GB2184619A

GB2184619A - Method of detecting maximum cylinder pressure in internal combustion engine

Info

Publication number: GB2184619A
Application number: GB08628829A
Authority: GB
Inventors: Akira Tanaka
Original assignee: Honda Motor Co Ltd
Current assignee: Honda Motor Co Ltd
Priority date: 1985-12-02
Filing date: 1986-12-02
Publication date: 1987-06-24
Also published as: GB2184619B; GB8628829D0

Abstract

The output of a cylinder pressure sensor is digitally sampled and the rate of change of the sampled data is successively calculated. Each calculated rate of change is compared with that one sample earlier, and when the former is found to exceed the latter by more than a predetermined amount it is determined that knocking has occurred. When knocking is not found, the sampled data at the time when the rate of change becomes zero or negative is defined to be the maximum cylinder pressure Pmax. At the same time, the maximum cylinder pressure angle is determined by measuring the time lapse between a reference point such as piston TDC and the point of time at which the rate of change in the sampled data becomes zero, and converting the measured time lapse into the corresponding change in crankshaft angle. <IMAGE>

Description

SPECIFICATION Method of Detecting Cylinder Pressure in Internal Combustion Engine Field ofthe Invention This invention relates to a method of detecting cylinder pressure in an internal combustion engine and more particularly to a method for detecting the maximum cylinder pressure as well as, if desired, the crankshaft angle at which the maximum cylinder pressure occurred.

Description of the Prior Art Recent years have seen extensive adoption of methods for detecting cylinder combustion pressure in internal combustion engines and for using the detected values to control the ignition timing and other operating parameters of the engine. For the required determination of the combustion pressure, and more specificallyforthe determination of the maximum cylinder pressure (Pmax), it has been the practice to employ a method as disclosed in Japanese Patent Publication No.

59(1984)-2796 in which a maximum pressure detection circuit is constituted using an operational amplifier, a capacitor and the like, the outputs of a pressure sensor are added and input to the detection circuit, whereby the maximum pressure value is determined, and the determined value is stored at the capacitor. Further, for detecting the crankshaft angle at which the maximum pressure was reached, i.e. the maximum cylinder pressure angle 6pumas, there has conventionally been used a method such as that disclosed in Japanese Patent Publication No. 58(1 983)-33394 in which the output of a pressure sensor is differentiated using a differentiation circuit and the result of the differentiation is applied to a comparison circuit for comparison with a predetermined value.However, as these prior art methods both rely on analog circuits for the detection, they are not totally satisfactory as regards accuracy or as regards tracking performance during high-speed engine operation. Another disadvantage of the conventional methods is that they require complicated circuitry.

As prior art technology aimed at overcoming the shortcomings of the aforesaid methods there can be noted Japanese Unexamined Patent Publication No. 57(1982)-i 73565 which relates to a method of using an A/D converterto A/D-convert the output of a cylinder pressure sensor once per a prescribed angle of crankshaft rotation and defining the crankshaft angle at the time the converted value reached maximum as the maximum cylinder pressure angle epmax. Although this method discloses methods for detecting the cylinder pressure and determining the maximum pressure value and the crankshaft angle at which the maximum pressure value occurs, it is incapable of simultaneously determining the presence or absence of knocking on the basis of the detected cylinder pressure and in this respect leaves much to be desired.

Summary of the Invention An object of the invention is to provide a method of detecting cylinder pressure in internal combustion engines which is free of the abovementioned defects of the prior art methods and which enables high accuracy and outstanding tracking performance in cylinder pressure detection.

Another object of the invention is to provide such a method which is capable not only of detecting the maximum cylinder pressure and the maximum cylinder pressure angle but also of simultaneously determining the presence or absence of knocking on the basis of the output of a pressure sensor.

The present invention accomplishes these objects by providing a method of detecting cylinder pressure in an internal combustion engine comprising the steps of digitally sampling the output data of a cylinder pressure sensor, calculating the rate of change of the sampled digital data and determining that knocking has occurred when the rate of change alters by more than a predetermined value, and defining the maximum value of the sampled digital data to be the maximum cylinder pressure when it is not determined that knocking has occurred. The maximum cylinder pressure angle, i.e. the crankshaft angle at which the maximum cylinder pressure occurred, can also be detected.

Brief Description of the Drawings These and other objects and advantages of the present invention will be more apparent from the following description and drawings in which: Figure 1 is a block diagram showing the structure of an exemplary device for carrying out the method of the present invention; Figure 2 is a flowchart showing the general idea of the detection method according to the invention; Figure 3 shows two diagrams for explaining the detection method according to the flowchart of Figure 2; Figure 4 is a flowchart showing the detection method of the flowchart of Figure 2 in greater detail; Figure 5 is a flowchart illustrating a second embodiment of the detection method according to the present invention; Figure 6 is a diagram for explaining the detection method according to the flowchart of Figure 5; and Figures 7 and 8 are diagrams for explaining the case where a blind zone is provided.

Description of the Preferred Embodiments A detection method according to the invention will be explained in conjunction with the device for carrying it out shown in Figure 1.

In the detection device of Figure 1, a piezoelectric pressure sensor 10 is provided to face into a combustion chamber of an internal combustion engine. The output of the pressure sensor 10 is applied to a charge amplifier (not shown) for charge-voltage conversion and, after being adjusted to an appropriate pulse width, is sent to a control unit 12, where it is first passed through a low pass filter 14. The cutoff of the low pass filter is set at a higher frequency than that expected to arise upon the occurrence of knocking so that when the sensor output passed through the low pass filter 14 is forwarded to a microcomputer 16 it still contains any knocking frequency component of the original output. In the microcomputer 16, the sensor output is first applied to an A/D converter 18 for conversion into a digital signal.On the other hand, a crankshaft angle sensor 22 for detecting the crankshaft angle of a crankshaft (not shown) to which a piston 20 is connected is provided at an appropriate location as, for example, within a distributor (not shown) associated with the engine. Once per prescribed angles of crankshaft rotation, the crankshaft angle sensor 22 produces a cylinder identification signal, piston TDC signals and unit angle signals, all of which are input to the microcomputer 16 via an I/O board 24 thereof.

The microcomputer 16 comprises a central processing unit (CPU) 26, a read-only memory (ROM) 28 and a random access memory (RAM) 30.

The CPU 26 computes the proper ignition timing from the input data using the ROM 28 and the RAM 30 and outputs a command to ignite the air and fuel mixture within the cylinder combustion chamber via an igniter 32, a distributor (not shown) and a spark plug 34.

The method of detecting cylinder pressure according to the present invention will now be explained. As shown by the flowchart outlining the method in Figure 2, the output of the pressure sensor 10 is digitally sampled (step 50), the rate of change of the so-obtained digital data is calculated (step 52), knocking is determined to have occurred when the rate of change alters by more than a predetermined value (steps 54, 56), and the maximum value of the sampled digital data is defined to be the maximum cylinder pressure when it is not determined that knocking has occurred (step 58). As will be shown later, the maximum cylinder pressure angle can be detected at the same time.

The method of detection according to the flowchart of Figure 2 will now be explained with reference the diagrams in Figure 3. Afterthe pressure sensor output signal has passed through the low pass filter 14, it is converted into digital values at a predetermined sampling rate (defined in terms of the sampling interval "tsamp").

Considering that the engine will be operated at high speed, a sampling interval in the range of 10 u-25 us is preferable. Defining the digital values as "pn-i, pn ", . . .", and the difference therebetween as "ps" (=pn-pn-1 ), the absolute rate of change "an" is IPSI an tsamp Therefore, defining the absolute rate of change one sampling earlier as "and", it is possible to discriminate the change in gradient of the pressure waveform by comparing an and an-i.

Referring to the flowchart in Figure 4, which shows the detection method of the present invention illustrated in Figure 2 is greater detail, in step 70 the difference "ps" between two adjacent digital data "pn-l" and "pn" is calculated. Next, in step 72, the absolute change rate "an" is calculated by dividing the difference "ps" by the predetermined sampling interval "tsamp" and in step 74 is compared with the absolute change rate an-i one time earlier. (Forthefirst calculation, an appropriate initial value is set for an-i and from the second calculation the value calculated for an one calculation earlier is used as the value of and.) In step 74, when the result of the comparison is that the current absolute change rate an exceeds the absolute change rate an-i one time earlier (or the initially set value) by not less than a predetermined amount, this is deemed to constitute an abnormally large variation in the change rate and it is decided that a jagged waveform indicative of knocking has occurred. In this case, the procedure moves to a separate knocking control operation involving compensatory ignition retarding or the like (step 76).

At the same time, the current absolute change rate an is redefined as an-l for use in the next calculation (step 78). The procedure then returns to step 70.

If knocking occurs as shown by Figure 3(A), this can be determined from the fact that the consequent sharp rise in sensor output will cause the difference between the absolute change rates an and an-l to exceed the predetermined value. Therefore, the predetermined value may be set at any appropriate level or, put differently, at any level sufficient for discriminating between the knocking and nonknocking states.

In step 74, if the difference between an and an-l is found to be less than the prescribed value, the procedure moves to step 80 where it is decided whether the difference "ps" is zero or negative.

Since the result of this decision will be YES only if the sampling has been made at or after the peak of the waveform (in which case "ps" will be zero and negative, respectively), it can be presumed that the peak has not been reached yet if the result of the decision is NO. In the case of a NO decision, therefore, after resetting the flag, which will be explained with reference to step 84, OFF at step 82, the current absolute change rate an is redefined as an-1 for use in the next calculation (step 78) and the procedure returns to step 70.

When the decision is YES in step 80, it is determined in step 84 whether this is the first affirmative decision made in the step 80 by checking whether or not the flag is set. If the flag bit is not on, the flag is now raised (step 86) and, since the occurrence of this first YES decision in step 84 means that the peak of the waveform has been reached, the largest of the data pn, pn-i . . . . up to this time (pnmax) is in step 88 deemed to approximate the maximum cylinder pressure (Pmax). (See Figure 3(B).) Then the current absolute change rate an is redefined as an-1 for use in the next calculation (step 78) and the procedure returns to step 70. Even after the peak of the waveform has been passed, the procedure oftheflowchart is continued for the purpose of determining whether knocking occurs.If knocking is found to occur, a knocking control operation is initiated through step 76, and if no knocking is found, the procedure returns to step 70 without further calculation of a maximum cylinder pressure since the flag bit has already been once set (steps 80, 84, 86). Although the detection of the maximum cylinder pressure is based on an approximation, the result can be made substantially equal to the actual value by making the sampling interval "tsamp" adequately small.

Figure 5 shows another embodiment of the detection method of the present invention. In this embodiment the maximum cylinder pressure angle (Gpmax) is detected together with maximum cylinder pressure (Pmax). The calculation method is shown in Figure 6.

In step 100 of Figure 5, starting from the instant of arrival of a piston TDC signal from the crankshaft angle sensor 22, time lapse measurement is started by counting the clock pulses of a clock (not shown) built into the microcomputer 16. In the succeeding step 102, when it has been confirmed that the crankshaft has rotated by a first predetermined angle following the arrival of the piston TDC signal, the rate of change is calculated in the same manner as in the first embodiment and discrimination and decision are conducted with respect to the calculated result (steps 104--112). The reason for carrying out the calculation, discrimination and decision operations after the crankshaft has rotated by a predetermined angle is to limit the region to be investigated to the minimum required.

Next in step 114, it is judged whether the peak of the waveform has been reached on the basis of whether or not the difference "ps" is zero. When it is found that the cylinder pressure has peaked, the maximum value of the data, i.e. pnmax, is deemed to approximate the maximum cylinder pressure Pmax (step 116), in the same manner as in the first embodiment. Then the number of times that "ps" was found to equal zero is counted and the sum "N" is stored in the RAM 30 (step 118). The reason for this operation is that since the sampling interval is made shorter than that in the first embodiment, the peak of the sensor output waveform will include a flat region over which the relation "ps=0" will hold over several successive samplings, making it desirable to select the midpoint of this flat region as corresponding to the maximum cylinder pressure angle Opmax.

That is to say, when the termination of the flat region has been confirmed in step 120 from the fact that the difference "ps" has turned negative, the aforesaid time lapse measurement is discontinued and the period of time up to the discontinuation of time lapse measurement is defined as "tstop" (step 122). Here, defining the time lapse up to the position corresponding to the maximum cylinder pressure angle ûpmax as "tpmax", the lapsed time "tpmax" can be calculated as tpmax=tstop-((tsampx1/2 N)+At) where At is the time required to confirm that the flat region has ended. (Step 124) ûpmax is now obtained by multipiying the lapsed time "tpmax" by a time/angle conversion factor "k" (step 126).The conversion factor "k" is calculated as k=(engine rpmx360 (degrees))/60 (sec) After completion of the steps 110,118,120 and 126, the current change rate an is redefined as an-l for use in the next comparison (step 112) and repetition of the procedures oftheflowchart is continued until the second predetermined crankshaft angle has been reached (step 128).

When an ignition device or the like employing high-voltage switching is used in conjunction with the first or second embodiments described above, there is a possibility that the output of the cylinder pressure sensor will be affected by noise, giving rise to fine ripples in the output waveform as shown in Figure 7(A). In such a case, if a point at which difference "ps" becomes zero or turns negative is defined as the point of maximum cylinder pressure, it is possible that, as shown in Figure 7(B), a minor peak"pi" present in the waveform prior to the actual peakthereofwill be misidentified as the point of maximum cylinder pressure. Therefore, it is preferable to establish a blind zone of predetermined width with respect to the value of "ps".That is, it is preferable to define that all values of "ps" within a predetermined range of values are equal to a single constant value, in this way avoiding the possibility of a minor peak being misidentified as the point of maximum cylinder pressure. Figure 8 shows an example of how the blind zone can be set in relation to the engine rpm or the engine load.

As the present invention provides a method for detecting cylinder pressure in which the output of a pressure sensor is subjected to high-speed A/D conversion and the rate of change in the digital data is determined, it enables not only the determination of the maximum cylinder pressure and the maximum cylinder pressure angle but also the detection of whether or not knocking has occurred.

Moreover, it permits accurate determination of the maximum cylinder pressure, the maximum cylinder pressure angle and the presence or absence of knocking without need for storing large quantities of data in memory, and exhibits excellent tracking performance even during high-speed engine operation.

While the above description discloses preferred embodiments of the invention, it is to be understood that numerous modifications or alterations may be made without departing from the scope of the invention as set forth in the following claims.

Claims

1. A method of detecting cylinder pressure in an internal combustion engine comprising the steps of: a. digitally sampling the output data of a cylinder pressure sensor, b. calculating the rate of change of the sampled digital data and determining that knocking has occurred when the rate of change alters by more than a predetermined value, and c. defining the maximum value of the sampled digital data to be the maximum cylinder pressure when it is not determined that knocking has occurred.

2. A method as claimed in Claim 1 further including a step following step c of determining the crankshaft angle at which the maximum cylinder pressure occurred.

3. A method as claimed in Claim 1 or 2 comprising calculating the rate of change (ps=pn-pn- 1) in the cylinder pressure sensor output between consecutive points (pn, pn- 1) separated by a sampling time interval (tsamp).

4. A method as claimed in Claim 3 wherein a blind zone is established with respect to said rate of change.

5. A method of detecting cylinder pressure in an internal combustion engine, substantiaily as hereinbefore described with reference to the accompanying drawings.