EP0127510A1 - Method of interrupting fuel injection into an internal-combustion engine during deceleration - Google Patents

Abstract

According to this method, the fuel injection is interrupted in response to the detection of a deceleration and the injection is commanded again when a threshold linked to the engine speed is reached.

For this purpose, the speed (N) or the rotation period (T) of the motor is measured over consecutive equal intervals or angles of rotation, the difference (Δ T, Δ N) between the value of the last speed or period measured and the value seen through a high-speed speed filter or period measured at the previous interval, the said difference (Δ T, Δ N) is compared to a fixed threshold (S) and, if a deceleration condition of the engine is detected, the fuel injection is again ordered when the difference (Δ T, Δ N) is greater than the threshold (S).

Application: to internal combustion engines with positive ignition for motor vehicles.

Description

The present invention relates to a method for cutting off the injection of fuel in the deceleration phases of an internal combustion engine.

With electronically controlled fuel injection systems, whether of the single-point or multi-point type, it is possible to improve the average consumption of the engines by cutting the fuel injection in areas of strong deceleration .

This fuel injection cutoff can be triggered in response to the closing of an electrical switch indicating the closed position of the engine air flow control throttle, upon detection of a pressure threshold prevailing at the manifold. intake, upon detection of a predetermined limit injection time or the like.

The fuel economy achieved will be greater the longer the injection cut-off. It is therefore desirable to put the injection system back into service, that is to say injecting fuel again, at an engine speed which is as low as possible.

This is not without difficulties because it is noted that the drop in speed of an unloaded engine is much greater when it is not powered. It therefore results from the deceleration cut-off procedure a high risk of engine stalling when the injection return to service regime, hereinafter called "revalorization speed or threshold", is too low. This phenomenon is noticeable after an acceleration when idle (particularly in the presence of a power-assisted steering) or when the driver disengages after a long deceleration under load.

This difficulty can be encountered with multi-point injection systems (one injector per cylinder), but it is greatly amplified with a single-point injection system comprising a single injector positioned above a throttle arranged in a body similar to that of a carburetor to control the air flow of the engine. There is indeed a significant delay between the order to resume injection and the time when a cylinder has received enough fuel to ensure proper combustion.

Patent GB-A-2,062,295 attempts to provide a solution to this problem by means of a process according to which the injection of fuel is interrupted in response to the detection of a deceleration, the period of rotation of the engine over consecutive equal intervals or angles of rotation, calculate the difference between the period measured over the last interval and the period measured over the previous interval, compare this difference to a fixed threshold to determine whether the engine is in slow or rapid deceleration and the fuel injection is again commanded at one or other of the two fixed thresholds of engine speed depending on whether the engine is decelerating slowly or rapidly.

However, such a method is not reliable due to the existence of rapid variations in the instantaneous period of the engine which are often fleeting and of mechanical origin, such as play in the transmissions, problems of engine suspension, or even any other phenomenon generating instantaneous and brief speed variations. This can result in false detections of slow or rapid deceleration conditions which can result, either in a premature "re-coupling" if a rapid deceleration has been detected, or on the contrary by an engine stalling in the contrary case . Furthermore, the use of two fixed "re-coupling" thresholds instead of one is a compromise solution which does not completely eliminate the risks of stalling the engine and does not allow the engine cut-off process to be optimized. bus injection burant in deceleration.

The invention aims to eliminate these drawbacks by means of a fuel injection cut-off method for an internal combustion engine with electronically controlled injection which ensures satisfactory engine operation in terms of quality and driving pleasure, while making it possible to maximize savings in consumption by anticipating the re-coupling phenomenon when there is a risk of engine stalling.

To this end, the subject of the invention is a method of cutting off the fuel injection in the deceleration phases of an internal combustion engine, according to which the fuel injection is interrupted in response to the detection of a deceleration and fuel injection is re-ordered when a threshold linked to the engine speed is reached, characterized in that the speed (N) or the period of rotation (T) of the engine is measured at intervals or angles of consecutive equal rotations, we calculate the difference (Δ N, Δ T) between the speed or period (X, T) measured on the last intern angular valle and the value ( _n , _n ) seen through a pass filter at the top of the speed or of the period (N _n-1 , T _n-1 ) measured over the previous angular interval, the said difference (AN, ΔT) is compared to a fixed threshold (S) and, if a deceleration condition of the engine is detected, fuel injection is again ordered when the difference (6 N, Δ T) is greater than the threshold ( S).

• - la figure 1 est un graphique montrant la variation du régime d'un moteur à injection d'essence en fonction du temps dans le cas d'une décélération en charge suivie d'une décélération à vide ;
• - la figure 2 est un schéma-bloc d'un système d'injection mono-point pour la mise en oeuvre du procédé de coupure de l'injection suivant l'invention ; et
• - la figure 3 est un algorithme illustrant le procédé suivant l'invention.

Other characteristics and advantages of the invention will emerge from the description which follows of an example of implementation illustrated by the appended drawings in which:

• - Figure 1 is a graph showing the variation of the speed of a petrol injection engine as a function of time in the case of a deceleration under load followed by a no-load deceleration;
• - Figure 2 is a block diagram of a single-point injection system for implementing the injection cut-off method according to the invention; and
• - Figure 3 is an algorithm illustrating the method according to the invention.

The graph in FIG. 1 shows an example of evolution of the speed N (t _{r /} min) of an internal combustion engine with petrol injection and with controlled ignition of a motor vehicle. Between instants t ₀ and t ₁ the engine speed is constant and, from time t _{1 ′} , a deceleration under load occurs following a lift of the accelerator pedal from the driver of the vehicle. At time t ₂ it disengages and it is noted that the deceleration of the motor is then much more accentuated than between the instants t ₁ and t ₂ .

The injection of petrol having been cut off at the instant t of the foot lift of the accelerator pedal, this injection is returned to service at a speed N _S called "re-coupling threshold", at the instant t ₃ . The accelerator pedal being supposed to be always released, that is to say the throttle closed, the engine speed then decreases gradually until the engine idle speed N _R.

This graph in FIG. 1 shows that the re-coupling threshold N _S must be both high to avoid stalling the engine during large idle decelerations and as low as possible to maximize consumption gains during the lowest decelerations under load where the risk of stalling is not as high.

This leads to the adoption of a variable N _{S re-} coupling threshold as a function of the instantaneous engine operating conditions. One solution would consist in comparing the derivative of the speed or the period of the engine calculated over an angular rotation interval with a threshold which itself is a function of the speed. However, this solution is relatively complex and cumbersome to implement.

To solve this problem, the invention consists, when a deceleration condition has been detected and the fuel injection cut off, continuously comparing to a fixed threshold S the difference between the speed or period of rotation of the engine measured on the last angular interval and the speed or period measured at the previous angular interval seen through a high-pass filter, and to order fuel injection again when the difference is greater than the threshold S. The angular interval chosen will be preferably equal to that which separates the passage of the pistons through top dead center, namely a half-turn in the case of a four-cylinder engine.

FIG. 2 describes a single-point injection system for implementing the method according to the invention. A similar injection system is described in more detail in French patent application 82-13996 to which reference may be made, it being understood that the invention is not limited to such a system.

Briefly, the single-point injection system shown comprises a programmed microcomputer 1 such as, for example, the microprocessor MC 6801, which receives on an input the value of the pressure in the intake manifold of the engine supplied by a sensor of pressure 2 and converted into digital form by an analog / digital converter 3. A position sensor 4 detecting the passage of the teeth which is provided at its periphery with a target fixed on the crankshaft of the engine and rotating in synchronism with it delivers a signal which is shaped and processed in a block 5. Block 5 produces a synchronization signal S identifying the passage of each cylinder through a top dead center and used by the microcomputer 1 to calculate the speed of the engine. The acquisition and calculation of these "pressure" and "speed" parameters and a map which it has in memory allow the microcomputer 1 to calculate the injection time "t. Calculated" according to an appropriate algorithm. Thanks to the synchronization signal Sy and its internal clock, the microcomputer 1 controls, in the vicinity of the top dead center, the opening of an injector 7 via a power circuit 6.

In addition, the computer 1 is connected by one of its inputs to an electrical throttle switch 8 which closes when the engine throttle valve is in the closed position, that is to say that the driver has his foot lifted from the accelerator pedal.

We will now refer to FIG. 3 which represents an algorithm which can be inserted into the main program of the injection computer 1 to implement the injection cut-off method. As known per se, this program as well as the cut-off algorithm takes place with a periodicity depending on the number of cylinders of the engine, that is to say at each half-turn in the case of a four-cylinder engine, which will be assumed to be the case later.

The first step 10 is a test on the speed N of the engine which is compared to a limit speed N ₁ . In fact, the injection is never cut off if the rotation speed is lower than this limit speed N ₁ , for example 1100 rpm; in this case, the injection time t _i is then calculated by a conventional subroutine 11 as a function, in particular, of the speed and the pressure, as indicated above.

If the speed N is greater than the limit speed N ₁ , a test is carried out at 12 to determine whether or not the engine is decelerating. The response to this test is given by the state of the contactor 8 (opening or closing), or by any other appropriate means.

If a deceleration is not detected, we go to subroutine 11 and, if not, we calculate and store the time T n taken by the motor to make the last half-turn (step 13).

expression dans laquelle :

• - P est l'opérateur de Laplace
• -
  
  est une constante de temps de filtrage de valeur prédéterminée.

The next two steps 14a and 14b consist of performing calculations to compare the threshold S the difference between the period measured over the last angular interval and the period measured at the previous angular interval seen through a high-pass filter, which allows to get rid of problems of parasitic speed variations in a simple manner. Considered as a Laplace transform, this comparison is expressed in the form:

expression in which:

• - P is the Laplace operator
• -
  
  is a filter time constant of predetermined value.

dans laquelle :

• - T_n-1 est la période mesurée à l'intervalle angulaire précédent ;
• - T_n-1 est la valeur de la période vue à travers un filtre passe-haut à l'intervalle angulaire précédent ;
• - α est une constante programmable de valeur prédéterminée ;

Step 14a consists in calculating the value T _n seen through a high-pass filter of the period measured over the previous angular interval and expressed by the relation:

in which :

• - T _n-1 is the period measured at the previous angular interval;
• - T _n-1 is the value of the period seen through a high-pass filter at the previous angular interval;
• - α is a programmable constant of predetermined value;

A l'étape 15, on compare la différence Δ T avec un seuil fixe S :

• - si A T est inférieur au seuil S on maintient la coupure de l'injection (t. _i = 0 à l'étape 16) et on passe à la suite du programme représentée sous forme du bloc 17 ;
• - si Δ T est supérieur ou égal au seuil S, on reprend l'injection de carburant. Pour cette première injection suivant une coupure, on engendre de préférence un temps d'injection constant K (étape 18), nettement supérieur, par exemple de trois à cinq fois, à celui qui serait normalement calculé par le microcalculateur en fonction des paramètres instantanés du moteur, sachant que le papillon est alors fermé et que le régime de réattelage N_S, bien que variable, est compris en pratique dans une certains fourchette, de l'ordre de 1 300 à 1 500 tr/mn.

In step 14b, the above-mentioned difference Δ T is calculated which is such that:

In step 15, the difference Δ T is compared with a fixed threshold S:

• - if A T is less than the threshold S, the cut-off of the injection is maintained (t. _i = 0 in step 16) and we move on to the program shown in the form of block 17;
• - if Δ T is greater than or equal to the threshold S, the fuel injection is resumed. For this first injection following a cut-off, preferably a constant injection time K (step 18) is generated, clearly greater, for example three to five times, than that which would normally be calculated by the microcomputer as a function of the instantaneous parameters of the engine, knowing that the throttle valve is then closed and that the N _S realignment speed, although variable, is included in practice in a certain range, of the order of 1,300 to 1,500 rpm.

It is known that, during a cold start or after a cut in deceleration, part of the injected fuel tends to deposit on the walls of the intake manifold. The injection of a quantity of fuel significantly greater than that theoretically necessary to ensure combustion for the given engine conditions, during the first injection stroke following a cut, makes it possible to remedy this phenomenon of wetting of the walls of the manifold d 'admission.

dans laquelle X représente soit la période T, soit la vitesse ou régime N.Of course, the invention is not limited to the embodiments which have just been described and numerous modifications can be made thereto without departing from its scope. Thus, for example, we can calculate the variation in speed AN instead of the variation in period Δ T from one angular interval to the next and compare AN to a fixed threshold. It is also possible to carry out tests 10 and 12 on the limit regime N ₁ and the detection of a deceleration after the calculation of the period T or of the regime N at the preceding interval, or after the calculation of A T or Δ N. The relation (1) can therefore be expressed in the general form:

in which X represents either the period T, or the speed or speed N.

Finally, it should be noted that when the injection is resumed following a cut in deceleration, it is possible to generate several shots of injection of duration greater than that theoretically necessary for ensuring combustion for the given conditions of the engine.

Claims (4)

1. Method for cutting off the fuel injection in the deceleration phases of an internal combustion engine, according to which the fuel injection is interrupted in response to the detection of a deceleration and the injection is commanded again of fuel when a threshold linked to the engine speed is reached, characterized in that the speed (N) or the period of rotation (T) of the engine is measured at consecutive equal intervals or angles of rotation, the difference is calculated (Δ N, Δ T) between the speed or period (N, T) measured over the last angular interval and nn the value ( NOT _not, T _n ) viewed through a high-pass filter of the speed nn or of the period (N _{n-1 '} , T _n-1 ) measured over the previous angular interval, the said difference (Δ N, A T) is compared to a fixed threshold (S) and, if an engine deceleration condition is detected, fuel injection is again ordered when the difference (Δ N, Δ T) is greater than the threshold (S). 2. Method according to claim 1, characterized in that the value ( NOT _n , T _n ) viewed through a high-pass filter of the speed or the period (N _{n-1 '} , T _n-1 ) measured over the previous angular interval is calculated in accordance with the relation:

in which : - X _n-1 is the speed (N) or the period (T) measured at the previous angular interval; - X _n-1 is the value of the speed or the period seen through a high-pass filter at the previous angular interval; and - α is a programmable constant of predetermined value. 3. Method according to any one of claims 1 and 2, characterized in that a condition of deceleration of the engine is detected by reading the state of an electrical contactor (8) of the throttle valve and the switching off of fuel injection when, on the one hand, the contactor is in a state representative of the closed position of the throttle and that, on the other hand, the engine speed is greater than a limit speed (N ₁ ). 4. Method according to any one of claims 1 to 3, characterized in that, when said difference (Δ T, Δ N) is greater than the fixed threshold (S), a first injection shot of fixed duration is ordered and greater than the injection time strictly necessary to theoretically ensure combustion under instantaneous engine operating conditions.

Images