DE10360834B4 - Method and control device for controlling an output of signals for an internal combustion engine - Google Patents

Abstract

A method of controlling the output of a signal (70, 62; 72) at a start in the duty cycle of a direct fuel injection internal combustion engine (10) comprising a controller program having a functional level (52) and a driver level outputting the signal (70,62; 72) (56), wherein the duty cycle extends over a fixed angular range of a shaft of the internal combustion engine (10), and periodically discrete angular values (64) of the angular position of the shaft are detected, and wherein in each case an angular distance (80) between a detected discrete angle value ( 64) and a reference angle (79), and wherein at the driver level (56) a time interval (68) of the output of the signal (70, 62, 72) to at least one periodically detected discrete angle value (64) as a starting value for a Timer, which upon reaching a final value, an output of the signal (70, 62; 72) triggers, and wherein in case of deviations of the reference angle (79) from the initial angle d wherein the time interval (68) at the driver level (56) is related to the reference angle (79), and wherein at least a plurality of successive discrete angle values (64) determines the time interval (68) of the output of the signal; the output of the signal (70, 62; 72) takes place after expiration of the time interval (68) when the time interval (68) corresponds to a predetermined distance.

Description

State of the art

The invention relates to methods for controlling the output of a signal at a start in the duty cycle of a direct fuel injection internal combustion engine with a controller program having a functional level and a signal output driver level, the duty cycle extending over a fixed angular range of a shaft of the internal combustion engine and being periodically discrete Angle values of the angular position of the shaft are detected and wherein in each case an angular distance between a detected discrete angle value and a reference angle is determined and wherein at the driver level, a time interval of the output of a signal to at least one periodically detected discrete angle value is determined as a start value for a timer when an end value is reached, an output of the signal is triggered.

The invention further relates to a control device which carries out such a method.

From the EP 1 165 962 B1 For example, a method of controlling the output of the end of charge time of an ignition coil is known. The duty cycle extends over a fixed angular range of a shaft of the internal combustion engine. Periodically discrete angular values of the angular position of the crankshaft are detected. An angular distance between a segment start and a firing angle is determined. The time required for the angle remainder to an angle encoder pulse is determined as a start value for a timer. In case of deviations of the end of the charging time of the ignition coil from the closing start angle, the time required for the angular rest is related to the end of the charging time of the ignition coil.

Such methods and control devices are used in modern motor vehicles in series and are thus known per se. Examples of signals are injection pulses, with which fuel injection valves are controlled, and signals that trigger a spark ignition of a fuel / air mixture in a combustion chamber of the internal combustion engine by a spark.

In this case, a functional level is understood to mean a partial program and an associated computer structure which exercises a higher-level function in the controller. Examples of superordinate functions are the evaluation of input signals and the transfer to other subprograms as well as the coordination of subprograms.

A driver level is understood to mean a subprogram and an associated computer structure which is assigned directly to an actuator, for example an injection valve, an injection valve arrangement, an air mass actuator (eg a throttle valve) or a spark plug.

In an internal combustion engine operating according to the four-stroke principle, a duty cycle related to a specific combustion chamber of the internal combustion engine extends, for example, over two revolutions of a crankshaft or one revolution of a camshaft.

The angular position of one or both of these waves is usually determined by inductive scanning ferromagnetic markers, which are arranged on a rotatably connected to one of these shafts encoder wheel. For example, conventional donor wheels have about sixty ferromagnetic markers in the form of teeth. The resulting 6-degree subdivision is insufficient for accurate control of injection and ignition. For this reason, signals for controlling the injection and ignition are set to specific angular positions defined by specific teeth.

It is understood by a touchdown that when passing through the defined angular position, a timer is activated, which triggers the signal output at a desired time interval for passing through the defined angular position. In other words: At the time of touchdown is switched from an angle measurement to a time measurement.

At a constant angular velocity of the crankshaft, the desired time interval is equivalent to a desired angular distance. However, if the speed of the crankshaft fluctuates greatly, the signal triggering may deviate from the desired angle value. These deviations were tolerable in combustion engines with external mixture formation by carburetor or port injection, because these internal combustion engines are characterized by a certain insensitivity and robustness to fluctuations in the timing of injections and ignition.

With increasing demands on the exhaust quality and the fuel consumption of internal combustion engines but also internal combustion engines with direct injection of fuel into combustion chambers have established, in which more sensitive combustion processes can be realized. In particular, in spark-ignition internal combustion engines with direct injection, a plurality of combustion processes can be distinguished from one another, wherein the classification is based on the mechanisms that dominate the mixture formation. Thus, the methods are classified as radiant, wall or air-guided, depending on whether predominantly the beam dynamics itself, the beam deflection on a combustion chamber wall or the Charge movement is used for targeted charge stratification.

The direct injection systems of the first generation introduced today in series work with wall and air-guided combustion processes. Compared to conventional intake manifold injection, these systems allow fuel savings of a few percent.

By introducing spray-guided combustion processes in a second-generation gasoline direct injection, further consumption advantages can be achieved, since spray-guided direct injection is the most thermodynamically favorable.

In order to realize this savings potential, long-term stable and reliable ignition and combustion must be guaranteed for all fuel qualities available on the market. It is problematic that have shown high misfire rates in spray-guided combustion, which are due to insufficient accuracy of the coordinated control of the output of injection pulses and ignition signals.

A more finely tuned to the angle signal reproducible output of control signals but also for other internal combustion engines of interest, for example, to meet further increasing legal requirements for the exhaust gas quality in the future.

Against this background, the object of the invention is to provide a method and a control device for reproducing finely tuned output of control signals for an internal combustion engine that is tailored to the angle signal. In this case, a temporally reproducibly finer resolution is understood to mean a smaller fluctuation range of deviations from actual angular positions to nominal angular positions when statistically considered, in which signal outputs are to take place or take place. In other words, the invention aims at reducing errors in the angular positioning of signals to be output, such errors occurring as a result of a speed change (angular velocity change) after switching from the angle measurement to the time measurement.

This object is achieved by a method having the features of claim 1 and a control device having the features of claim 11.

In the known methods and control devices, the driver level can process either only a start angle or an end angle, which is determined as a time interval to a discreetly detected angle value. If the reference angle deviates from the starting angle, a suitable starting angle must be calculated starting from the reference angle. Deviations of the starting angle from a reference angle can occur, for example, when a well-defined end of an injection is more important than a precisely defined start. This situation occurs, for example, in a spray-guided combustion process, in which the end of an injection must be very accurately related to an ignition timing to enable a reliable ignition of the charge of fuel vapor and air passing by an injection jet accelerated at a spark plug.

In the known methods and control devices, a starting angle was calculated for deviations between the reference angle and the starting angle from the ignition point as reference angle and the injection duration on the functional level. The calculation on the functional level has the disadvantage that there are only rotational speed values which have been determined in a so-called synchro-grid. The synchro-grid has a comparatively large angular period which, for example, extends over one third of a crankshaft revolution. As a result, only comparatively old speed values and thus angular speed values are available for the conversion of the reference angle into a starting angle. With rapid speed changes over time, this results in an undesirably large spread of the starting angle determined with old values. In the known methods, deviations between desired starting angles and actual starting angles occur with very dynamic speed fluctuations.

By contrast, the processing according to the invention of arbitrary reference angles at the driver level permits determination of time intervals of the signal output, which relates directly to the reference angle, with the effect that a remaining scattering width is smaller by one order of magnitude than in the prior art. In other words, the invention makes it possible to reduce the inaccuracy of the signal output related to a desired angle by an order of magnitude.

As a result, arbitrary reference angles can be used when adapting ECU programs to different combustion engine projects and different combustion methods, without having to accept the aforementioned inaccuracies.

It is preferred that in the case of a plurality of successive discrete values, in each case at least one time interval of a signal output is determined and that the output of the signal takes place after the time interval has elapsed, when the signal is output time interval or its angular equivalent corresponds to a predetermined distance.

As a result of the redetermination of the time interval to a signal output, which may be repeated repeatedly in the time grid of the periods of the comparatively coarse angle signal, and the comparison of the distance with the predetermined distance, a maximum of actuality can be achieved in the determination of the distance. As a result, for example, a respectively last possible calculation event, which is triggered by a falling or rising edge of the described angle sensor system, can be systematically used to determine the distance. The distances between setting up a signal output and the signal output are thereby significantly shorter compared to the prior art. This contributes to the aforementioned improvement in the timing accuracy of the output of control signals by an order of magnitude improved.

This effect is particularly important in internal combustion engines with jet-guided combustion process in which a charge of ignitable mixture formed from the injection jet due to the induced by the injection charge movement is present only a short time in sufficient proximity and sufficiently slow speed at the spark plug.

It is also preferred that the reference angle serves as the basis for the output of an injection train consisting of a plurality of individual injection pulse widths.

In this case, the reference angle can mark any position of an injection train, which consists of several individual injections. Thus, that point of the injection train, whose angular position for the combustion process is to be kept particularly accurate, are freely defined and therefore also in the middle of a single injection pulse width or outside a single injection pulse width or outside of the injection train.

It is further preferred that an angular distance between the discretely detected angle value and the start of a signal to an injection request must meet a predetermined condition.

By this configuration, the relative position of individual injections of an injection train can be controlled each other very accurately and with little effort.

Furthermore, it is preferred that the predetermined condition is satisfied when a time interval between the time of detection of the discrete angle value and a triggering of the output of the signal falls below a predetermined threshold value.

The threshold is preferably predetermined to be greater than (or equal to) the period of time to detect the subsequent discrete angle value and to include the time required for any calculations and hence delays so as to ensure that the triggering of the output of the Signals for all possible value combinations are made on time.

Alternatively, the following sequence of steps is preferable: determining an angle difference between a discrete angle value and a reference angle, determining a difference of angle difference and predetermined maximum angle distance, converting the difference into a time equivalent, comparing a sum of a pre-storage time and a delay time with the time equivalent, and setting the start time if the said sum is greater than the time equivalent.

By these embodiments, a smallest possible distance between the last determination of the distance (each triggered by successive tooth flanks) and the start of the signal for injection request is achieved, resulting in the highest possible up-to-date calculation of the distance between a calculation event on a tooth edge and a triggering of a control signal leads. A calculation event is an event that triggers the calculation. This can be, for example, a rising or falling edge in the signal of the angle sensor. The threshold is preferably set such that an already calculated distance is discarded if the distance determination can still be based on a more recent angle and a more recent angular velocity determined from a subsequent discrete value.

By means of this configuration, the most recent distance value can be identified in each case in a simple manner, so that the determination of the distance takes place with a timeliness which can be controlled by the dimensioning of a time span or an angular span.

A further preferred embodiment provides that the reference angle corresponds to an ignition point. However, the reference angle does not have to correspond to the ignition timing. It is essential that the reference angle is freely selectable.

By means of these measures which can be used both individually and in combination, the reproducible control of burns of combustion chamber fillings can be substantially improved. As a result, the potential of spray-guided combustion can be better exploited. Also at other combustion processes result in better combustion quality, which has a positive effect on a targeted reduction of pollutant emissions.

It is also preferable that the discrete values are detected by scanning discrete markings on a donor wheel rotatably connected to a shaft of the internal combustion engine.

Furthermore, it is preferred that rising and / or falling edges in a signal which depicts the rotation of the markings are assigned to an actual angular position of the shaft and thus to a position within the work cycle as discrete values.

These embodiments are already widely used without the signal processing according to the invention. It is a particular advantage of the invention that the mentioned improvements can be achieved without a costly refinement of the angle sensor, so that inexpensive existing systems for angle detection can be used.

A further preferred embodiment provides that the predetermined condition is satisfied if, in the current assessment, no signal output is active.

In this embodiment, the injection signal is preferably applied to a plurality of consecutive discrete angle values. If the signal is not yet active when acquiring a discrete angle value, the touchdown at the previous angle value is discarded and reset after a current recalculation. The term of touchdown here also means to start a timer (hardware timer) with the time interval between the current position and the beginning of a signal to be output. The signal is only actively output when the timer has counted down from its start value to a final value, for example zero. As long as the final value has not yet been reached, the process of starting up (from the past) can be discarded by deactivating the Timers timer via software.

Here, too, the calculation of the signal output used for the signal output, for example of an injection start, takes place as late as possible and thus as up-to-date as possible before the start of a simple injection or multiple injection as injection train consisting of a plurality of individual injections. An essential difference to the prior art, as in the configuration with the threshold for the distance, is that the time interval between a reference angle for the signal output and the signal output directly at each detection of a discrete value, so for example at each tooth flank enters into the calculation. In the prior art, however, in a first step, the time interval is converted on the basis of a less current angular velocity into an angle having tolerances as a result.

It is also preferable that, in determining the time intervals, a desired angular distance is determined from an angular velocity or time period determined from a duration of a period or sub-period of a signal representative of the rotation of the markers.

With regard to embodiments of the control unit, it is preferred that the control unit carries out at least one of the above-described embodiments of methods.

Further advantages will become apparent from the description and the accompanying figures.

drawings

Embodiments of the invention are illustrated in the drawings and are explained in more detail in the following description. Show it:

1 schematically an internal combustion engine in partial section;

2 a partial structure of a control unit;

3 an assignment of signals to be output to signals of a rotational speed sensor of the internal combustion engine;

4 an assignment as in 3 with the difference that the signal to be output defines an injection train;

5 a flowchart as a first embodiment of a method according to the invention; and

6 a further flowchart as a second embodiment of a method according to the invention.

Description of the embodiments

In the 1 is an internal combustion engine 10 with actuators and sensors in a simplified form shown in partial section. The internal combustion engine 10 has at least one combustion chamber 12 on top of a piston 14 is sealed movable. The combustion chamber 12 is via an inlet valve 16 with air from an intake pipe 18 filled. To fill the combustion chamber 12 with air is via an injection valve 20 metered a suitable amount of fuel. A spark plug 22 ignites the resulting mixture of fuel and air. After burning the filling of the combustion chamber 12 the burnt residual gases are via an outlet valve 24 from the combustion chamber 12 ejected and then the combustion chamber 12 again with air over the inlet valve 16 filled. The control of the injection valve 20 and the control of the spark plug 22 be from a controller 26 of the internal combustion engine 10 controlled.

The sensors of the internal combustion engine 10 includes inter alia a camshaft sensor 28 and / or crankshaft sensor technology 30 , The camshaft sensors 28 can be an inductive pickup 32 comprising the ferromagnetic markers 34 on a non-rotatably coupled to the camshaft sender wheel 36 scans. Similarly, the crankshaft sensor 30 an inductive pickup 38 comprising the ferromagnetic markers 40 on a crankshaft sensor wheel 42 scans. In this way, the respective angular positions of the camshaft and / or the crankshaft of the internal combustion engine 10 be recorded. These quantities characterize the position of the piston 14 in the internal combustion engine 10 and thus ultimately enable one to the respective piston position and the respective cycle of the working cycle of the internal combustion engine 10 coordinated control of the injection valve 20 and the spark plug 22 ,

The camshaft sensors 28 and the crankshaft sensor system 30 provide both together and individually examples of an angle sensor for periodically detecting discrete values of angular positions of a shaft of the internal combustion engine 10 It is understood that the position of the piston 14 and thus the angular position of a shaft of the internal combustion engine 10 can also be detected with other sensors, such as magnetic or optical angle sensors on the camshaft and / or crankshaft.

As additional sensors, the internal combustion engine 10 an air mass meter 44 on top of that via the suction tube 18 to the internal combustion engine 10 flowing air mass detected. Next, the sensors of the engine 10 at least one more driver's wish 46 on which the driver's torque request to the controller 26 transmitted.

It is understood that modern internal combustion engines 10 can be provided with a variety of other sensors that can detect, for example, cooling water temperatures, oil temperatures, intake air temperatures, exhaust gas temperatures, pressures in the intake, in the combustion chamber or in the exhaust gas area. The in connection with the 1 said sensors 38 . 32 . 44 and 46 should therefore not represent a conclusive enumeration.

From the signals of said sensors and possibly from signals of other sensors forms the control unit 26 Control signals for actuators, the behavior of the internal combustion engine 10 determine. The behavior of the internal combustion engine 10 For example, it is significantly characterized by its torque output from the control unit 26 via a change in the combustion chamber filling with air and / or a change in the ignition timing and / or a change in the via the injection valve 20 metered amount of fuel can be done. It can be used for a duty cycle of a piston 14 metered amount of fuel to be metered by a single injection or by an injection train, which consists of several successive partial injections.

In the 1 is a throttle valve actuator as an actuator for changing the combustion chamber filling with air 48 shown, the angular position of a throttle 50 in the intake pipe 18 adjusted. Alternatively or in addition to such a filling actuator, the filling of combustion chambers 12 of the internal combustion engine 10 For example, by intervening on a fully variable valve control can be realized, as it is already in series in some motor vehicles today.

Such a quantity control of the combustion chamber charge usually takes place in the homogeneous operation of an internal combustion engine 10 with direct injection instead. In shift operation, the internal combustion engine 10 however, usually unthrottled, so for example with largely open throttle 50 operated. The torque is determined in this mode on the metered amount of fuel, which, so to speak, not the quantity, but the quality of the filling of combustion chambers 12 is changed. Alternatively or additionally, the torque is adjusted via a variation of the ignition timing.

2 illustrates essential basic functions of the control unit 26 in which input signals from the said sensors 32 . 38 . 44 and 46 be provided to drive signals for said actuators 48 . 20 and 22 are processed. For processing, the controller indicates 26 first a functional level 52 on. Where is the functional level 52 not as a physical component of the control unit 26 but as part of the engine control program that is in the control unit 26 is processed using data stored there, to understand. In the functional level 52 For example, torque requirements are coordinated with options for setting the required torque. torques can from the driver via the driver request generator 46 or functions of the control unit 26 or from other control devices (eg from an ESP control device). So can the controller 26 For example, torque requirements for implementing an idle speed control, for controlling switching operations in a downstream transmission, to maintain a desired speed of the motor vehicle (cruise control) or to stabilize the vehicle in a driving situation (electronic stability program ESP, traction) set.

The torque requirements can be met by various Steep interventions, the combustion in the at least one combustion chamber 12 of the internal combustion engine 10 influence. In this case, further requirements must be met, for example, requirements for the exhaust gas quality. To realize the torque requirements, the controller 26 including an air system 54 , a fuel system 56 and an ignition system 58 on. Again these are systems 54 . 56 and 58 not as physical components of the control unit 26 but rather as program structures to be understood with hardware structures of the control unit 10 interact. Thus, the fuel system, which also includes the injection driver, has one (or more) dedicated event-controlling timers and computational capacity to coordinate injections in the piston's duty cycle 14 , The timer is dependent on the crystal frequency of the computer and controls, for example, events such as injections during a working cycle of the piston 14 , About the air system 54 For example, the filling of the combustion chamber is set. The air system 54 therefore controls the filling actuator 48 at. About the fuel system 56 become injection valves 20 controlled and via the ignition system 58 the activation of the spark plug takes place 22 ,

Torque changes that must be made very quickly, for example, in the context of idle control, are preferred via the ignition system 58 adjusted while longer-acting torque requirements over the air system 54 and / or the fuel system 56 be set. The systems 54 . 56 and 58 do not work independently but have to work in a coordinated way. Thus, for example, the transition of the internal combustion engine from a homogeneous operation in a shift operation of a quantity control in a quality control switches, for example, the throttle 50 fully open and the fuel system 56 Injection pulse widths outputs their position relative to positions of the piston 14 is changed compared to the homogeneous operation.

3 shows a control of the output of an injection signal, in which as input parameters, inter alia, a tooth flank event 67 , a period 69 or a partial period of the tooth signal 66 , a desired starting angle 61 , an injection pulse width 62 , a delay time 70 and a maximum angle distance are processed as a predetermined condition. In this embodiment corresponds to the Sollbeginnwinkel 61 a reference angle. As an actual angular position, for example, an interrupt or a calculation trigger comes into question, for example by a rising edge 64 in the signal 66 a speed sensor 30 is produced. In the known method, the start of the injection is at the rising edge 64 placed. The term touchdown here means that a timer has a starting value Ta 68 is set from which is counted down. For the calculation of the starting value Ta 68 First, the Sollbeginnwinkelabstand 60 from the angular velocity (angle / time) or the angular time duration (time / angle) of the crankshaft converted into a time interval, of which a delay time TI 70 is subtracted, for example, a signal delay in a power amplifier module of the control unit 26 considered. The angular distance caused by the starting value 68 is compared with the maximum angle distance. If the angular distance 68 is smaller than the maximum angle distance, the predetermined condition is considered satisfied and the injection is started. The angular velocity or the angular time duration becomes, for example, in the driver of the fuel system 56 for the control of the injection valves 20 formed from current values for the angular velocity or the angular time duration.

Current values for the angular time duration or the angular velocity can be, for example, the time interval between two rising edges (or two falling edges or one rising and one falling edge) of the signal 66 and the associated angular distance, by the geometry of the encoder wheel 42 or 36 is predetermined, determined. Preference is given to two immediately consecutive flanks 65 . 67 (in the example of rising flanks) uses the shortest possible distance to determine the starting value 68 have and a period duration 69 the signal 66 to ensure the greatest possible up-to-dateness.

After the time Ta 68 that is, after counting up or down the start time Ta 68 through the timer, the start of the output of the injection pulse width 62 including the delay time 70 triggered and after expiry of the delay time TI 70 opens the injector 20 and it gets fuel into the combustion chamber 12 injected. As an alternative to counting up or down the time TA 68 you can also use a hardware-triggered "match" for which the current absolute time with the absolute time programmed in the timer (Ta 68 ) is compared.

Known drivers for gasoline direct injection become one at the functional level 52 used in a synchro grid certain angular velocity. The synchro-grid is defined by larger angular spreads, typically of the order of one-third of a crankshaft revolution. As a result, deviations in the actual angle of a signal output from the associated setpoint angle occurred during rapid changes in speed.

Due to the small distance Ta plus TI between a touchdown and a start of the injection, however, a more recent, at the driver level of the fuel system angular velocity formation (or angle duration formation) occurs even with the occurrence of large speed gradients only small deviations of the actual injection start angle of the target injection start angle.

The same applies to deviations of the actual injection end angle from the target injection end angle. The injection duration or injection pulse width 62 may generally not be lengthened or shortened in the injection output, for instance to achieve a smaller angular deviation, because otherwise the fuel quantity decisive for the torque in stratified operation would be changed.

For new combustion processes, such. As the spray-guided combustion method injection trains are generally used from a plurality of separate partial injections. It is of great importance that the reference angle of the injection train is output as accurately as possible. Since the previously known methods only process the starting angle according to the method presented above, the conversion from the reference angle to the starting angle on the functional level 52 made, creating large tolerances between the reference reference angle and the actual reference angle.

The tolerance is essentially caused by the fact that the engine speed calculated in the synchro grid is used in the conversion from the desired reference angle to the injection start angle. Since between the time of determining the engine speed and the time of reaching the reference angle, a large period of time, in which the engine speed resp. can change the angular velocity greatly, this conversion is subject to large tolerances.

In the 4 These relationships are exemplified by the output of an injection train 72 from three single injections 74 . 76 and 78 explained.

Between the reference angle span 80 and the starting angle distance 82 to a rising edge 64 the signal 66 is the pre-storage time Tv 73 with Tv = Ti_1 + T_p + Ti_2. Here, Ti_1 is the injection pulse width of the single injection 74 , Ti_2 the injection pulse width of the single injection 76 and T_p a predetermined time interval 84 between said individual injections 74 and 76 , The reference number 85 analogously denotes a distance between the following single injection pulse widths 76 and 78 ,

When issuing injection trains 72 becomes a reference angle for each injection train 79 as angular distance 80 for detecting a discrete angle value, for example a rising edge 64 , certainly. The reference angle 79 may have any position to the injection train and before, lie within or after the injection. In other words, the time Tv 73 between the beginning 61 the first injection and reaching the reference angle 79 becomes a freely selectable parameter that can be optimally adapted to the requirements of new combustion processes.

After converting the angular distance 80 At a time interval, all other distances of the injection train 72 as time intervals to the time equivalent with expiration of the angular distance 80 certainly. This allows, after the output of the reference angle 79 already by using a current angular velocity (or angular duration) with reduced tolerances, all other significant points in time of an injection train are issued with increased accuracy.

In other words, the injection driver of the fuel system 56 Processes the parameters reference angle and pre-storage time, so that a conversion of a reference angle to a start angle, which takes place on the function level, is eliminated. The reference angle 79 replaces the starting angle. The storage time Tv 73 is an additional parameter. The reference angle 79 is freely definable on the injection train and can generally also be outside, respectively before the first or after the last injection. The start trigger for the start of the injection train 72 is placed in the smallest possible distance before the injection train. Thus, the calculation of Ta associated with touch-up is based on up-to-date angular velocity information.

The following are with reference to the 5 and 6 two embodiments of inventive method described in terms of the way of putting the first injection 74 differ.

In the first embodiment, in the 5 is shown, for example, on each tooth flank (eg. 64 ) calculates the time until the beginning of the first injection, if the signal output has not already started. step 100 corresponds to the detection of a triggering event for performing the calculation at an angle base value, for example at a tooth edge 64 , The placement takes place in this embodiment exactly once. Therefore, in step 102 checks whether the injection has already been started. If this is the case, then step 103 branched back to a main program. If an injection has not yet been set up, it will be in step 104 the angular distance of the tooth flank to the reference angle (target angle) determined. In step 106 the angular distance is converted to a time interval and in the step 108 the value becomes Ta 68 certainly. Ta 68 can be calculated as the difference between the time interval and the pre-storage time 73 the signal to be output and the delay time 70 be determined. In step 110 is checked if Ta 68 within a defined angle range after the current tooth flank 64 lies. If not, the calculation is done by branching into the step 112 canceled, which leads to the main program. For example, in step 110 Ta 68 compared with a threshold SW having a predetermined maximum tooth flank distance 64 to the starting angle after expiry of Ta 68 defined as a given condition. To assess the current angular velocity is used. The defined angle range is converted into a time and with Ta 68 compared. For the touchdown is the in step 104 determined angular distance between the tooth flank and the reference angle in the step 106 recalculated into a time equivalent using recalculated angular velocity (or angular time duration), and then in step 108 the time Ta 68 calculated. The threshold is preferably set such that an already calculated distance is discarded if the distance determination can still be based on a more recent angle and a more recent angular velocity determined from a subsequent discrete value.

If the maximum distance is greater than Ta 68 is in step 114 the start of the signal output is set. The signal output is made by starting a timer in the injection driver. For example, the timer runs from a start value Ta to a final value (eg, zero), thereby defining a predetermined period of time Ta 68 associated with the detection of one of the discrete values (for example, a tooth flank 64 ) begins.

In step 116 a flag is set that represents an already mounted start. In step 118 it is checked if the time Ta has expired. Only if this is the case takes place in the step 120 an output of an injection request signal. This is done in step 118 a count t of the timer with the time Ta 68 compared. As long as the end of Ta is not reached, there is no transition to the step 120 , When exceeding the time Ta 68 on the other hand takes place in the step 120 a triggering of the signal output. If the time Ta is less than or equal to zero, the signal output is triggered immediately. This may be the case if the injection data is available too late. The specified condition is fulfilled.

By means of this embodiment, the most recent distance value can be identified in each case in a simple manner, so that the determination of the distance takes place with an up-to-dateness controllable by the dimensioning of the maximum angle span. As a result, for example, an injection is started when a current tooth flank 64 is within a predefined angular distance to the start of injection. If this is not the case, the touchdown is eliminated. By determining the angular distance, an exact assessment basis for the calculation of the time equivalent is initially provided, which is used as a finely resolvable substitute variable for the angular distance. The consideration of a current angular velocity (or angular time duration) provides an improved correspondence between time equivalent and angular distance.

It is particularly important in spray-guided combustion processes that ignition be coordinated with the end of an injection. Therefore, this embodiment, which effectively uses the end of an injection as a reference angle in one example, is advantageous in connection with a spray-guided combustion process. A further preferred embodiment provides that the difference is formed from the first time interval and a sum from the duration of the signal train to be output including the pause times and a further period of time, wherein the further time period at least one delay time TI 70 a signal output represented by a power amplifier module, which is the interface between the injection system 56 and one or more injectors 20 forms. This embodiment allows the consideration of physically caused delays in the signal triggering, as they can occur for example by signal propagation times or charging time constants of capacitors in power amplifiers of the control unit.

In a second embodiment is in step 109 checked whether an injection train signal is already active on a tooth flank, that is, whether the start of the signal for injection request has already been triggered. If this is not the case, the time until the beginning of the first injection minus the delay time is calculated and the injection is started. In the other case, in step 111 branched back to a main program. At each tooth flank where the injection is not active, the calculation on the previous tooth flank is discarded and the start of injection is restarted. This embodiment is described by the flowchart of 6 shown. 6 differs from 5 mainly by the fact that for the assessment of the composition of the injection is not the marker for an already done touchdown 116 in step 102 and checking Ta for the predefined area in the step 110 is used, but the test of the step out 120 resulting pin state of the injection request signal in step 109 is used, and that the setting of a flag is omitted (no step 116 ).

Further alternatively to the embodiments according to the 5 or 6 can also be done as follows:
Calculating the angular difference between a tooth flank and the reference angle,
Determining a difference between angular difference and predetermined maximum angular distance,
Converting the difference into a time equivalent,
Compare the sum of pre-storage time 73 and delay time 70 with the time equivalent, and
Setting the starting time when the said sum is greater than the time equivalent.

In both embodiments according to the 5 and 6 For example, the calculation of the start of injection actually used later for the signal output thus takes place as late as possible before the beginning of the injection train. A significant advantage over the prior art is that the time interval between the reference angle and start of injection is received directly on each tooth flank in the calculation, whereas in previously known injection drivers for gasoline direct injection in a first step, the time interval based on an "old" angular velocity is converted into a tolerance-affected angle.

The theoretical optimum with respect to the tolerance of the reference angle in compliance with the time Tv would be reached if the injection would be set up exactly at the time of its beginning with the current angular velocity (time t = 0 at the beginning of the injection) or tooth time Tz. As already described, in the application of the invention under real conditions tolerances are smaller by an order of magnitude than in known injection drivers for gasoline direct injection, in which the drivers process only the initial angle. In the known injection systems, the larger tolerance is essentially caused by the fact that the engine speed calculated in the synchro-grid is used in the conversion from the desired reference angle to the injection start angle. Since between the time of determining the engine speed and the reference angle then a large period of time, in which the engine speed resp. can change the angular velocity greatly, this conversion is subject to large tolerances.

By redetermining the time interval to a signal output, which may be repeated repeatedly in the time grid of the calculation times of the comparatively coarse angle signal, and comparing the distance with the predetermined condition (or the maximum angle distance for touchdown), a maximum of actuality can be achieved in the determination of the distance. As a result, for example, a respectively last possible calculation event, which is triggered by a falling or rising edge of the described angle sensor system, can be systematically used to determine the distance.

Claims (10)

Method for controlling the output of a signal ( 70 . 62 ; 72 ) at a start in the working cycle of an internal combustion engine ( 10 ) with direct fuel injection, with a control unit program with a functional level ( 52 ) and one the signal ( 70 . 62 ; 72 ) issuing driver level ( 56 ), wherein the duty cycle over a fixed angular range of a shaft of the internal combustion engine ( 10 ) and periodically discrete angle values ( 64 ) of the angular position of the shaft are detected, and wherein in each case an angular distance ( 80 ) between a detected discrete angle value ( 64 ) and a reference angle ( 79 ) and at the driver level ( 56 ) a time interval ( 68 ) the output of the signal ( 70 . 62 ; 72 ) to at least one periodically detected discrete angle value ( 64 ) is determined as the start value for a timer which, when a final value has been reached, receives an output of the signal ( 70 . 62 ; 72 ) and in the case of deviations of the reference angle ( 79 ) from the start angle the time interval ( 68 ) at the driver level ( 56 ) to the reference angle ( 79 ) and wherein at a plurality of consecutive discrete angle values ( 64 ) at least the time interval ( 68 ) of the output of the signal is determined and the output of the signal ( 70 . 62 ; 72 ) after expiration of the time interval ( 68 ) takes place when the time interval ( 68 ) corresponds to a predetermined distance. Method according to claim 1, characterized in that the reference angle ( 79 ) as the basis for the output of one of several Single injection pulse widths ( 74 . 76 . 78 ) existing injection train ( 72 ) serves. A method according to claim 2, characterized in that an angular distance between the discreetly detected angle value and the start of the output of a signal to an injection request must meet a predetermined condition. A method according to claim 3, characterized in that the predetermined condition is fulfilled when an angular distance ( 68 ) between the discreetly detected angle value ( 64 ) and the start of the output of a signal falls below a predetermined threshold value. Method according to claim 4, characterized by the following steps; Determining an angular difference between a discrete angle value ( 64 ) and the reference angle ( 79 ), - determining a difference between angular difference and predetermined maximum angular distance, - converting the difference into a time equivalent, - comparing a sum of storage time 73 and delay time 70 with the time equivalent, and - setting the start time, when said sum is greater than the time equivalent. Method according to one of the preceding claims, characterized in that the reference angle ( 79 ) is freely selectable. Method according to at least one of the preceding claims, characterized in that the discrete angle values ( 64 ) by scanning discrete markings ( 34 ; 40 ) on a non-rotatably connected to a shaft of the internal combustion engine encoder wheel ( 36 ; 42 ). Method according to claim 7, characterized in that rising and / or falling edges ( 64 ) in a signal ( 66 ), the rotation of the markings ( 34 ; 40 ), an actual angular position of the shaft and thus a position within the duty cycle are assigned as discrete angle values. Method according to one of the preceding claims, characterized in that, in the determination of the time intervals, a desired angular distance is calculated from an angular velocity or angular time period on the basis of a period or partial period of a signal representing the rotation of the markings ( 34 ; 40 ) maps. Control unit ( 26 ) which is adapted to carry out at least one of the methods according to claims 1 to 9.

Images