DE102016220765A1 - Diagnostic procedure for parallel electromagnetic injectors or valves - Google Patents

Abstract

Method (100) for operating a fuel injection system (2) having a plurality of injectors or valves (21a, 21b, 21c) which are each actuable by associated electromagnets (22a, 22b, 22c), the electromagnets (22a, 22b, 22c) in a parallel circuit (23) are combined and wherein the parallel circuit (23) with a common voltage source (61) is connectable, wherein a diagnostic quantity D, which is a measure of the inductance L _{P of} the parallel circuit (23) is determined (110) , and that from the diagnostic quantity D, the number n of those electromagnets (22a, 22b, 22c) participating in the parallel circuit (23) is determined (120).
Associated control unit (6) and computer program product.

Description

The present invention relates to a method with which a parallel connection of a plurality of electromagnetic injectors or valves for a fuel injection system can be examined for defects.

State of the art

In fuel injection systems for natural gas-powered passenger cars, the fuel is usually metered in via injectors or valves, which can be actuated via associated electromagnets. Corresponding injectors or valves are technically mature and available on the market.

Fuel injection systems for commercial vehicles require a significantly higher mass flow. In order to resort to the well-proven in the car field injectors or valves and at the same time not unnecessarily drive the circuit complexity, several such injectors or valves per engine cylinder are provided and controlled in parallel via a common voltage source.

Essentially two types of errors can occur in such circuits. On the one hand, there may be short circuits that can be detected easily via the standard diagnostic functions of ECUs. On the other hand, there may be errors that exclude individual injectors or valves from the parallel circuit. For example, the electrical contacting of an electromagnet can be interrupted, or the magnetic coil of an electromagnet can be burnt out. Errors of this kind are much more difficult to recognize. This is especially true when the parallel circuit is operated in normal operation in current control. The corresponding controller then keeps the current I flowing through the parallel circuit at its usual level. However, this current I is distributed to fewer electromagnets, which are thereby possibly overloaded and may also fail in the manner of a domino effect. Furthermore, the amount of fuel that the inoperable injector or the inoperative valve should have metered is missing for combustion in the engine. As a result, misfires and errors in the lambda control can occur.

First approaches for the detection, which of several jointly controlled injectors due to a faulty electrical connection is not functional, are from the US Pat. No. 7,168,413 B2 known.

Disclosure of the invention

Within the scope of the invention, a method for operating a fuel injection system with a plurality of injectors or valves has been developed. The electromagnets are combined in a parallel circuit, and the parallel connection can be connected to a common voltage source.

According to the invention, a diagnostic quantity D is determined which is a measure of the inductance L of the parallel circuit. From the diagnostic quantity D, the number n of those electromagnets is determined that participate in the parallel connection.

It has been recognized that in this way a particularly clear signal can be obtained as to how many of the existing electromagnets actually participate in the parallel connection, i. be traversed by current and the associated injector, or the associated valve actuate.

For the total inductance L _{P of} a parallel connection of n inductances L ₁ to L _{n, the} following applies: $$\frac{1}{L_P}={\displaystyle \underset{i=1}{\overset{n}{\Sigma }}\frac{1}{L_i},}$$

If all n inductances L ₁ to L _{n have} the same value L _E , as is generally the case when using a plurality of similar injectors or valves per engine cylinder, then: $$\frac{1}{L_P}=\frac{n}{L_e}\iff L_P=\frac{L_e}{n},$$

Typically, two or three injectors or valves are connected in parallel (n = 3). Failure of an injector or valve significantly changes the total inductance L _{P of} the parallel connection. Conversely, since there is no other cause from which the total inductance L _P should change comparably strongly, it can be concluded from the change in L _P that at least one injector or valve is not functional.

In this case, a particularly large signal swing is achieved if L _{P is} measured directly as a diagnostic variable D. Corresponding dedicated measuring means are, however, not normally provided in control units for fuel injection systems and would have to be retrofitted. However, there are a number of diagnostic variables D which depend on L _P , ie are a measure of L _P , and which can be detected at the same time as the measuring means already present in typical control devices.

In a particularly advantageous embodiment of the invention, the diagnostic variable D is formed from at least one measured value for the current I flowing through the parallel circuit. In this case, the measured value is recorded at a predetermined time after the application of a constant voltage U to the parallel circuit. This measurement can be carried out, for example, in the boost phase, starting from the unactuated state of the injectors or valves, a high constant voltage U is applied to drive the current I through the parallel connection of the electromagnets as quickly as possible to a high level. The main advantage of a measurement in the boost phase is that any current regulation is not yet active so that the relevant tolerances do not apply.

For example, the current I can be measured by the parallel circuit about 250 μs after the application of the voltage U.

In order to further reduce the tolerances associated with a measurement at only one point in time, in a further particularly advantageous embodiment of the invention, a plurality of measured values for the current I are taken in a predetermined time interval T, while the voltage U is applied to the parallel circuit. A diagnostic variable D is then selected, which is a measure of the slope ΔI of the current I during the time interval T. An example of a tolerance associated with the single-shot measurement of current I is the offset error of the analog-to-digital converter used to digitize the current I reading for further processing.

For example, from a time point of about 50 μs to 250 μs after the application of the voltage U on, further measured values can be recorded at periodic intervals of, for example, 4 μs, and the gradient ΔI can be determined, for example, with a linear regression.

In fuel injection systems for vehicles are usually only one or more fixed voltages U available, which are derived from the vehicle electrical system voltage and can vary with this. Nevertheless, it is desirable to be able to maintain the current I at a predetermined level I _P by means of the parallel connection in the starting phase, that is to say during the actuation process of the injector or valve, at least in the time average. It is also desirable to be able to current I after successful actuation process stabilize at least in time average at a lower level I _H which holds the injector or the valve just securely in the actuated state. For this purpose, in a further particularly advantageous embodiment of the invention, the parallel circuit is fed via a two-point controller from a voltage source with a constant voltage U.

The two-point controller is given the current level to be maintained in the time average as a setpoint. At the same time the current I is fed as feedback to the two-point controller. If the actual current I is less than the nominal value, the parallel circuit is connected to the voltage source via a switch; if the actual current I is higher than the setpoint value, the parallel connection is disconnected from the voltage source again, and the current I begins to decay. In the steady state, a periodic time program u (t) sets with which the two-point controller applies the voltage U to the parallel circuit.

From this time program u (t), the diagnostic quantity D is evaluated. The inventors have recognized that the time program u (t) depends, on the one hand, on known or easily obtainable constants, namely on the voltage U and on the desired value of the current level to be kept in the time average. On the other hand, the time program u (t) is largely determined by how fast the current I in the parallel circuit responds to a change in the applied voltage (in each case between the two values zero and U). Since the electromagnets are essentially inductive loads, the parallel connection is also essentially an inductive load. The speed at which the current I follows a change in the applied voltage is thus determined largely by the inductance L _{P of} the parallel connection.

The determination of the diagnostic quantity D from the time program u (t) eliminates, on the one hand, the tolerances which are associated with the single-shot measurement of the current I, since the measurement is averaged over a plurality of individual measurements and over a longer period of time. On the other hand, it is also self-consistent to check whether a steady state exists, by evaluating how exactly the time program u (t) adheres to a fixed periodicity.

For example, the diagnostic quantity D can be evaluated from the control frequency f, which is equivalent to the period with which the switching on and off of the voltage U is repeated. Also, the duty cycle R and the duration t _ON or t _{OFF of} the on or off phase are significantly determined by how fast the current I responds to the switching on and off of the voltage U. Thus, the diagnostic size D can also be evaluated from these variables.

In a further particularly advantageous embodiment of the invention, the diagnostic quantity D from the temporal decay behavior of the current I by the parallel circuit after disconnecting the parallel circuit from the voltage source, i. after switching off the voltage U and the transition of the electromagnets in a so-called freewheeling phase, evaluated.

An example of a freewheeling phase which can be used in this way is the transition from the boost phase, in which the voltage U is applied continuously to the parallel circuit, into the starting phase, in which the current I is to be kept at the level I _P on average over time ,

The end of the boost phase is heralded by the fact that the voltage U is no longer applied constant to the parallel circuit, but the current control is activated. If, for example, the already mentioned two-position controller is used for this, then at the end of the boost phase it will first register that the actual current I is higher than the setpoint value I _P. Accordingly, the voltage U remains initially off. Only when the current I falls below the level I _P , if necessary. Plus a switching hysteresis, the voltage U is turned on again.

The speed at which the current I decays after switching off the voltage U is determined by the number n of electromagnets actually participating in the parallel connection, and thus by the total inductance L _{P of} the parallel circuit: the more electromagnets in the boost phase actually have been magnetized, the more magnetic energy is stored in the parallel circuit and the longer the decay of the current in the RL circuit formed from the inductances of the electromagnet on the one hand and the ohmic resistors on the other hand. Thus, the temporal decay behavior of the current I is also a variable which is a measure of the inductance L _{P of} the parallel connection.

During the boost phase, instead of the voltage U, a higher voltage U _B can optionally be applied in order to accelerate the build-up of the current I into the electromagnet. This does not change the described basic functionality.

In a further particularly advantageous embodiment of the invention, the diagnostic quantity D is evaluated from the speed with which the magnetic energy in the parallel circuit can be dissipated as electrical energy from the parallel circuit. This can be done, for example, in the phase of so-called quick extinguishing, when the injector or the valve is to be returned from the actuated to the unactuated state. During fast erase, the parallel circuit is short-circuited with a capacitor in order to transfer and thus recover the magnetic energy stored in the electromagnets in electrical form into the capacitors. The electromagnets and the capacitor then form an LC circuit, and the time constant, with which the energy transport proceeds into the capacitor and the current I decays through the parallel circuit, is again decisively determined by the inductance L _{P of} the parallel circuit.

In order to perform the method, only an already existing electrical access to the parallel connection of the electromagnets is needed, on the one hand to a voltage U applied to the parallel circuit and on the other hand, the current flowing through the parallel circuit current I can be measured. All other tasks arising in the context of the method can be done in the control unit for the fuel injection system. This control unit may be a dedicated control unit, which is assigned solely to the fuel injection system, or an engine control unit, which additionally assumes other tasks of engine management.

The invention therefore also relates to a control unit for a fuel injection system having a plurality of injectors or valves, which are each actuated by associated electromagnets, wherein the electromagnets are combined in a parallel circuit and wherein the parallel circuit is connectable by the controller to a common voltage source.

According to the invention, the control unit comprises a measuring unit for determining a diagnostic quantity D, which is a measure of the inductance L of the parallel circuit, and an evaluation unit for evaluating the number n of those electromagnets which participate in the parallel circuit from the diagnostic variable D.

In a further particularly advantageous embodiment of the invention, a two-step controller is provided in the control unit in order to keep the current I flowing through the parallel circuit at a constant level I _P , I _{H over} the time average. The time program u (t), with which the two-position controller applies the voltage U to the parallel circuit, is guided into the measuring unit, and / or into the evaluation unit.

In this case, the measuring unit, and / or the evaluation unit, and / or the two-position controller can be implemented in hardware. However, each of these components may also be implemented in whole or in part in software or firmware. In particular, the functionality of an existing controller may be modified or extended by modifying its software or firmware to enable it to carry out the method according to the invention. A corresponding software or firmware is thus an independently usable and salable product.

Therefore, the invention also relates to a computer program product with machine-readable instructions which, when executed on a computer and / or on a control unit, upgrade the computer and / or the control unit to a control unit according to the invention, and / or cause it to perform a method according to the invention.

Further measures improving the invention will be described in more detail below together with the description of the preferred embodiments of the invention with reference to figures.

list of figures

• 1 Anwendung des Verfahrens 100 in einem beispielhaften Kraftstoffeinspritzsystem 2 eines Kraftfahrzeuges 1;
• 2 zeitlicher Verlauf U(t) der Spannung an der Parallelschaltung 23 und I(t) des Stroms durch die Parallelschaltung 23 bei drei funktionsfähigen Injektoren 21a-21c (2a), bei zwei funktionsfähigen Injektoren 21a-21b (2b) und bei nur einem funktionsfähigen Injektor 21a (2c);
• 3 Auswertung (120) des vom Zweipunktregler 62 generierten Zeitprogramms u(t) über die Regelfrequenz f (3a), über das Tastverhältnis R (3b) und über die Einschaltzeit t_ON (3c) nach der Anzahl n der funktionsfähigen Injektoren 21a-21c;
• 4 Auswertung (120) der Dauer t_F der Freilaufphase zwischen dem Ende der Boost-Phase und dem Einsetzen des Zweipunktreglers 62 nach der Anzahl n der funktionsfähigen Injektoren 21a-21c;
• 5 Auswertung (120) des Gradienten dl/dt bei Schnelllöschung nach der Anzahl n der funktionsfähigen Injektoren 21a-21c.

It shows:

• 1 Application of the method 100 in an exemplary fuel injection system 2 of a motor vehicle 1;
• 2 time course U (t) of the voltage at the parallel circuit 23 and I (t) of the current through the parallel circuit 23 at three operational injectors 21a-21c ( 2a ), with two operational injectors 21a-21b ( 2 B ) and with only one functional injector 21a ( 2c );
• 3 Evaluation (120) of the time program u (t) generated by the two-point controller 62 via the control frequency f ( 3a ), via the duty cycle R ( 3b ) and on the switch-on time t _ON ( 3c ) according to the number n of operational injectors 21a-21c;
• 4 Evaluating (120) the duration t _{F of} the freewheeling phase between the end of the boost phase and the onset of the two-level controller 62 according to the number n of the operational injectors 21a-21c;
• 5 Evaluation (120) of the gradient dl / dt in the case of rapid quenching according to the number n of functional injectors 21a-21c.

To 1 includes the exemplified engine 3 in the motor vehicle 1 a single cylinder 31 , This cylinder 31 is via a fuel injection system 2 with gaseous fuel 5 , here compressed natural gas, from a reservoir 51 provided. The storage tank 51 is over the shut-off valve 52 with the fuel injection system 2 connectable.

The fuel injection system 2 includes a suction tube 4 that the cylinder 31 controlled by a throttle 41 combustion air 42 supplies. At the three bungs 43a . 43b and 43c whose distances from each other in 1 are greatly exaggerated, the combustion air 42 each through an injector 21a . 21b respectively. 21c fuel 5 from one with the shut-off valve 52 coupled distribution line 53 admixed. The injectors 21a . 21b respectively. 21c are each by associated electromagnets 22a . 22b and 22c actuated. These are normally closed valves, ie the spring force of the respective valve spring 24a . 24b . 24c acts closing, and the magnetic force of the respective electromagnet 22a . 22b . 22c works opening. The electromagnets 22a . 22b and 22c are in a parallel circuit 23 summarized and therefore only together by the control unit 6 controllable.

The control unit 6 includes a fixed voltage source 61 with the voltage U, which optionally has a two-position controller 62 or via a boost circuit 65 to the parallel connection 23 can be created. Each by the parallel connection 23 flowing current I is using the ammeter 66 measured.

The respective measured value for the current I becomes the two-position controller 62 fed as feedback and at the same time the measuring unit 63 fed. The measuring unit 63 additionally receives the time program u (t), with which the two-position controller the voltage U to the parallel circuit 23 turns on, and delivers in step 110 of the procedure 100 a diagnostic quantity D, which is a measure of the inductance L _{P of} the parallel circuit 23 is. The evaluation unit determines from diagnostic quantity D 64 turn in step 120 of the procedure 100 the number n of electromagnets 22a . 22b . 22 connected to the parallel circuit 23 participate, and thus the number n of functioning injectors 21a . 21b . 21c ,

In the 2a to 2c are for one 1 ajar prototypical test arrangement each at the parallel circuit 23 voltage applied U (t) and by the parallel circuit 23 flowing current I (t) plotted as functions of time t, the scale of 2a to 2c is identical. The time course U (t) is not identical with the time program u (t), which is the two-position controller 62 but a trace taken with an oscilloscope. The time course U (t) therefore also contains overshoots and other deviations from the nominal rectangular shape of the applied voltage pulses.

The activation of the parallel connection 23 runs in the 2a to 2c each according to the same scheme. First, a boost voltage U _{B is} applied in order to start from the unactuated state of the injectors 21a to 21c to increase the current I (t) by the parallel connection as quickly as possible. This boost voltage U _B is higher than the fixed voltage U of the voltage source 61 , For this purpose contains the boost circuit 65 a (in 1 not separately drawn) voltage-increasing DC-DC converter. As soon as the maximum boost current I _{B is} reached, the boost circuit is activated 65 disabled, and the control of the on the parallel connection 23 applied voltage U is sent to the two-position controller 62 to hand over. The two-position controller 62 Now has the task, the current I through the parallel circuit 23 in a pull-in phase to stabilize at a first, high setpoint I _P , thus allowing the armatures of the injectors 21a . 21b and 21c be tightened as soon as possible in their actuated end position. When this is done, the setpoint is lowered to a lower value I _H , which is sufficient only for holding the armature in the actuated end position. To return to the unactuated state, the setpoint for the current I is reduced back to zero. This results in the voltage U (t) in each case by mutual induction in the electromagnet 22a to 22c a spike S.

In the boost phase, the current I (t) is in each case measured several times in the time interval T, and the slope ΔI of the current over the time interval T is measured on the basis of a linear regression.

Here comes the first major difference between the 2a . 2 B and 2c to day. The more electromagnets 22a . 22b . 22c at the parallel connection 23 participate, the lower the total inductance L _{P of} the parallel connection 23 , and the faster the current I (t) responds to the boost voltage U _B. Accordingly, the larger the slope .DELTA.I and the shorter the duration of the boost phase, the more electromagnets 22a . 22b and 22c function.

Furthermore, from the comparison of 2a to 2c it can be seen that the control frequency f of the two-point controller 62 the larger, the more electromagnets 22a . 22b . 22c function, the lower the total inductance L _{P of} the parallel connection 23 is. Because the electromagnets 22a . 22b . 22c then generally react faster, a lower duty cycle R and a lower duty cycle t _{ON is} necessary to after a decay of the current I (t) under the respective setpoint I _P and I _H the current I (t) back again on this Setpoint I _P or I _H to drive.

These relationships are in the 3a to 3c once again clarified. In 3a is in the holding phase with the setpoint I _H, the control frequency over the number n of the functioning electromagnets 22a . 22b . 22c applied. In 3b is the duty cycle R of the two-point controller 62 plotted against n, wherein between the pull-in phase with the setpoint I _P and the hold phase with the setpoint I _{H is} differentiated. In 3c the on-time t _{ON is} plotted per cycle, again differentiating between the pull-in phase with the setpoint I _P and the hold phase with the setpoint I _H. The signal swing varies between the different evaluations according to the 3a to 3c , and partly between the pull-in phase and the holding phase, considerably. However, in all examples it is large enough to clearly indicate the failure of one or more electromagnets 22a . 22b . 22c determine and initiate appropriate countermeasures.

In 4 is the duration t _{F of} the freewheel phase from the end of the boost phase to the onset of the two-step controller 62 applied over n. The larger n is, the more magnetic energy must be dissipated in the freewheeling phase at ohmic resistances, and the longer is the duration t _{F of} the freewheeling phase.

In 5 the gradient dl / dt of the current I (t) is plotted against n. The larger n, the smaller the total inductance L _{P of} the parallel connection 23 and the faster the energy transport from the solenoid coils of the electromagnets 22a . 22b and 22c in the (in 1 not shown) capacitor for energy recovery.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 7168413 B2 [0005]

Claims (10)

Method (100) for operating a fuel injection system (2) having a plurality of injectors or valves (21a, 21b, 21c) which are each actuable by associated electromagnets (22a, 22b, 22c), the electromagnets (22a, 22b, 22c) are combined in a parallel circuit (23) and wherein the parallel circuit (23) with a common voltage source (61) is connectable, characterized in that a diagnostic quantity D, which is a measure of the inductance L _{P of} the parallel circuit (23) is determined (110), and that from the diagnostic quantity D, the number n of those electromagnets (22a, 22b, 22c) participating in the parallel circuit (23) is determined (120). Method (100) according to Claim 1 , characterized in that the diagnostic quantity D is formed from at least one measured value for the current I flowing through the parallel circuit (23), the measured value being recorded at a predetermined time t _M after the application of a constant voltage U to the parallel circuit (23) , Method (100) according to Claim 2 , characterized in that a plurality of measured values for the current I are successively recorded in a predetermined time interval T, while the voltage U is applied to the parallel circuit (23), wherein the diagnostic quantity D is a measure of the slope .DELTA.I of the current I during the time interval T. , Method (100) according to one of Claims 1 to 3 , characterized in that the current I flowing through the parallel circuit (23) by supplying the parallel circuit (23) from a voltage source (61) with a constant voltage U via a two-position controller (62) on averaged over time at a constant level I _P , I _H is held, wherein the diagnostic quantity D from the time program u (t), with which the two-point regulator (61) applies the voltage U to the parallel circuit (23), is evaluated (110). Method (100) according to Claim 4 , characterized in that the diagnostic quantity D from the control frequency f, from the duty cycle R, and / or from the duration t _ON or t _{OFF of} the _ON or _OFF phase, the time program u (t) is evaluated (110). Method (100) according to one of Claims 1 to 5 , characterized in that the diagnostic quantity D is evaluated from the temporal decay behavior of the current I by the parallel circuit after disconnecting the parallel circuit from the voltage source (110). Method (100) according to one of Claims 1 to 6 , characterized in that the diagnostic quantity D from the speed at which the magnetic energy in the parallel circuit as electrical energy from the parallel circuit is dissipated, is evaluated (110). Control unit for a fuel injection system with a plurality of injectors or valves, which are each actuated by associated electromagnets, wherein the electromagnets are combined in a parallel circuit and wherein the parallel circuit is connectable by the control unit with a common voltage source, characterized in that the control unit comprises a measuring unit ( 63) for determining a diagnostic quantity D, which is a measure of the inductance L of the parallel circuit, as well as an evaluation unit (64) for evaluating the number n of those electromagnets which participate in the parallel circuit, from the diagnostic quantity D. Control unit after Claim 8 , characterized in that a two-step controller is provided to keep the current flowing through the parallel circuit current I at a constant level I _P , I _H , the time program u (t), with the two-position controller, the voltage U to the Applies parallel circuit, in the measuring unit, and / or in the evaluation, is performed. A computer program product comprising machine-readable instructions which, when executed on a computer and / or on a controller, transfer the computer and / or the controller to a controller according to any one of Claims 8 to 9 upgrade, and / or cause, a method (100) according to one of Claims 1 to 7 perform.

Images