SE537390C2 - Combustion engine diagnosis - Google Patents

Abstract

A combustion engine (100) includes at least three cylinders (C1, C2, C3, C4, C5, C6). A diagnostic system for the engine (100) comprises at least one sensor (s7) configured to register a feed signal (rpm) representing a respective torque contribution from each of the cylinders; and a processing unit (110) configured to check on the basis of the feed signal (rpm) whether or not the motor (100) is operating accurately. The processing unit (110) forms a first test variable (a) on the basis of a first pair of food values representing a respective torque contribution from two of the cylinders during a food period. The processing unit (110) also forms a second test variable (b) on the basis of a second pair of food values representing a respective torque contribution from two of the cylinders during the food period, wherein said first and second pairs of food values are from at least three different cylinders in the engine (100). . The first and second test variables (a; b) are tested against an alarm criterion, and if the alarm criterion is met, the processing unit (110) generates an alarm code (A) indicating an error in at least one of the cylinders.

Description

BACKGROUND OF THE INVENTION AND PRIOR ART The present invention relates generally to the testing of internal combustion engines. In particular, the invention relates to a diagnostic system according to the preamble of claim 1 and a method according to the preamble of claim 8. The invention also relates to a computer program according to claim 13 and a computer-readable medium according to claim 14.

Modern internal combustion engines have advanced fuel injection systems in order to optimize the fuel utilization rate and thereby save energy and mitigate the engine's environmental impact. However, if errors occur in the fuel injection system, this can have highly undesirable consequences. For example, there may be a blockage in a fuel injector to a certain cylinder, which leads to this cylinder not receiving any fuel. This results in the engine producing an uneven torque, which in turn causes increased wear on the engine and an associated driveline. If the problem is not remedied, it can eventually lead to engine failure. In all transactions, the uneven running of the engine gives rise to vibrations, which, if the engine is engaged in a vehicle, provide a reduced comfort for the driver of the vehicle.

US 2012/0299051 describes a fault detection method, wherein so-called the cylinder power density value is determined during operation of an internal combustion engine. By comparing the cylinder power density values with a threshold value, one checks whether there is an imbalance in the engine, which may indicate a fault in one of its cylinders.

JP 7103047 shows a diagnostic method where an abnormal engine condition is detected which is due to a cylinder in the engine not consuming any fuel. A signal representing the crankshaft speed is recorded to transmit a time interval corresponding to the combustion cycle of the vane cylinder. A possible deviation between these time intervals and a reference value is considered to indicate an anomaly in a cylinder US 2002/0148441 reports a solution in which speed fluctuations in an engine speed are studied in order to detect any faults in its fuel injection system. Specifically, bandpass filters were used to analyze amplitude variations at certain frequencies, which have been found to be particularly characteristic for detecting faults in the fuel injection function.

US 2008/0228341 describes a method for determining a fault condition in the fuel system of an internal combustion engine. One or more sensors provide have a signal representing combustion operations in the cylinders. This signal is integrated, producing a first and a second gradient for each cylinder, respectively. By comparing the gradients for the same cylinder, conclusions are drawn as to whether or not a particular cylinder works normally.

PROBLEMS RELATED TO PRIOR ART Common to the above-mentioned solutions is that in one way or another the speed signal of an internal combustion engine is analyzed in order to determine which torque contribution each cylinder in the engine gives. If the torque contribution of the cylinders varies unacceptably much in relation to each other, this is interpreted as a fault condition. However, the known solutions for this type of analysis have similarities in drawing correct conclusions when the engine has been exposed to, for example, transient processes. Namely, in these cases, the torque contribution of the cylinders may appear to vary relatively much with each other even if the engine and its injection system work perfectly. An alarm signal and / or error code may therefore be generated in the onodan, which leads to unjustified downtime.

SUMMARY OF THE INVENTION The object of the present invention is therefore to provide a solution which alleviates the above-mentioned problems and provides a more reliable detection of faulty cylinders in an internal combustion engine.

According to one aspect of the invention, the object is achieved by the initially described diagnostic system, wherein the processing unit is configured to form a first test variable based on a first pair of food values representing a respective torque contribution from two of said at least three cylinders during a food period. The processing unit is further configured to form a second test variable on the basis of a second pair of food values representing a respective torque contribution from two of said at least three cylinders during the food period. It has been assumed that the first and second pairs of food racks graze from at least three different cylinders in the engine. The processing unit is then configured to test the first and second test variables against an alarm criterion, and given that the alarm criterion is met, the processing unit is configured to generate an alarm code indicating tel in at least one of said engine cylinders.

This system is inconvenient because it can specify whether an identified fault meant too high or too high a torque contribution from a certain cylinder. In addition, the system can handle load and speed drops where many of the previous known solutions indicate tel, even though there is no tel.

According to an embodiment thereof, the aspect of the invention has the first test variable representing a difference between the largest torque contribution and the second largest torque contribution. Furthermore, the second test variable advantageously represents a difference between the second largest torque contribution and the lowest torque contribution. In this way, stable detection criteria are obtained for detecting cylinders which give a too high torque contribution.

According to another embodiment of the same aspect of the invention, the processing unit is configured to test the first and second test variables against the alarm criterion by forming a first test variable representing a ratio between the first test variable and the sum of the first test variable and the second test variable; and forming a second test variable representing the sum of the first test variable and the second test variable. The processing unit is further configured to examine whether the first test quantity exceeds a first threshold value, and to examine whether the second test quantity exceeds a second threshold value. The alarm criterion is considered to have been met if the first test quantity exceeds the first threshold value and at the same time the second test quantity exceeds the second threshold value.

This test method has proven to be advantageous when the engine is subjected to transient operation. In these cases, the torque contributions of the cylinders tend to fluctuate considerably (especially if the torque contributions are not specifically measured, but are only estimated). Typically, however, both the largest and the second largest moment contribution represent the largest value, which is why the risk of false alarms is relatively small.

According to a further embodiment of it, the aspect of the invention indicates the alarm code error in that of the engine cylinders which gives the largest torque contribution when the alarm criterion is met.

This type of deviation can in fact be associated with too much fuel being fed into the cylinder, for example due to a fault in a fuel injection means.

According to other embodiments thereof, the aspect of the invention is the first test variable formed on the basis of a minimum torque contribution from the cylinder of said at least three cylinders which during a feeding period gives a lowest torque contribution and a second lowest torque contribution from the cylinder of said at least three cylinders which during the feeding period gives a second lowest torque contribution. The second test variable is formed on the basis of the next lowest torque contribution and a maximum torque contribution from the cylinder of said at least three cylinders which during the feeding period gives a highest torque contribution. In this case, the first test variable advantageously represents a difference between the second lowest torque contribution and the lowest torque contribution, 4 while the second test variable represents a difference between the highest maximum torque contribution and the second lowest torque contribution.

Furthermore, the processing unit is suitably configured to test the first and second test variables against the alarm criterion by: forming a first test variable representing a ratio between the first test variable and the sum of the first test variable and the second test variable; forming a second test variable representing the sum of the first test variable and the second test variable; examine whether the first test quantity exceeds a third threshold value, and examine whether the second test quantity exceeds a fourth threshold value. The alarm criterion is considered to have been met if the first test quantity exceeds the third threshold value, and the second test quantity exceeds the fourth threshold value. The alarm code in this case indicates a fault in that of the engine cylinders, which gives the lowest torque contribution when the alarm criterion is met. Thus, for example, a defective fuel injection means can be identified in an efficient manner.

According to yet another embodiment of it, the aspect of the invention comprises said sensor for detecting the food signal at least one accelerometer and / or at least one cylinder pressure sensor. Thereby, adequate data can be registered which reflects the torque contribution of each cylinder.

According to a further embodiment thereof, the aspect of the invention includes a speed sensor, which is configured to generate a speed signal representing the speed of the internal combustion engine. The processing unit is configured to derive a respective torque contribution from each of the engine cylinders from the speed signal. An advantage of this construction is that one and the same sensor can be used to transmit signals from two or more cylinders.

According to another embodiment thereof, the aspect of the invention comprises the diagnostic system comprising at least one filtering unit configured to receive and line filter (for example layer pass or bandpass filter) at least one original feed signal from the internal combustion engine. The filtering unit is configured to, in response to the received signal, generate at least one resulting signal representing a filtered version of the food signal, where, for example, extreme signal values during the food period have been deposited. This further reduces the risk of false alarms.

According to another aspect of the invention, the object is achieved by the method initially described, wherein a first test variable is formed on the basis of a first pair of food values representing a respective torque contribution from two of said at least three cylinders during a food period; and a second test variable is formed on the basis of a second pair of the food values representing a respective torque contribution from two of the at least three cylinders during the food period, the first and second pairs of the food values harrowing from at least three different cylinders in the engine. The first and second test variables are tested against an alarm criterion, and given that the alarm criterion is met, an alarm code indicating an error is generated in at least one of the engine cylinders. The advantages of this method, said choice as with the preferred embodiments thereof, appear from the discussion above with reference to the proposed diagnostic system.

According to a further aspect of the invention, the object is achieved by a computer program downloadable to the internal memory of a computer, comprising software for controlling the steps according to the method proposed above when said software cars on a computer.

According to another aspect of the invention, the object is achieved by a computer-readable medium having a darpa stored program, the program being adapted to form a computer to control the steps according to the method proposed above.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention will now be explained in more detail by means of embodiments, which are described by way of example, and with reference to the accompanying drawings.

Figure 1 shows a schematic view of an engine and a diagnostic system according to a first embodiment of the invention; Figure 2 shows a schematic view of an engine and a diagnostic system according to a second embodiment of the invention; Figure 3 contains a diagram illustrating a first example of the torque contribution of different cylinders as a function of time; Figure 4 contains a diagram illustrating a second example of the torque contribution of different cylinders as a function of time; and Figure shows a flow chart illustrating a preferred embodiment of the general method of the invention.

DESCRIPTION OF EMBODIMENTS OF THE INVENTION We first refer to Figure 1, which shows a schematic view of an internal combustion engine 100 and a diagnostic system according to a first embodiment of the invention. The internal combustion engine 100 is assumed to include at least three cylinders. The engine 100 illustrated in Figure 1 is equipped with six cylinders C1, 02, 03, 04, 05 and 06 in a straight configuration. However, the invention is also applicable to other cylinder numbers3 and types of cylinder configurations, such as V, W, boxer and Wankel engines.

The proposed diagnostic system comprises at least one sensor s1, s2, s3, s4, s5 and s6, respectively, and a processing unit 110. According to the embodiment of Figure 1, a sensor s1, s2, s3, s4, s and s6 are configured to register a respective feed signal T1, T2, T3, T4, T5 and T6, which represent a respective torque contribution from the usual cylinders C1, 02, 03, C4, 05 and 06. Each sensor s1, s2, s3, s4, s5 and s6 can thus include an accelerometer and / or a cylinder pressure sensor.

The processing unit 110 is configured to examine, on the basis of the feed signals T1, T2, T3, T4, T5 and T6, whether or not the internal combustion engine 100 is operating accurately. Generally, the processing unit 110 is configured to form a first test variable a on the basis of a first pair of food values representing a respective one of two of the cylinders C1, 02, C3, C4, 05 and 06 during a food period. The processing unit 110 is also configured to form a second test variable b on the basis of a second pair of the food values representing a respective torque contribution from two of the cylinders C1, 02, C3, C4, 05 and 06 during the feeding period. According to the invention, the food values may partially overlap, but the first and second pairs of food values must be from at least three different cylinders of the engine 100 cylinders C1, 02, 03, 04, 05 and 06, respectively. Furthermore, the processing unit 110 is configured to test the first and the second test variable against an alarm criterion. If the alarm criterion is met, the processing unit 110 is configured to generate an alarm code A indicating error in at least one of the cylinders C1, 02, 03, C4, C5 or C6.

We now refer to Figure 3, which shows a diagram illustrating a first example of the torque contributions T of the various cylinders C1, C2, C3, 04, 05 and C6 as a respective function T1, T2, T3, T4, T5 and T6 of the time t By the torque contribution T is meant either the maximum torque that a cylinder generates during a combustion cycle (that is, two crankshaft revolutions on a four-stroke engine), or an average torque that a cylinder generates during a combustion cycle.

According to one embodiment of the invention, the processing unit 110 is specifically configured to form the first test variable a on the basis of (i) a maximum torque contribution x from the cylinder which during a feeding period (e.g. corresponding to a certain number of sampling intervals for a respective digital sensor in the sensors s1). s2, s3, s4, s5 and s6) give a maximum torque contribution and (ii) 8 a second largest torque contribution y from the cylinder, which during the feeding period gives a second largest torque contribution. The processing unit 110 is also configured to form the second test variable b on the basis of (i) the second largest torque contribution y and (ii) a minimum torque contribution z from the cylinder, which during the feeding period gives a lowest torque contribution.

Figure 3 shows how the assignment of the variables x, y and z varies over time t. Has a feed signal T6 represents the variable x up to a first time t1. Until a second time t2, a food signal Ti represents the variable z. Between the first time t1 and a third time t3, a food signal T5 represents the variable x. Between the second time t2 and a fourth time t4, the food signal T6 represents the variable z, after which the food signal Ti anyo the variable z represents. Between the third time t3 and a fifth time t5 a food signal T3 represents the variable x, and between the fifth time t5 and a sixth time t6 the food signal 16 represents the variable x. From and including the sixth time t6 then the food signal T3 variable x.

The assignment of the variable y also varies over the time t. For example, the food signal T3 represents the variable y between the fourth time t5 and the sixth time t6, while the food signal T6 represents the variable y between and the sixth time t6 and an eighth time t8. Throughout the process, the processing unit 110 tests the first and second test variables a and b, respectively, against an alarm criterion; and if the alarm criterion is met, then the processing unit 110 generates an alarm code A. In the example shown in Figure 3, the alarm criterion is fulfilled at a seventh time between the sixth time t6 and the eighth time ts.

Namely, at the seventh time t7, a ratio between the first test variable a and the second test variable b meets the alarm criterion.

According to embodiments of the invention, the first test variable a represents a difference between the largest torque contribution x 9 and the second largest torque contribution y; and the second test variable b represents a difference between the second largest torque contribution y and the lowest torque contribution z.

The processing unit 110 is advantageously configured to test the first and second test variables a and b, respectively, against the alarm criterion by: forming a first test quantity representing a ratio between the first test variable a and the sum of the first test variable a and the second test variable b, the viii saga a. a + b and form a second test variable representing the sum of the first test variable a and the second test variable b, the viii saga a + b.

The processing unit 110 then examines if the first test quantity a1 (a + b) exceeds a first threshold value, and at the same time the second test quantity a + b exceeds a second threshold value. Given that case, the processing unit 110 generates an alarm code A, which indicates a fault in at least one of the cylinders C1, C2, C3, C4, 05 and C6. At the seventh time t7, therefore, the current conditions are assumed to be met.

In this case t7, the feed signal 13 from a third cylinder 03 in the internal combustion engine 100 represents the largest torque contribution x. According to a preferred embodiment of the invention, the alarm code A thus indicates an error in the third cylinder 03. In addition, the alarm code A further indicates that the third cylinder C3 receives too much fuel, because its torque contribution 13 is too high.

Figure 4 shows a diagram, which illustrates a second example of how the torque contributions Ti, T2, T3, 14, T5 and T6 of the different cylinders C1, C2, 03, 04, 05 and C6 vary over time t.

According to it, the embodiment of the invention has the processing unit 110 configured to form a first test variable a on the basis of (i) a minimum torque contribution z from the cylinder, which during a feeding period (e.g. corresponding to a certain number of sampling intervals for a respective digital sensor in the sensors s1, s2, s3, s4, s5 and s6) gives a minimum torque contribution and (ii) a second lowest torque contribution w from the cylinder, which during the feeding period gives a second lowest torque contribution. The processing unit 110 is also configured to form a second test variable b on the basis of (i) the next lowest torque contribution w and (ii) a maximum torque contribution x from the cylinder which during the feeding period gives a lowest torque contribution.

Figure 4 shows how the assignment of the variables x, w and z varies over time t. For example, the torque contribution T1 of the first cylinder C1 represents the variable z initially while the torque contribution 14 of the fourth cylinder C4 represents the variable w. At a time t9, however, the third cylinder C3 represents torque contribution 13 variable z, and at a subsequent time tic) the alarm criterion is met.

According to this embodiment of the invention, the first test variable a represents a difference between the second lowest torque contribution z and the lowest torque contribution w; and the second test variable b represents a difference between the highest torque contribution x and the second lowest torque contribution w.

Also, the processing unit 110 is advantageously configured to test the first and second test variables a and b, respectively, against the alarm criterion by: forming a first test quantity representing a ratio between the first test variable a and the sum of the first test variable a and the second test variable b, i.e. saga aa + b. and form a second test variable representing the sum of the first test variable a and the second test variable b, i.e. 11 saga a + b.

The processing unit 110 then examines if the first test variable a1 (a + b) exceeds a third threshold value, and at the same time the second test variable a + b exceeds a fourth threshold value. Given that this is the case, the processing unit 1 generates an alarm code A, which indicates an error in at least one of the cylinders C1, C2, 03, 04, C5 and 06. At time ten, all these conditions are assumed to be fulfilled.

At time tic) the feed signal 13 from the third cylinder 03 in the internal combustion engine 100 represents the lowest torque contribution z. According to a preferred embodiment of the invention, the alarm code A thus indicates an error in the third cylinder 03. Preferably, the alarm code A also indicates that the third cylinder 03 receives too little fuel, because its torque contribution T3 is too low.

According to one embodiment of the invention, the diagnostic system includes at least one filtering unit 120, which is configured to receive and line filter the original feed signals from the internal combustion engine 100, and in response, generate resulting signals representing the feed signals T1, 12, 13, 14, and T6. The filtering unit 120 is arranged to deposit extreme food values, the viii fair values which deviate strongly from other food values during the food period (so-called outliers), for example by team-pass or band-pass filtering of the incoming signals.

Thus, the analysis of the processing unit 110 can be based on data with higher reliability and relevance.

It is advantageous if the processing unit 110 includes, or is communicatively connected to, a memory unit M, which includes a computer program including software for controlling the processing unit 110 to operate as described above.

Figure 2 shows a schematic view of an internal combustion engine 100 and a diagnostic system according to a second embodiment of the invention. All male reference numerals in Figure 2 which correspond to male reference numerals in Figure 1 indicate the same units, variables and signals as those which have been described above with male reference to Figure 1.

The embodiment of Figure 2 differs primarily from that of Figure 1 in that the diagnostic system illustrated in Figure 2 includes a speed sensor s7 configured to generate a speed signal rpm representing the speed of the internal combustion engine 100. The processing unit 110 is thus configured to derive a respective torque contribution T1, T2, T3, T4, T5 and T6 from the cylinders C1, C2, C3, C4, C5 and C6 from the speed signal rpm. Examples of how this can be effected are described in the above-mentioned patents.

As an alternative to the speed sensor s7, a torque sensor can be arranged on the crankshaft of the engine 100 between the cylinders and the flywheel, which torque sensor is configured to receive the torque contribution of the various cylinders C1, C2, 03, 04, 05 and 06, respectively.

Advantageously, the diagnostic system also includes a filtering unit 220, which is configured to receive and line filter the original feed signal from the speed sensor, and in response, generate resultant signals representing the feed signals rpm. The filtering unit 220 is also designed to remove extreme food values, i.e. the values which deviate strongly from other food values during the food period (so-called outliers), for example by team-pass or band-pass filtering of the incoming signals.

For the purpose of summarizing, the general method according to the invention will now be described with reference to the flow chart in Figure 5.

In a first step 510, at least one feed signal T1, T2, T3, T4, T5 and T6 or rpm is registered, which represents a respective torque contribution from the usual cylinder C1, 02, 03, 04, 05 and 06, respectively, in an internal combustion engine. The engine is assumed to be equipped 13 with at least three cylinders.

Then, in a step 520, a first test variable a is formed on the basis of a first pair of the food values representing a respective torque contribution from two of said cylinders during a food period.

For example, the first pair of feed values may represent a maximum torque contribution x from the cylinder which during a feeding period gives a largest torque contribution and a second largest torque contribution y from the cylinder which during the feeding period gives a second largest torque contribution. Alternatively, the first pair of feed values may represent a minimum torque contribution z from the cylinder which during a feeding period gives a lowest torque contribution and a second lowest torque contribution w from the cylinder which during the feeding period gives a second lowest torque contribution.

In a step 530, which is preferably executed in parallel with step 520, a second test variable b is formed on the basis of a second pair of the food values representing a respective torque contribution from two of said cylinders during the food period. The first and second pairs of food utensils are provided with harrows from at least three different cylinders of said cylinders. For example, the second pair of feed values may represent the second largest torque contribution y and a lowest torque contribution z from the cylinder, which during the feeding period gives a lowest torque contribution. Alternatively, the second pair of food values may represent the largest moment contribution x and the second lowest moment contribution w during the meal period.

A subsequent step 540 examines whether or not the internal combustion engine is considered to function accurately. Specifically, in step 540, the first and second test variables a and b, respectively, are tested against an alarm criterion. If the alarm criterion is met, a step 550 follows, and otherwise the procedure loops back to step 510.

In step 550, an alarm code indicating an error is generated in at least one of the engine cylinders C1, C2, C3, C4, 05 or C6, preferably the engine (s) of the engine which are associated with the 14 most deviating torque contribution (s). when the alarm criterion is met. Then the procedure ends. Alternatively, it is conceivable that the procedure after step 550 returns to step 510 for further testing.

The method steps described with reference to Figure 5 can be controlled with the aid of a programmed computer apparatus. In addition, although the embodiments of the invention described above with reference to the figures include a computer and processes challenging in a computer, the invention extends to computer programs, especially computer programs on or in a bar adapted to practically implement the invention. The program may be in the form of cold code, object code, a code which constitutes an intermediate between cold and object code, as in partially compiled form, or in any other form which is convenient to use in implementing the process according to the invention. The bar can be any entity or device which is capable of just the program. For example, the bar may comprise a storage medium such as a flash memory, a Read (Only Memory) ROM, for example a CD (Compact Disc) or a semiconductor ROM, EPROM (Electrically Programmable ROM), EEPROM (Erasable EPROM), or a storage memory. netic recording medium, such as a floppy disk or hard disk. In addition, the carrier may be a transmitting carrier such as an electrical or optical signal, which may be conducted by an electrical or optical cable or by radio or otherwise. Since the program is formed by a signal which can be conducted directly by a cable or other device or means, the bar can be constituted by such a cable, device or means. Alternatively, the bar can be an integrated circuit in which the program is embedded, where the integrated circuit is adapted to perform, or to be used in performing, the current processes.

The invention is not limited to the embodiments described with reference to the figures but can be varied freely within the scope of the following claims.

Claims (14)

Patent claims A diagnostic system comprises an internal combustion engine (100) including at least three cylinders (C1, C2, C3, C4, C5, C6), the system comprising: at least one sensor (s1, s2, s3, s4, s5, s6; s7 ) configured to record at least one feed signal (Ti, 12, T3, T4, T5, T6; rpm) representing a respective torque contribution from each cylinder of said at least three cylinders, and a processing unit (110) configured to on the basis of said at least a food signal examines whether or not the internal combustion engine (100) is operating accurately, characterized in that the processing unit (110) is configured to: form a first test variable (a) based on a first pair of food values representing a respective torque contribution (x, y; 15 w , z) from two of said at least three cylinders during a food period; form a second test variable (b) on the basis of a second pair of food values representing a respective torque contribution (y, z; x, w) from two of said at least three cylinders during the food period, where said first and second pairs of food grains from at least three different cylinders of said at least three cylinders; test the first test variable (a) and the second test variable (b) against an alarm criterion, and provided that the alarm criterion is met, generate an alarm code (A) indicating an error in at least one of said at least three cylinders, wherein: the first test variable (a) is formed on basis of a maximum torque contribution (x) from the cylinder (C3) of said at least three cylinders which during a feeding period gives a largest torque contribution and a second largest torque contribution (y) from the cylinder of said at least three cylinders which during the feeding period gives a second largest torque contribution, and the second test variable (b) is formed on the basis of the next largest torque contribution (y) and a minimum torque contribution (z) from the cylinder of said at least three cylinders which during the feeding period gives a minimum torque contribution. 16 The diagnostic system of claim 1, wherein: the first test variable (a) represents a difference between the largest torque contribution (x) and the second largest torque contribution (y), and the second test variable (b) represents a difference between the second largest the torque contribution (y) and the lowest torque contribution (z). The diagnostic system of claim 2, wherein the processing unit (110) is configured to test the first and second test variables (a; b) against the alarm criterion by: forming a first test quantity representing a ratio between the first test variable (a) and the sum of the first test variable (a) and the second test variable (b), form a second test variable representing the sum of the first test variable (a) and the second test variable (b), examine whether the first test quantity exceeds a first threshold value, and examine whether the second the test quantity exceeds a second threshold value, the alarm criterion being met if the first test quantity exceeds the first threshold value, and the second test quantity exceeds the second threshold value. The diagnostic system according to any one of the preceding claims, wherein the alarm code (A) indicates a fault in that of said at least three cylinders (03) which gives the largest torque contribution (x) when the alarm criterion is met. The diagnostic system of any preceding claim, wherein said at least one sensor (s1, s2, s3, s4, s5, s6) includes at least one of an accelerometer and a cylinder pressure sensor. The diagnostic system according to any one of claims 1 to 5, wherein: said at least one sensor includes a speed sensor 17 (s7) configured to generate a speed signal (rpm) representing the speed of the internal combustion engine (100), and the processing unit (110) is configured to derive a respective torque contribution (Ti, 12, T3, T4, T5, T6) from each of said at least three cylinders from the speed signal (rpm). The diagnostic system of any preceding claim, comprising at least one filtering unit (120; 220) configured to receive and line filter at least one original feed signal from the internal combustion engine (100) and in response generate at least one resultant signal representing said at least one feedstock. signal (T1, T2, 13, T4, T5, 16; rpm). A method of diagnosing an internal combustion engine (100) including at least three cylinders (C1, 02, 03, 04, 05, 06), the method comprising: recording at least one feed signal (T1, T2, T3, T4, T5, T6; rpm) representing a respective torque contribution from a conventional cylinder of said at least three cylinders, and examining, on the basis of said at least one feed signal (11, 12, T3, T4, T5, T6; rpm), whether or not the internal combustion engine is operating accurately, can be characterized by the formation of a first test variable (a) on the basis of a first pair of food values representing a respective torque contribution 24. y; w, z) from two of said at least three cylinders during a feeding period; forming a second test variable (b) based on a second pair of the food values representing a respective torque contribution 25. z; x, w) from two of said at least three cylinders during the food period, wherein said first and second pairs of food hosts are from at least three different cylinders of said at least three cylinders; testing the first test variable (a) and the second test variable (b) against an alarm criterion, and provided that the alarm criterion is met generating an alarm code (A) indicating error in at least 18 of said at least ten cylinders, wherein: the first test variable (a ) is formed on the basis of a largest torque contribution (x) from the cylinder (C3) of said at least ten cylinders which during a feeding period gives a largest torque contribution and a second largest torque contribution (y) from the cylinder of said at least ten cylinders which during the feed period gives a second largest torque contribution, and the second test variable (b) is formed on the basis of the second largest torque contribution (y) and a minimum torque contribution (z) from the cylinder of said at least ten cylinders which during the feeding period gives a minimum moment contribution. The method of claim 8, wherein: the first test variable (a) represents a difference between the largest torque contribution (x) and the second largest torque contribution (y), and the second test variable (b) represents a difference between the second largest the torque contribution (y) and the lowest torque contribution (z). The method of claim 9, wherein the first and second test variables (a; b) are tested against the alarm criterion by: forming a first test variable representing a ratio between the first test variable (a) and the sum of the first test variable (a) and the second test variable (b), forming a second test quantity representing the sum of the first test variable (a) and the second test variable (b), examining whether the first test quantity exceeds a first threshold value, and examining whether the second test quantity exceeds a second threshold value, whereby the alarm criterion is met if the first test quantity exceeds the first threshold value, and the second test quantity exceeds the second threshold value. The method according to any one of claims 8 to 10, wherein the alarm code 19 (A) indicates a fault in that of said at least three cylinders (03) which gives the largest torque contribution (x) cla the alarm criterion is met. The method of any of claims 8 to 11, comprising linear filtering at least one initial feed signal from the internal combustion engine (100), wherein at least one resulting signal is generated which represents said at least one feed signal (T 1, T 2, T 3, 14, T 5, T6; rpm). A computer program downloadable to the internal memory (M) of a computer, comprising software capable of performing the steps according to any one of claims 8 to 12 when said program [is run on the computer. A computer readable medium (M) having a darpa stored program, wherein the program is adapted to shape a computer to perform the steps of any of claims 8 to 12.