MXPA00008401A - Process for detecting a misfire in an internal combustion engine and system for carrying out said process - Google Patents

Abstract

A process for detecting a misfire in one or more cylinders (3, 3') of an internal combustion engine, including the following operative steps:sampling the exhaust gas pressure values during at least one engine cycle, the sampling frequency being proportional to the crankshaft rotational speed;analyzing the sampled signal in the frequency domain;calculating a misfire index as a function of the results of said analysis;comparing said index with one or more threshold values. Said frequency domain analysis preferably includes a Fourier transform of the sampled signal. The present invention also relates to a system carrying out said process.

Description

PROCEDURE TO DETECT A FAILURE OF IGNITION IN AN INTERNAL COMBUSTION ENGINE AND SYSTEM TO CARRY OUT SAID PROCEDURE DESCRIPTIVE MEMORY The present invention relates to a method for detecting an ignition failure in an internal combustion engine and in particular, a method that can be used to detect an ignition failure in one or more cylinders of an internal combustion engine. The present invention also relates to a system for carrying out said method. It is known that to monitor the performance of an internal combustion engine, in particular a racing engine with a large number of cylinders, it is convenient to detect the occurrence of the ignition failure of the fuel mixture in one or more cylinders. A method for carrying out said detection, which is known from US-5576936 and which currently plays an important role with respect to the increasingly stringent rules for controlling the escape of contaminants, consists in measuring the sudden fluctuations in the rotational speed of the crankshaft by means of an electronic sensor located near the steering wheel. This sensor is connected to the control unit placed inside the car, which receives all the information related to the engine and transmitted through suitable sensors. When calculating the fluctuations in the speed according to the torque applied, it is possible to identify a possible misfire on an engine cylinder. However, this procedure does not allow to identify precisely in which cylinder the ignition failure occurred and also has a high probability of error, particularly in the case of the automobile that moves and is subjected to intense variations, for example, caused by defects on the road surface, which temporarily affects the rotational speed of the crankshaft. In order to overcome these advantages, US-5109825 devised to measure the fluctuations in time of the exhaust gas pressure of the engine. Although the pressure sensors available on the market are very accurate and provide an almost real-time response, the known procedures to detect the ignition failure based on the measurement of pressure fluctuations in the exhaust gas, remain very imprecise and poorly reliable, particularly when applied to engines with a high number of cylinders. Therefore, the object of the present invention is to provide a method for detecting ignition failure that is free from the aforementioned disadvantages. Another object of the present invention is to provide a system which carries out said procedure. These objects are achieved by means of a procedure and a system whose main characteristics are described in the first and eighth claim, respectively.

Thanks to the sampling and subsequent frequency analysis of the pressure values detected in the exhaust pipes, the method according to the present invention provides superior precision and reliability with respect to prior art processes. In fact, if the ignition of the engine is regular, the periodic openings of the exhaust valves of the cylinder generate pressures of pressure in the exhaust pipes that have the same periodicity and similar waveforms. On the contrary, in the case of misfire on one of the cylinders, the corresponding pressure pulsation is modified, thereby changing the periodic pattern of the pressure values. The reference for the synchronization with the pulse frequency is easily deduced from the sensors that detect the rotational speed of the crankshaft and / or camshaft. Another advantage of the method according to the present invention is that through the sequence analysis of the sample signal it is possible to determine if only one or more ignition failures occurred during a single motor cycle. In fact, the amplitude of the module of the different harmonics of the sample signal depends on the number of cylinders where the ignition failure occurred. An additional advantage of the method according to the present invention is that through the frequency analysis of the sample signal it is possible to determine not only the ignition failure, but also the position of the cylinder where it occurred. In fact, knowledge of the firing sequence of the cylinder and comparison of the phase of the first harmonic of the sample signal with the phase of the first cylinder, provides a phase difference which indicates the position of the cylinder where the power failure. These and other advantages and features of the method and system according to the present invention will be clear to those skilled in the art from the following detailed description of one embodiment thereof, with reference to the accompanying drawings in which: 1 shows a schematic view of the system according to the present invention; Figure 2 shows a flow diagram of the method according to the present invention; Figures 3a, 3b and 3c show three diagrams of the pressure as a function of the rotation of the crankshaft; Figures 4a, 4b and 4c show three further diagrams of the pressure as a function of the rotation of the crankshaft; Figures 5a, 5b and 5c show three diagrams of the ignition failure index as a function of the number of engine cycles; Figure 6 shows a Fourier transform of the diagram of Figure 3a; and Figures 7a, 7b and 7c show three polar coordinate diagrams of the main harmonic of the pressure in the diagrams in Figures 3a, 3b and 3c.

With reference to Figure 1, it is noted that the system according to the present invention includes in a known manner a control unit 1 (indicated by a dotted line) which in turn includes a pair of mutually connected electronic controllers 2 , 2 'each of which provides control over one of the two rows of cylinders 3, 3' of the engine. In this embodiment, a V12 engine having two rows of six cylinders 3, 3 'each is described, but in other embodiments, the number of cylinders and / or rows can obviously change. The controllers 2, 2 'are connected in a known manner to a pair of cooling temperature sensors 4, 4' and to two pairs of sensors 5, 5 'and 6, 6' respectively, which detect the temperature and pressure of the air in the admission multiples 7, 7 '. The controllers 2, 2 'are also connected to a pair of lambda sensors 8, 8' to analyze the oxygen content in the exhaust pipes 9, 9 ', to two series of injectors 10, 10' which inject the fuel towards the intake pipes 11, 11 'of the cylinders 3, 3', as well as a pair of ignition coils 12, 12 '. The exhaust pipes 9, 9 'are preferably further provided with a pair of temperature sensors 13, 13' connected to the controllers 2, 2 '. The system according to the present embodiment of the invention suitably includes a sensor 14 which detects the rotational speed of the steering wheel 15 forming an assembly with the crankshaft and an additional pair of sensors 16, 16 'that detect the rotation of the camshaft 17. These sensors 14, 16 and 16 'are connected to the controllers 2, 2' so that the latter, based on the received data, can calculate in real time the speed and angle of rotation of the crankshaft during a motor cycle. . The presence of the sensors 14, 16 and 16 'is made necessary by the fact that the flywheel 15 in a four-stroke engine makes two revolutions (720 °) per cycle, so the reference provided by the sensors 16, 16 'allows to distinguish the first revolution from the second. In order to carry out the process according to the present invention, in the two exhaust pipes 9, 9 'there are two high-precision pressure sensors duly arranged 18, 18' connected to the controllers 2, 2 ', said sensors transmitting in real time an electrical signal whose voltage is proportional to the measured pressure. In addition, the controllers 2, 2 'are connected to a pair of warning lights 19, 19' placed inside the automobile, to a port 20 for connection to an external processor, as well as to a sensor 21 which detects the position of the regulator. 22. Referring now to Figure 2, it is noted that the process according to the present invention includes, after a certain period from the start of the engine, a first step of periodic verification, for example every second, of the operating status of the engine. In fact, in order to obtain reliable results from this procedure, it is preferred that it only be performed if some parameters of the motor are within a predetermined scale of values. In particular, the method according to the present invention is activated only when the cooling temperature measured by the sensors 4, 4 ', the air temperature measured by the sensors 5, 5' and the air pressure measured by the sensors 6, 6 'in the multiples 7, 7' are above certain thresholds stored in the memory of the controllers 2, 2 '. In addition, these controllers verify that the revolutions per minute (rpm) detected by the sensor 14 are within a predetermined scale of values. Table 1 below shows an example of the values that meet the conditions for the start of the procedure.

TABLE 1 Start conditions An additional condition for starting the procedure may be to reach a certain opening of the regulator 22 as detected by the sensor 21. If the above conditions are met, at the start of a motor cycle, corresponding to a certain position of the camshaft 17 as detected by the sensors 16, 16 ', the controllers 2, 2' begin to sample the electrical signals transmitted by the sensors 18, 18 ' and proportional to the pressure inside the exhaust pipes 9, 9 '. These analog signals are converted in a known manner into a digital form and then stored in a buffer within each controller 2, 2 '. The sampling sequence is suitably synchronized with the rotational speed of the handwheel 15 as detected by the sensor 14, so that at the end of the motor cycle, detected through the sensors 16 and 16 ', a predetermined number is stored, for example 64, of pressure samples. Although the response of the pressure sensors 18, 18 'is almost immediate, in order to accurately synchronize with the motor, the controllers 2, 2' take into account the almost constant delay caused by the time required by the pressure pulsation to move from the exhaust valves of the cylinders 3, 3 'to the pressure sensors 18, 18' along the exhaust pipes 9, 9 '. Thanks to the temperature sensors 13, 13 ', it is possible to compensate for small fluctuations in said delay caused by fluctuations in the temperature inside the tubes 9, 9'. After having been taken as a sample, the pressure values corresponding to a motor cycle are processed by the controllers 2, 2 'which, at the same time, sample another series of pressure values which are stored in an additional buffer for a subsequent processing. This processing performed by each processor of the controllers 2, 2 'suitably includes an analysis in the frequency domain and in particular a Fourier transform of the sample signal, through which two series of coefficients corresponding to the real part and the imaginary part of the first harmonics of the signal. In particular, in this mode the coefficients of the first 32 harmonics of the sample signal are calculated, but in other modalities, it is obviously possible to calculate a different number of harmonics according to the needs. These coefficients are used to calculate in a known manner the module of the first harmonics, for example the first three and then, when combining the values of these modules, obtain an index which allows detecting an ignition failure in one or more of the cylinders 3, 3 '. This ignition failure rate can be calculated in several ways, for example, by adding or multiplying the harmonic modules. Before this addition or multiplication, the modules can possibly be multiplied or raised to a driving force with a different coefficient for each harmonic, in order to obtain a proven addition or multiplication. In this mode, the ignition failure rate is calculated simply by adding the modules of the first three harmonics. Once said index has been calculated, it is compared with the predetermined threshold values stored in the controllers 2, 2 '. Table 2 below shows an example of threshold values of the ignition failure index obtained experimentally as a function of the engine rpm detected by the sensor 14 and the pressure in the manifolds 7, 7 'detected by the sensors 6, 6 '.

TABLE 2 Threshold values of the ignition failure index The controller 2 or 2 'which detects the excess of said threshold, indicates through the warning light 19 or 19', that an ignition failure occurred in the corresponding row of the cylinders 3 or 3 '. At this time, controller 2 or 2 'which detected the ignition failure, preferably compares the module of each of the first three harmonics with the predetermined threshold values also stored as a function of engine rpm and pressure in the corresponding manifold 7 or T. If the three modules are within a scale of values between a minimum threshold and a maximum threshold, a single ignition fault is detected, that is, a misfire occurred only in one of the cylinders 3 or 3 ', otherwise a multiple ignition fault is detected, ie an ignition failure occurred in at least two of the cylinders 3 or 3' belonging to a row. The following tables 3.1, 3.2, 4.1, 4.2, 51 and 5.2 show examples of minimum values and amplitudes of the threshold scales for the modules of the first three harmonics.

TABLE 3.1 Minimum threshold values for the first harmonic module TABLE 3.2 Amplitude of scale for the first harmonic module TABLE 4.1 Minimum threshold values for the second harmonic module TABLE 4.2 Amplitude of scale for the second harmonic module TABLE 5.1 Minimum threshold values for the third harmonic module TABLE 5.2 Amplitude of scale for the third harmonic module If a misfire is detected in only one of the six cylinders 3 or 3 ', the corresponding controller 2 or 2' can determine the position of the cylinder where the ignition failure occurred by first calculating in a known manner the phase of the first harmonica. Consequently, by subtracting the phase of the first harmonic from the phase of the first cylinder of the motor cycle, stored in the controllers 2, 2 'by means of a frame as a function of the motor rpm, a phase difference is obtained which corresponds approximately to the phase of the cylinder where the ignition failure occurred. For example, if at a given engine rpm, the phase of the first cylinder of the engine cycle is 210 °, an ignition failure occurred in the first, second, third, fourth, fifth or sixth cylinder in order of ignition when the phase of the first harmonica is respectively between 180 ° C and 240 °, 120 ° and 180 °, 60 ° and 120 °, 0 ° and 60 °, 300 ° and 360 ° or 240 ° and 300 °. Table 6 below shows the relationship between the engine rpm and the first cylinder phase in order to determine the position of the cylinder where the ignition failure occurred.

TABLE 6 Relationship between motor rpm and phase of the first cylinder Each detection of an ignition failure in one of the engine cylinders, as well as the position of the corresponding cylinder in the event of a single ignition failure, is stored in suitable counters in the memory of the controllers 2, 2 '. This memory can be read through port 20 by means of an external processor during the operation of the car, in order to diagnose possible engine failures. Referring now to Figures 3a to 3c, it is observed through measurements made in experimental tests where the ignition faults in the test engine originated, how the signal transmitted by the sensors 18, 18 'changes as a function of the ignition failure in one of the cylinders 3, 3 '. In particular, Figure 3a shows that at approximately 2000 rpm with an engine load of around 15%, the voltage (given in Volts) at the terminals of the pressure sensors 18, 18 'proportional to the pressure in the tubes of Exhaust 9, 9 'is almost regular with six periodic oscillations during one engine cycle (indicated by the angle of rotation of the crankshaft from -180 ° to 540 °). This voltage is indicated with a thin line, while a thick line indicates the voltage in the case of misfire on the first cylinder. In this case, it is clearly observed that the voltage pattern has a first irregularity around 240 ° and a second irregularity around 480 °. However, Figure 3b shows that at approximately 4000 rpm with an engine load of approximately 100%, the voltage pattern in case of regular ignition is more complicated with respect to the previous case. However, the voltage pattern in case of misfire on the first cylinder (still indicated with the thick line) moves away around 400 ° from the regular ignition voltage pattern (still indicated with the thin line). Furthermore, Figure 3c shows that at approximately 6000 rpm with a motor load of approximately 100%, the voltage pattern of the pressure sensors 18, 18 'is different in the case of misfire on the first cylinder, in particular around 470 °. Similarly, with reference to Figures 4a to 4c, it is observed, even through measurements made in experimental tests, how the signal transmitted by the pressure sensors 18, 18 'changes as a function of the ignition failure in one. of cylinders 3, 3", without considering that the ignition failure was caused by a lack of fuel injection or ignition in the cylinder.In fact, it is observed that the voltage pattern in case of lack of injection (indicated with the thick line) is substantially equal to the voltage pattern in case of lack of ignition (indicated by dotted line) This correspondence can be found both at a low rpm, ie at approximately 2000 rpm with an engine load of approximately 15% (figure 4a), at an intermediate rpm, ie at approximately 4000 rpm with an engine load of approximately 55% (figure 4b), and at a high rpm, ie at approximately 6000 rpm with an engine load of approximately 100% (figure 4c). Referring now to Figures 5a to 5c, it is observed that the ignition failure rate measured as a function in the engine cycles (indicated on the horizontal axis) shows easily detectable peaks, which correspond to the moments in which it originated an ignition fault at experimental level in one of the cylinders of the engine. This can be found both at a low rpm, ie at approximately 1000 rpm with an engine load of approximately 15% (figure 5a), at an intermediate rpm, ie at approximately 3000 rpm with an engine load of approximately 55% (figure 5b) and at a high rpm, ie at approximately 5000 rpm with a motor load of approximately 100% (Figure 5c).

With reference to figure 6, it is observed that the module of the first ten harmonics of the signal (in Volts) transmitted by the sensors 18, 18 'evidently changes from the case of regular ignition in all the cylinders (indicated by the white bars) to the case of misfire on the first cylinder (indicated by the gray bars). The figure shows the module of the first ten harmonics calculated with the engine at 2000 rpm and a load of approximately 15%, that is, the case shown in figure 3a and figure 4a. The figure clearly shows that in the case of regular ignition, the module of the sixth harmonic is much larger than the other modules, while in the case of misfire on the first cylinder there is also a significant contribution of the modules of the first harmonics, in particular of the first three. It is clear that the contribution of the module of each harmonic depends on some factors which must be considered when establishing the threshold values of the ignition failure index. These factors include, for example, the shape of the exhaust pipes 9, 9 ', the number and the ignition sequence of the cylinders 3, 3' of each row. Finally, referring to figures 7a to 7c, it is observed that the phase of the first harmonic changes as a function of the position of the cylinder where the ignition failure occurred. In fact, it is possible to identify six separate areas, each area corresponding to a motor cylinder, where the polar coordinates of the module and phase of the first harmonic are concentrated at the moment of the ignition failure. In particular, it is observed that said coordinates are concentrated in six sectors that have an extension of 60 ° each, whose sequence is defined by the cylinder firing sequence, which in this modality is 1-4-2-6-3 -5 for the rows of cylinders 3. Considering the engine phase, this correspondence can be found both at a low rpm, ie at approximately 2000 rpm with an engine load of approximately 15% (figure 7a), at an intermediate rpm , ie at approximately 4000 rpm with a motor load of approximately 100% (FIG. 7b), and at a high rpm, ie at approximately 6000 rpm with a motor load of approximately 100% (FIG. 7c). Those skilled in the art can make possible additions and / or modifications to the embodiment illustrated and described above, without departing from the scope of the invention. In fact, it is obvious that the type of sampling, frequency analysis and particularly the method to calculate the ignition failure rate, can change according to the type of engine being monitored. Similarly, threshold values can also change according to the experimental tests carried out on each type of engine. Finally, it is obvious that the method according to the present invention can be used in combination with one or more prior art methods.

Claims (15)

NOVELTY OF THE INVENTION CLAIMS

1. - A method for detecting an ignition failure in one or more cylinders (3, 3 ') of an internal combustion engine, characterized in that it includes the following operative steps: taking a sample of the exhaust gas pressure values for at least a motor cycle, the sampling frequency being proportional to the rotational speed of the crankshaft; analyze the sample signal in the frequency domain; calculate an ignition failure index as a function of the results of that analysis; compare said index with one or more threshold values.

2. The method according to the preceding claim, further characterized in that said frequency domain analysis includes a Fourier transform of the sample signal.

3. A method according to the preceding claim, further characterized in that the calculation of the ignition failure index includes the combination of the module of some harmonics of the sample signal.

4. The method according to the preceding claim, further characterized in that the ignition failure index calculation includes the addition of the module of at least the first three harmonics of the sample signal.

5. - The method according to one of the preceding claims, further characterized in that the sampling of the pressure values begins at the start of a motor cycle.

6. The method according to one of the preceding claims, further characterized in that it includes the comparison of the module of at least one harmonic of the sample signal with one or more threshold values.

7. The method according to the preceding claim, further characterized in that it includes the calculation of the phase of the first harmonic of the sample signal and the calculation of the difference between said phase and the phase of at least one motor cylinder ( 3, 3 ').

8. A system for carrying out the method according to one of the preceding claims, characterized in that it includes at least one sensor (18, 18 ') that detects the pressure in the exhaust pipes (9, 9') and the less a sensor (14) that detects the rotation of the crankshaft, said sensors (14, 18, 18 ') being connected to at least one control unit (1, 2, 2') that includes means for the analogous to digital conversion of the electrical signal transmitted by the sensor (18, 18 ') which detects the pressure in the exhaust pipes (9, 9'), means for sampling the converted signal in digital form, memory means for storing the sample signal, as well as as means for analyzing the sample signal in the frequency domain, calculating an ignition failure index as a function of the results of said analysis and comparing said index with one or more threshold values.

9. - The system according to the preceding claim, further characterized in that it includes at least one sensor (16, 16 ') that detects the rotation of the camshaft (17).

10. The system according to claim 8 or 9, further characterized in that it comprises means for controlling the sampling frequency of said sampling means according to the signal transmitted by the sensor (14) that detects the rotation of the crankshaft.

11. The system according to one of claims 8 to 10, further characterized in that it comprises at least one sensor (4, 4 ') that detects the refrigerant temperature and at least two sensors (5, 5', 6, 6 ' ), respectively, which detect the temperature and air pressure in the intake manifolds (7, 7 '), said sensors (4, 4', 5, 5 ', 6, 6') being connected to said control unit ( 1, 2, 2 ').

12. The system according to one of claims 8 to 11, further characterized in that it comprises at least one warning light (19, 19 ') indicating an ignition failure in at least one engine cylinder, said warning light (19, 19 ') being connected to said control unit (1, 2, 2').

13. The system according to one of claims 8 to 12, further characterized in that it comprises a sensor (21) that detects the position of the motor controller (22), said sensor (21) being connected to said control unit ( 1, 2, 2 ').

14. - The system according to one of claims 8 to 13, further characterized in that it comprises at least one sensor (13, 13 ') that detects the temperature in the exhaust pipes (9, 9'), said sensor (13, 13). ') being connected to said control unit (1, 2, 2').

15. A car characterized in that it includes a system according to one of claims 8 to 14 for detecting an ignition failure in one or more cylinders of the engine (3, 3 ').