DE4316759A1 - Directional valve with a proportional function - Google Patents

Abstract

The invention relates to a digital controller structure with whose aid standard directional valves having pulse magnets can be used as proportional valves. By means of suitable dimensioning of the correction factors used in the control loop, in the case of a defined desired value preset by the electronics, the flow in the directional valve can be kept constant over large pressure ranges, that is to say this arrangement corresponds to a proportional valve with an integrated pressure balance.

Description

The invention relates to a hydraulic / pneumatic directional valve, the without design changes by using a special tax electronics can be operated as a proportional valve.

Proportional valves require in comparison to directional valves with so-called inexpensive shock magnets are equipped, special expensive proportional magnets.

The object of the invention is to use a special rule process directional valves with inexpensive shock magnets also as Pro to operate proportional valves.

The invention is explained with reference to FIGS. 1 to 8 listed below. It shows:

Fig. 1, the identifier of a typical proportional magnet,

Fig. 2, the interaction of the proportional solenoid valve and path identifier,

Fig. 3 shows the typical identifier of a shock-magnet and its interaction with the directional control valve identifier,

Fig. 4 measures to "treatment" of the shock-magnet for use as "Rule Magnet"

Fig. 5 shows a possible structure of the control electronics for directional valves with proportional function using shock-magnet as an actuator,

Fig. 6 is a representation of the stable and unstable regions of the valve-actuating portion,

Fig. 7 shows a possible course of the magnetic correction factors Fl, Fr,

Fig. 8 shows a possible structure of the control electronics with integrated load-independent flow control.

In the case of proportional valves, so-called proportional magnets ( Fig. 1) are used. These are expensive special magnets, in which the complex design of the magnet core and the winding ensures that the magnet plunger generates a constant force corresponding to the magnet current I over a larger working stroke. This is the basic prerequisite for achieving stable operating points with proportional valves with spring centering ( Fig. 2).

Each position of the valve piston corresponds to a defined one Restoring force, which in turn defines a stable working point to magnetic force characteristics with defined magnetic current Ix forms (I1, X1; I2, X2; I3, X3; I4, X4 etc.).

Using current-controlled magnets can be stable Set up control identifiers for proportional valves.

As a rule, one is used for each of the two valve halves Magnet used, d. H. works out in the left valve travel area finally the right prop. magnet and in the right valve path Area of the left prop. Magnet.

To improve the control accuracy, proportional Valve displacement sensors with position control electronics used. Regardless of this, however, the working area of each of the two remains Magnets (due to its design) on each half of the way Valve limited.

If you place an impact magnet in place of the proportional magnet ( Fig. 3), you can see the following:

• - There are no defined stable operating points (I) as with the proportional magnet ( Fig. 2) when the current (I) is used as the controller output variable Y.
• - Current characteristics (e.g. I4) can be seen that are far lie above the working characteristic of the valve piston (for Directional control valve particularly cheap!), Which is not a static one Allow adjustment of the valve piston.
• - At least one characteristic (e.g. I3) touches the work characteristic as tangent. This causes static Regulating a so-called unstable balance in one small work area and does not offer stable control quality.
• - Many magnetic identifiers (e.g. I1, I2) form two static intersections with the valve identifier,
  - a stable intersection,
  - an unstable intersection.

Overall, the combination of directional valve and current-controlled shock magnet identification shown in FIG. 3 leads to a statically unstable control characteristic that appears unsuitable for practical use.

By using correction factors ( FIG. 4) one could “linearize” the function of the non-linear shock magnets. This would lead to the use of oversized shock magnets with a large space requirement and an additional cost requirement.

The following shows how a dynamically stable system identifier can be generated from the statically unstable conditions ( FIG. 3) by "dynamizing" the characteristic of the impact magnet. With this method, the static assignment of a magnetic current characteristic curve to a valve position ( FIG. 3) is basically dispensed with.

Basically, control magnets are nowadays based on pulse width modulation (PWM) driven, i.e. H. the magnets are operated by a switching mechanism controlled with a mostly constant high frequency, the Pulse-pause ratio, i.e. the ratio of on to off Time is used as a "quasi-analog" control signal. The middle Current consumption of the magnet is roughly proportional to the on Switching ratio of the PWM power signal. During the start-up Phase, the full operating voltage is applied to the magnet. A pure control of the shock magnet via a from the controller output Y generated PWM signal (as usual with prop valves) leads here to no stable usable control behavior.

The use of shock magnets as actuators for a proportion valve is possible if you:

• - As in some cases also a displacement control loop with proportional valves builds up, i.e. the position of the magnets or the directional valve piston detected by a displacement sensor and the deviation between Setpoint and actual value are fed to a (digital) controller that is off the control deviation generates an actuating signal Y, which affects both Magnets act in such a way that a stable position is regulated leaves,
• - one magnet with the positive signal + Y, the other Drives magnets with the negative signal Y,
• - builds up the controller as a PULSE controller, d. H. the controller (via a PWM power output stage) generates current pulses, the size of which on the control deviation and its pulse duration depending on the 1st and 2. Derivation of the control deviation xw is
• - The variable pulse duration and frequency over the electrical eigen frequency of the shock magnets,  
• - Ensures that the current pulses emitted by the controller last finally statically as small as possible control deviations in one dynamically stable balance of forces between magnetic force 1, Magnetic force 2, the spring restoring forces, the mass forces and generate the frictional forces involved.

The interaction here is problematic with this pulse control of the two magnets in both working areas of the valve because everyone of the two magnets

• - a very progressive (destabilizing KRAFT current identifier,
• - In the working area of the other magnet only low forces Size.

For your understanding it should be said here that with a digital controller high clock frequency, which should work as a pulse controller,

• - the proportional part of the controller, the pulse height,
• - the differential part of a mathematically exact differentiation (via the speed dx / dt) the pulse duration,
• - The integral part of the zero position of the signal pulse represents.

Characteristic of this pulse control is

• - That with comparatively small proportional factors (smaller Pulse height),
• - that with very large differential factors (small pulse lengths)

is worked.

Due to the non-linearity of the magnetic characteristics and their inequality the forces in interaction are promising Pulse control without influencing the signal output Y of the pulse Controller not possible.

In addition, the controller output signal Y itself must have a characteristic value Table Y1 = F (Y) (saved as a table in the digital computer) the non-linear characteristic of the actuators (impact magnets) be adjusted.

The adjusted controller output signal Y1, which acts as a pulse, must also be processed separately for each of the two magnets by a higher-level correction function ( Fig. 6):

Ylinks = Y1 (controller) _* Fl (Xventil)
Yrechts = Y1 (controller) _* Fr (Xventil)

Both functions F1 (X) and Fr (X) ( Fig. 6), which are dependent on the valve travel X-valve, are stored as characteristic value tables in the digital computer.

The determination of the functions Fl (X), Fr (X) for a specific magnet type is very easy if you have the following stability criterion note:

The pulse controller only works stably when the position of the valve piston (actual value curve) resulting from the sum of the forces is within the stability range ( FIG. 6) of the setpoint curve.

So you choose a controller with

• - a weak pulse height (= small proportional factor),
• - the shortest possible pulse duration (large differential factor),
• - no zero positioning of the pulse height (integral factor = 0),
• then must
• the resulting control characteristic curve (actual value characteristic curve) of the valve lies in the stable region of the characteristic diagram (setpoint value characteristic curve) ( FIG. 6),
• - with constant pulse amplification of the controller structure (linearized Controller output signal) the distance between the actual value curve and setpoint curve must be constant.

To achieve this optimized characteristic, set the factor Fl, Fr = 1 in the "weak working area" ( Fig. 7) of the magnet (ie use the maximum possible impulse force of the magnet) and adjust the identifier in the "strong area" by a factor Fl, Fr <1 in such a way ( Fig. 7) that the actual value curve of the valve piston runs within the stable range (with the controller setting parameters defined above) as exactly as possible parallel to the setpoint curve ( Fig. 6).

After completing this system optimization, the proportional range (pulse height) can now be increased to the stability limit of the control. Then the actual and setpoint curve in the representation according to FIG. 6 practically coincide.

Manufacturing tolerances of magnet and are problematic Valve piston forces (spring force, friction force) in series operation. This applies particularly to the area

• - the end positions of the magnets,
• - The spring-centered central position of the valve piston.

To ensure system stability there, the whole remaining work area with reduced (the control accuracy ver deteriorating) control factors.

This is prevented by the fact that the control parameters (KP, KI, KD) are stored in the digital controller as characteristic diagrams ( FIGS. 5, 9a, 9b, 9c) which are dependent on the valve position X.

The control parameters for each valve position X can be set individually be optimized, d. H. only in the area of larger manufacturing tolerances The valve is operated with control parameters that are larger Guarantee distance to system stability.

The control structure is explained with reference to FIG. 5.

A direction sensor ( 3 ) is attached to the directional valve ( 1 ) equipped with shock magnets ( 2, 102 ) and sends the valve piston position ( 5 ) as an actual value signal ( 105 a) to a digital pulse controller ( 9 ). The control parameters (KP, KI, KD) of this digital pulse controller ( 9 ) are taken from a characteristic value table ( 9 a, 9 b, 9 c), in which the path signal ( 5 ) shows as a table pointer. The output signal Y ( 8 ) of the controller shows a characteristic value table (not shown), from which a non-linear control signal ( 8 a) is taken. This signal ( 8 a, 108 a) is multiplied by an individual factor Fl ( 7 ), Fr ( 107 ) for each of the two magnets and fed to a PWM power output stage ( 4, 104 ). Each of these two output stages sends current-power pulses to each of the two shock magnets ( 2 ), ( 102 ) and acts on the spring-loaded valve piston with opposing magnetic forces.

The correction factors ( 6, 106 ) ensure that the pulse-controlled forces of the magnets ( 2, 102 ) in each phase of the (stabilized) control process the valve piston in the stable area of the controller detection ( Fig. 6).

With valves with small manufacturing tolerances, the path-dependent correction of the controller parameters ( 9 a, 9 b, 9 c) can be omitted.

In the case of hydraulic directional control valves in particular, depending on the size of the flow rate Q and the valve opening cross section A, flow forces occur which attempt to close the valve. These forces cause the actual value curve ( Fig. 6) to shift into the stable working range of the valve.

An increase in the pressure difference dp at the directional control valve with the same control setpoint W ( FIG. 5) causes two effects:

• a. The flow of oil flowing through the valve increases due to the greater pressure difference.
• b. Due to the larger flow forces, the control deviation xw ( Fig. 5) increases, and the valve piston performs a closing movement ( Fig. 6). This reduces the flow through the valve. The size of the closing movement of the valve piston depends on the size of the correction factor F1, FR ( FIG. 7).

The influences a. and b. work opposite.

With suitable dimensioning of the factors Fl, Fr ( FIG. 7), which can be changed via the valve piston travel x, the influences a. and b. over a larger pressure range, ie the directional control valve with the control strategy shown in FIG. 5 becomes a load-independent flow control valve. Its function is comparable to that of a proportional valve with upstream or downstream pressure compensators for generating load flows that are independent of load.

If necessary, the load-independent flow control can also be carried out by such. B. Supplement to the electronic control shown in FIG. 8.

There, the respective valve opening cross section ( 17 ) is recorded for each valve position x ( 105 ) via a Rome table ( 13 ) and a correction value ( 12 ) is taken into account via a further calculation rule ( 14 ) taking into account the changed control deviation xw ( 16 ) determined, which is added to the original target value W ( 10 ). From this, a new setpoint W '( 11 ) is formed, which now corrects the valve opening when the pressure difference changes and thus keeps the valve flow constant.

Claims (8)

1. Hydraulic / pneumatic control valve with shock magnets as actuators, characterized in that a continuous control function can be achieved by controlling the control magnets via a suitably dimensioned digital controller. 2. Control valve according to claim 1, characterized in that the over Control solenoids controlled by PWM signals via a pulse control signal variable pulse height, pulse length and pulse frequency become. 3. Control valve according to claim 1 and 2, characterized in that the output signal Y of the pulse controller via a correction Function adapted to the non-linear behavior of the actuators becomes. 4. Control valve according to claim 1 and following, characterized in that that the corrected output signal Y1 of the pulse controller for each magnet separate correction factors to the individual Magnetic identifier is adjusted. 5. Control valve according to claim 3 or 4, characterized in that the correction factors Fl, Fr of the magnets from position X of the Valve piston are influenced. 6. Control valve according to claim 1 and / or following, characterized in that the control parameters of the pulse controller ( 9 ) are influenced by the position of the actuating magnets or the valve piston. 7. Control valve according to claim 3 and / or following, characterized shows that the influencing factors Fl, Fr are chosen so that given setpoint W at the input of the controller despite fluctuating Pressure difference at the directional control valve the flow rate remains constant. 8. Control valve according to claim 3 and / or following, characterized records that in addition the flow Q in the directional control valve with fluctuating Pressure difference kept constant via a calculation circuit is, the measured value additionally by the flow forces movement of the valve piston caused on the valve piston is used becomes.