WO2022090395A1

WO2022090395A1 - Method for determining an opening time of an injector having a solenoid valve, computer program, control apparatus, internal combustion engine and motor vehicle

Info

Publication number: WO2022090395A1
Application number: PCT/EP2021/079986
Authority: WO
Inventors: Philipp Hackmann; Daniel Leineweber; Daniel AUGUST
Original assignee: Volkswagen Aktiengesellschaft
Priority date: 2020-10-30
Filing date: 2021-10-28
Publication date: 2022-05-05
Also published as: DE102020213705A1; CN116324149A; EP4237675A1

Abstract

The invention relates to a method for determining an opening time of an injector having a solenoid valve. The method comprises: determining a time of contact of the armature, at which an armature of a solenoid valve comes into contact with a valve needle of the solenoid valve; determining an opening delay time of the injector, which corresponds to a time span between the time of contact of the armature and an opening time of the solenoid valve; and determining an opening time of the injector based on the time of contact of the armature and the opening delay time.

Description

description

Method for determining an opening time of an injector with a solenoid valve, computer program, control unit, internal combustion engine and motor vehicle

The invention relates to a method for determining an opening time of an injector with a solenoid valve and a computer program, a control device, an internal combustion engine and a motor vehicle.

In internal combustion engines, injectors are used to inject fuel directly into a combustion chamber. An engine control unit controls the switching valve integrated in the injectors, which causes an injection nozzle to open and close again. The quantity of fuel injected can be determined by way of an opening duration of the switching valve.

When solenoid valve injectors are actuated electrically, opening and closing of these valves can only take place with a delay. The distortion of the individual injectors is subject to tolerances, with the result that the injectors have different opening durations with the same activation duration. This results in an undesirable uneven distribution of the fuel mass.

WO 2011/012518 A1 describes a method for operating a solenoid valve of an injector. The solenoid valve has a valve element with a valve needle and an armature that can be moved by means of an electromagnet. The solenoid valve is closed, for example, in the de-energized state. After the activation of the electromagnet begins--after the lift-off delay--the opening movement of the valve element begins, which is limited by a stroke stop. The stroke stop is used to determine the end of movement of the valve element. The lift-off delay can be determined from the start of actuation, the end of movement and the previously determined movement time ("flight time") of the valve element. The lift-off delay corresponds to a period of time between the start of energizing an armature winding (“start of control”) and the lift-off of the valve needle from its seat.

DE 10 2009 045 469 A1 describes a method for operating a fuel injection valve of an internal combustion engine, in which a first delay time is determined, which is a temporal che difference characterized between a point in time of a first change in a control signal for the valve and a point in time of a first change in the operating state of the valve that corresponds to the first change in the control signal. From the first delay time, at least a second delay time of the valve is deduced, which characterizes a time difference between a point in time of a second change in the control signal that differs from the first change and a point in time of a second change that corresponds to the second change in the control signal Change in operating status of the valve.

EP 2 685 074 A1 describes a method for detecting an opening of an electromagnetically actuated fuel injection valve, which is actuated by applying a control signal. A coil voltage of the fuel injection valve is monitored from the time the injection valve closes, and a length of a curve segment with the same sign is determined from the second derivation of the coil voltage. If the length of the curve segment exceeds a calibrated threshold, it is concluded that the injector is open.

The object of the present invention is to provide an improved method for determining an opening time of an injector with a solenoid valve, an improved computer program, an improved control unit, an improved internal combustion engine and an improved motor vehicle.

This object is achieved by the method according to claim 1, the computer program according to claim 12, the control unit according to claim 13, the internal combustion engine according to claim 14 and the motor vehicle according to claim 15.

Further advantageous refinements of the invention result from the dependent claims and the following description of preferred exemplary embodiments of the present invention.

A first aspect of the invention relates to a method for determining an opening time of an injector with a solenoid valve. The procedure includes:

- Determining an armature impact time at which an armature of the solenoid valve strikes a valve needle of the solenoid valve;

- Determining an opening delay time of the injector, which corresponds to a period of time between the armature impact time and an opening time of the valve; and

- Determination of an opening time of the injector based on the armature impact time and the opening delay time. The injector with the solenoid valve, also known as the solenoid valve injector, is used to inject fuel into a combustion chamber of an internal combustion engine. The injector is actuated electromagnetically. For this purpose, the injector has a coil for generating a magnetic field, so that the coil can be used as an electromagnet. When the solenoid valve is in a resting state, in which the coil is not energized and there is therefore no magnetic field, the valve needle is pressed into a valve seat by a pretensioning element, e.g. a spring, and a valve opening is thus closed, as a result of which the solenoid valve is in a closed ( valve) position is pushed or held. To open the solenoid valve, a current can be applied to the coil, which creates the magnetic field. During this opening phase, the magnetic force exceeds a prestressing force of the prestressing element. Thus, during the opening phase, the armature arranged on the valve needle can be moved by the magnetic force in such a way that the armature entrains the valve needle and moves it against the prestressing force. This lifts the valve needle out of the valve seat, releases the valve opening and thus opens the solenoid valve. To close the solenoid valve, the current applied to the coil is switched off so that there is no longer a magnetic field. As a result, the valve needle is pressed back into the valve seat by the pretensioning element and the valve opening is blocked, so that the solenoid valve is again in its closed valve position.

The opening time of the injector is when the valve needle lifts out of the valve seat and exposes the valve opening so that fuel can be injected into the combustion chamber. In other words, the opening time is the time when the valve needle is initially lifted off the valve seat.

The opening time of the injector or the solenoid valve is determined with the method. Consequently, a driving current for generating the magnetic field of the coil is applied to start the opening phase of the solenoid valve.

In the method, the armature impact time is determined at which the armature strikes the valve needle. For explanation, an arrangement of the armature on the valve needle is first described. When the coil is in a currentless state, the armature is in a rest position in which it rests on a rest seat arranged on the valve needle. In an energized state of the coil, the armature lifts off the resting seat due to the magnetic force and moves in the direction of an armature stop on the valve needle. The armature is thus arranged on the valve needle in such a way that it is located between the resting seat and the armature stop. is removable. The armature impact time is therefore the point at which the armature strikes the armature stop of the valve needle (during the opening phase of the solenoid valve). An armature free path can be derived from the armature impact time, which corresponds to a movement distance of the armature from the rest position to impact with the stroke stop of the valve.

In the method, the opening delay time is also determined, which corresponds to the time span between the armature impact time and the opening time of the solenoid valve. During the opening phase, the anchor moves in the direction of the anchor stop and finally strikes there. In order to lift the valve needle out of the valve seat, the pretensioning force must be overcome. This opening delay time describes the time between the impact of the armature on the armature stop and the actual opening of the solenoid valve. The opening delay time depends significantly on the prestressing force of the prestressing element in the injector. The opening delay time also depends on the magnetic force generated by the energized coil. In other words, the opening delay time is dependent on an electromagnetic actuator of the injector, which includes the pretensioning element, the coil, the valve needle with the rest seat and the armature stop, and the armature.

In some embodiments, the influence of the prestressing element can be taken into account with a model, for example. This can be an empirical and/or mathematical model. In other embodiments, the opening delay time can be approximated with a constant time period.

Furthermore, in the method, the opening time of the solenoid valve is determined based on the armature impact time and the opening delay time. In some embodiments, the opening time can be determined by adding the opening delay time to the armature impact time.

The opening time for a solenoid valve injector can be determined comparatively easily and precisely with the method. The method takes into account the design and/or dimensioning of the elements of the solenoid valve injector, such as the electromagnetic actuator, which depends on the inductance of the coil (which results from the number of turns in the coil, the dimensions of the coil and, if applicable, one of the coil results in trapped material), the armature free travel on the valve needle, etc. may depend. In some embodiments, the armature impact time can be determined by evaluating a voltage profile. The voltage curve means the curve of the voltage applied to the coil. The voltage curve can be recorded particularly easily and precisely using appropriate measurement technology. In this way, for example, the raw voltage signal can be recorded. Consequently, the armature impact time can be determined particularly easily and precisely by evaluating the stress profile.

In further specific embodiments, the evaluation of the voltage profile can include an evaluation of a first time derivative of the voltage profile. When the injector opens, a change in the course of the first derivative of the voltage course can be observed due to a change in speed and a moving mass of the armature and/or the valve needle. The armature impact time can thus be determined comparatively precisely and simply by evaluating the first time derivative of the voltage profile.

In other embodiments, the armature impact time can correspond to an extreme value in the first time derivative of the voltage profile. When the injector opens, the voltage curve flattens out when the armature hits the valve needle and then carries it with it. “Extreme value” means that the first time derivative shows a maximum or a minimum at the time of anchor impact. The maximum and the minimum are a maximum and a minimum value. By determining the extreme value, the armature impact time can be calculated and thus easily determined.

In further embodiments, the determination of the anchor impact time can also include:

- determining a start time and an end time of an evaluation period for evaluating the first time derivative of the voltage profile;

- forming a first difference quotient curve over the evaluation period based on the starting time;

- forming a second difference quotient profile over the evaluation period based on the end time; and

- Forming an auxiliary function that includes a quotient from the first difference quotient curve and the second difference quotient curve or a difference between the first difference quotient curve and the second difference quotient curve; and

- Determination of an extreme value in the auxiliary function, the extreme value corresponding to the anchor impact time. With the steps above, I can determine a sudden drop in gradient in the first time derivative of the voltage curve. The point in time of the abrupt drop in gradient corresponds to the point in time of the armature impact. The above steps for determining the armature impact time result from the assumption that the first derivation of the voltage profile can be approximated using two straight lines, namely a first straight line coming from the start time and a second straight line coming from the end time. On the one hand, the incline of the first straight line is constant until the abrupt drop in incline ("kink") is exceeded during the forward pass. "Forward passage" means the consideration of the first straight line starting from the start point in the direction of the end point. On the other hand, the gradient of the second straight line is constant until the abrupt drop in gradient is reached during the reverse pass. "Reverse passage" means the consideration of the second straight line starting from the end point in the direction of the start point. In order to compare the gradients of the first and second straight lines, difference quotients (courses) can be formed and evaluated for the first and second straight lines.

By determining the evaluation period for determining the moment of impact of the armature, periods of time in the voltage curve in which an instant of impact of the armature is implausible can already be sorted out in advance. The method can thus be carried out in a more resource-efficient manner, for example on a control unit.

A difference quotient describes the ratio of the change in a first variable to the change in a second variable, with the first variable being dependent on the second variable. The difference quotient can be used, for example, to determine the slope of a linear function. In order to determine a difference quotient, the ratio between a distance between measured values that are determined at a first point in time and a second point in time and a distance between the first and second point in time can be determined. The times can be chosen arbitrarily. In the present case, the first point in time can be a current point in time within the underlying evaluation period, with the current point in time always being a current journal of a predetermined time-discrete evaluation grid (computing grid). The measured values are determined by the evaluation grid at (measurement) times that are (essentially) equidistant from one another. The second point in time can be constant, e.g. For example, the second point in time can be the start point in time or the end point in time of the evaluation period. The term "difference quotient" explicitly includes both positive and negative values, ie geometrically, the gradient triangle, which is well known for determining a difference quotient, can be oriented in the direction of the x-axis (usually a time axis) or in the opposite direction of the x-axis . "Difference quotient curve" means the curve of the difference quotient over a certain period of time. According to the above embodiments, the first difference quotient curve is formed over the evaluation period and is based on the start time. This means that the difference quotient over the entire evaluation time - space for each (measuring) point in time the difference quotient is determined in relation to the starting time.The same applies to the second difference quotient profile.

For example, the first difference quotient curve and the second difference quotient curve can be determined as follows:

whereby

I ₁ t()=first difference quotient curve l ₂ (t)=second difference quotient curve t=time in the evaluation period _A =start time of the evaluation period t _E =end time of the evaluation period

U'(t) = first derivative of the voltage curve at time t

U'(t _A ) = first derivative of the voltage profile at the beginning of the

evaluation period

U'(t _E )=first derivation of the voltage profile at the end of the evaluation period

Furthermore, an auxiliary function is formed. The auxiliary function is formed in such a way that the first and the second difference quotient profile can be compared directly with one another.

The auxiliary function can thus include the quotient (the ratio) from the first and the second differential quotient profile. In one example, the auxiliary function can correspond to the quotient from the first and the second difference quotient profile. For example, the auxiliary function can be the second difference quotient profile divided by the first difference quotient profile or vice versa. In further examples, the auxiliary function can also include further parameters. If the auxiliary function is formed as a quotient, in some examples the first differential quotient curve and the second differential quotient curve are formed in such a way that the auxiliary function always supplies positive values. Alternatively can the auxiliary function can also be formed in such a way that it includes an absolute value of the quotient. For this purpose, the auxiliary function can, for example, correspond to an absolute value of the quotient of the first differential quotient profile and the second differential quotient profile. The auxiliary function can be formed as follows, for example:

with a(t) = auxiliary function

Alternatively, the auxiliary function can include a difference between the first difference quotient curve and the second difference quotient curve. For example, the auxiliary function can correspond to the difference between the first difference quotient curve and the second difference quotient curve. The auxiliary function can be, for example, the second difference quotient course minus the first difference quotient or vice versa.

The first and the second difference quotient course can be evaluated with the help function. The auxiliary function can be used to identify the point in time at which the distance between the first difference quotient curve and the second difference quotient curve is greatest. At this point in time there is then an extreme value in the auxiliary function. The anchor impact time corresponds to the extreme value in the auxiliary function. In other words, the armature impact time is present when the auxiliary function has the extreme value. Depending on how the auxiliary function is formed, the extreme value can be a maximum or minimum value. Thus, the armature impact time (based on the first time derivative) can be calculated and thus easily determined.

In other embodiments, the auxiliary function can be formed as follows:

With

The symbols used here are already described above. If the auxiliary function is formed in this way, the armature impact time corresponds to a peak in the form of an extreme value.

In some embodiments, the voltage profile can include the voltage profile during a boost phase. The boost phase is a phase during an opening control of the solenoid valve, in which a high voltage, the so-called booster voltage, is applied to the solenoid valve, which can be up to 100 volts, for example. As a result, the current in the coil rises many times more steeply than when battery voltage is applied. The anchor impact time falls in this boost phase. In some embodiments, the voltage profile can be evaluated when the boost phase is present. The method can thus be carried out in a particularly resource-efficient manner, for example, on a control unit.

In further embodiments, the opening delay time can be determined using a model that simulates an operating behavior, in particular an opening behavior and a closing behavior, of the solenoid valve. As mentioned above, the opening behavior depends (among other things) on the electromagnetic actuator of the injector and largely on the pretensioning element of the electromagnetic actuator. Accordingly, the closing behavior also depends on it. There is thus a strong correlation between the opening behavior and the closing behavior. In this way, conclusions can be drawn about the opening behavior of the magnetic valve from the closing behavior of the magnetic valve. Since the closing behavior can be determined comparatively easily using known methods and the opening behavior has a strong correlation to the closing behavior, the opening delay time can be determined particularly easily using the above-mentioned operating behavior model.

In some embodiments, the opening delay time can be determined as a function of a closing delay time of the injector. In other words, the opening delay time can be determined based on the closing delay time. The closing delay time is the time between switching off the current to the coil and closing the solenoid valve. Since there is a strong correlation between the opening behavior and the closing behavior of the solenoid valve, the opening delay time also correlates strongly with the closing delay time.

In some embodiments, the opening delay time can be determined using an opening delay time model. In some embodiments, the above performance model may include the open delay time model. In some embodiments, the above performance model may be the open delay time model. The opening delay time model can, for example, be a characteristic curve or a characteristic map. The opening delay time model is constructed in such a way that the closing delay time is used as the input variable and the opening delay time for the injector is the output variable. In some embodiments, an opening delay time characteristic curve can be used in that the opening delay time is plotted against the closing delay time. The closing delay time is usually particularly easy to determine with the help of measurement technology and appropriate evaluation. Consequently, with knowledge of the closing delay time, the opening delay time can be inferred particularly easily with the aid of the opening delay time model.

In some embodiments, the opening delay time model can be determined as follows. In a first step, the closing delay time of the injector for different activation times is determined for an injector using a large number of tests. A control time is a duration of the energization of the coil. Methods that are already known can be used to determine the closing delay time, such as, for example, the evaluation of a second time derivative of the voltage profile at the coil after the activation current has been switched off. In the next step, an average closing delay time for the injector is formed from the test results. A prestressing force indicator (prestressing element model) for the injector can be derived from the mean closing delay time, the prestressing force indicator corresponding at least to a measure of the prestressing force of the prestressing element. In the next step, the opening delay time of the injector for different activation times is determined with a large number of tests on the test bench and an average opening delay time is formed from this. The average closing delay time and the average opening delay time can be compared with an actual flow rate of the injector. The actual flow depends on the actual closing delay time and the actual opening delay time. By comparing the average values for the opening delay time and closing delay time with the actual flow rate of the injector, the relationship between the average opening delay time and the average closing delay time can be adjusted. This relationship can finally be stored in the opening delay time model. With the opening delay time model, the (mean) opening delay time can be determined as a function of the preload force or the preload force indicator and the (mean) closing delay time.

The procedure for determining the opening delay time model can be carried out for different injectors, so that the model can depict a large number of injectors. As shown above, in some embodiments the opening delay time may depend on the biasing force of the biasing element of the solenoid valve. The biasing member urges the solenoid valve to the closed position.

In some embodiments, the preload force of the preload element can be approximated by a preload element model. The prestressing force can be determined particularly easily, at least approximately, via the prestressing element model. The prestressing element model can be determined mathematically and/or empirically, for example. In some embodiments, as mentioned above, the biasing element model may be derived from the mean closing delay time. The prestressing element model is used to determine the opening delay time. The influence of the prestressing element on the opening delay time can be taken into account in a particularly simple manner by means of the prestressing element model.

In some embodiments, the open delay time may be a predetermined constant amount of time. The predetermined constant period of time can depend on the injector and can be determined by preliminary tests on the test bench. The predetermined constant time period can be stored in the control unit for the internal combustion engine. As a result, the use of the predetermined period of time as the opening delay time is particularly resource-efficient, since no calculations or evaluations (in the control unit) have to be carried out to determine the opening delay time.

A further aspect of the invention relates to a method for determining the opening time of the injector. The method according to the further aspect comprises:

- Evaluation of the time derivation of the voltage profile that is present during the boost phase at the injector, wherein the evaluation of the time derivation includes an evaluation of a second derivation of the voltage profile and the opening time corresponds to an extreme value of the second derivation, which is present as the second extreme value during the boost phase.

The alternative method is suitable for solenoid valve injectors that are designed without an armature and/or an armature freewheel.

It was recognized that during the boost phase for opening the solenoid valve there is a double peak structure in the second derivation of the voltage curve.

In the alternative method, the opening time corresponds to the time at which the second extreme value is present in the second derivation. Thus, the relevant second extreme value is the second peak in the above double peak structure. In other words, the second extreme value corresponds to the second maximum (in terms of time) in the second derivation of the voltage profile during the boost phase.

As mentioned above, the voltage curve on the coil can be recorded using suitable measurement technology. The second derivation can thus be determined based on the voltage profile. With the alternative method, it is possible to calculate the opening time by determining the extreme value and thus determine it precisely.

A second aspect of the invention relates to a computer program which includes instructions which, when the program is executed by a computer, cause the latter to execute one of the methods described above. The computer program can be stored on an electrical storage medium.

A third aspect of the invention relates to a control unit that is set up to carry out one of the methods described above.

A fourth aspect of the invention relates to an internal combustion engine. The internal combustion engine can have the injector described above and can be controlled via the above control unit. The internal combustion engine is set up and designed to carry out one of the methods described above.

A fifth aspect of the invention relates to a motor vehicle with the control device described above. The motor vehicle is set up and designed to carry out one of the methods described above.

Embodiments of the invention will now be described by way of example and with reference to the accompanying drawings. It shows:

1a, 1b schematically show a solenoid valve injector;

2 shows schematic diagrams for an armature lift, for a control current and for a valve lift;

3 schematically shows a method for determining an opening time according to an embodiment;

4 shows a schematic of a method for determining an opening delay time model;

FIG. 5 schematically shows two curves for a first derivation of a voltage curve; FIG. 6 schematically shows a method for determining an armature impact time;

Fig. 7a-c explanatory examples for the method according to Fig. 6;

FIGS. 8a, 8b show an example of a result from the method according to FIG. 6;

9a, 9b show an example of a further result from the method according to FIG. 6; and

10 schematically shows a motor vehicle with a control device according to one embodiment.

FIG. 1a schematically shows a solenoid valve injector (injector) 100 in a closed valve position and FIG. 1b shows the injector 100 in an open valve position. The injector 100 has a solenoid valve that includes a valve needle 5 and a valve seat 15 . The injector 100 has an electromagnetic actuator for actuating the solenoid valve, which includes a coil 1 , an armature 11 and a biasing element 13 .

The solenoid valve is a normally closed valve. This means that when the coil 1 is not energized, the valve needle 5 is arranged on the valve seat 15 in such a way that an injection opening 17 is closed by the valve needle 5 .

In this case, the prestressing element 13 is designed to hold the solenoid valve in the closed position. For this purpose, the prestressing element 13 applies a prestressing force to the valve needle, so that the valve needle is moved in the direction of the valve seat 15 and thus in the closing direction. In the example shown, the biasing element 13 is designed as a spring.

The valve needle 5 has a resting seat 7 and an armature stop 9 for the armature 11, between which the armature 11 can be moved. The rest seat 7 and the armature stop 9 thus define an armature lift or an armature free travel for the armature 11 relative to the valve needle 5. The injector 1 also has a lift stop 3 which limits a lift of the valve needle 5 (valve lift). In the closed valve position, the armature 11 sits on the rest seat 7 and in the open valve position, the armature 11 rests against the armature stop 9 and the stroke stop 3 . The armature 11 can be moved from the rest seat 7 to the armature stop 9 by applying a drive current I to the coil 1 by magnetic force. The armature 11 is held on the armature stop 9 by the magnetic force, so that the armature 11 carries the valve needle 5 along against the prestressing force of the prestressing element 13 and thus lifts the valve needle 5 out of the valve seat 15 until the armature 11 strikes the stroke stop 3 . As a result, the injection opening 17 is uncovered, so that fuel can be injected through the injection opening 17 into a combustion chamber of an internal combustion engine. 2 shows a control current diagram 20 for a time profile of the control current I at the coil 1, an armature lift diagram 30 for a time profile of the armature lift and a valve lift diagram 50 for a time profile of the valve lift. The armature stroke is the stroke of the armature 5 between the resting seat 7 and the armature stop 9. Diagrams 20, 30, 50 show the progression over time very schematically.

The control current diagram 20 shows the application of the control current I at the control time 21 for opening the solenoid valve. In this case, the course over time has a steep edge immediately after activation time 21 , so that activation current I reaches a corresponding value for a pull-in current phase 22 (pick-up current) at pull-in time 23 comparatively quickly. The time between activation time 21 and pull-in time 23 corresponds to a boost phase during an opening phase of the solenoid valve. The steep edge makes it possible to achieve a low tolerance and high reproducibility of the fuel injection quantity. The steep edge is achieved by applying what is known as a booster voltage to the coil operation until the corresponding current value for the inrush current phase 22 is reached. At the holding time 25, the drive current I is reduced from the inrush current to a value for a holding current phase 24 (holding current). The control current I is switched off at the switch-off point in time 27 and then reaches the value zero at the control end point in time 29 .

The armature stroke diagram 30 shows the course of the armature stroke over time. After the drive current I is applied at the drive time 21 , the armature 5 only lifts off the rest seat 7 at the armature lift-off time 31 . There is therefore a lift-off delay 33 between the application of the drive current I and the lift-off of the armature 5 . The armature 5 strikes the armature stop 9 at the armature impact time 35 . The period of time between the anchor lift-off time 31 and the anchor impact time 35 corresponds to a flight time 37 of the armature 5.

After the impact of the armature 11 on the armature stop 9, the armature 11 is held on the armature stop 9 for a period of time 39 due to the magnetic force of the coil 1. The armature stroke relative to the resting seat 7 no longer changes due to the armature stop 9 and is therefore constant for the period of time 39 . After the drive current I has been switched off at the switch-off point in time 27, the armature 11 is no longer held on the armature stop 9 by the magnetic force of the coil 1. Consequently, the armature 11 moves again in the direction of the resting seat 7 at the armature dropout time 41 . At the armature rest point 43, the armature 11 rests against the rest seat 7 again. The valve lift diagram 50 schematically shows the course of the valve lift of the solenoid valve. This means a lift or a deflection of the valve 5 relative to the valve seat 15 . There is an opening delay time 52 between the armature impact time 35 and an opening time 51 of the solenoid valve. This is because the armature 11 hits the armature stop 9 at the armature impact time 35 and carries the valve needle 5 along against the prestressing force of the prestressing element 13 . In this case, the armature 11 must above all overcome the prestressing force of the prestressing element 13 until the valve needle 5 is initially lifted out of the valve seat 15 at the opening time 51 .

At time 53, armature 11 strikes a stroke stop 3, which represents a maximum valve stroke for valve needle 5 and thus the open valve position. After the impact of the armature 11 on the armature stop 9, there is an opening delay time 52, which the valve needle 5 reaches before the maximum valve lift is reached. The solenoid valve is held in this open position for a period of time 55 to inject fuel into the combustion chamber.

It can also be seen from the valve lift diagram 50 that there is a closing delay time 61 between switching off the control current I at the control point in time 27 and reaching the closed valve position at the valve closing point in time 59 . After switching off the control current I, the previously built up magnetic field also decreases continuously. The magnetic field is then weakened in such a way that the pretensioning element 13 moves the valve needle 5 in the direction of the valve seat 15 at the point in time 57 until the solenoid valve is in the closed valve position at the valve closing point in time 59, i.e. the valve needle 5 closes the injection opening 17 .

FIG. 3 shows a method 200 according to an embodiment, with which the opening time 51 of the magnet valve can be determined.

In 210 the armature impact time 35 is determined. According to one embodiment, the armature impact time 35 can be determined by evaluating the voltage applied to the coil 1 . In this case, the voltage or a voltage signal at the coil 1 can be detected by means of suitable measurement technology. According to one specific embodiment, the armature impact point in time 35 is determined by analyzing the first derivative of the time profile of the voltage applied to the coil 1 . For example, the armature impact time 35 can correspond to the time at which an extreme value is present in the first derivation of the voltage profile. The relevant extreme value can be the extreme value that occurs first during the boost phase be the derivative. The analysis of the voltage profile can be carried out at least or only during the boost phase (which occurs between the activation time 21 and the pull-in time 23).

In 220 the opening delay time 52 is determined. An opening movement (opening behavior) and a closing movement (closing behavior) of the valve needle 5 and thus the opening delay time 52 and the closing delay time 61 depend decisively on the pretensioning element 13 . Therefore, based on the closing delay time 61 of the solenoid valve, the opening delay time 52 of the solenoid valve can be determined. An opening delay time model, which is described below with reference to FIG. 4, can be used to determine the opening delay time from the closing delay time.

The closing delay time 61 can be determined by determining the valve closing time 59 and the switch-off time 27 . The switch-off time 27 can be determined by detecting the switch-off time of the control current I. The valve closing time 59 can be determined by known methods. With knowledge of the valve closing point in time 59 and the switch-off point in time 27 , the closing delay time 61 can then be determined, which corresponds to the time span between the switch-off point in time 27 and the valve closing point in time 59 .

In some embodiments, the open delay time 52 determined in 220 may be a predetermined constant time period. This approach makes a previous determination of the closing delay time 61 superfluous, as a result of which the determination of the opening delay time 52 is particularly simplified.

In 230, the opening time 51 is determined based on the armature impact time 35 and the opening delay time 52. According to one embodiment, the opening delay 52 is added to the armature impact time 35 in order to obtain the opening time 51.

Fig. 4 shows a method for determining the opening delay time model 300.

In 310 the closing delay time 61 of the injector 100 is determined for different activation times. For this purpose, tests can be carried out on a test bench. A mean closing delay time for the injector 100 is formed from the plurality of closing delay times 61 for different activation times. In 320 , a prestressing force indicator (prestressing element model) for the injector 100 is derived from the mean closing delay time, the prestressing force indicator corresponding to at least one measure of the prestressing force of the prestressing element 13 .

In 330, the opening delay time 52 of the injector is determined for different activation times on the test bench. A mean opening delay time for the injector 100 is formed from the plurality of opening delay times 52 for different control times.

In 340, the average closing delay time and the average opening delay time are compared with an actual flow of the injector 100 in order to adjust or specify a relationship between the average opening delay time and the average closing delay time.

In 350 the relationship between the mean opening delay time and the mean closing delay time is stored in the opening delay time model. In other words, in 250 the opening delay time model is determined or created. With the opening delay time model, the (mean) opening delay time can be determined as a function of the preload force or the preload force indicator and the (mean) closing delay time.

The method 300 can be carried out for different injectors 100 in order to depict a large number of injectors or closing delay times 61 with the opening delay time model.

FIG. 5 shows a diagram in which the first derivation of the voltage curve 71 according to a first example is shown. The first derivation of the voltage curve 71 is determined based on a raw signal of the voltage (detected by a voltage sensor) and is therefore very noisy. Consequently, a subsequent evaluation of the raw signal and thus the first derivation of the voltage profile 71 determined therefrom can lead to inaccurate results. The raw signal can therefore be smoothed using appropriate methods for subsequent evaluations. For example, using orthogonal polynomials, a smoothed curve 73 of the first derivation of the voltage curve can be determined from the raw signal, which is comparatively less noisy and comparatively smooth. Other methods of smoothing the raw voltage signal are also possible. For a subsequent evaluation, for example in a control device, it can be checked whether the supplied voltage signal, ie the raw signal, can be used or whether the raw signal or the first time derivation has to be smoothed beforehand. FIG. 6 shows a method for determining the armature impact time 400 according to a further embodiment for the block 210 from the method 200 according to FIG.

In 410 the start time t _A and the end time t _E of the evaluation period for evaluating the first time derivation of the voltage profile are determined. As described above, the first time derivative can be used based on the voltage profile according to a raw signal or based on a smoothed first time derivative of the voltage profile.

In 420, a first difference quotient profile I ₁ is formed over the evaluation period based on the start time t _A .

In 430, a second difference quotient profile l ₂ is formed over the evaluation period based on the end time t _E .

In 440, an auxiliary function a(t) is formed, which includes a quotient from the first differential quotient curve and the second differential quotient curve. Alternatively, the auxiliary function a(t) can include a difference between the first difference quotient curve and the second difference quotient curve.

In 450 the auxiliary function is evaluated by determining an extreme value in the auxiliary function a(t). The extreme value then corresponds to the anchor impact time.

7a-c show diagrams which are used to derive the method for determining the time 400 of the anchor impact. As can be seen from FIG. 5 , the first derivation based on the raw signal 71 shows a kink-like drop 75 . The first derivation of the voltage curve 71 can be divided into a region before the kink-like drop 75 (to the left of the kink-like drop 75) and into a region after the kink-like drop 75 (to the right of the kink-like drop 75). The same applies to the smoothed curve 73 for the first derivation of the voltage signal.

In the two areas, the first derivation of the voltage curve 71 can be approximated by linear functions. For example, the following functions can be used for the left area 77 and the right area 79 .

Where f ₁ (t) is the first linear function for the left area, f ₂ (t) is the second linear function for the right area, mi and m2 are the corresponding gradients of the straight lines resulting from the functions, and n2 are the ordinate intercepts ( or displacement constant) of the straight line.

FIG. 7a shows an exemplary approximation function f(t) for an exemplary first derivation of a voltage curve (which differs from the voltage curve from FIG. 5) by the linear functions f ₁ (t), f ₂ (t). The approximation function f(t) is shown for an evaluation period between the (selected) start time t _A and the (selected) end time t _E . The example approximation function f(t) is as follows:

If one now considers a left-hand area between the starting point in time t _A and a bend 81 or the point in time of the bend, the following applies: and

If one considers a right-hand area between the bend 81 and the end time t _E , then the following applies: and

7b shows the above relationships of the gradients mi, m2 for the left and the right area by showing a first difference quotient curve I ₁ and a second difference quotient curve L ₂ , where the following applies:

It can be seen from FIG. 7b that the kink 81 in the approximation function f(t) occurs at the point in time t at which the distance between the first difference quotient curve I ₁ and the second difference quotient curve I ₂ is greatest. This point in time t of the greatest distance between the first difference quotient curve I ₁ and the second difference quotient curve I ₂ can be determined, for example, by forming an auxiliary function a(t) as follows:

Alternatively, the auxiliary function a(t) can also be formed as the difference between the first difference quotient curve I ₁ and the second difference quotient curve I ₂ .

Since the first derivation of the voltage curve and thus the approximation function f(t) decreases continuously, the straight lines of the linear functions f ₁ (t), f ₂ (t) have a negative gradient and thus m ₁ , m ₂ have a negative value . Consequently, a (positive) peak 87 can be seen in the curve of the auxiliary function a(t), which is shown in FIG. In other words, the time of peak 87 corresponds to anchor impact time 35.

A sudden drop in the first derivation of the voltage curve can thus be determined with the auxiliary function a(t). Consequently, a point in time of the buckling drop and thus the point in time of the armature impact can also be determined by calculation and thus easily.

8a and 8b show an exemplary evaluation of a voltage profile according to a second example. A smoothed curve 93 for the first derivation of the voltage profile according to the second example and the corresponding auxiliary function a(t) is shown. It can be seen that the smoothed curve 91 has a sharp drop 93 at time t=250 ps (microseconds) and the auxiliary function a(t) has a peak (maximum) 95 at the same time t=250 ps.

FIG. 9a shows a diagram in which a first derivative over time for a voltage profile according to a third example is shown. It can be seen that a flattening of the course 103 follows a first bend-like drop 101 . The flattening out 103 is in turn followed by a second kinked drop 105. FIG. 9b shows a diagram with an auxiliary function a(t) which is formed for the voltage profile shown in the diagram from FIG. 9a. In this case, the auxiliary function a(t) has a first peak (maximum) 113, which can be assigned to the first sharp drop 101, and a second peak 115, which can be assigned to the second sharp drop 105. The armature impact time 35 corresponds to the first peak 113. In other words, in order to determine the armature impact time 35 using the auxiliary function a(t), in some embodiments only the first peak (maximum) 113 can be determined, which occurs within the evaluation period between Start time t _A and the end time t _E occurs. As an alternative to this, in other embodiments the evaluation period can be selected in such a way that only one peak occurs.

FIG. 10 schematically shows an exemplary control unit 170 that is set up to execute the methods/models described above. Control unit 170 is arranged in a motor vehicle 180, shown schematically, and can control an internal combustion engine 179, shown schematically. The control unit 170 includes a processor 172, a memory (electronic storage medium) 174 and an interface 178. Furthermore, software (a computer program) 176 is also stored in the memory 174, which is designed to execute the methods described above. The processor 172 is configured to execute software 176 program instructions. The interface 178 is also designed to receive and transmit data. For example, it can be an interface to a CAN bus of motor vehicle 180, via which control unit 170 receives signals and transmits control commands.

List of reference symbols 1 coil 3 stroke stop 5 valve needle 7 rest seat 9 armature stop 11 armature 13 prestressing element 15 valve seat 17 injection opening 20 activation current diagram 21 activation time 22 pull-in current phase 24 holding current phase 25 holding time 27 switch-off time 29 activation end time 30 armature lift diagram 31 armature lift-off time 33 lift-off delay 35 armature impact time 37 (anchor) flight time 39 period of maximum armature lift 41 armature release time 43 armature rest time 45 release delay 50 valve lift diagram 51 opening time 52 opening delay time 53 time of the armature impact on the stroke stop 55 time span of maximum valve lift 57 start of movement of the valve needle in the direction of the valve seat 59 valve closing time 61 closing delay time 71 first derivation of a first voltage curve 73 smoothed curve of the first derivation of the first voltage curve

75 kinky waste

81 kink

91 first derivative of a second voltage profile

93 kinky waste

95 peaks(maximum)

101 first kinky descent

103 Flattening of waveforms of first derivatives of voltage waveforms

105 second kinky descent

111 first peak (maximum)

115 second peak (maximum)

170 controller

172 processor

174 memory (electronic storage medium)

176 Software

178 interface

179 internal combustion engine

180 motor vehicle

100 injector

200 methods for determining the opening time

210 Determination of anchor impact time

220 Determination of the opening delay time

230 Determining the opening time

300 Method for determining/creating the opening delay time model

310 Determining the mean closing delay time

320 Determination of the prestressing element model

330 Determination of the mean opening delay time

340 Comparison with the actual flow of the injector

350 Creation of the opening delay time model

400 method for determining anchor impact timing

410 Determining the start time and end time of an evaluation period

420 forming a first difference quotient curve

430 forming a second difference quotient curve

440 Forming an auxiliary function

450 Determination of an extreme value of the auxiliary function a(t) auxiliary function I ₁ First difference quotient curve I ₂ Second difference quotient curve t _A Start time of the evaluation period t _E End time of the evaluation period

Claims

Claims 1. A method for determining an opening time (51) of an injector (100) with a solenoid valve comprising: - determining an armature impact time (35) at which an armature (11) of the solenoid valve strikes a valve needle (5) of the solenoid valve; - Determining an opening delay time (52) of the injector (100), which corresponds to a period of time between the armature impact time (35) and an opening time (51) of the solenoid valve; and - determining an opening time (51) of the injector based on the armature impact time (35) and the opening delay time (52).

2. The method according to claim 1, wherein the armature impact time (35) is determined by evaluating a voltage profile.

3. The method according to claim 2, wherein the evaluation of the voltage curve comprises an evaluation of a first time derivative of the voltage curve.

4. The method according to claim 3, wherein the armature impact time (35) corresponds to an extreme value in the first time derivative of the voltage curve.

5. The method according to claim 3, wherein the determination of the armature impact time further comprises: - determining a start time and an end time of an evaluation period for evaluating the first time derivative of the voltage profile; - forming a first difference quotient curve over the evaluation period based on the starting time; - forming a second difference quotient profile over the evaluation period based on the end time; and forming an auxiliary function which includes a quotient from the first difference quotient curve and the second difference quotient curve or a difference between the first difference quotient curve and the second difference quotient curve; and - determining an extreme value in the auxiliary function, the extreme value corresponding to the armature impact time.

6. The method according to claim 5, wherein the auxiliary function (a(t)) is formed as follows:

With

where t = time in the evaluation period t _A = start time of the evaluation period t _E = end time of the evaluation period

U'(t) = first derivative at time t

U'(t _A ) = first derivative at the beginning of the evaluation period

U'(t _E ) = first derivative at the end of the evaluation period

7. The method according to any one of the preceding claims, wherein the voltage curve includes the curve of the voltage during a boost phase.

8. The method according to any one of the preceding claims, wherein the opening delay time (52) is determined with a model that simulates an operating behavior, in particular an opening behavior and a closing behavior, of the solenoid valve.

9. The method according to any one of the preceding claims, wherein the opening delay time (52) as a function of a closing delay time (61) of the injector (100) is determined.

10. The method according to claim 7, wherein the opening delay time (52) is determined with an opening delay time model.

11. The method according to any one of claims 1 to 5, wherein the opening delay time (52) is a predetermined constant time period. Computer program (176), comprising instructions which, when the program is executed by a computer, cause it to carry out a method according to one of the preceding claims. Control unit (170) set up to carry out a method according to one of Claims 1 to 11. Internal combustion engine with a control unit (179) according to claim 13, wherein the internal combustion engine (179) is set up and designed to carry out a method according to one of claims 1 to 11. Motor vehicle (180) with an internal combustion engine (179) according to Claim 14, wherein the motor vehicle (180) is set up and designed to carry out a method according to one of Claims 1 to 11.