EP1121511B1 - Method and device for electromagnetic valve actuating - Google Patents

Description

The invention relates to electromagnetic actuators intended to move in translation a valve to bring it alternately in an open position and a position of closing. It finds a particularly important application in the control of the valves of an internal combustion engine, spark or compression ignition.

We know (US-A-4 614 170) an electromagnetic actuator, having a ferromagnetic material frame or pallet valve stem drive, return means elastics designed to keep the valve at rest in a middle position between the fully open and closing, and electromagnetic means for bringing alternately the valves in the two positions. Ways electromagnetic described in document US-A-4 614 170 have a first electromagnet with ferromagnetic core placed on one side of the palette and whose excitement attracts the palette in one direction tending to close the valve and a second electromagnet, placed the other side of the palette, whose excitement tends to bring the valve in the fully open position.

Another electromagnetic actuator is described in document FR-A-2 784 222. The electromagnetic means of this last actuator have a single coil mounted on a circuit ferromagnetic construction as it presents, in combination with frame, two stable flow paths magnetic corresponding both to a low value, generally zero, air gap between the armature and the circuit ferromagnetic.

The mode of actuation of these electromagnetic actuators is the next. Electromagnetic means allow to exercise forces for bringing the armature into a “high” position that we will suppose to correspond to the closing of the valve, and a so-called "low" position, corresponding to the opening, and maintain the armature in these positions. In the "up" position the crew compresses a mechanical energy storage spring as long as a suitable current in a coil or the single coil retains the frame. When the holding current is removed, the spring propels the moving part towards the “low” position. Rod attached to frame pushes valve stem and compresses the valve closing spring. At the end of the race the armature, a holding current is established in the coil or a suitable coil to keep the valve open. The valve closing spring in turn stores energy and in turn propels the valve and armature upward when the holding current is cut.

Part of the mechanical energy is lost by friction, shocks, eddy currents and resistant work of the forces of back pressure, especially on the exhaust. As a result, must use an additional force, called appeal, in addition to the force exerted by the springs to compensate for the losses of energy with each passage of the armature from an extreme position to the other.

The additional energy to be supplied must be sufficient to guarantee a complete stroke of the armature, but not be excessive in order to avoid a terminal shock which would cause noise and wear. In practice, the speed at impact should not exceed a few hundredths of a meter per second to maintain wear and noise to an acceptable level.

Existing electromagnetic processes and devices find it difficult to fill both simultaneously above conditions in a simple way. Or else they have to tolerate a high impact speed, or they require the presence of a position and / or speed sensor which complicates the process and the device and which increases the cost of production.

We know however (JP-A-09 320841) an actuator in accordance with the preamble of claim 1.

The present invention aims in particular to provide a method and an electromagnetic valve actuator allowing satisfactory control of the applied energy, without however require a sensor.

For this, the invention uses the fact that one can constitute the ferromagnetic circuit of electromagnetic means so such that there is an almost linear relationship between the reluctance R (x) and the air gap x during the last fractions of millimeter of the path before bonding the reinforcement against the ferromagnetic circuits. This property is encountered in particular in the case of electromagnetic means at a single coil of the kind described in patent application No. 98 12489 already mentioned. In such an actuator, the inductance L (x) of the coil also varies almost linearly within a range starting immediately beyond the central position of the armature if the notches in the ferromagnetic circuit have substantially the same length as the thickness of the reinforcement. Gold R (x) and L (x) can be calculated from the current i in the coil (or two coils in series), which allows to calculate x almost at any time after passing the central position and deduce the speed.

Consequently, the invention proposes in particular an actuator The electromagnetic valve according to claim 1.

The calculation means can deduce the variation of reiuctance of the current measurement during the terminal phase of the approach of the reinforcement, i.e. for weak air gaps and we can deduce the variation of x over time from the variations reluctance.

In an advantageous embodiment of the invention, the calculation means are also provided for calculating so repetitive the value of the inductance of electromagnetic means while the air gap exceeds a determined value, this which allows you to determine, for example by tabulations, the values corresponding to x. Under these conditions, regulation can order a speed variation profile on most of it the path of the pallet. In the interval for which the laws of variation of L and R as a function of the air gap x are not very linear, an approximate value of x can be obtained at any instant by barycentric average of the values of x given by supposed linear interpolations of L and R in function of x.

Thanks to this constitution, we can make a real regulation, as opposed to simple open loop control or as opposed to simply adjusting the current supplied during of a cycle from the results obtained during a cycle previous armature oscillation.

The invention also provides a method for controlling valve using such an actuator, according to the claim 10.

In practice, the area in which there is no linear variation of L or R can be very narrow. R in fact varies almost linearly as a function of x as long as the air gap x does not exceed a value x ₁ of about 0.5 mm, in the case of an actuator whose electro-mechanical parts have the constitution shown in patent application n ° 98 12489. The inductance L varies almost linearly as a function of x as soon as x exceeds a value x ₂ which is typically about 1 mm, when the thickness of the armature is substantially equal to that of the notches in the magnetic circuit.

The characteristics of L and R as a function of x can be obtained on other types of magnetic circuit, derived from that described in application No. 98 12489.

The above features as well as others, advantageously usable in conjunction with the above but can be independently, will appear better on reading the following description of particular embodiments, given by way of nonlimiting examples.

• la figure 1 montre un actionneur de soupape selon un mode de réalisation, en coupe suivant un plan passant par l'axe de la soupape ;
• la figure 2 est une vue de détail destinée à montrer les paramètres qui interviennent ;
• la figure 3 est un diagramme de fonctionnement ;
• la figure 4 montre une variante de la figure 3.
• la figure 5 montre une variante de circuit magnétique utilisable.

The description refers to the accompanying drawings, in which:

• Figure 1 shows a valve actuator according to one embodiment, in section along a plane passing through the axis of the valve;
• Figure 2 is a detail view intended to show the parameters involved;
• Figure 3 is an operating diagram;
• FIG. 4 shows a variant of FIG. 3.
• FIG. 5 shows a variant of a usable magnetic circuit.

The actuator 10 shown in Figure 1 consists of a assembly intended to be mounted on the cylinder head 12 of an engine. he has a housing made up of several parts 14 and 16 stacked and assembled by means not shown, such as screws. These parts are made of non-ferromagnetic material, for light alloy example. The housing can be fixed on the cylinder head 12 by means of a shim 20 also made of non-material ferromagnetic.

The actuator has a movable armature in the form of a pallet 22. This frame is made of ferromagnetic material, advantageously laminated to reduce losses. It is fixed on a rod 24 for driving the valve 25. In general, several valves are mounted side by side and we only have of small width for each actuator in the direction perpendicular to that of Figure 1. This leads to give the palette a rectangular shape. Pallet cannot rotate in the part 16. The rod 24 can be guided by a ring 26 fixed to an annular extension of the part 16.

Two return springs 28a and 28b are provided to maintain the valve at rest in a substantially middle position between the closed position and the full open position. A springs 28a is compressed between a plate 30 fixed to the rod 24 and the extension of part 16. The other spring 28b is compressed between a plate 31 fixed to the valve stem and the bottom valve well in the cylinder head. The distribution game between the lifted rod and the closed valve guarantees tightness. The actuator can just as easily be used with a spring single working in traction / compression and / or supplemented by elastic shock absorber ensuring sealing when closing the valve, as indicated in French patent No. 98 11 670, this which makes it possible to constitute the stem and the stem of valve of a single piece.

The housing contains a core of ferromagnetic material 36, advantageously laminated, delimiting a ferromagnetic circuit with the pallet, and a coil 38 placed in the core. The core represented can be in two complementary parts, in support one against the other or in one piece. The constituent sheets of each half of the nucleus are E-shaped. The branches upper 42 engage in the coil 36 which they support via a mandrel 44. The other two branches of each half delimits a volume of movement of the pallet. The support of the pallet against the bottom 46 of the volume defines the valve fully open position. The ceiling 48 of volume is in a location relative to the valve seat such that the support of the armature does not prevent the valve from closing. A middle notch 49, corresponding to the rest position of the pallet 22, can be provided in the chamber, length slightly greater than the thickness of the pallet. Above and below the notch, the volume wall leaves only the clearance necessary for travel to reduce reluctance.

The assembly consisting of the pallet, the valve and the spring constitutes an oscillating system having a natural frequency. In steady state, the coil is fed to bring the crew mobile in its extreme position then under a holding current weaker, until the displacement of the moving crew in the other direction.

In an actuator as shown, the reluctance R (x) of the magnetic circuit varies in a substantially linear manner as long as the value x of one of the air gaps is less than a value x ₁ which is generally about 0.5 mm. The inductance L (x) also varies substantially linearly as a function of x when the air gap exceeds a value x ₂ of approximately 2 mm.

R(t)=n2 L(t) où :

• r est la résistance connue de la bobine (éventuellement corrigée en fonction de la température),
• L'instant T_o est choisi tel que L(T_o) *i (T_o) soit connu. On choisira souvent T_o tel que i (T_o) = 0,
• n est le nombre de spires de la bobine.

To implement the invention, the actuator comprises a supply circuit (FIG. 2) having a sensor 50 of the current i in the coil. The output is used by a calculation circuit 52 which controls the voltage applied by a generator 54. A convenient solution, since it allows digital calculations, consists in sampling the signal i. A sampling frequency of 20 kHz will generally give satisfactory results. If the coil is supplied with a voltage u, the inductance L (t) and the reluctance R (t) can be obtained by an integral calculation program:

R (t) = not 2 L (t) or :

• r is the known resistance of the coil (possibly corrected as a function of temperature),
• The instant T _o is chosen such that L (T _o ) * i (T _o ) is known. We often choose T _o such that i (T _o ) = 0,
• n is the number of turns of the coil.

The current is slaved, using a regulation loop, by comparing the current i measured with a set value. The gap observed allows the command to be corrected analogically.

It is advantageous to use a mounting of the kind illustrated in Figure 3, where the computer consists of several modules and regulates the voltage u applied to the coil, in the form of pulses at a fixed frequency fe, by command of a power switch, constituting the generator 54, by a width modulator pulse 58. Modulator 58 provides an output signal periodic, at a frequency fe of a few tens of kHz having an RCO opening cyclic report.

où KO est tel que TO = KO*te K est tel que T = K*te In the illustrated embodiment, the integral giving L (t) is calculated in a total flux estimator, from the knowledge, through a sampler-blocker 60, of the duty cycle RCO (k) applied to the sampling period k, of duration T _e = i / fe and of the nominal voltage Un applied by the switch 54:

where KO is such that TO = KO * te K is such that T = K * te

The voltage Un is known by construction of the circuit of switching and does not necessarily have to be acquired in real time.

The current I representative of i (t) is presented to the computer after the sampler-blocker 66 has it sampled at an instant undisturbed by the commutations of the modulator and that an anti-aliasing filter has attenuated the harmonics beyond fe / 2. The integral of r * i (T) is calculated numerically by a method of the cumulative, simple or trapezoidal, integration surfaces.

• Une cartographie donnant R en fonction de x et i permet de tirer x par interpolation linéaire, connaissant R et i pour x inférieur à x1.
• Une cartographie donnant L en fonction de x et i permet de tirer x par interpolation linéaire, connaissant L et i pour supérieur à x2.

L and R can thus be calculated at successive times t; then

• A mapping giving R as a function of x and i makes it possible to derive x by linear interpolation, knowing R and i for x less than x1.
• A mapping giving L as a function of x and i makes it possible to derive x by linear interpolation, knowing L and i for greater than x2.

Between x1 and x2, we can estimate x by an average barycentric linear interpolations from R and at from L.

After filtering the estimated value of x and x by a filter digital with a cutoff frequency of a few kHz, a derivation operation in a speed calculation module 65 provides an estimate v and of the speed during the whole course of a transition, without the need for a sensor specific. For an automobile engine, a frequency fe of 20 kHz and a cutoff frequency of 75 kHz generally give good results.

Before further describing the assembly shown in Figure 3, we will set out the principles adopted for its operation.

The comparison of the estimated speed v and with a profile of set speed determined by simulation and experimentally allows to develop, through a corrector 62, the profile of the calling force F (t) by achieving a compromise between power consumed and delay in applying correction forces.

To improve the control of the force F (t) applied, it is advantageous that the current setpoint results from a servo loop which digitally slaves the flow  (t) contributing to the realization of the magnetic force. This approach makes the force control robust against positioning uncertainties, in particular for small air gaps. Thus, the force setpoint that the corrector 62 calculates is translated by a module 63 into a setpoint on the magnitude of the total flux  (t) which is none other than the flux useful at the force of the leakage flux linked to the inductance Lf leakage. The useful flow  _u can be written:  (t) = αF (t) 1/2 where α is a scale factor linked to the geometry of the magnetic circuit and determined by simulation and tests.

The total flow setpoint, developed in a module 63, is given by the formula:  vs =  u + L p .i (T)  vs =  u + L p xi (t)

The set flow  _c will be compared to the estimated total flow

This formula uses terms already available from the position estimator 64 and involves the inductance of Lf leakage, determined by simulation and tests.

Such a system thus works with three closed loops nested; the first involves speed, the second the useful flux, and the third the current i in the coil.

In the diagram of FIG. 3, the position estimator 64 receives a digitized signal I representing the current i measured. From I, from the flow  and and from stored tables R (x, i) and L (x, i), it develops, at each sampling instant from the start T ₀ of a cycle, information of position which is transmitted to the module 65 for calculating the estimated real speed. The corrector 62 compares the real speed profile with the speed profile v _c setpoint and provides a signal representative of the force to be exerted F (t) to the module 68 for calculating a setpoint flow  _c taking account of the inductance leakage Lf and the coefficient α. The setpoint current i _c necessary to create the flow  _c is calculated in a module 72 from the difference between  _c and the estimated total actual flow  and (t). This estimated total real flux is given by a module 70 on the basis of stored values r and Te, of the signal representative of the measured current i, of the nominal voltage of the generator and of the RCO.

The digital signal representing the setpoint current is supplied to the pulse width modulator 58 by the digital to analog converter 60, and it compares this signal to I.

The corrector 62 can be provided to take into account also, on certain fractions of the course, a force profile setpoint fc.

• la boucle de courant constituée du modulateur 58, du capteur 50 et du commutateur 54 ;
• la boucle de flux, constituée de l'estimateur de flux total, du module de calcul du flux de consigne et du module 72, se refermant sur la précédente ;
• la boucle de vitesse comprenant l'estimateur de position 64 et le module de calcul de force, se refermant sur la précédente.

We find, in Figure 3, the three nested loops mentioned above:

• the current loop consisting of the modulator 58, the sensor 50 and the switch 54;
• the flow loop, consisting of the total flow estimator, the setpoint flow calculation module and the module 72, closing on the previous one;
• the speed loop comprising the position estimator 64 and the force calculation module, closing on the previous one.

The modules can be made up of micro-electronic components or program elements.

In an alternative embodiment, shown in FIG. 4, the current loop is deleted. The current calculation module setpoint is replaced by a module 74 which calculates directly RCO (t) and applies it to modulator 58 which controls the intervals of application time of the voltage u.

In yet another variant, shown in FIG. 5, the pallet is advantageously laminated and with bevelled edges parallel to the poles of the nucleus. The inductance tabulation and reluctance as a function of current and air gap allows accurately find the position and speed of the pallet the fact that the armature is not magnetically saturated in its operating range and the flow closes in passing mainly by the reinforcement thanks to the shape of the pole pieces of the nucleus.

The natural asymmetry of the upper flow circuit by relation to the lower flow path can be accentuated (for shorten start-up time) by tilting different to the upper 80 and lower polar surfaces 82, the facing polar surfaces remaining parallel between they.

Claims (10)

1. An electromagnetic valve actuator comprising a valve drive armature (25), resilient return means (28a, 28b) provided to hold the valve at rest in a determined position, substantially halfway between two extreme positions including a valve closed position, electromagnetic means (38) having a ferromagnetic core placed on both sides of the armature, and a power circuit for applying power in alternation to said electromagnetic means, the actuator being characterized by means for calculating the velocity with which the armature approaches each of its extreme positions, and by means for applying a current to the electromagnetic means for controlling said velocity, said calculating means performing the calculation from a measure of the excitation current in the electromagnetic means and said applying means servo controlling the variation of the velocity to a determined reference profile.
2. An actuator according to claim 1, characterized in that the calculation means are also designed to calculate repetitively the inductance and the reluctance of the electromagnetic means while the airgap exceeds a determined value, and the power supply circuit controls a velocity variation profile over the major portion of the armature stroke.
3. An actuator according to claim 2, characterized in that the electromagnetic means comprise a single coil mounted on a ferromagnetic circuit of structure such that in combination with the armature it presents two stable magnetic flux paths each corresponding to a small, generally zero, value for the airgap between the armature and the ferromagnetic circuit.
4. An actuator according to claim 1, 2, or 3, characterized in that the calculation means are designed to determine the inductance L(t) and the reluctance R(t) by calculating:

   

   R(t) = n2 L(t) where:

   r is the known resistance of the coil (possibly corrected as a function of temperature);

   To is such that L(To)*i(To) is known; and

   n is the number of turns of the electromagnetic means.
5. An actuator according to claim 2 or 3, characterized in that the electromagnetic means are powered by a switch (54) controlled by a pulse width modulator (58) which supplies a periodic output signal at a frequency fe of a few tens of kHz having a duty ratio that is controlled to cancel the difference between an estimated velocity profile v and and a reference profile.
6. An actuator according to any one of claims 2 to 5, characterized in that the calculation means are designed to transform the current reference into a flux reference and to compare the total flux created by the estimated real current with the leakage flux.
7. An actuator according to claim 6, characterized in that the calculation means constitute a current loop, a useful flux loop, and a velocity loop that are mutually interleaved.
8. An actuator according to any one of claims 1 to 7, having a single coil, a laminated armature, and pole surfaces on the armature that are chamfered and parallel to the poles of the core.
9. An actuator according to claim 8, characterized in that, in order to increase the natural asymmetry of the top flux circuit relative to the bottom flux circuit, the pole areas of the armature are of different values.
10. A method of controlling a valve using an actuator according to claim 1, in which the excitation current passing through the electromagnetic means is sampled, variations in the inductance L(t) and the reluctance R(t) are deduced from the current by calculation, followed by variations in the airgap (x) by referring to stored tables, the velocity of variations in airgap over time is deduced and the application of a voltage to the electromagnetic means is controlled in such a manner as to servo-control variations in time of the airgap (x) and of the velocity (v) to a predetermined profile.

Images