GB2310458A

GB2310458A - Leakage fault recognition in the fuel supply of an i.c. engine with high-pressure fuel injection

Info

Publication number: GB2310458A
Application number: GB9623407A
Authority: GB
Inventors: Bernhard Bronkal; Juergen Biester
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1995-11-09
Filing date: 1996-11-08
Publication date: 1997-08-27
Anticipated expiration: 2016-11-08
Also published as: JP3995118B2; GB2310458B; GB9623407D0; FR2741113A1; JPH09170518A; FR2741113B1

Description

2310458 FAULT RECOGNITION IN AN ENGINE FUEL SUPPLY SYSTEM The present

invention relates to a method of and means for recognising leakage in a fuel supply system in an internal combustion engine with high-pressure injection.

In a motor vehicle with a fuel-injected engine, fuel is conveyed with the aid of a fuel pump from a fuel tank and fed by way of ducts to injection valves. Excess fuel is usually returned by way of a return duct back to the fuel tank. In the case of an engine with a high-pressure injection, in particular in an engine with compression ignition, a further pump follows a first pump and produces a very high pressure in a high-pressure region connected with the injection valves.

With such a fuel supply system there is the danger that fuel in the case of a fault with a valve or injection nozzle is constantly injected into the associated combustion chamber. Moreover an outward leakage is also possible, in which case fuel escapes under high pressure into the engine compartment. It is therefore proposed, for example in DE- PS 31 26 393, to provide means which continuously measure the pressure in the high-pressure region of the fuel supply system, wherein a falling of the pressure in a storage device below a predetermined value leads to a fault recognition. Since fuel would be injected constantly into the engine in such a case, the engine is switched off or the fuel conveying terminated after the recognition of a fault.

According to a first aspect of the invention there is provided a method of recognising leakage in a fuel supply system of an internal 2 - combustion engine with high-pressure injection, in which system fuel is conveyed from a low-pressure region into a high-pressure region by at least one pump, the pressure in the high-pressure region is controllable by at least one pressure control means and a pressure sensor detects a pressure value in the high-pressure region, the method comprising the steps of so driving said at least one pressure control means during starting of the engine that the pressure in a fault-free state rises to a predetermined expected value and recognising the presence of a fault if the detected pressure value does not reach the expected pressure value.

Preferably, the fault is recognised when the detected pressure value does not reach the expected pressure value within a presettable time period.

Expediently, a second expected pressure value is presettable on reaching the expected pressure value and a fault is recognised when the detected pressure value does not reach the second expected pressure value within a second presettable time period.

If so desired, the check can take place on each start or only when a fault was recognised during the preceding operation. A check can also take place on the first putting into operation and/or on each new start after a service. Different expected pressure values and time periods can be presettable in dependence or when the check takes place. The expected value is preferably presettable in dependence on the properties of the fuel and/or the high-pressure pump, the volume of a common fuel rail of the system and/or the rotational speed of the engine.

According to a second aspect of the invention there is provided 3 - means for the recognition of a leakage in a fuel supply system of an internal combustion engine with high-pressure injection, wherein fuel is conveyed from a low-pressure region into a high-pressure region by at least one pump, the pressure in the high-pressure region is controllable by at least one pressure control means and a pressure sensor detects a pressure value in the high-pressure region, characterised in that means are provided which on starting of the engine drive at least one of the pressure control means in such a manner that the pressure in the fault-free state rises to an expected value and which recognise a fault when the detected pressure value does not each the expected pressure value.

A method exemplifying and recognition means embodying the invention may have the advantage over the known system that the entire high-pressure fuel supply system is monitored for tightness and it is recognised not only whether an injection valve is constantly open, but also whether a leakage to the outside is present. In that case, the simplicity of the procedure' is particularly advantageous.

An example of the method and embodiment of the recognition means according to the present invention will now be more particularly described with reference to the accompanying drawings, in which:

Fig. 1 is a schematic block diagam of a fuel system with recognition means embodying the invention; Fig. 2 is a flow chart of the steps of a first method exemplifying the invention; and Fig. 3 is a flow chart of the steps of a second method exemplifying the invention.

- a - Referring now to the drawings there is shown in Fig. 1 essential parts of a fuel supply system of an internal combustion diesel engine with high- pressure injection. The illustrated system is usually known as a common rail system. A fuel tank 10 is connected with a rail 35 by way of a fuel feed duct with a filter 15, a preliminary conveying pump 20, a shut-off valve 25 and a high-pressure conveying pump 30. A pressure-regulating valve 40 or pressure-] imiting valve is arranged in the fuel feed duct between the pump 30 and the rail 35. The feed duct is connected by means of this valve with a return duct 45, which serves to return fuel to the tank 10.

The shut-off valve 25 is actuable by means of a coil 26 and the valve 40 by means of a coil 41. A sensor 50 is arranged at the rail 35. The sensor 50 is preferably a pressure sensor, which provides a signal corresponding to the fuel pressure in the rail. The rail 35 is connected by way of respective ducts with individual injectors 61 to 66. The injectors comprise electromagnetic valves 71 to 76, by means of which the fuel flow through the injectors is controllable.

In addition, the injectors communicate by way of respective connections with the return duct 45.

The output signal of the pressure sensor 50 as well as the output signals of further sensors 80 are applied to a control unit 100, which in turn controls the valves 71 to 76, the coil 26 of the preliminary conveying pump 20, the coil 41 of the pressure-regulating valve 40 and the highpressure conveying pump 30. The pressure in the high-pressure region can be controlled by the pressure-regulating valve 40 and/or the high-pressure pump 30. The high-pressure pump and the pressure-regulating valve therefore both represent pressure control means. The region between the fuel tank 10 and the high-pressure pump 30 is a low-pressure region and the region between the pump 30 and the injectors is a high-pressure region. 5 In operation of the system the pump 20, which can be an electrical pump or a mechanical pump, conveys the fuel, which is in the tank 10, by way of the filter 15 to the pump 30. The pump 30 conveys the fuel into the rail 35 and causes a build-up of pressure therein. Usually, pressure values of about 30 to 100 bars are achieved in systems for an applied ignition engine and pressure values of about 1000 to 2000 bars are achieved for a compressionignition engine.

The shut-off value 25, which is drivable by the control unit 100, is arranged between the high-pressure pump 30 and the preliminary pump 20 in order to interrupt the fuel flow.

Starting from the signals of the sensors 80, the control unit 100 determines control signals for action on the valves 71 to 76 of the injectors 61 to 66. The beginning and the end of fuel injection into the engine is controlled by opening and closing of the valves 71 to 76.

The pressure of the fuel in the rail 35 and thereby in the highpressure region is detected by means of the pressure sensor 50. Starting from this value, the control unit 100 computes a signal for action on the pressureregulating valve 40. Preferably, the pressure is regulated to a presettable value, which inter alia depends on operating conditions of the engine as detected by means of the sensor 80, by driving of the pressure-regulating valve 40.

The fuel feed can be stopped by the valve 25 in the case of a f aul t. The valve 40 is so controlled on recognition of a fault that the pressure in the rail 35 decays. In addition, the valves 71 to 76 are controlled so as to remain closed and thus no injection takes 5 place.

Leakage can occur in these systems. In that case, fuel escapes from the high-pressure region, which is disposed under high pressure, by way of the injectors and into the engine and/or by way of a leakage into the engine compartment of the vehicle. Such leakages into the engine compartment or from a faulty injector must be able to be recognised. It is particularly important that a leakage is recognised reliably on a restarting of the engine after having been switched off.

For this purpose, after actuation of the starter, it is checked whether the fuel pressure reaches a first threshold value within a presettable first time period. If this is not the case, a major system fault is present or the fuel system must be ventilated. Subsequently, it is awaited whether the fuel pressure rises to a second threshold value within a second time period. If this is also not the case, a fault bit is set, which indicates that a leakage is present. If the pressure does rise, the normal starting procedure can be returned to.

It is particularly advantageous if the check takes place only when the fault bit, which indicates that a leakage is present, is set. This bit is set during normal operation when a leakage is recognised. In this case, a check can take place only if a fault was recognised during the preceding operation. Thus, restarting can be prevented if the leakage is still present, but made possible if the leakage has been eliminted.

It is advantageous if the fault bit is set before first putting into operation of the engine or after a service has taken place, so that during the service or during the first putting into operation a test routine will be carried out to check whether a leakage is present. Special data, i.e. specific time thresholds and pressure thresholds, are chosen in this case.

If the check takes place on each start, i.e. independently of the fault bit, the time thresholds can be chosen to be correspondingly shorter.

In summary, the or at least one pressure control means is controlled in such a manner on starting of the engine that the pressure rises in the fault-free state. If the pressure does not rise as expected, a fault is recognised.

The steps of a method exemplifying the invention are illustrated in the flow diagram of Fig. 2. In a first step 200, a first time counter Z1 and a second time counter Z2 are set back to 0. In a second step 205, it is checked whether a fault bit F is set to 1. If the fault bit F is set to 1, this indicates that a leakage was recognised during the last operation of the engine and the engine was turned off by way of the shut-off valve 25, the pressure-regulating valve 40 and/or by driving of the injectors at the quantity zero. If the interrogation in step 205 recognises that this fault bit F is not set, the checking program ends with a step 270 and the usual start routine follows so that the engine can be started. If, however, this bit is set to 1, a test program for leakage recognition is started.

An interrogation step 210 checks whether engine rotational speed N is greater than a value N1. If this is not the case, the interrogation of step 210 is repeated at a later time. If the interrogation step 210 recognises that the rotational speed N is greater than N1, an interrogation step 215 follows, which checks whether the rotational speed N is smaller than a value N2. If this is not the case, i.e. if the rotational speed is greater than N2, the normal starting program is likewise passed over to in the step 270. The interrogation steps 210 and 215 thus check whether the rotational speed is in a window defined by the speeds N1 and N2. For example, a value of 100 revolutions per minute can be chosen for the speed N1 and a value of 300 revolutions perminute for the speed N2.

If it is recognised that the speed lies in the window represented by N1 and N2, a certain keying ration M for the driving of the valve 40 is preset in a step 200. This keying ratio is chosen so that a pressure of about 200 bars prevails in the case of a faultfree system. Subsequently, the time counter Z1 is increased by 1 in a step 225. An interrogation step 235 then follows, which checks whether the pressure P in the rail is greater than or equal to a first threshold value Pl. If this is not the case, an interrogation step 230 follows, which checks whether the content of the time counter Z1 is greater than or equal to a first threshold value S1. If this is the case, a serious fault is recognised in a step 240 and a corresponding fault report can be delivered. If the time counter has not yet reached its end value S1, the step 200 is repeated.

If the interrogation step 235 recognises that the pressure P has risen to a value greater than or equal to the first pressure threshold value P1, then a step 245 follows, in which a second keying ratio Tv2 is preset. This keying ratio for the value 40 is chosen so that a second pressure value of about 1000 bars sets in. Subsequently, the second time counter Z2 is increased by 1 in a step 5 250.

A following interrogation step 260 checks whether the pressure P in the rail is greater than or equal to a second pressure threshold value P2. If this is not the case, an interrogation step 255 checks whether the content of the time counter Z.9 is greater than or equal to a second threshold value S2. If this is not the case, the step 245 is repeated. If this is the case, i.e. the time threshold value S2 is exceeded, a fault of the sytem is recognised in a step 265. If the interrogation step 255 recognises that the pressure P is greater than or equal to the pressure threshold value P2, the normal starting program follows in the step 270.

The check takes place only when the fault bit is set, i.e. a fault was recognised during the previous operation. In this case, for example, the following values are chosen for the threshold values. For example, a time of 2 seconds is chosen for the first time threshold S1 and a time of one second for the second time threshold S2. The pressure threshold P1 is chosen to be, for example, 20 bars and the second threshold value to be 1000 bars.

It is particularly advantageous when the check takes place on each start, which means that the interrogation step 205 is omitted.

In this case, shorter time threshold values and lower pressure threshold values are to be chosen so that the start is not delayed.

It is particularly advantageous if the fault bit F is set to 1 before the first putting into operation of the engine and more strict threshold values are presettable at the same time. This means that the time thresholds are chosen to be smaller and/or the pressure threshold values are chosen to be higher. Thus, test conditions can be adapted to the conditions in service. This means that the check takes place on the first putting into operation and/or on each new start after a service.

If an emergency switching-off of the engine is performed due to a defect, a new start of the engine is possible in the presence of certain conditions. If the emergency switching-off took place due to an internal leakage, restarting could lead to appreciable engine damage. In this case, restarting of the engine is not possible. If the emergency switching-off took place due to a fault in the lowpressure region, restarting of the engine should be made possible.

For that purpose, during the starting the pressure-regulating valve 40 and/or the high-pressure pump 30 is driven so that the rises. During the starting operation, the control unit 100 the pressure build-up. The control unit compares the pressure build-up with a theoretical threshold build-up pressure observes measured computed by the unit. The computation of the theoretically reached pressure value takes place on the assumption that the most unfavourable preconditions are present in respect of the efficiency of the high- pressure pump and the modulus of elasticity of the fuel.

If the fuel pressure reached within a preset waiting time is greater than the computed one, no leakage is present. If no pressure build-up takes place, then a leakage is present or the pressure build-up was delayed by reason of a fault in the low-pressure region.

To distinguish whether a leakage or an empty tank is present, the rail pressure is observed further and after the time at which a pressure rise was recognised a new computation of the pressure rise is performed and a renewed comparison with the expected pressure value is made.

Additionally or alternatively, the rotational speed is observed. Since no injection takes place, no further rise in rotational speed may be observed. Alternatively, the rotational speed can be monitored for the maximum rotational speed of the starter.

A corresponding procedure is illustrated by the flow diagram of Fig. 3. On starting, the high-pressure conveying pump 30 and/or the pressure regulating valve 40 is driven in such a manner in a first step 300 that a maximum possible pressure build-up is achieved. In a following step 302, a time counter t is set to 0. Subsequently, the time counter t is increased by 1 in a step 304. An interrogation step 306 checks whether a waiting time tS has elapsed. If this is not the case, the step 304 is repeated.

After elapsing of the waiting time tS, the expected value PS for the pressure in the rail is computed in a step 308. This takes place according to the formula:

ts PS = Q E / V.

t = 0 In this case, the modulus of elasticity of the fuel is denoted by E, the volume flow corresponding to the conveying volume of the highpressure pump by Q and the volume of the rail by V. The volume flow Q results from the conveying performance of the high-pressure pump multiplied by the number of the revolutions of the pump, which depend on the rotational speed N.

The expected value to which the pressure PS rises in the rail is presettable in dependence on the properties of the fuel and/or the rotational speed N of the engine. In that case, only one and/or all 5 of the mentioned magnitudes could be taken into consideration.

Subsequently, the actual pressure PI in the high-pressure region is detected in step 310. An interrogation step 312 checks whether the pressure PI is greater than the expected pressure PS. If this is the case, fault-free operation is recognised in step 314 and the usual program takes place.

If this is not the case. i.e. the pressure PI is smaller than or equal to the expected pressure PS, this can have different causes. On the one hand, it is possible that a leakage is present and on the other than it is possible that the fuel tank is empty and the expected pressure value is not reached due to the empty tank.

In order to distinguish these cases, the following procedure is carried out: Subsequent to the interrogation step 312, an old pressure value PIA is written over in a step 316 by the actual value PI of the pressure. Subsequently, the actual rotational speed NA is filed in a step 318.

The new values for the pressure PIN and for the rotational speed NN are then detected in a step 322. A following interrogation step 324 checks whether the pressure PIN is greater than the old value of the pressure PIA. If this is not the case, this means that no pressure rise has taken place since the last program run, and an interrogation step 326 follows. The interrogation step 326 checks whether the new rotational speed NN is greater than the old rotational speed NA. If this is the case, this means that the rotational speed has risen, although no pressure rise has taken p l ace. This is only possible when an injection of fuel into the combustion chambers took place due to a leakage. In this case, a 5 fault is recognised in a step 330.

If the interrogation step 326 recognises no rise in rotational speed. an interrogation step 328 follows, which checks whether the engine rotational speed NN is greater than the rotational speed NS of the starter. If this is the case, a fault is recognised inthestep 330 since this is not possible in the case of fault-free operation. In the case of starting before combustion has taken place, the rotational speed NN cannot be greater than the rotational speed NS at which the starter drives the engine.

The interrogation steps 326 and 328 can take place one after the other or alternatively.

If no rise in rotational speed took place and/or the engine rotational speed NN is not greater than the rotational speed NS of the starter, then, in a step 332, the old value for the rotational speed NA is written over by the new value NN and the old value for the pressure PIA is written over by the new pressure value PIN. Subsequently, the step 322 is repeated.

If the interrogation step 324 recognises that a rise in pressure has taken place, the integrator for computation of the expected pressure PS is set to 0. At the same time, the time counter t is set back to 0. Subsequently, the time counter is increased by 1 again in a step 342. An interrogation step 344 checks whether a preset waiting time tS has elapsed. If this is not the case, the step 342 is repeated. If this is the case, then the expected pressure PS for the pressure in the rail is determined in a step 350 in corresponding manner to the step 308.

Subsequently, the detection of the actual rail pressure P11 takes place in a step 355. An interrogation step 360 checks whether the pressure PI is greater than the expected pressure PS. If this is the case, fault-free operation is recognised in the step 314. Otherwise, leakage is recognised in a step 370.

Claims

1. A method of recognising leakage in a fuel supply system of an internal combustion engine with high-pressure injection, in which system fuel is conveyed by conveying means from a low-pressure region to a high-pressure region in which the fuel pressure is controllable by pressure control means, the method comprising the steps of influencing the pressure control means during starting of the engine to cause the fuel pressure in the high-pressure region to rise to a predetermined expected value, detecting the value of the fuel pressure in the high-pressure region, and recognising a fault if the detected value fails to reach the expected value.

2. A method as claimed in claim 1, wherein said fault is recognised if the detected value fails to reach the expected value within, a predetermined time period.

3. A method as claimed in claim 1 or claim 2, comprising the further steps of influencing the pressure control means to cause the pressure in the high-pressure region to rise to a further expected value after attainment of the first-mentioned expected value, detecting the value of the fuel pressure in the high-pressure region, and recognising a fault if the detected value fails to reach the further expected value within a predetermined time period.

4. A method as claimed in any one of the preceding claims, wherein the steps of influencing, detecting and recognising are carried out on each occasion the engine is started.

5. A method as claimed in any one of claims 1 to 3, wherein the steps of influencing, detecting and recognising are carried out on each occasion the engine is started following recognition of a fault 5 during a last preceding performance of the steps.

6. A method as claimed in any one of claims 1 to 3 and 5, wherein the steps of influencing, detecting and recognising are carried out when the engine is first put into operation.

7. A method as claimed in any one of claims 1 to 3, 5 and 6, wherein the steps of influencing, detecting and recognising are carried out on each occasion the engine is started following servicing.

8. A method as claimed in claim 2 or claim 3, wherein the steps of influencing, detecting and recognising are carried out on predetermined occasions of starting the engine and said expected value and said time period, or each said expected value and each said time period as the case may be, are predetermined in dependence on when the steps are carried out.

9. A method as claimed in any one of the preceding claims, wherein said expected value, or each said expected value as the case may be, is predetermined in dependence on at least one of the characteristics of the fuel, characteristics of the conveying means, volume of a fuel feed rail of the system and engine feed.

10. A method as claimed in claim 1 and substantially as hereinbefore described with reference to the accompanying drawings.

11. Recognition means for recognising leakage in a fuel supply system of an internal combustion engine with high-pressure injection, in which system fuel is conveyed by conveying means from a low pressure region to a high-pressure region in which the fuel pressure is controllable by pressure control means, the recognition means comprising means for influencing the pressure. control means during starting of the engine to cause the fuel pressure in the highpressure region to rise to a predetermined expected value, means for detecting the value of the fuel pressure in the high-pressure region and means for recognising a fault if the detected fault fails to reach the expected value.