EP2072820B1 - Method of analysing the operation of a dosing pump for liquid, especially a fuel dosing pump for a vehicle heater - Google Patents

Abstract

The method involves determining a starting time of a movement of a piston (26) as an analysis variable by formation of a temporal conduction of an electric current flowing in an excitation time interval and comparison of the temporal conduction with an assigned threshold, and determining an end time of the movement of the piston as another analysis variable. The two analysis variables are compared with respective assigned references. A presence of an error condition is detected when the analysis variables deviates from the references.

Description

The present invention relates to a method for analyzing the operation of a metering pump for liquid, in particular Brennstoffdosierpumpe for a vehicle heater, which metering a clocked between two end positions reciprocating piston and an associated thereto, by applying a voltage during energization time intervals in respective working strokes of the piston electrically excitable drive unit comprises.

From the DE 10 2005 024 858 A1 a method for operating a metering pump according to the preamble of claim 1 is known. In this method, a metering pump is basically operated by a reciprocating piston for conveying liquid by an electrically energizable drive unit is moved from a first end position to a second end position, including during a energization time interval to the electrically excitable drive unit, a voltage in the form a clocked voltage signal is applied. By this pulsed, generally pulse width modulated voltage results in an example represented by forming the arithmetic mean voltage, which essentially determines how fast the piston moves from its first end position to the second end position, so for example to minimize the volume of a Pump chamber is moved. In association with the design of such a pump specified by certain data, it may be determined that, ideally, the time of reaching the second end position, generally reaching an attack, is a predetermined time after the beginning of a respective energization time interval or after the beginning of the movement Piston should lie. Will be recognized, that this second end position is reached too early or too late can be adjusted by varying the voltage applied to the drive unit, ie by varying the duty cycle of the pulsed voltage, the average applied voltage and thus trying to the time at which the second end position , That is, the stop position, is reached, to move so that it is at or near the predetermined and to be considered as a reference target time.

It is the object of the present invention to provide a method by which the operation of a fluid metering pump can be analyzed to detect the presence of fault conditions.

According to the invention, this object is achieved by a method for analyzing the operation of a metering pump for liquid, in particular Brennstoffdosierpumpe for a vehicle heater, which metering a clocked between two end positions reciprocating piston and associated with this, by applying a voltage during energization time intervals in each of the working cycles of According to claim 1. The method comprises the steps of: determining a start time of movement of the piston as a first analysis quantity and determining an end time of movement of the piston as a second analysis quantity, comparing at least one analysis variable with a reference associated therewith, and based on on the result of the comparison, detection of the presence of a fault condition if the analysis variable deviates from the reference.

In the present invention, therefore, based on the time at which a piston of a metering pump begins to move, and / or based on the time at which the piston reaches its end position, for example, a stop position, and based on respectively assigned reference values determines whether the piston has moved in a normal manner, that is, correctly, or whether the course of movement deviates from the normal expected, which is characterized makes it clear that the respective analysis variable deviates from the assigned reference. It should be pointed out here that the deviation of a respective analysis variable from a reference is to be understood such that, for example, a reference time point to be considered as a reference does not correspond to the respective analysis value or the analysis variable deviates from such a time point by a predetermined extent or not within a time window defined at such time.

According to the invention, it is further provided that the first analysis variable is only determined if, in one or more preceding work files, the comparison of the second analysis variable with its associated reference indicates the presence of a fault condition. This procedure is based on the knowledge that a second analysis variable can only occur if the piston has started to move, ie if a first analysis variable would also be available. Since, if there is a second analysis variable for evaluation, the respectively assigned first analysis variable does not necessarily have to be evaluated, it can be dispensed with, which reduces the processing outlay.

The error state that is used as the trigger for subsequently determining also the first analysis variable for following work cycles is a state in which no second analysis variable was detected in an energization time interval.

In order to have sufficient information in particular for later analysis processes, which can be carried out for example in a laboratory or a workshop, it is proposed that in association at least one analysis variable and / or the comparison result of the comparison of at least one analysis variable with the reference assigned to it is stored for a respective working cycle of the piston.

In the case of the method according to the invention, it can be provided that, when the second analysis variable lies in front of the reference assigned to it, it is detected that there is an error state comprising the conveying of air.

Furthermore, it is possible that when the second analysis variable is in an excitation time interval and after the reference assigned to it, it is concluded that a fault condition comprising a difficult conveying is present.

According to a further aspect, it can be provided that when no second analysis variable is detected in an excitation time interval and a first analysis variable is detected, it is concluded that a fault condition comprising a difficult conveying is present.

Furthermore, if no first analysis variable was detected in an excitation time interval, it is possible to detect the presence of an error state comprising a movement blocking of the piston.

In a particularly advantageous development of the method according to the invention, it is proposed that if the second analysis variable is in an excitation time interval before or after the reference assigned to it, the second analysis parameter is attempted in at least one subsequent working cycle by varying the excitation voltage for the drive unit Reference, and that when a variation of the excitation voltage does not lead to a sufficient shift of the second analysis size, is detected on the presence of a fault condition. Thus, according to this aspect, it may be first attempted to generate an indication indicative of the presence of an error condition or to avoid information. Only when it becomes apparent that a variation of the excitation voltage does not lead to the desired result, is then detected on the presence of a fault condition.

In order to determine in a precise manner when a piston starts or has begun to move, it is proposed that the first analysis variable be formed by forming the first time derivative of the electrical current flowing in an energization time interval and comparing it to an associated first threshold is determined.

Furthermore, for the precise determination of the reaching of the second end position, ie the stop position, it may be provided that the second analysis variable is determined by forming the second time derivative of the electrical current flowing in an energization time interval and comparing it with an associated second threshold.

Fig. 1

eine Prinzipdarstellung eines Fahrzeugheizgeräts mit einer durch elektrische Erregung betreibbaren Dosierpumpe;

Fig. 2

ein Diagramm, welches den zeitlichen Verlauf des Erregungs- stroms der Dosierpumpe in Zuordnung zu einem Erregungszeitin- tervall bei korrektem Betrieb der Dosierpumpe wiedergibt;

Fig. 3

in prinzipieller Darstellung die erste zeitliche Ableitung des Erre- gungsstroms;

Fig. 4

in prinzipieller Darstellung die zweite zeitliche Ableitung des Erre- gungsstroms;

Fig. 5

ein der Fig. 2 entsprechendes Diagramm für einen Fehlerzustand, in welchem die Dosierpumpe Luft fördert;

Fig. 6

ein der Fig. 2 entsprechendes Diagramm für einen Fehlerzustand, in welchem ein Kolben der Dosierpumpe klemmt;

Fig. 7

ein der Fig. 2 entsprechendes Diagramm für einen Fehlerzustand, in welchem der Kolben der Dosierpumpe sich zu langsam bewegt.

The present invention will be described below in detail with reference to the accompanying drawings. It shows:

Fig. 1

a schematic diagram of a vehicle heater with an operable by electrical excitation metering pump;

Fig. 2

a diagram which shows the time course of the excitation current of the metering pump in association with an excitation time interval during correct operation of the metering pump;

Fig. 3

in a schematic representation the first time derivative of the excitation current;

Fig. 4

in a schematic representation, the second time derivative of the excitation current;

Fig. 5

one of the Fig. 2 corresponding diagram for a fault condition, in which the metering pump promotes air;

Fig. 6

one of the Fig. 2 corresponding diagram for a fault condition in which a piston of the metering pump is stuck;

Fig. 7

one of the Fig. 2 corresponding diagram for a fault condition in which the piston of the metering pump moves too slowly.

In Fig. 1 is a fuel heater to be operated with liquid fuel, such as can be used for example as a heater or heater in a vehicle, generally designated 10. The heating device 10 comprises a burner region 12 with a combustion chamber 16 formed in a housing 14. Combustion air is conducted into the combustion chamber 16 via a combustion air blower 18. A metering pump 20 conveys the fuel required for combustion with the combustion air in liquid form from a reservoir 22 to the combustion chamber 16, where, for example, a porous evaporator medium can take up this fuel and deliver it into the combustion chamber 16 in vapor form.

A drive device 24 controls the operation of the heater 10 by generating corresponding excitation signals for the combustion air blower 18 and the metering pump 20 or other system areas not shown here, such as an ignition device or an electrically energizable heating element serving to heat an evaporator medium.

The metering pump 20 is in principle constructed with a piston 26 which is in a cylinder 28 between an end position with maximum volume of a pump chamber 30 and an end position with minimum volume of the pump chamber 30 back and forth. In general, the piston 26 is biased by a biasing arrangement, so for example a spring, in the direction of its first end position, ie the maximum volume of the pump chamber 30. An electrically excitable drive unit 32, that is, for example, an electromagnet arrangement, shifts in electrical Exciting the piston 26 to promote the liquid contained therein by reducing the pump chamber volume 30 in the direction of the combustion chamber 16. In this case, a stop is provided for the piston 26 which defines the second end position of the piston 26 with a minimum volume of the pump chamber 30.

In order to move the piston 26, the drive device 24 outputs a pulsed voltage signal U for each operating cycle of the piston 26, that is to say generally a pulse width modulated (PWM) signal. This signal U can be tapped from the supply voltage, wherein by adjusting the duty cycle during the application of the voltage signal U adjusting average voltage, for example, represented by the arithmetic mean, can be adjusted. The electrical current I flowing when the voltage signal U is present can be detected by an ammeter 34, which inputs a corresponding signal into the control device 24. Of course, the flow meter 34 may also be part of the drive device 24 itself.

The Fig. 2 shows in principle the time course of the current I during a power stroke and normal operation. A working stroke of the piston 26 is defined by a complete reciprocation and begins, for example, each at a time t _e , ie the time of the beginning of an energization time interval I _e , during which the in the Fig. 2 symbolically indicated by dashed line pulsed voltage signal U is applied to the drive unit 32. A power stroke I _{A of} the piston 26 ends with the beginning of the next energization time interval I _e , ie the next time t _e .

Thus, if the pulsed voltage signal U is applied to the drive unit 32 at time t _e , the current I initially increases until the magnetic force acting on the piston 26 or an armature or the like coupled thereto is so great that at a time t _s the piston 26 begins to move. By the movement of the piston 26 and the occurring Back induction, the current flow goes into a shallower section. At a time t _on , the stop position, ie the second end position, is reached, beyond which the piston 26 can not move on. Since from the time t _on the piston 26 so no longer moves, no mutual induction, so that the current I initially increases again, until the time t _a the end of the energization time interval I _{e is} reached and the current I then decays, for example, exponentially , From the time t _a , the piston 26 then begins to move back to its first end position.

Such metering pumps are operated so that working at a working frequency in the range of 3-10Hz, ie 3 to 10 work cycles per second. Within such a working cycle I _A , the excitation time interval I _{e can take} a period of about 40 ms. Ideally, the point in time t _at which the second end position, ie the movement stop, is reached, should be about 35 ms after the start t _{e of} the excitation time interval I _e , so that the time span over which the drive unit stops when the piston 26 is no longer moving 32 is energized as short as possible, however, however, it can be ensured that the piston 26 reaches this second end position.

The Fig. 3 shows in association with Fig. 2 the first time derivative of the in Fig. 2 represented current flow. At time t _e , the current initially rises abruptly. The slope of the current waveform then decreases until reaching the time t _s again. From this point in time t _s , the gradient continues to fall or fall, so that a negative gradient, ie a negative first time derivative, results. At the time of the end stop, so to Zeitpung _T, the gradient again increases suddenly and then decreases again, until the time t _a, that at the timing at which the energization is actually completed, the current drops again, and thus the Gradient takes a negative value.

In association with in Fig. 2 shown current curve shows the Fig. 4 the second temporal derivative, ie the first time derivative of the in Fig. 3 shown gradients. Here it is characteristic in relation to the two at the times t _e and t _at the sudden increase of the second time derivative.

This behavior of the first derivative and the second time derivative can be used to the two times t _s and t _on, thus to determined the start of the movement of the piston 26 and the reaching of the stop position of the piston 26th In association with the first time derivative and the second time derivative respective thresholds S ₁ and S _{2 are} given. If the first time derivative of the current profile falls below the first threshold S ₁ , this is indicated as an indication of the in Fig. 2 recognizable transition into a much flatter current flow, so the beginning of movement evaluated. The point of time t _s determined in this way can, as explained below, then be taken into account as the first analysis variable for further evaluation. Accordingly, the time t can be recognized _to when the second time derivative exceeds the associated threshold S _2, namely from the beginning of the excitation time interval I _e exceeds a second time. The thus determined time t _an can then be used as a second analysis variable for further processing.

It should be noted here that in principle these two analysis variables or times can also be determined in another way. For example, it would be possible to work only with the gradient, that is to say the first time derivative, wherein the time t _on could be determined as the time in which the gradient crosses the associated threshold S ₁ for the second time within an excitation time interval or for the first time after Time t _s exceeds.

During operation of the metering pump 20, the two times t _s and t can be determined _in each case as a first analysis variable and as a second analysis variable, for example in the manner described above. Based on these sizes can then, as in the following with reference on the Fig. 5 to 7 explained, various error conditions during operation of the metering pump 10 are detected.

The Fig. 5 shows the course of the current I, plotted against time, in the event that, for example, due to shocks or due to an almost complete emptying of the reservoir 22, the metering pump 20 at least temporarily promotes air instead of the liquid fuel. The time t _s at which the piston 26 begins to move will in this case occur approximately at the same time as it is in the correct mode of operation. Due to the fact that a lesser movement resistance will be present already from the beginning, it is possible that the piston 26 starts to move slightly earlier.

Since in the course of this movement, the resistance against which the piston 26 must be moved, is significantly lower, since no liquid must be displaced from the pump chamber 30, the piston 26 will reach its second movement end position much earlier, so that used as the second analysis size time t _made with respect _to normal functionality as in Fig. 2 is represented, will occur much earlier. Here, for example, based on the in Fig. 2 represented diagram for this time t _to a to be expected for normal operating conditions reference value t _an 'or a reference value range to be specified, in which normally, so with correct operation, the end position should be reached. Lies like this Fig. 5 shows the time t actually detected _in front of or an excessive extent prior to the reference value t _on ', so this is a clear indication of an over-rapid movement of the piston 26, which can in principle only occur if air is promoted. By comparing the second analysis variable t _an with its associated reference t _an ', therefore, this error state can be detected. In a data memory present, for example, in the control device 24, a code representing this error state "conveying air" can be stored. Likewise, it is of course possible in association with the successive work cycles not only such a fault condition or to store the codes indicating normal operation, but also, for example, the actual respectively recorded analysis variables.

Occurs in such a system then a malfunction, for example in the form of a flame, because not enough fuel is available, then the reason can be recognized in a subsequent evaluation, since already before the occurrence of the flame burst working cycles have occurred associated with the flameout indicative error code and is stored. If only the second analysis variable is stored per se, it can be recognized by subsequent evaluation, that is to say by later comparison of the same with the associated reference, that the occurring flameout was conditioned or at least favored by the conveyance of air.

Since such error conditions generally lead to malfunctions within a comparatively short time, the most efficient use of the existing storage volume can be achieved by storing this data only in association with a specific number, for example 100, in the past working cycles I _A that in each case for the last example, 100 work file I _A this information is present.

In order to increase the accuracy in the error indexing, it is further possible, for example, to proceed, if it is initially recognized that the stop indicated by the second analysis variable, ie the time t _an , to occur too early, first by varying the duty cycle of the Voltage signal U to be attempted to move this time t _an in the direction of its reference t _an '. Too early occurrence of the second analysis size may possibly be compensated for by shifting the duty cycle in the direction of decreasing the average applied voltage. Moves on this measure out the second analysis variable t _to not or not sufficiently fast or sufficiently close to the reference _at t 'zoom, so as could by reducing the Stress and thus reducing the piston 26 driving force whose too fast movements are not sufficiently slowed down, so that then, for example, the corresponding error code can be set or stored, for example, also in conjunction with information that characterizes the applied voltage.

Another error condition is in Fig. 6 shown. One recognizes in Fig. 6 in that, within the excitation time interval I _e , ie during the period during which the pulsed voltage signal U is applied, the current I increases continuously. Thus, neither a first analysis variable, that is to say the time t _s , nor a second analysis variable, that is to say the time t _on , can be determined here. In fact, this condition can only occur if the piston 26 does not begin to move within the excitation time interval I _e . This means that the piston 26 is blocked from movement due to any defect. Thus, if neither the first analysis variable nor the second analysis variable is detected, then a code indicating this error state, that is to say movement blocking of the piston 26, can be set or stored or a numerical value indicating the non-occurrence of the respective analysis variable can be stored as the corresponding time.

The Fig. 7 shows an error condition in which within the excitation time interval I _e, although the first analysis value t _s occurs, so at time t _s, the piston 26 begins to move until the end time t _{a of} the excitation time interval _, the occurrence or reaching the end stop could not be detected , so no second analysis size could be determined. So, this is a condition in which, upon excitement, the piston 26 has begun to move, but obviously moves too slowly. This can be caused, for example, by the fact that the fuel to be pumped is too tough, or that in the conveying path of the fuel downstream of the metering pump 20, a delivery backlog has occurred, for example due to the blockage of a delivery line. Also in this case, a corresponding error code can be set, which indicates that there is a problem in conveying the fuel through the lines is present, that in principle, however, the metering pump 20 would be able to promote the fuel. Again, the actual applied voltage can be evaluated or stored as a further analysis variable.

In the in the 6 and 7 existing error conditions can be seen that in each case no second analysis size can be determined equally. Due to the simplest possible data processing, it may therefore be advantageous, in principle, to determine only the second analysis variable t _an during the operation of the metering pump 20 in the individual working cycles I _A. If a second analysis variable can be determined, then a first analysis variable must necessarily be present. However, since with regard to the error states described the temporal position of the first analysis variable is of subordinate importance and only the question of whether or not within a excitation time interval I _e a first analysis variable can be determined, it is then if a second analysis variable t _{an are} determined could, sufficiently, to determine their position or to compare with the associated reference t _an '. However, if no second analysis variable has occurred within an excitation time interval I _e , it is then possible to proceed in such a way that in one or more subsequent work cycles I _{A an} attempt is also made to determine the first analysis variable. If a first analysis size can be determined, although a second analysis size could not be determined, this would indicate that in Fig. 7 illustrated case of too slow movement of the piston 26 out. If no first analysis size can be determined, this points to the in Fig. 6 case shown. The processing effort can thus be kept low by actually determining the first analysis variable only if this is also helpful for the further evaluation.

As set forth above, taking into account the determined analysis quantities or the error codes generated in association therewith, when a malfunction has occurred, it can be reconstructed which problem has led to the malfunction, possibly in a repair operation make appropriate corrections. For this purpose, the memory already described above can be read out and evaluated with regard to the data stored therein. In principle, however, it is also possible to detect immediately during operation if fault conditions occur which could lead to a malfunction in the near future. Shows, for example, that the in Fig. 5 shown case, so apparently air is promoted, so the working clock frequency at which the metering pump 20 is operated, can be increased to even before it can lead to a flame breakage nachzufördem faster fuel, if there is still fuel in the reservoir 22 is available. In this way, the occurrence of the malfunction may then be completely prevented. Also in the in the Fig. 6 7 and 7 can immediately, if it has been detected in one or more consecutive work cycles, for example, by increasing the average applied voltage to be tried to move the piston faster or at all. If the system is in a startup phase, in which case the fuel line must first be refilled, the information can also be used to extend this start phase accordingly, until it is ensured that there is sufficient fuel in the line to feed in to begin in the combustion chamber 16.

Claims (9)

1. A procedure for analyzing the operation of a dosing pump for fluid, in particular a fuel dosing pump for a vehicle heating device, with the dosing pump (20) comprising a piston (26), moveable in a clocked manner between two final positions, and in association to the piston comprising a drive unit (32), excitable by applying a voltage (U) during excitation time intervals (I_e) in respective duty cycles (I_A) of the piston (26), the procedure comprising the following steps:

   - determining a starting time (t_s) of the movement of the piston (26) as a first parameter of analysis
   and
   determining a final time (t_an) of the movement of the piston (26) as a second parameter of analysis,

   - comparing at least one parameter of analysis (t_s, t_an) with an associated reference (t_an'), characterized in that

   an error condition is recognized on the basis of the result of comparison, if the parameter of analysis (t_s, t_an) deviates from the reference (t_an'), and that the first parameter of analysis (t_s) is only determined if in one or more preceding duty cycles the comparison of the second parameter of analysis (t_an) with the associated reference (t_an') indicates the presence of an error condition, wherein in an excitation time interval (I_e) no second parameter of analysis (t_an) is recognized.
2. A procedure according to claim 1, characterized in that in association to a respective duty cycle (I_A) of the piston (26), at least one parameter of analysis (t_s, t_an) or/and the reference result of the comparison of at least one parameter of analysis with the associated reference (t_an') is stored.
3. A procedure according to claim 1 or 2, characterized in that, if the second parameter of analysis (t_an) precedes the reference (t_an'), an error condition is recognized indicating the transport of air.
4. A procedure according to one of claims 1 to 3, characterized in that, if the second parameter of analysis (t_an) lies within an excitation time interval (I_e) and is succeeding its reference (t_an'), an error condition is recognized indicating an impeded transport.
5. A procedure according to one of claims 1 to 4, characterized in that, if within an excitation time intervals (I_e) no second parameter of analysis (t_an) is recognized and a first parameter of analysis (t_s) is recognized, an error condition is recognized comprising an impeded transport.
6. A procedure according to one of claims 1 to 5, characterized in that, if within an excitation time intervals (I_e) no first parameter of analysis (t_s) is recognized, an error condition is recognized comprising obstructed movement of the piston (26).
7. A procedure according to one of claims 1 to 6, characterized in that, if the second parameter of analysis (t_an) precedes or succeeds the associated reference (t_an') in an excitation time interval (I_e), the second parameter of analysis (t_an) should, if possible, be adjusted towards the reference (t_an) during at least one following duty cycle (I_A) by varying the excitation voltage (U) for the drive unit (32) and in that, if a variation of the excitation voltage (U) does not result in efficiently adjusting the second parameter of analysis (t_an), an error condition is recognized.
8. A procedure according to one of claims 1 to 7, characterized in that the first parameter of analysis (t_s) is determined by establishing the first derivation in time of the current (I) flowing during an excitation interval (I_e) and by comparing the latter with an associated first threshold (S₁).
9. A procedure according to one of claims 1 to 8, characterized in that the second parameter of analysis (t_an) is determined by establishing the second derivation in time of the current (I) flowing during an excitation interval (I_e) and by comparing the latter with an associated second threshold (S₂).

Images