EP0662697A1 - Method for controlling a bistable electromagnetic actuator and device to carry out such method - Google Patents

Abstract

The procedure enables control of an electromagnetic actuator (1) having two active stable positions and a rest position. The actuator includes a mobile armature (3) linked to the element to be moved, for example an internal combustion engine valve. The armature is arranged in the air gap (e) between two identical electromagnets (11,12). It is held in this rest position by the balanced forces of two identical springs (31,32). Switching is achieved by activating the second electromagnet a predetermined time interval (delta-T) after the cutting of current to the first electromagnet, and vice versa. The excitation of the second electromagnet may be advanced w.r.t. this time (delta-T), with the timing being adjusted for optimum effect according to the speed of operation.

Description

The invention relates to a method for controlling a bistable electromagnetic actuator intended for example for controlling a valve of an internal combustion engine. It also relates to a control device and an actuator specially designed for the implementation of this method.

The actuators concerned are of the known type with two stable states respectively obtained by the excitation of the coil of two electromagnets (FIGS. 1 a to 1 c ) in the air gap of which a movable armature is arranged.

In the known methods, one passes from the first stable state to the second stable state and vice versa by establishing the current of the coil of the second electromagnet a certain time Δ T after the interruption of the current of the coil of the first electromagnet. For the capture of the armature released by the first electromagnet to be carried out in an optimal manner, the time difference Δ T must be suitably chosen. If this time is too short, the energy accumulated by the movable armature can cause rebounds, if it is too large the capture may not take place.

More particularly, in the use for the control of valves of an engine, many disturbing factors can intervene and modify the time difference which would be optimal. This is how the charging current of the coils depends on the condition and temperature of the vehicle battery. The mechanical resistances are also a function of the temperature and in particular of the oil temperature, which introduces large differences between cold start and hot engine. The choice of a constant Δ T therefore leads to faulty operations.

The invention proposes to overcome the above drawbacks in a method for controlling an electromechanical actuator with two stable active positions and a rest position, actuator of the type comprising a movable armature linked to a member to be actuated, for example a valve of an internal combustion engine, this armature being arranged, in the air gap separating first and second identical electromagnets placed opposite, equidistant at rest from the two electro -magnets thanks to the opposite return forces of two identical springs, process in which the switching from one to the other active position is carried out by controlling the excitation of the second electromagnet when a time ΔT, predetermined for the normal operating conditions, has elapsed since the command to cut off the excitation current of the first electromagnet and vice versa.

• a)- la commande de l'excitation du deuxième électro-aimant est avancée par rapport au temps ΔT prédéterminé pour un fonctionnement optimal dans les conditions normales;
• b)- on coupe le courant d'excitation du deuxième électro-aimant à un moment déterminé par la vitesse de déplacement de l'organe mobile:
• c)- on rétablit le courant d'excitation après une durée de coupure fonction décroissante de la vitesse de déplacement de l'organe mobile.

The process according to the invention is characterized by the following stages:

• a) - the control of the excitation of the second electromagnet is advanced with respect to the predetermined time ΔT for optimal operation under normal conditions;
• b) - the excitation current of the second electromagnet is cut off at a moment determined by the speed of movement of the movable member:
• c) - the excitation current is restored after a cut-off time that decreases as a function of the speed of movement of the movable member.

Preferably, the breaking of the excitation current of the second electromagnet is controlled during the passage of the movable armature by a reference position which advantageously corresponds to the rest position of the armature.

According to another important characteristic, the duration of the cut is calculated as a function of the elapsed time T _C - T _P between the instant T _C of the start of the cut and the instant T _P of the normal end of the previous command.

Preferably in this case, the duration ΔT _C of the cut-off is an increasing function of the elapsed time T _C - T _P.

A device for implementing the method according to the invention is of the type comprising an electromechanical actuator, with two stable active positions and a rest position, the movable armature of which is linked to a member to be actuated is disposed in the air gap of two identical electromagnets placed opposite one another in a housing and means for switching the control currents of the electromagnets, and is characterized in that it comprises means for generating and delivering to the switching means a signal during the passage of the movable frame to a predetermined position of its travel.

Preferably, the means for generating and delivering the signal comprise a position sensor linked to the actuator housing which can advantageously be arranged, to detect the passage of the movable armature, equidistant from the two electromagnets.

• les figures 1a à 1c montrent un actionneur connu dans ses positions stables;
• le graphique de la figure 2a montre la position P de l'armature en fonction du temps;
• les graphiques des figures 2b et 2c montrent les signaux de commande d'un procédé connu pour commander un tel actionneur;
• le graphique de la figure 2d montre l'intensité du courant d'une bobine en fonction du temps;
• les figures 3a et 3b montrent les défauts de capture de l'armature dans le procédé connu;
• la figure 4 est un schéma synoptique d'un dispositif pour la mise en oeuvre de l'invention.
• les figures 5a à 5e sont des diagrammes montrant le fonctionnement du procédé conforme à l'invention;
• la figure 6 montre en coupe axiale un actionneur conforme à l'invention appliqué à la commande d'une soupape.

The invention will be better understood in the detailed description which follows and refers to the appended drawings given solely by way of example, and in which:

• Figures 1 a to 1 c show an actuator known in its stable positions;
• the graph in FIG. 2 a shows the position P of the armature as a function of time;
• the graphs of FIGS. 2 b and 2 c show the control signals of a known method for controlling such an actuator;
• the graph in FIG. 2 d shows the intensity of the current of a coil as a function of time;
• Figures 3 a and 3 b show the capture defects of the armature in the known method;
• Figure 4 is a block diagram of a device for implementing the invention.
• Figures 5 a to 5 e are diagrams showing the operation of the method according to the invention;
• Figure 6 shows in axial section an actuator according to the invention applied to the control of a valve.

A known electromechanical actuator with two stable positions and a rest position is shown in FIGS. 1 a , 1 b and 1 c and designated by the general reference 1.

It comprises, fixed in a housing 10, two identical electromagnets 11 and 12 each comprising an excitation coil 21,22. A movable frame 3 consisting of a plate made of magnetic sheet is placed in the air gap e comprised between the two electromagnets arranged opposite. Two identical helical springs 31 and 32 are arranged along the axis XX of the actuator on either side of the plate 3 so as to keep it at rest, that is to say when none of the coils is excited, equidistant from the two electromagnets (Figure 1a). The rod 5 disposed axially and fixed to the center of the plate 3 is connected to the member to be controlled, not shown.

Figure 1b shows a first stable position obtained when the coil 21 of the electromagnet 11 is energized while the coil 22 of the electromagnet 12 is not. The armature 3 is in this case bonded to the armature of the electromagnet 11. FIG. 1 c shows the second stable position obtained when the coil 22 is energized while the coil 21 is not. The armature 3 is in this case bonded to the armature of the electromagnet 12.

The method usually used for switching from the first stable position to the second stable position will now be described with reference to FIGS. 2 a to 2 d . The first position of the armature 3 is represented from O to T1 by the ordinate P1 and the second position by the ordinate P2 reached at time T2 . The current control signals in the coils 21 and 22 are designated respectively by A1 and A2 . At state 1 of these signals the corresponding coil is energized. To pass the armature 3 from P1 to P2 , the current of the coil 21 is first cut by passing the sienal A1 from state 1 to state O. A certain time ΔT after this cut-off, the current I of the coil 22 is controlled by passing the signal A2 from state O to state 1 and the charge of the coil 22 changes as shown in FIG. 2 d .

In order for the armature 3 to be captured by the electromagnet 12 in an optimal manner, the time difference ΔT must be suitably chosen to avoid non-sticking of the armature (FIG. 3 a ), if the magnetic energy is insufficient, or rebounds (Figure 3b ), if there is excess energy.

In addition, in the application to the electromagnetic control of the valves of an internal combustion engine, the operating conditions linked to the vehicle cause variations in the inrush current shown in FIG. 2 d , for example during a start-up at cold and / or if the battery charge is not optimal.

We are therefore led to anticipate the transition from state O to state 1 of signal A2 (or of signal A1 ), with respect to the ΔT determined optimally under normal conditions (hot engine, normal battery charge) so as to have enough energy at the time of contact of the armature 3 with the magnetic circuit of the electromagnet 12 (or 11) in all conditions, which leads to the disadvantage of rebounds if the conditions are normal.

According to the invention, the procedure is to remedy these drawbacks in the manner which will be explained below with reference to FIGS. 5 a to 5 e .

In these figures, the origin of the abscissa represents the instant when the coil of the electromagnet corresponding to the stable state previously produced is de-energized by passage of the corresponding control signal from state 1 to state O to l 'instant T _P. The control of the excitation of the other electromagnet is advanced with respect to the predetermined time Δt (FIG. 5 b ) for optimal operation under normal conditions, as already explained, at time T _a (FIG. 5 c ) .

The excitation current of this electromagnet is then cut at an instant T _c determined by the displacement of the movable armature 3, for example during its passage through a point P _R intermediate of its stroke from P1 to P2 (figure 5 a ). The excitation current is then restored after a cut-off duration calculated as a function of the movement of the armature 3, for example as a function of the time elapsed between the cut at T _P and the passage of the armature 3 through the reference position. P _R.

If the bistable actuator used is perfectly symmetrical like that described in FIGS. 1 a to 1 c , the reference position P _R will preferably be chosen equidistant from the two electromagnets and more generally it will correspond to the rest position of the movable frame 3 (Figure 1 a ).

The duration of the cutoff as a function of the speed of movement of the armature 3 is determined experimentally as a function of the operating characteristics of the member to be controlled.

For example, when ordering an intake valve with a diameter of 32 mm and a mass of 0.14 kg, the valve corresponding to the smaller of the two intake valves of a four-cylinder engine close to 1.9 l , and whose 6.5 mm stroke must be covered in a constant time of 5 milliseconds, the instant T _P of the interruption of the signal A1 being taken as origin, the orders of magnitudes can be the following:
Normal order ΔT = 0.8 to 1.4 ms Advanced order TA = 0.5 ms Controlled interruption T _C = 3 ms Duration of the interruption ΔT _C = 0.4 ms

In general, to obtain a cut-off duration that decreases as a function of the speed of movement of the movable member, ΔT _C will be an increasing function of T _c - T _P. The magnetic force is thus properly metered to avoid rebounds of the plate 3 while leaving a sufficient margin to compensate for disturbances in the movement of the plate.

où K est une constante telle que O < K < 1 et T _P une donnée du contrôle moteur.In practice we can retain an increasing linear function of the form

${\text{ΔT}}_{\text{VS}}\text{= (}{\text{T}}_{\text{VS}}\text{-}{\text{T}}_{\text{P}}\text{-}\underline{\text{Init}}\text{)}\text{x K}$

where K is a constant such that O < K < 1 and T _P a motor control data.

I nit is the time which elapses between the end of the command of the previously excited coil and the moment when the plate 3 actually takes off from the magnetic circuit of this coil.

We will now describe a device designed for implementing a method according to the invention. The device (FIG. 4) comprises an actuator 1 of the type described above and further comprising a position sensor 6 capable of emitting an electrical signal when the armature 3 passes to its rest position. A control system 7 receives, in addition to the usual information for example from the control electronics of a motor to internal combustion, the actuator 1 of which controls a valve, the signal emitted by the sensor 6 when the frame 3 passes.

The system 7 produces the logic signals A1 and A2 controlling the coils 21 and 22 respectively, by means of a power stage 8.

We will now describe a method which can be used by the control system 7 to trigger the dip at time T _C and calculate its duration ΔT _C. To determine the duration T _C - T _P - Init , one could of course use an analogical method but we know that the numerical methods are more precise and more reliable, in particular by using a clock and a down counter. To multiply T _C - T _P - Init by the factor K , it is then advantageous to use two time bases, one to count, the other to count.

Il est intéressant d'utiliser le même circuit pour la commande de deux bobines en multiplexant les deux entrées et les deux sorties de commandes A1 et A2.The end of command A1 of the previously energized coil must first be identified, for example by carrying out edge detection. The instant T _P thus identified is chosen as the origin. An up-down counter is then initialized to the value Init and then counted down, when the counter reaches zero, up to the time T _C of passage of the plate 3 in front of the detector 6 is counted using a first clock. We thus obtain T _C - T _P - Init at the output of the up-down counter. The start of the dip is simultaneously triggered at time T _C and a countdown is then made with a second clock. The frequency of the two clocks being in the K ratio, we obtain a duration:

${\text{ΔT}}_{\text{VS}}\text{= (}{\text{T}}_{\text{VS}}\text{-}{\text{T}}_{\text{P}}\text{-}\text{Init}\text{) x}\text{K}$

It is interesting to use the same circuit for the control of two coils by multiplexing the two inputs and the two control outputs A1 and A2 .

According to a preferred embodiment of an actuator according to the invention shown in FIG. 6 at scale one, the sensor 6 is inserted into the outer wall of the housing 10 equidistant from the armatures of the coils 11 and 12. A valve V and its seat S are shown in the intermediate position corresponding to the state of rest of the actuator. The valve stem coincides with the rod 5 driven by the movable frame 3. The equilibrium position P1 corresponds to the closed valve and the position P2 when it is fully open.

The springs 31 and 32 bear respectively, by their end opposite the plate 3, on the housing 10 and on the cylinder head C of the engine.

The actuator comprises, in the usual way, means for adjusting the closed position of the valve V resting on its seat S. These means may include shims or washers 102 arranged under the flange 101 of the housing 10 before fixing to the cylinder head C by screws such as 103. It also usually comprises means for adjusting the zero position corresponding to the state of rest. These means can be constituted by a screw 104 cooperating with a corresponding tapping of the housing 10 and on which the end of the spring 31 opposite the plate 3 bears. The axial displacement of the support formed by the screw 104 thus makes it possible to adjust the plate 3 rest position. After adjustment, the screw 104 is locked by a lock nut 105.

So that the position of the sensor 6 on the housing wall 10 corresponds as exactly as possible to the zero position of the tray 3, the sensor 6 is centered in a bore of a cylindrical spacer 106 which determines the width of the air gap e between the armatures of the electromagnets 11,12, this air gap defining the total stroke of the plate 3 between its two extreme stable positions P1, P2 .

Claims (10)

Method for controlling an electromechanical actuator (1) in. two stable active positions and a rest position, actuator of the type comprising a movable frame (3) linked to a member to be actuated, for example a valve (V) of an internal combustion engine, this frame being arranged in the air gap (e) separating a first and a second identical electromagnet (11,12) placed vis-à-vis, equidistant at rest of the two electromagnets through the opposing return forces of two identical springs (31,32 ), method in which the switching from one to the other active position is carried out by controlling the excitation of the second electromagnet when a time ΔT , predetermined for normal operating conditions, has elapsed since the command to cut off the excitation current of the first electromagnet and vice versa, characterized in that it comprises the following steps: a) - the control of the excitation of the second electromagnet is advanced with respect to the predetermined time ΔT for optimal operation under normal conditions; b) - the excitation current of the second electromagnet is cut off at a moment determined by the speed of movement of the movable member (3,5); c) - the excitation current is restored after a cut-off time ΔT _C decreasing function of the speed of movement of the movable member. Method according to claim 1, characterized in that the breaking of the excitation current of the second electromagnet is controlled by the passage of the movable armature by a reference position ( P _R ). Method according to claim 2, characterized in that the reference position ( P _R ) corresponds to the rest position of the armature (3). Method according to any one of Claims 1 to 3, characterized in that the duration of the cut is calculated as a function of the elapsed time T _C - T _P between the instant T _C of the start of the cut and the instant T _P from the end of the previous command. Method according to claim 4, characterized in that the duration ΔT _C of the cut-off is an increasing function of the time T _C - T _P. Device for implementing the method according to claims 1 to 5, comprising an electromechanical actuator (1) with two stable active positions and a rest position, the movable armature of which is linked to a member to be actuated (V) is arranged in the 'gap ( e ) of two identical electromagnets (11,12) placed opposite one another in a housing (10) and means for switching the control currents of the electromagnets, characterized in that it comprises means (6) for generating and delivering to the switching means a signal when the movable armature passes to a predetermined position ( P _R ) of its travel. Device according to claim 6, characterized in that the means for generating and delivering the signal comprise a position sensor (6) linked to the housing of the actuator. Device according to claim 7, characterized in that the sensor (6) is arranged to detect the passage of the movable armature (3) equidistant from the two electromagnets. Device according to claim 8, characterized in that the sensor (6) is inserted in the outer wall of the housing (10). Device according to claim 8 or claim 9, characterized in that the sensor (6) is centered in a spacer (106) delimiting the total stroke of the plate (3) between its two extreme stable positions ( P1 , P2 ).

Images