A method for counteracting pre-ignition events in a spark-ignition internal combustion engine

Abstract

There is described a method for counteracting a pre-ignition event in a spark-ignition internal combustion engine, the method comprising: - detecting a pre-ignition event (MK) in at least one cylinder of said internal combustion engine, - controlling, for each cylinder of said internal combustion engine for which a pre-ignition event (MK) has been detected, an increase of a combustion relative air-to-fuel ratio value (λ L) with respect to a combustion relative air-to-fuel ratio value (A c) in force at the time of the pre-ignition event (MK).

Description

[0001] A method for counteracting pre-ignition events in a spark-ignition internal combustion engine

[0002] TEXT OF THE DESCRIPTION

[0003] Field of the Invention

[0004] The present invention refers to internal combustion engines . More speci fically, the invention was developed with reference to combustion control in spark-ignition engines .

[0005] Known Art

[0006] In spark-ignition engines it is not unusual to encounter soiling of the combustion chamber due to long periods of idling operation and / or repeated cycles of cold start and stop, this not being followed by a period of medium / high load operation, the latter being a condition which burns the deposits and eliminates or reduces soiling . Soiling may cause the appearance of one or more hot spots which may generate multiple preignition events in the combustion chamber .

[0007] With reference to Figure 1 , there is shown a multiple diagram illustrating di f ferent operating parameters of a spark-ignition internal combustion engine , including : a maximum pressure in each cylinder PC, MAX ( overlapping curves in a number corresponding to the number of cylinders ) ;

[0008] - a boost pressure PBOOS ;

[0009] - a current relative air-to- fuel ratio ( air-to- fuel ratio / stoichiometric air-to- fuel ratio ) Ac, in combination with a reference " 1" indicating a unit value ;

[0010] - a time axis ( time ) with data shown in seconds ( s ) ;

[0011] - a signal D indicative of the occurrence of a preignition event ; the diagram in Figure 1 indicates a logic state of the signal : 0 = false (no pre-ignition) , 1 = true (pre-ignition) . The signal D is based, i . a . , on a signal from one or more knock sensors ( generally accelerometers ) on board the engine .

[0012] As can be observed in the diagram of Figure 1 , at the detection of one or more pre-ignition events MK - corresponding to maximum pressure peaks PC, MAX in the cylinder af fected by the pre-ignition event , and being associated to peaks MK_D in the signal D, which correspond to the switching of the logic state from 0 to 1 - the engine electronic control responds by controlling an enrichment of the mixture , reducing the relative air- to- fuel ratio to the value XR, which is lower both than one unit and than the air-to- fuel ratio values which are in force in operating conditions without anomalies .

[0013] However, such an intervention is often insuf ficient to eliminate all the pre-ignition events . In the latter case , the further intervention performed by the engine electronic control comprises a fuel cut-of f , which is exempli fied - in terms of air-to- fuel ratio , which approaches values higher than a unit - by reference CT , which is generally accompanied by a reduction in the air flow rate into the engine .

[0014] The most evident problem in the known strategy for counteracting pre-ignition, shown and commented with reference to Figure 1 , is that some of the interventions are clearly perceived by the driver . More speci fically, i f on one hand the mixture enrichment is not perceived by the driver, the fuel cut-of f and the air flow rate reduction are instead clearly perceived, resulting in an unpleasant driving feeling .

[0015] Obj ect of the Invention

[0016] The invention aims at solving the technical problems outlined in the foregoing . More speci fically, the invention aims at ef fectively counteracting the pre- ignition events in an internal combustion engine , while preventing such counteracting interventions from being perceived by the driver .

[0017] Brief Description of the Figures

[0018] The invention will now be described with reference to the annexed Figures , which are provided merely by way of non-limiting example and wherein :

[0019] - Figure 1 , already described in the foregoing, is a multiple diagram showing various operating parameters of an internal combustion engine on the occurrence of a pre-ignition event , and the consequent counteracting intervention performed according to the known art , and

[0020] - Figure 2 is a multiple diagram showing various operating parameters of an internal combustion engine on the occurrence of a pre-ignition event , and the consequent counteracting intervention performed according to the invention .

[0021] Detailed Description

[0022] Referring to Figure 2 , a method for counteracting a pre-ignition event in a spark-ignition internal combustion engine generally comprises :

[0023] - detecting a pre-ignition event in at least one cylinder of the internal combustion engine ,

[0024] - controlling, for each cylinder of the internal combustion engine for which a pre-ignition event has been detected, an increase of a combustion relative air- to- fuel ratio Xc with respect to the combustion relative air-to- fuel ratio in force at the time of the pre- ignition event .

[0025] Figure 2 shows the method according to the invention by means of time diagrams of the parameters PC, MAX , PBOOS , AC, D . They are complemented by a time diagram of a further value INJ_D, corresponding to a duration in time of a fuel inj ection stage into the engine cylinders . The diagram in Figure 2 shows , in thicker lines , a curve INJ_D, 3 of a trial engine with a 6-cylinder V configuration and boosting via a turbocharger, and referred to cylinder no . 3 , within which a pre-ignition event is taking place . The diagram also includes the duration values DI and D2 , with Dl> D2 .

[0026] At time tl a pre-ignition event takes place , which is detected by the engine electronic control . The detection of a pre-ignition event is preferably performed on the basis of knock sensors ( accelerometers ) installed on the engine crankcase , but it is possible - in combination therewith or as an alternative - to detect a pre-ignition event on the basis of pressure information within the cylinders . The two detection modes will be described separately in the following, it being understood that the mode based on the use of pressure sensors ( and thus on pressure data ) within the cylinder may be used both individually and in test bench conditions , for calibrating the knock sensors and the related determination model of the knock and pre- ignition events (which corresponds to the first mode described in the following) , and also in combination with the strategy based only on the knock sensors , as a measure for making the data more robust .

[0027] Detection mode / I : Knock sensors

[0028] According to this mode , the detection of the occurrence of a pre-ignition event in a cylinder of the engine is possible thanks to a knock sensor ( accelerometer ) which is generally mounted on the engine crankcase .

[0029] The vibrations triggered by the pre-ignition event are transmitted across the material of the crankcase to the knock sensor, the knock sensor converts the (mechanical ) signal resulting from the vibrations into an electrical signal which is acquired by an engine control unit .

[0030] However, the knock sensor also collects and converts all the accelerations due to the normal operation of the engines , such as valve impacts against their respective seats , regular combustion, the operation of inj ectors , pumps , etc .

[0031] In order to correctly associate the presence and the severity of the combustion with pre-ignition, there is performed a post-processing of the raw signal exiting the knock sensor .

[0032] The signal post-processing is burdensome both from the point of view of the parameter calibration, which may be assisted - on the test bench - by a validation through the pre-ignition detection according to the following mode 2 , and as regards the computational burden for the engine control unit , since it is necessary to define observation windows and the knock sensor signal must undergo filtering and integration as a function of the rotational speed and of the load of the engine .

[0033] The pre-ignition check based on the signal of the knock sensor, which has been post-processed by the engine control unit , provides as output data an index obtained on the basis of a ratio between the real instantaneous combustion noise and the background combustion noise .

[0034] The index is compared with a threshold for detecting a combustion with pre-ignition, and the correction is based on the intensity of the knock sensor signal which is post-processed by the engine control unit . Of course , this applies both to the case of a single knock sensor and to the case of a plurality of knock sensors ( e . g . one for each stand) .

[0035] To sum up, mode 1 envisages operating according to the following sequence of operations :

[0036] - acquiring a signal from one or more knock sensors of the engine , filtering the signal from one or more knock sensors of the engine by eliminating the components thereof originating from phenomena other than combustion, obtaining, from the filtered signal , a first signal representative of a background combustion noise and a second signal representative of an instantaneous combustion noise ,

[0037] - defining a pre-ignition index on the basis of a ratio between the second signal and the first signal ( in other words , instantaneous combustion noise / background combustion noise ) ,

[0038] - indicating the occurrence of a pre-ignition event (MK) when the ratio between the second signal and the first signal is greater than a corresponding threshold value .

[0039] Detection mode / 2 : Pressure sensors / information within the cylinder ( each cylinder )

[0040] In this mode (which, for brevity, will be named "mode 2" in the following ) , the detection of the occurrence of a pre-ignition event in a cylinder of the engine ( at least one cylinder , i . e . in each cylinder wherein the event takes place ) comprises :

[0041] - acquiring, as a function of time or of a crank angle , combustion pressure values within each cylinder,

[0042] - comparing a maximum absolute value PC, MAX of the combustion pressure in the cylinder with a first threshold value , in particular a maximum admissible value of the combustion pressure in the cylinder,

[0043] - calculating a maximum absolute value of a filtered pressure signal , ( the indicator is referred to as MAPO, Maximum Absolute Pressure Oscillation) , and comparing the maximum absolute value (MAPO) with a second (maximum) threshold value ,

[0044] - determining a crank angle MFB10% (wherein MFB means "Mass of Fuel Burnt" ) at which combustion of an amount equal to 10% of the mass of fuel supplied to the cylinder has occurred, and comparing said crank angle MFB10% with a third (minimum) threshold value .

[0045] The occurrence of a pre-ignition event , which is summari zed in the logic state of signal D, is indicated when the first threshold value and the second threshold value are exceeded, and when the value of MFB10% is less than the third threshold value ( this is due to the fact that , in the case of pre-ignition, the value of the angle where the 10% combustion of the mass of fuel supplied to the cylinder takes place is very small ) .

[0046] I f the determination of the occurrence of a pre- ignition event takes place by means of a combination of the modes 1 and 2 , the occurrence of a pre-ignition event is determined when - with reference to mode 2 - the first and the second threshold values are exceeded, and when the value MFB10% is less than the third threshold value , and when - with reference to mode 1 - the ratio between the second signal and the first signal is greater than a corresponding threshold value ( in other words , when all conditions (AND) are met which are envisaged by both mode 1 and mode 2 for detecting a pre-ignition event ) .

[0047] As already stated with reference to Figure 1 , the pre-ignition event - which in this case regards cylinder no . 3 of said engine - manifests itsel f with a maximum pressure peak MK PC, MAX which occurs at time tl , and which is accompanied by a peak in the profile of the signal D ( logic state switching from " 0" to " 1" ) . Once the pre- ignition event has been detected, it may be observed that an increase is controlled of the relative air-to- fuel ratio with respect to the value Xc in force at the time of the pre-ignition event . Always referring to cylinder no . 3 , the relative air-to- fuel ratio is increased up to a value XS , L , which is preferably higher than the value 1 . 1 . Generally speaking, the value Xc in force at the time of the pre-ignition event is a value lower than a unit (which therefore indicates a rich mixture) , especially in a supercharged engine such as the engine mentioned herein. Generally speaking, moreover, if the engine is supercharged the increase of the relative air-to-fuel ratio XS,L will have a value outside the interval 0.95-1.1, since said combustion air-to-fuel values would lead to abnormal combustion events, which would jeopardize the structural integrity of the engine components.

[0048] Always referring to Figure 2, the increase of the relative air-to-fuel ratio up to the value XS,L preferably takes place by reducing the duration of the injection stage with respect to the duration in force at the time when the pre-ignition event takes place. In the diagram of Figure 2, the duration decreases from a value DI, in force at the time tl, to a value D2 lower than the value DI. This corresponds to the supply of a smaller amount of fuel to the engine which, with the same amount of supplied air, results in a leaner fuel / air mixture. In this regard, it will be observed that the boost pressure PBOOS remains substantially unchanged across the preignition event, and therefore the air flow rate also remains substantially unchanged.

[0049] After an interval having a duration exceeding a threshold duration (which may be calibrated in terms of engine cycles) without the occurrence of new preignition events in the cylinder (cylinder no. 3 in the present case) , the value of the combustion relative air- to-fuel ratio is restored to the value of combustion relative air-to-fuel ratio Xc in force in the cylinder at the time of the pre-ignition event. The time interval associated to restoring the combustion air-to-fuel ratio value in force at the time of the pre-ignition event is exemplified by a time t2, which follows the time tl (i.e. by a time interval t2-tl) : the relative air-to-fuel ratio is reduced by prolonging the duration of the fuel inj ection stage , and consequently by enriching the mixture . The air flow rate supplied to the cylinder remains , also in this case , substantially constant , since the boost pressure PBOOS remains constant as well .

[0050] The main technical advantages achieved by the method according to the invention may be summari zed in the combination of imperceptibility for the driver and of a high ef ficiency in contrasting pre-ignition events . Moreover, the method may be implemented individually for each cylinder, without modi fying the combustion air supply system, since it is possible to simply act on the amount of inj ected fuel . Of course this does not exclude , based on the features of the combustion air supply system ( including the presence of variable timing or variable li ft systems of the intake valves ) , performing the increase of the combustion relative air-to- fuel ratio by acting on the flow rate of the air supplied to the engine , speci fically by increasing it .

[0051] As a comparison, let us consider the following Tables no . 1 and no . 2 relating to tests performed on the reference engine described herein ( such Tables correspond to the diagrams of Figures 1 and 2 ) . The reference XL must be construed as generic with respect to the cylinder, and the meaning thereof is equivalent to XL, 3 •

[0052] Table 1

[0053] Table 2 Thus , the test data show that the method according to the invention, which substantially consists in an opposite strategy with respect to what is implemented in the known art , has a largely superior ef ficiency than the known methods based on enriching the mixture . In the case of the test provided herein, which consists in an acceleration ramp ( starting from soiling conditions ) to reach a full-load operating point at 5000 rpm, while maintaining the engine point achieved ( 5000xWOT ) for 120 seconds , the number of pre-ignition events counteracted only by an intervention on the air-to- fuel ratio was equal to the total number approximately in all the cylinders of the engine , as opposed to a maximum counteracting ef ficiency that only reaches 75% (with an average of 42 % ) in the known methods based on mixture enrichment .

[0054] Of course, the implementation details and the embodiments may amply vary from what has been described and illustrated without departing from the extent of the present invention, as defined by the annexed claims .

Claims

CLAIMS1 . A method for counteracting a pre-ignition event in a spark-ignition internal combustion engine , the method comprising :- detecting a pre-ignition event (MK) in at least one cylinder of said internal combustion engine ,- controlling, for each cylinder of said internal combustion engine for which a pre-ignition event (MK) has been detected, an increase of a combustion relative air-to- fuel ratio ( XL ) with respect to a combustion relative air-to- fuel ratio (Xc) in force at the time of the pre-ignition event (MK) .2 . The method according to claim 1 , wherein said controlling an increase of a combustion relative air-to- fuel ratio ( XL ) comprises increasing the combustion relative air-to- fuel ratio value ( XL ) to a value equal to or greater than 1 . 1 .3 . The method according to claim 1 or claim 2 , wherein said detecting a pre-ignition event (MK) in at least one cylinder of said internal combustion engine comprises :- acquiring a signal from one or more knock sensors of said engine , filtering said signal by eliminating the components thereof originating from phenomena other than combustion obtaining, from the filtered signal , a first signal representative of a background combustion noise and a second signal representative of an instantaneous combustion noise- defining a pre-ignition index on the basis of a ratio between said second signal and said first signal , and- indicating the occurrence of a pre-ignition event (MK) when the ratio between said second signal and saidfirst signal is greater than a corresponding threshold value .4 . The method according to any one of the preceding claims , wherein said detecting a pre-ignition event (MK) in at least one cylinder of said internal combustion engine comprises :- acquiring, as a function of time or a crank angle , pressure values within each cylinder, in particular combustion pressure , comparing an absolute value of combustion pressure in the cylinder with a first threshold value , in particular a maximum admissible value of said combustion pressure in the cylinder,- calculating a maximum absolute value of a f iltered pressure signal (MAPO) , and comparing said maximum absolute value with a second threshold value , determining a crank angle MFB10% at which combustion of an amount equal to 10% of the mass of fuel supplied to the cylinder has occurred, and comparing said crank angle MFB10% with a third threshold value , indicating the occurrence of a pre-ignition event (MK) when the first threshold value and the second threshold value are exceeded, and when the crank angle MFB10% is less than the third threshold value .5 . The method according to claim 4 , comprising indicating the occurrence of a pre-ignition event (MK) when the first threshold value , the second threshold value and when the crank angle MFB10% is less than the third threshold value , and further when the ratio of said second signal to said first signal is greater than the corresponding threshold value .

6. The method according to claim 3 , wherein said one or more knock sensors comprise one or more accelerometers installed on an engine crankcase .7 . The method according to any one of the precedingclaims, comprising restoring the value of the combustion relative air-to-fuel ratio to the value in force in the cylinder (XL) at the time of the pre-ignition event after an interval having a duration exceeding a threshold duration without the occurrence of new pre-ignition events (MK) in the cylinder.

8. The method according to any one of the preceding claims, wherein said controlling an increase in a combustion relative air-to-fuel ratio (XL) comprises reducing a duration of a fuel injection stage (INJ_D) .

Images