EP0077996B1 - Method and apparatus to control the idling speed of a combustion engine - Google Patents

Description

State of the art

The invention relates to a method and a device for regulating the speed of an internal combustion engine according to the preamble of the main claim and the first device claim.

The article "New mixture formation system for petrol engines" in "Automobil Technische Zeitschrift 83 (1981) 5 _" on page 219 ff discloses an electronically controllable carburettor system with regulation of the idling speed as well as electronic control during start, warm-up and acceleration. The idle speed _- control system includes an adjustable in position throttle stop plate. When idling, the measured engine speed is continuously compared with the desired setpoint and a PI control algorithm uses the deviation to calculate the required throttle valve position change.

Furthermore, it is provided to bring the throttle valve into a position during the overrun operation, in which no fuel is also sucked out via the idling nozzle.

In the subject of DE-A-2 049 669, a speed-sensitive electrical circuit acts on an electromagnetically actuated actuator, with which the amount of intake air can be changed in the idle position of the throttle valve. To this end, the electromagnetically actuated actuator acts in a cross-sectional control manner on a bypass channel parallel to the throttle valve.

In this known device, the exclusive control of the amount of intake air could be problematic, since it will probably not be possible in this way to take all influencing factors into account; in particular, it is not possible to actively influence the position of the throttle valve and thus to carry out an effective change in the filling intervention.

The arrangement for idle speed control known from DE-A-2 546 076 acts on a throttle valve arranged in the intake pipe of the internal combustion engine. A setpoint generator and an actual value generator are provided for the speed, the output voltages of which are fed to the two inputs of a differential amplifier. An output signal characterizing the control deviation acts on an actuator designed as a solenoid. The actuator is continuously connected to the throttle valve and adjusts it according to the control deviation. This circuit is also not able to introduce boundary conditions into the control and take external factors into account and thus ensure under all circumstances that the idle speed of an internal combustion engine remains safely within a predetermined range, even if transition conditions that take effect quickly have to be absorbed. In particular, this known circuit is not suitable to be used at the same time to influence the overrun operation, namely for fuel-saving overrun cutoff.

The object of the invention is to optimally further develop the system known from the article in "Automobil Technische Zeitschrift".

See also EP-A-77 997. There, a "method for controlling the speed of an internal combustion engine ..." is claimed, in which the integral part of the control amplifier, bypassing a dead zone range, is proportional to the actual value of the speed in order to center the idle speed operating point against long-term influences Signal is supplied. In addition, proportional and integral parts of stair-shaped and asymmetrical types are provided.

Advantages of the invention

The method according to the invention and the device according to the invention, each with the characterizing features of the main claim or the first device claim, have the advantage, in contrast, that any external boundary conditions are introduced in addition, interfering influences and precise positioning of the idling speed, in particular also while avoiding long-term influences such as temperature and air pressure can be realized.

It is particularly advantageous that the transitions between the various operating states, which constantly occur during operation of an internal combustion engine, can be smoothly absorbed and smoothed by the invention, for example from thrust to part load, from part load to idling, idling to thrust, etc.

The invention works with regard to the setting of the idle speed in fully regulated operation; If the idling speed or the speed in the area close to idling is exceeded, the throttle valve can be switched over to control and partial tracking of the actuator.

The regulation and control according to the invention reacts quickly and reliably to all possible disturbance variables.

Advantageous developments and improvements of the invention are possible through the measures listed in the subclaims. It is particularly advantageous here that the technical implementation of the overall concept according to the invention can be carried out by circuits and systems which operate on an analog and / or digital basis, computers or the known microprocessors also being able to be used to perform certain subfunctions.

drawing

Embodiments of the invention are shown in the drawing and are explained in more detail in the following description. 1a and 1b show the control behavior of the main controller for idling or the area close to idling with P component and I component, in each case above the speed deviation, based on an idling speed setpoint, FIG. 2 in the form of a diagram the transition behavior from thrust to idling, the path of the actuator being plotted over time, FIG. 3 in the form of a diagram the transition behavior from thrust to partial load, the actuator path being plotted against time, FIG. 4 the transition behavior from part load in Idling in the form of a diagram and FIG. 5 the intervention of the control with respect to the actuator control in accordance with a pulse length modulation in the form of a diagram. 6a to 6d show, in the form of diagrams over time, the detection of the actual position of the actuator via thresholds and the resultant control, similar to pulse length modulation, for example of valves in the actuator by the electronic control circuit, while FIGS. 7 and 8 in Form of block diagram representations Realization options for the electronic control circuit shows in an essentially digital representation.

Description of the embodiments

For a better understanding of the present invention, the procedural basic concept with regard to the control sequence and the control behavior is first explained in detail below with reference to the diagrams in FIGS. 1 to 6, the central electronic control circuit 1 shown in FIG. 7 essentially being explained with regard to its main functions and is taken into account, in each case in connection with the peripheral sensors and circuits which are to be used and which mainly serve to obtain actual value signals.

As shown in FIG. 7, the central electronic control circuit 1 works on the output side via an output stage 1a on an actuator 2, which in the exemplary embodiment shown is preferably designed as an electropneumatic actuator and has an evacuating valve 2a and an aerating valve 2b. The control of the valves 2a, 2b takes place electrically via assigned relays 3a, 3b, specifically, as will be explained further below, according to a process similar to a pulse length modulation via output stage transistors 4a, 4b connected to the respective relays. The actuator 2 actuates with its valves 2a, 2b a push) 10, which rests on the main throttle, not shown, so that when the evacuating valve is actuated, the plunger 10 is retracted and thus the main throttle is closed more tightly, while when the ventilating valve is actuated, the tappet 10 adjusted more and accordingly the main throttle is opened more. However, the main throttle or a mechanical part connected to it, for example a throttle valve lever, can be lifted off the tappet 10 at any time by actuating the accelerator pedal and is therefore also only under e.g. Spring pressure on.

The overall control concept distinguishes four functional areas, which are dependent on speed thresholds and a characteristic signal which is derived from the system of the main throttle on the tappet 10. The central electronic regulating and control circuit controls the electropneumatic actuator 2 differently in accordance with the functional range detected in each case. The four functional areas can be characterized as follows:

1. Starting the internal combustion engine

Distinguishing feature: The engine speed is less than a certain starting speed (n≤n _Anl ).

2. Idling of the internal combustion engine

Distinguishing feature: The speed lies between the starting speed and a thrust speed (n _Anl <n <n _thrust ) and the main throttle is applied to the tappet 10.

3. The internal combustion engine is in the partial load range

Distinguishing feature: The speed is between the starting speed and the thrust speed (n _Anl <n <n _thrust ) and the main throttle is not on the tappet 10 (or on a mechanical stop).

4. Overrun operation of the internal combustion engine

Identifier; The effective actual speed of the internal combustion engine is greater than or equal to the overrun speed (n ≥ n _Schun ).

To obtain and supply a speed signal to the central control circuit 1, the time between two ignition pulses 5, which are supplied to the terminal 6 of the circuit of FIG. 1, is most conveniently measured and the time interval (period duration) thus obtained is used for speed detection. It goes without saying that, instead of the ignition pulses, other signals can also be used, which can occur synchronously with the engine speed, for example dead center sensor e. the like

For period measurement, the central control circuit 1 can have a clock generator or oscillator, which is not initially shown separately; If the central control circuit is a so-called microcomputer circuit at least in some areas - preferably a 4-bit microcomputer that has neither a timer nor an interrupt option, then the circuit sequence of the controller (namely the controller program in this case) organized in a loop. This program _loop has a constant running time T _loop and forms the time base of the control program. In order to detect an ignition pulse, a flip-flop 7 is always set by the flip-flop, which works as a buffer and can be, for example, a monoflop or a bistable element, ie a flip-flop. The timing between two ignition pulses as a measure of the period of the speed is measured by the oscillator or clock generator of the central control circuit 1, which works with a constant oscillation period, so that with the oscillations of the clock generator with a significantly higher frequency, based on the highest, due to the occurrence of Ignition pulses characterized speed frequency, a counter is acted upon, the current counter reading then in each case in a memory cher loaded and the counter is reset when an ignition pulse occurs. The occurrence of an ignition pulse can be determined in each case by the fact that the flip-flop 7 has been switched to its other state. The flip-flop 7 as a buffer is then, if it is a bistable element, either reset by the next clock pulse - which means that the current counter reading is taken over into the memory at the same time - or the reset takes place automatically if the flip-flop is a monoflop. The memory content is then a measure of the period and thus of the speed of the internal combustion engine, the resolution being determined by the frequency of the clock generator or oscillator in the control circuit. If the program loop frequency of a microcomputer used in this case is used for the derivation of the clock frequency, then the buffer 7 is queried as a monoflop or as a flip-flop once per loop pass and then either reset by the program (with two stable states of the buffer) or automatically reset (with one Monoflop). If a monoflop is used, the program can wait for a reset to wait for the monoflop to reset after an ignition pulse occurs, thereby achieving yitter-free speed detection. Then T _Mono t _{! O} p ^> Ts _ch i _e it _e . In this case, too, the counter is incremented each time the loop is run, so that when the next ignition pulse occurs, a current counter reading corresponding to the period of the speed is located in it and can be loaded into the memory. In any case, there is always a current speed signal in the memory, which can be evaluated accordingly by the control circuit 1 and is available.

The detection of the position of the main throttle (system) on the tappet 10 or on a mechanical stop corresponding to the throttle valve closed or throttle valve opened in order to distinguish between the functional areas mentioned further above takes place with the aid of a throttle valve switch designated by 8 in FIG. 7. The throttle valve switch 8 can be designed such that a signal log 1 results, for example, when the main throttle is present on the actuator or on the tappet 10 and a signal log 0 when the main throttle is not present. It is possible to evaluate these two signal states directly by the control circuit 1; for debouncing this encoder signal, which is a switch, it can also be advantageously carried out in such a way that in the case of signal log 1 from the encoder or switch 8, a counter via the clock generator or oscillator of the control circuit 1 or by means of the loop frequency of Microprocessor is increased and decreased at signal log 0. This counter can be counted up or down between a maximum and a minimum value. When the counter limits are reached, a buffer is set or reset. For example, the buffer can be set to the log 1 signal when the maximum value is reached and to the log 0 signal when the minimum value is reached. It can be seen that this buffer is most appropriately a bistable element, the output signals of which indicate whether the main choke is present or not.

By evaluating the two information about the speed and the "main throttle system" fed to the control circuit 1 in this way, the following workflow is obtained which represents the functional areas which are further characterized. The control circuit is given a target idling speed, which can also be a counter content, for example, or in the case of an analog configuration, for example, a constant voltage which is compared with the content of the memory already mentioned above, or with an analog voltage which is derived from the memory content in a known manner The central control circuit 1 is designed such that it has at least one This PI control behavior applies to the idling functional area and to the resulting special control structure of the plunger 10 via output stage 1a and actuator 2. The PI control behavior works with a proportional component en and integrator running speeds, but with a preferably analog design of the control circuit 1, however, a certain asymmetry in the PI control behavior is provided, so that, for example, if the speed falls below the setpoint speed in idle, the reaction can be faster and / or stronger, possibly with an additional D component, to save the engine from going out, so to speak.

If the control circuit 1 is constructed with digital components or implemented in the form of a microcomputer, then e.g. To simplify the program structure, constant proportions and integrator running speeds are not used, but the speed range is divided into several ranges, also and preferably different ranges, around the target idling speed according to the diagrams in FIGS. 1a and 1b, these ranges then each having constant P- Parts and constant I running speeds included. This then results in stepped platform curves for given control deviations for both the P and the I component.

It can be seen from the illustrations in FIGS. 1a and 1b that, apart from a zero range (dead zone range) arranged on both sides symmetrically around the target idling speed, with respect to the P component and the I component of the control amplifier, deviations from this target idling speed up and down also P components of different sizes or I running speeds can be worked, so that overall there is a non-linear control behavior with stronger intervention options and reactions if desired, if certain Speed ranges around the target idle speed can be reached from the actual speed.

The P component and the integrator level are added and used for actuator control. With a digital structure of the control circuit, the PI sum of the P and I components can be stored in an output memory and in an intermediate memory. A possible variant here is to load the buffer store with a value in the output store averaged over a certain time. As long as the functional range idling prevails, the circuit operates fully in the regulated mode according to the overall concept, whereby, as will be discussed further below, the plunger position 10 caused by the actuator 2 is also detected and compared with the setpoint value, which is the PI sum at the output of the control amplifier.

In the presence of the partial load functional area, that is to say when the main throttle is actuated and no longer applied, the integrator is first stopped by this transition of the main throttle identification signal from log 1 to log 0; A suitable blocking signal eliminates the further evaluation of the P component. The PI sum in the buffer, which corresponds to the last value in the idle range, is still used for actuator control, so that it remains in the last position before the main throttle is actuated (initially).

Arise during operation overrun phases, the control circuit 1 is configured so that the actuator is moved back for rotational speeds n≥n _thrust. Therefore, when the main throttle is not actuated, thrust cut-off is possible via the accelerator pedal if the mixture generator (for example, carburetor) is designed accordingly.

If the functional area of starting the internal combustion engine results, then the speed _signal present in this case of n <n _Anl is evaluated in such a way that the actuator is fully extended in order to clearly determine the position. In an advantageous embodiment, a temperature detection is carried out in addition to setting the initial values for the integrator level and the PI sum in the output memory and buffer.

In the case of transitions between the functional areas, the following special functions result, which will be discussed first with reference to FIGS. 2 to 4. The respective transitions can be easily detected and carried out by the central control circuit 1 by changing the sensor signals supplied to them with regard to the speed and the identification signal with respect to the main throttle system.

If there is a transition from thrust to idling, the PI sum is output in the buffer for actuator control for a specific time t _vs (t _vs can be, for example, 2 seconds) (compare the course of the diagram indicating the actuator's travel over time) Fig. 2). The actuator therefore occupies the last position in the idle range before changing to another range. There is the possibility and can preferably also be used to add a constant for a short time t _a (eg t _a = 0.2 s) in addition to the output value for actuating the actuator from the buffer. This results in a brief increase in filling after thrust cut-off to avoid speed drops. The control is released again after tvs (and t _a ) have expired.

If there is a transition from overrun to partial load, then the function can proceed according to the diagram in FIG. 3 as with the transition from overrun to idle; After the transition period ty _s , however, the control is not released, but the actuator remains in the last position of the idle range in accordance with the PI sum stored in the buffer. The transition period t _vs shown in FIG. 3 is therefore only given for better understanding and is of no importance for this transition function.

If there is a transition from part load to idling as shown in FIG. 4, the central control circuit 1 controls the actuator in such a way that the actuator remains in the position in the part load range for a predetermined period of time tt (for example also 2 s) has shown the last value before leaving the idle range, i.e. again corresponds to the PI sum in the buffer. The regulation is then released.

• a) die Hauptdrossel liegt nicht an. In diesem Fall wird Schubbereich erkannt, d.h., das Stellglied fährt zurück auf eine Ausgangs-Nullposition.
• b) Liegt die Hauptdrossel an, dann wird weiterhin auf Leerlaufbereich erkannt; mit anderen Worten, die Steuerschaltung regelt die Position des Stössels weiter. Hierdurch lässt sich ein Fehlverhalten der Regelung verhindern, wenn die Pl-Summe im Zwischenspeicher einen zu grossen Wert enthält (bei Übergang Schub in Leerlauf würde das Stellglied zu weit anstellen).

When changing from idle to thrust, two different conditions could be recognized, namely

• a) the main throttle is not present. In this case, the thrust range is recognized, ie the actuator moves back to an initial zero position.
• b) If the main throttle is present, then the idle range is still detected; in other words, the control circuit further regulates the position of the plunger. This can prevent the control from malfunctioning if the PI sum in the buffer contains too large a value (the actuator would move too far when the thrust changes to idle).

It has already been pointed out earlier that the valves of the electropneumatic actuator are actuated electrically, with discontinuous behavior being achieved. First of all, it should be pointed out that, as will be explained in more detail below, the central control circuit also receives a position signal or position signal relating to the tappet position and thus, what applies to the control area, also to the position of the throttle valve. This position signal is an actual value signal and is compared by a suitable comparator or comparator of the control circuit with the target value, which results as a PI sum in the buffer.

The illustration in FIG. 5 shows in the form of a diagram how the control circuit controls the actuator 2 with a method approximating the pulse length modulation in order to achieve the desired one Position or location of the plunger 10 to achieve.

In area 0, which extends symmetrically around the respective position setpoint S _so n, for example, the actuator is not activated and both valves are closed. In other words, the plunger retains its current actual value position, so that there is a certain symmetrical dead zone range.

In the area marked 11 in FIG. 5, that is to say if the actual value of the actuator is too great, the evacuating valve 2a is actuated with a pulse duty factor a ₂ , which increases with increasing distance from the target value from a ₂₁ = 0 to a ₂₂ = 100% quasi steadily. The same applies to area I, in which the actual value of the respective position of the actuator is too low and therefore pulse-length-modulated actuation of the venting valve 2b takes place.

According to a preferred embodiment of the present invention, the position detection of the plunger position is carried out either by simply returning a tapped potentiometer potential, which in turn is adjusted by the plunger position. In the case of a digital design of the control circuit, the position detection can be carried out with the aid of a digital-to-analog converter, which is designated 9 in FIG. 7. The digital-to-analog converter queries the potentiometer voltage that can also be used here (according to the actuator position) via thresholds. The potentiometer, which is adjusted by the actuator or the plunger, is designated by 15 in FIG.

The representation of FIG. 6 shows in diagram form more precisely what is meant. Once per clock cycle or when using a microcomputer once per program loop run, two thresholds, namely an upper threshold A and a lower threshold B, are used to query whether the potentiometer voltage is above, within or below the thresholds; the two valves 2a, 2b of the actuator are then activated accordingly. The two thresholds, as shown in Fig. 6, are overlaid with a sawtooth or triangular shape, i.e. they are sawtooth-shaped, so that a pulsed output differential value is obtained directly by comparison with the actual value on the potentiometer, which leads to pulsed control signals with different pulse durations for different actual value positions and, if necessary, corresponding deviation from the setpoint.

6a shows the position setpoint corresponding to the PI sum at C approximately in the buffer. The two threshold curves originating from the digital-to-analog converter are expediently arranged symmetrically upwards and downwards around this setpoint value C, so that when the actual position value is identical to the setpoint value C, there is no overlap between the actual value initially assumed to be horizontal and the Sawtooth pulses of thresholds A and B result.

6a, three different actual value profiles are now assumed, namely an upper actual value profile D, a lower actual value profile E and an obliquely increasing actual value profile F, which thereby gradually approaches the desired target value C. In the event of a match between target value C and actual value, both valves 2a and 2b of actuator 2 are closed. The ram position is not influenced or changed.

If the actual value is constantly too high, as can be seen from the long dashed curve of curve D in the diagram in FIG. 6 (all curves are shown over time t), this leads to overlaps with the sawtooth curve of the upper threshold A and accordingly to in Fig. 6b shown control pulses for the evacuating valve. In this exemplary embodiment, these drive pulses each have a constant duration, which, however, cannot always be assumed.

The diagrams in FIGS. 6b to 6d each show two possible curve profiles over time one above the other, the upper diagram always being used for the control of the evacuating valve and the lower curve for the control of the ventilating valve.

If an actual value curve corresponding to curve D in FIG. 6a has to be assumed, then the associated representation in FIG. 6b shows that the venting valve is not activated at all, i.e. always remains closed, while the control pulses are supplied to the evacuating valve according to the upper curve, in the presence of which the evacuating valve is transferred to its open position. It can be seen that by actuating the evacuating valve, the plunger position is withdrawn (retracted), so that the actual value curve begins to decrease in accordance with curve D in FIG. 6a (this is not shown in the curves shown). However, it can also be seen that a gradual shifting of the actual value curve profile D downwards shortens the duration of the control pulses, that the overlap periods with the sawtooth threshold A become shorter.

If one assumes an actual value curve curve corresponding to E (short dashed curve), then the actual value is too low and the plunger 10 must be extended more. Since the actual value E deviates even more strongly from the predefined setpoint position than the previously assumed actual value D, in this case the time periods resulting from the covering with the lower sawtooth threshold B are also longer and, in accordance with the course of FIG. 6c, there are also longer control pulses for the venting valve, while the evacuating valve is not activated and therefore remains closed.

FIG. 6d is assigned to the gradually increasing actual value curve of the dash-dotted curve F in FIG. 6a; it can be seen that the actual value F is gradually approaching the desired setpoint C and therefore in this case also Control impulses for the ventilating valve become less and less.

For temperature detection, the digital-to-analog converter 9 already mentioned above can also be used, a counter being increased with each clock pulse from the oscillator or clock generator, which is part of the control circuit 1, or with each program loop run in a microcomputer; this counter reading is given to the digital-to-analog converter; Because of the double use, it goes without saying that this takes place at different times for the position detection of the ram and for the temperature detection, for example in a multiplex method. A comparator 11 is provided (see FIG. 7), the input of which is supplied with an analog temperature signal from a suitable resistance network. This resistance network contains at least one NTC or PTC resistor for temperature detection, which is in heat-conducting contact with suitable parts of the internal combustion engine, such as the cooling water. Since the comparator 11 is constantly supplied with a counter-proportional voltage, i.e. an increasing voltage from the digital-to-analog converter, the comparator 11 will then emit a signal when the counter-proportional voltage of the central control circuit 1 exceeds the temperature-dependent voltage. At this moment, the last counter reading corresponds to the temperature value at the comparator, so that it is a measure of the temperature range in which the internal combustion engine works. This counter reading is stored, which is easily possible as a transfer signal into a memory due to the comparator output signal, and is used for temperature evaluation. At the same time, the counter can be reset so that temperature changes can also be recorded.

• a) das evakuierende Ventil 2a ständig geöffnet wird, was jedoch dazu führt, dass sich infolge des hohen Unterdrucks im Stellglied beim Anstellen einer Ansprechverzögerung des Stellgliedes bei Verlassen des Schubbereichs ergibt, da das vorhandene Vakuum erst verringert werden muss, bevor eine Wegauslenkung überhaupt möglich ist. Es könnte sich daher eine zweite Variante dahingehend anbieten, dass
• b) das Stellglied lediglich in eine zurückgeführte Position gesteuert wird, die vor einem mechanischen Anschlag liegt und bei welcher daher das evakuierende Ventil nur so lange geöffnet ist, bis diese rückgenommene Position erreicht ist. Anschliessend ist das evakuierende Ventil wieder geschlossen, so dass sich die Umsteuerung in den Regelbereich dann mit kürzesten Erholzeiten durchführen lässt. Der sich hierdurch ergebende Verlust an Stellweg ist geringfügig.

With regard to the thrust positioning of the ram 10, two variants are possible. According to what was said above, the actuator is retracted at overrun speeds. This can be done in such a way that

• a) the evacuating valve 2a is constantly opened, which however results in the fact that, due to the high negative pressure in the actuator, when the actuator starts to delay a response when it leaves the thrust area, since the existing vacuum must first be reduced before a displacement is even possible . A second variant could therefore offer itself in such a way that
• b) the actuator is only controlled in a returned position, which lies in front of a mechanical stop and in which therefore the evacuating valve is only opened until this withdrawn position is reached. The evacuating valve is then closed again, so that the changeover to the control range can then be carried out with the shortest recovery times. The resulting loss of travel is minimal.

In an advantageous embodiment, the following variant can also be implemented for the functional sequence partial load or for the transition function from partial load to idling. The actuator moves the main throttle up to the system by setting the integrator, i.e. in this variant the integrator is not stopped when the main throttle is actuated in the partial load range. If there is then a transition from partial load to idling, the actuator is reset in a controlled manner until the idling speed is reached. This has the advantage that speed drops during the transition from partial load to idling are avoided if the load torque of the engine has previously been increased in the partial load range, for example by switching on consumers, air conditioning and the like; this solution is also advantageous for motor vehicles with automatic transmissions.

According to an advantageous embodiment of the present invention, a further possible variant of the transition behavior from thrust to idle results from the fact that the integrator is set via a D component and the control is then released. The D component is obtained by differentiating the speed signal, a large D component resulting when the speed drop rate is also high. In this way, an adjustment of the actuator via the integrator proportional to the speed of the sinking speed is achieved. The advantage lies in the fact that one can achieve a stronger setting of the actuator and thus a better interception of the speed via the higher speed sink rate at high load torques. This is also advantageous for motor vehicles with an automatic transmission, since the load torque of the torque converter strongly depends on the previous history.

Finally, according to a further variant for the partial load functional area and the transition area from partial load to idling, it is also possible to control the actuator in the partial load range in a speed-controlled manner, the speed setpoint also being influenced and the actual speed being able to be tracked. When changing to the idle range, the tracked speed setpoint is then reduced to the actual setpoint after a predetermined time function, with the result that the speed is reduced in a controlled manner via this time function.

Finally, FIG. 8 shows a possible embodiment from a large number of conceivable forms for realizing the central control circuit 1; in this embodiment, a predominantly digital mode of operation is required; wherein the circuit components already shown in Fig. 7 have the same reference numerals. In Fig. 8 the clock generator is designated 20; for speed detection, a counter 21 is acted upon by the counting pulses of the clock generator 20 and is then reset via the flip-flop 7 when an ignition pulse arrives; At the same time, a takeover pulse is emitted from the counter 7 via the connecting line 22 to a transfer gate 23 connected downstream of the counter 21, so that there is in each case a counter reading in the transfer gate or buffer store 23 which corresponds to the period of the actual speed. It is now possible to convert the counter reading in the buffer 23 into a voltage either by means of a digital-to-analog converter and then to compare and compare this voltage by comparing and forming a difference with a constant or, if appropriate, also by other boundary conditions and comparing the setpoint voltage resulting speed control deviation using a control amplifier with P and I components in accordance with the control behavior already shown in FIGS. 1a, 1b and explained further above to obtain a desired position value with respect to the actuator.

If you continue to work digitally here, the counter reading in the intermediate counter 23 is to be compared with a target counter reading in a register (not shown separately in FIG. 8), in which the target value of the speed is entered. This comparison can be made by counting the intermediate memory 23 and the register or a takeover counter connected downstream of this in each case with a high clock rate, so that there is a counter difference which is then processed separately with respect to the P component and the I component by supplying corresponding ones digital circuit components that function as shown in FIGS. 1a and 1b. In the illustration in FIG. 8, the block carrying out the difference formation with a target speed is generally designated by 24; the two downstream blocks 25 and 26 are each responsible for processing the speed difference. The binary words resulting at the outputs of the P block 25 and the I block are fed in parallel to a buffer memory 27 and the output memory 28, the counter reading of which, in addition of the P component and the I component, then corresponds to the PI sum of the target position of the actuator corresponds.

In order to form the thresholds A and B, with which the actual position value of the plunger 10 acquired at the potentiometer 10 and which is fed to one input of a downstream comparator 12, a first counter 29 and a second counter can be operated with a high clock rate 30 can be controlled for the formation of the upper threshold or the lower threshold. Different for the two counters 29 and 30, however, are the initial conditions that are set based on the PI sum for the target position in the output memory 28 via different upstream setting registers 31 and 32 with initial values that are symmetrical to the PI sum of the target position. Each time a predetermined counter reading of the counters 29 and 30 is reached, these are reset, for which purpose any means can be provided which need not be discussed further. This results in rapidly changing counter readings at the outputs of the counters, which are supplied via a downstream switching device 33 to the one digital-to-analog converter 9 already mentioned above, or, if desired, also separate analog-to-digital converters. The switch 33 can perform a multiplexed control. The output signal of the digital-to-analog converter is then the sawtooth-like threshold voltages A and B, with which the actual value of the actuator position at the comparator 12 that actually drops at the potentiometer 10 is compared. It can be seen that the pulse sequences shown in FIGS. 6b to 6d drop off directly at the output of the comparator, it being understood that this output is given selectively to the output stages 4a, 4b arranged on the output side by means of a switching device operating synchronously with the switching device 33, so that the downstream valves can be controlled correctly according to the sign.

In the same way, the temperature signal can be obtained with the aid of a further counter 34; the output signal of the comparator 11 is then fed back and triggers a take-over memory, which is not shown in FIG. 8 and takes over the current counter reading of the temperature counter 24. In this way, a temperature signal is also obtained, which can be used, for example, to effect corresponding setpoint changes. For example, this temperature signal, which is a binary word in the exemplary embodiment in FIG. 8, can be used to change the speed setpoint set in a register in accordance with a desired function, so that a cold speed setpoint can be used when the machine is cold.

Finally, another counter 35 is shown in FIG. 8 for obtaining the system signal of the main choke; this counter is supplied with up and down count signals at its inputs 35a, 35b in accordance with the position of the throttle valve switch, ie either log 0 or log 1. As soon as the counter 35 has reached a maximum or a minimum value, as already explained above hereby a downstream buffer 36 is set or reset. This buffer can be a bistable element and its output then shows the respective position of the throttle valve, whether it is on the actuator or not. The output signal of the buffer initially reaches the control amplifier with a P component and I component via a connecting line 37. If, for example, the main throttle is actuated, then as already mentioned above, the integrator is stopped by this output signal and the proportional control amplifier is blocked for the speed difference. With the circuit shown in FIG. 8, the functional sequences already mentioned above can be easily carried out; the main purpose of this is the output signal of the buffer 36 for the signal “choke system”. Regarding the processing of the effective actual value speed signal in the buffer 23 further comparison memories are provided, which are not shown in FIG. 8 and serve to fulfill the functions mentioned above with regard to the functional areas and the transition functions. Thus, when the actual speed in the memory 23 corresponds to the starting speed, the venting valve 2b in the actuator 2 can be controlled directly from the output of a first comparative memory, as a result of which the plunger 10 is fully extended. If there are overrun phases, then a further comparison memory, in which the limit speed for the overrun operation is set, is used to drive the actuator back here again, bypassing the control operation, namely, for example, by direct activation of the evacuating valve 2a. It is easy to see that all work areas and transition functions can be realized in this way. The transition function from thrust to idle is only considered as an example. If a system signal of the main throttle valve results from a corresponding switchover of the intermediate store 36, its output signal causes the inputs of the counters 29 and 30 or their set registers 31 and 32 to be switched over to the intermediate store 27 in order to take over the PI sum stored there for the actuator control . At the same time, a counter can be started which determines the delay time t _vs until its maximum value is reached and then effects the switch back to the output memory 28, simultaneously with the release of the control. In order to bring about the brief increase in filling if necessary (avoiding speed drops), different, increased initial values can be entered in the setting registers 31 and 32 for a transition period. Since this is a measure familiar to a person skilled in the art, there is no need to go into this further.

It is also pointed out that the path of the actuator is designated by S in FIGS. 2, 3 and 4, while the control is released in each case at point G of the curve in FIGS. 2 and 4. 2, the last value in idle is also denoted by H; at time t _{o there} is a transition from thrust to idling. Likewise, in the diagram of FIG. 4, at time t _1, the transition from part-load range to idling takes place via the further time delay tvr still provided there.

Claims (26)

1. Method for controlling the idling speed of a combustion engine through the position of a throttle flap in the induction manifold, also of influencing it in the speed range adjacent to and in overrun operation, with the evaluation of an actual value signal proportional to the speed, of a speed desired value signal to form the control deviation taking into consideration signals relative to the throttle flap position (throttle flap switch 8) and to the combustion engine temperature and corresponding modulation of a control element (10) acting upon the throttle flap, a PI-control of the speed deviation from the idling speed desired value and a position control of the control element for the throttle flap, characterized in that, taking into consideration a dead zone range round the desired idling speed, within which no output signals are generated by the P and I controller modules, respective proportional fractions and integrator run speeds (time constant, slope) are formed which are respectively constant only for at least three specific speed ranges about the desired idling speed, so that, in respect of the P and I-fractions, stepped respectively constant P-fractions and I-fractions starting from the desired idling speed are obtained for the PI overall control behaviour and the control amplification is additionally constructed asymmetrically relative to the possible downward or upward deviations from the idling desired speed in order to capture unsteady proportional and integral fractions (Fig. 1b).

2. Method according to Claim 1, characterized in that as a variation of the control system in the case of the idling function, the control system is switched off in the case of the start, partial load and overrun operating modes and the position of the part in physical contact with the throttle flap is determined solely by control of the control element.

3. Method according to Claim 1, characterized in that the Pi-sum corresponding to the position desired value of the control element which originates from the control system during idling is stored continuously in an intermediate memory.

4. Method according to Claim 3, characterized in that a value, averaged over a specific time, of the PI-sum in respect of the position desired value of the control element in an output memory is accepted in the intermediate memory.

5. Method according to any of Claims 1 to 4, characterized in that the actuation of the main throttle (through the accelerator pedal) is detected as main throttle contact signal and the integrator is stopped for the partial load functional range now existing and the last PI-sum present in the intermediate memory is accepted as position desired value for the further modulation of the control element.

6. Method according to one or more of Claims 1 to 5, characterized in that when overrun phases are detected (n-en_overrun) the control element is restored onto a stop.

7. Method according to one or more of Claims 1 to 6, characterized in that when the start operating state is detected the control element is totally extended and the combustion engine temperature is detected and is used to set initial values of the integrator state and of the PI-sum for the position desired value in the output memory and intermediate memory.

8. Method according to one or more of Claims 1 to 7, characterized in that for the speed detection the time between two impulses (ignition impulses; dead centre transducer impulses) appearing synchronously to the engine revolution are measured in that a counter is fed with counting impulses, whilst the current counter state is accepted into a memory with simultaneous resetting of the counter whenever an impulse synchronous with the engine revolution appears.

9. Method according to Claim 8, characterized in that where the controller program is organized in a loop of a central control device, the program loop is used with constant pass time and is used to inquire an intermediate memory triggered by impulses appearing synchronously with the engine revolution, whilst a counter is incremented at each loop pass.

10. Method according to any of Claims 1 to 9, characterized in that in order to cushion a throttle flap switch to detect the main throttle contact signal a counter is counted upwards or downwards according to the contact signal, whilst an intermediate memory is set or reset respectively each time the maximum or minimum value is attained.

11. Method according to one or more of Claims 1 to 10, characterized in that at the transition from the overrun to the idling operating state the PI-sum present in the intermediate memory for the position desired value is output for a prescribed time (t_vs) for the control element modulation, so that the control element assumes the last position in the idling range before the release of the control system after the time has expired.

12. Method according to Claim 11, characterized in that a constant value is added to the output value of the control element control system for a short time (t_a), so that a brief charge increase results in order to avoid speed breakdowns.

13. Method according to one or more of Claims 1 to 12, characterized in that at the transition from overrun to partial load the control element is maintained in the last idling range position, the blocking of the control system being maintained after expiry of the transition time (t_vs).

14. Method according to one or more of Claims 1 to 10, characterized in that at the transition from . partial load to idling of the combustion engine the control element is maintained for a prescribed time (t_vt) in the last partial load position corresponding to the last position desired value before leaving the idling range, before the control system is released after the time has expired.

15. Method according to one or more of Claims 1 to 10, characterized in that at the transition from idling to overrun, the overrun range is detected if the main throttle is not in contact, and the idling range continues to be detected if the main throttle is in contact.

16. Method according to one or more of Claims 1 to 15, characterized in that the control element constructed electropneumatically with a ventilating and an evacuating valve is modulated electrically in cadence after the manner of a pulse length modulation for each valve, whilst the cadence ratio of the modulating impulses increases virtually steadily with increasing distance of the position actual value of the control element from the position desired value formed by the PI-sum.

17. Method according to one or more of Claims 1 to 16, characterized in that threshold comparison voltages following a general sawtooth or delta shape are generated staggered upwards and downwards relative to the position desired value corresponding to the PI-sum and are fed to the one input of a comparator, to the other input of which the actual value voltage of the position actual value of the control element captured by means of a stroke/voltage converter (potentiometer) is fed so that modulation impulses for the control element valves are obtained at the output of the comparator from the superimposition of the sawtooth thresholds with the actual value.

18. Method according to one or more of Claims 1 to 17, characterized in that for the temperature detection a counter is incremented by counting impulses or at each program loop pass and the respective current counter state is fed after digital-analog conversion to a comparator, to the other input of which a voltage proportional to the combustion engine temperature is fed, so that an output signal then appears at the comparator when the counter-proportional voltage is equal to the temperature-dependent voltage, so that at the moment of the comparator output signal the counter state can be evaluated as a measure of the temperature.

19. Method according to one or more of Claims 1 to 18, characterized in that the overrun range operating state of the combustion engine the evacuating valve remains open by permanent modulation in order to effect a total retraction of the control element.

20. Method according to any of Claims 1 to 19, characterized in that in the overrun range the control element is retained in a position in front of the mechanical stop by interrupted modulation of the evacuating valve, so that short recovery times are obtained for the control element normal function.

21. Method according to one or more of Claims 1 to 20, characterized in that under partial load the control element of the main throttle is made to follow until contact by setting the integrator, so that at the transition from partial load to idling a controlled restoration of the control element is performed until the idling speed is attained.

22. Method according to one or more of Claims 1 to 21, characterized in that at the transition from overrun to idling operation the integrator is set through a D-fraction and the control system is then released so that an engagement of the control element proportional to the speed reduction velocity occurs through the integrator.

23. Method according to one or more of Claims 1 to 22, characterized in that in the partial load range the control element is guided controlled by the speed with simultaneous follow-up of the speed desired value with reference to the speed actual value, and that at the transition into the idling range the followed-up speed desired value is reduced to the normal desired value of the speed according to a time function.

24. Apparatus for performing the method according to one or more of Patent Claims 1 to 23, for controlling the speed of a combustion engine with a throttle flap in the induction manifold during idling, and of influencing it in the speed range adjacent to idling and in overrun operation, having a control amplifier which detects the control deviation between a speed-proportional actual value signal and a speed desired value signal and modulates a control element (10) in active association with the throttle flap, having a central electrical control circuit (1), to the inputs of which control signals (ignition impulses, dead centre transducer impulses) synchronous with the engine revolution, a main throttle contact signal from a throttle flap switch (8), a position actual value signal from a stroke/voltage converter (potentiometer 15) relating to the control element position and a combustion engine temperature signal from a temperature sensor of the combustion engine are fed, and to the output of which the control element (2) is connected, which moves a push rod in contact with the main throttle flap, characterized in that means for step-shaped asymmetrical P and I-control are provided taking into consideration a dead zone region around the desired idling speed.

25. Apparatus according to Claim 24, characterized in that in order to convert the digitally formed PI-sum position desired value a digital-analog converter (9) is provided, the output signal of which is fed to a comparator (12), the other input of which is connected to the stroke/voltage converter (potentiometer 10) for the position actual value, so that the output impulses of the comparator serve directly for the modulation of the control element (2).

26. Apparatus according to Claims 24 or 25, characterized in that the control element (2) is of electropneumatic construction with an evacuating valve to retract the push rod in contact with the throttle flap and a ventilating valve to engage the push rod.

Images