the predetermined detection period, the detection value of the valve characteristic at that time and the position where the drive member is displaced immediately before the change of the valve characteristic becomes impossible When the difference from the initial value of the valve characteristic corresponding to the mechanical limit position is larger than a predetermined amount, the initial value learning is prohibited, and it is determined that the driving sticking of the valve system has occurred. To detect abnormalities in the valve train.

Images