JPH03130545A - Wall flow meter of engine - Google Patents

Abstract

PURPOSE:To directly and quantitatively grasp a behavior of wall flow at a step variation time of supplied air-fuel ratio and thereby to eliminate a delay in response to an exhaust air-fuel ratio by measuring the fuel response at every cycle in the case that fuel injection pulse width is steppingly varied. CONSTITUTION:Timing of sampling is set by a means 34 based on a crank signal detected by a sensor 33. Intake negative pressure and air-fuel ratio detected by sensor 31, 32 are sampled by means 35, 36 in synchronization with the set timing thereby to memorize data by means 37, 38 only for a specified cycle number. In synchronization with the set timing, a fuel injection pulse width per one cycle is calculated by a means 39, and the data is stored only for the specified cycle. The fuel injection pulse width of an injector 30 is steppingly varied by a means 41. Fuel response is calculated by a means 42 at every cycle based on the data, and output by a means 43.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Field of Application) The present invention relates to an engine wall flow measuring device, and particularly to one for determining a fuel step response.

(Prior Art) It is known that the wall flow of fuel in the intake pipe of a Solin engine has a great effect on the acceleration/deceleration drivability and exhaust gas purification performance of a vehicle.

To improve these performances, it is necessary to reduce the wall flow and to make transient corrections to optimize the air-fuel ratio injected into the Schilling. For this purpose, it is necessary to measure the wall current, and various wall current measurements have been carried out in the past, such as "Internal Combustion Engine J Vol. 2" published by Sankaido Co., Ltd.
5 No. 317 (1986゜4)@page 67-No. 7
(See page 8).

(sM to be Solved by the Invention) By the way, it cannot be said that the method of analyzing the response delay and time constant of the amount of fuel supplied from the injector to shillings has not yet been established. For this reason, it is possible to grasp key points such as ■ predicting how fuel wall flow will affect engine performance when designing or changing parts of the fuel injection system, and ■ how to apply wall flow correction amounts due to transients. Because there is no such thing, it takes a lot of development man-hours.

The present invention has been made in view of such conventional problems, and an object of the present invention is to provide a device that measures the response delay of the exhaust air-fuel ratio when the supply air-fuel ratio is changed in steps.

(Means for solving !!1ljl!!) As shown in FIG. A sensor 32 outputs a corresponding output, and a sensor 3 outputs a crank angle signal.
3, means 34 for setting sampling timing so as to obtain data for each cycle in response to the crank angle signal, and means 34 for setting the sampling timing so as to obtain data for each cycle in response to the crank angle signal, and for controlling each of the suction negative pressure Pu and the air-fuel ratio A/F in synchronization with this timing signal. Means 35 and 36 for sampling the sensor output, and means 37 and 38 for storing each sampled data of suction negative pressure P2 and air-fuel ratio A/F for a predetermined number of cycles, are also synchronized with the timing signal. a means 39 for calculating the fuel injection pulse width T1 per cycle; a means 40 for storing data of the calculated fuel injection pulse width T for a predetermined number of cycles; and a means 39 for calculating the fuel injection pulse width T1 per cycle; a device 41 that can change the fuel injection pulse width in steps, and a predetermined number of cycles before and after the step change in the fuel injection pulse width.
The fuel response for each cycle (
For example, it comprises means 42 for calculating the fuel chamber ratio change rate Z) and a device 43 for outputting the calculated fuel response.

(Function) According to the present invention, since the fuel response when the fuel injection pulse width is changed in steps is measured for each cycle, the behavior of the wall flow at the time of step changes can be directly and quantitatively understood. Ru.

(Example) FIG. 2 is a block diagram of one embodiment.

In the figure, the sensor output from the suction negative pressure sensor (intake negative pressure PM downstream of the valve) and the sensor output from the air-fuel ratio sensor (exhaust air-fuel ratio A/F) are transmitted via each amplifier 11.12. The signal is input to the A/D converter 13. A/D converter 13
Then, each sensor output is A/D converted in synchronization with the timing signal from the control unit 21, and the suction negative pressure P and each r-9 of the air-fuel ratio A/F are controlled by the control unit 7).
Output to 21. Here, the A/D converter 13 functions as the sampling means 35 and 36 in FIG.

Reference numeral 19 denotes a pulse width step variable device that is interposed in the wiring that connects the injector installed in the suction port and the engine control unit that outputs a fuel injection pulse signal (INJ signal) to this injector. The pulse width of the fuel injection pulse can be increased or decreased by an arbitrary step width. This device 19 functions as the pulse width step variable device 41 of FIG. In addition,
The configuration of the pulse width step variable device 19 has been disclosed in another application by the present applicant approximately at the same time as the present application.

The fuel injection pulse signal from the pulse width step variable device 41 is passed through the waveform shaping circuit 14 to the pulse width calculation circuit 15.
is input. In the pulse width calculation circuit 15, the control a-
In synchronization with the timing signal from the fuel unit 21, the fuel injection pulse @T + per cycle is calculated and the data is output to the control unit 21. This pulse width calculation circuit 15 functions as the fuel injection pulse width calculation means 39 shown in FIG.

The unit angle signal and reference position signal from the crank angle sensor are input to the control unit 21 via the waveform shaping circuit 16. In addition, in the figure, in order to cope with the case where the engine type is different, a crank angle sensor selection circuit 17 is provided. Select.

Reference numeral 18 is an input device for engine data, and the value of the invalid pulse width Ts determined by the battery voltage and the intake valve closing angle.

The air-fuel ratio measurement angle, generous number, and engine data of the crank angle sensor are output to the control unit 21.

The control unit 21 is composed of a microcomputer consisting of a CPU, ROM, RAM, and I10 board.

Among these, the CPU takes in the required external data from the I10 boat according to the program written in the ROM, and performs arithmetic processing on the necessary values while exchanging data with the RAM. Output data to I10 port.

For example, based on the crank angle signal, the A/D converter 1
The timing of sampling performed in step 3 and the timing of calculation of the fuel injection pulse width per cycle are calculated, and the number of injections performed per cycle is counted. This is the timing setting means 34 in FIG.
8! It is Noh.

In addition, when the pulse width is changed in steps by the pulse width step variable device 19, the intake manifold pressure Pu + air-fuel ratio A is calculated for a predetermined number of cycles before and after the pulse width step change.
Using each data of /F and fuel injection pulse width TI,
Calculate the fuel chamber ratio change rate 2 for each cycle. This is the function of the fuel response calculation means 42 of FIG.

The ROM stores calculation programs for the CPU, and the RAM stores data used in calculations. For example, the sampled data of the suction negative pressure Pu, the air-fuel ratio A/F, and the fuel injection pulse width T are stored in the RAM for a predetermined number of cycles. This is the memory hand J in Figure 1.
3j37, 38 and 40 functions.

Various signals are input to the I10 boat, and the I1
From the 0 boat, a timing signal is sent to the A/D converter 13 and the pulse width calculation circuit 15, and the fuel chamber ratio change rate Z
The calculation results are respectively output to an output display device 25 consisting of a printer or a display.

FIG. 3 is a waveform diagram showing the timing of data sampling.

First, the intake negative pressure PIJ [mlllHg] is A/D converted in synchronization with the timing of closing the intake valve. In the figure, n-1+n and n+1 are added after the symbol P2 to distinguish each timing for each cycle.

Note that n represents the n-th intake count from immediately after the pulse width step change.

Regarding the air-fuel ratio A/F, the fuel chamber ratio MR [an anonymous number 1 is used] is actually its reciprocal. This is, as explained later,
This is because when considering the step response of the fuel under a condition where the air amount is constant, the fuel chamber ratio MR directly corresponds to the response of the fuel.

This fuel chamber ratio MR is determined by A/
D. Must be converted. This is because there is a predetermined time delay before the air-fuel mixture that has flowed into the cylinder at the position of Pyn shown in the figure is combusted, wandered, and reaches the air-fuel ratio sensor. Therefore, if numbers are assigned to the data of the fuel chamber ratio MR to distinguish each timing, the position delayed by a predetermined time will be M Rn, which corresponds to Pun.

Note that the predetermined time is set by a delay circuit attached to the A/D converter 13.

For the fuel injection pulse width T 1 [Ias], subtract the number of rising pulses x invalid pulse width Ts from the total fuel injection pulse width Ti from the previous intake valve closing timing to the current intake valve closing timing. The value shall be T
This is because s corresponds to the response delay time of the injector, and fuel is not injected during this Ts. For example, in the case of simultaneous injection (when the injectors are activated at the same time for each shilling, once per engine revolution), Py
1 cycle section from n-1 to Pyn (day in the figure)
The fuel injection pulse width Tln of ++) is calculated by the following formula.

T1n=TiA+TiB-TsX However, in the opening ceremony, T + A t T + B is each pulse width of the two injection pulses that rise in the d)n interval shown in the figure. Tln calculated in this way is Py
Corresponds to n.

In addition, in the case of sequential injection or group injection,
Calculate using the following formula.

TIn=Ti-Ts Note that T1 [arrival S] is the fuel equivalent pulse width required per cycle, and is expressed by the well-known formula Ti=TpXCOEFX.
It is calculated by a+Ts. However, Tp [ms] is the basic pulse width calculated from air 70 - meter output and engine speed N, C0EF [anonymous number 1 is the sum of 1 and various increase correction coefficients, α [anonymous number 1 is the air-fuel ratio sensor output is the air-fuel ratio feedback correction coefficient calculated based on .

Figure $4 is a diagram in which each data sampled in Figure 3 when the pulse width step variable device 19 increases the pulse width step is rewritten with the horizontal axis representing the number of cycles. As shown in the figure, each data of the fuel injection pulse TI, fuel chamber ratio MR, and suction negative pressure PM in the steady cycle before increasing the pulse width step is T , o , M RO and P,0,
Each data in a certain cycle in the middle of response is 'rl + MR and P2, and each data in a cycle that has settled down to steady state is T.
, E , M RE and PME. The same applies to the case where the pulse width is decreased in steps.

Here, the fuel chamber ratio change rate Z in a certain cycle during the response is calculated by the following equation (2).

Z=(MR-MRO)/(MRE-MRO)...■This ■formula is publicly known from Academic Lecture Kaiwari $872 (Published in October 1986 by Japan Automotive Technology Association), pages 553 and 554. .

However, the formula (2) is for the case where the rotational speed N and the suction chamber %ffi are constant (including the case where the fluctuation of Tl is small and the overshoot of N is small), so the pulse of the intake If the width fluctuation is large or the N overshoot is large (indicated by the broken line in Figure 14), the error will inevitably increase if the fuel chamber ratio is calculated using equation (2) only.

Therefore, formulas for the following cases (i) and (ii) are required.

(i) When Tl fluctuates greatly but the overshoot of N is small. (ii) When TI fluctuates widely and the overshoot of N is also large. Therefore, in this example, for each case of (rij(ii)), The fuel chamber ratio change rate Z[%J is calculated for each cycle using equations (2) and (2). However, since the measurement accuracy is in the following order: equation (0) > equation (2) > equation (2), equation (0) is used as much as possible.

In the case of (1), Z=(MR-MRO)/(MRE-MRO)X(Tr
E T "0)/ (T + - T IO)...■(
In the case of 11) Z = (Kl・MR−MRO)/(MRE−MRO)
×(ni2・TlE Tl0)/(Tl-Tl0)...
・■However, in the 0 formula, the coefficients 1 and 2 are given by the following formula.

Kl=(760+Pu GB) /(760+P, o-a B)...2=(760
+PME-G B) / (760+P, AO-GB)...■■,■ formula, G
B (mmHgJ is the offset amount (constant value) of the intake negative pressure P with respect to the cylinder intake air amount. This is compensation for the fact that the intake negative pressure Pv and the Schilling intake air amount are not proportional.

The case where the fuel injection pulse width TI fluctuates in (i) above is when the basic pulse width Tp is calculated using the F70-meter output, and the air 70-meter output is due to intake pulsation and inertia. It fluctuates under the influence. Therefore, in this case, it is necessary to correct the fluctuation that occurs in T, and in equation (2), (THE-Ti)/(T
l-TlO) is the pulse@sode positive term.

The reason why a large overshoot occurs in N in the above (ii) is that when the generated torque increases due to the pulse width step increase, the rotational speed N increases. In this case, fjS4
As shown by the broken line in the figure, when an overshoot occurs in N, the amount of intake air decreases due to this effect, so T
T+ will also be affected as I will decrease.

In FIG. 4, MRO,T, which is steady data.

0 and PνO, we used the simple average of 8 to 40 cycles of each data of pulse width step increment (Fig. 5 shows 8 cycles of fuel injection pulse width TI and fuel chamber ratio MR). An example of simple averaging is shown>
, MRE+T+E and PvE, which are also stationary data.
As shown in FIG. 6, the simple average of each data for a predetermined cycle (8 to 40 cycles) is used when the fuel chamber ratio MR has stably converged after increasing the pulse width step. As shown in Fig. 6, at least 100 cycles are set for the number of cycles until the fuel chamber ratio MR converges, but since the convergence of the fuel chamber ratio MR is slow at low temperatures,
The number of cycles is determined as appropriate depending on the temperature (for example, cooling water temperature). As a result, the calculation of the above formula 0 or formula (2) is performed sequentially using data that has been captured in advance after a predetermined cycle has elapsed after the pulse width is changed by a large step.

Furthermore, as shown in FIG. 7, there may be cases where the fuel chamber ratio MR is not stable for a very long time (shown by the solid line) or where it is reversed (shown by the broken line). In such cases, fluctuation factors in the fuel chamber ratio other than wall flow (for example, occurrence of misfire or partial combustion due to the air-fuel mixture becoming too lean, intake air temperature or cooling water temperature in the D-noetronic system or α-N system) Because the influence of changes in the fuel chamber ratio (such as changes) is considered, data from a portion where the rate of change in the fuel chamber ratio is relatively stable (near the peak value in the figure) is used.

After completing the data processing above, calculate the void ratio change rate Z for each cycle calculated using the above formula 0 or formula (■) on a prescribed paper (paper with the horizontal axis as the number of inhalations and the vertical axis as the logarithmic scale).
Plot to . An example of plotting is shown in Figure 18. In the figure, the number attached to Z is the number of times of inspiration.

Now, to explain the effect of this example, according to this example,
Since the step response of the fuel chamber ratio can be measured for each cycle, it is possible to evaluate the transient wall flow properties of the fuel injection system, resulting in the following effects.

(1) Understand the magnitude of fuel response delay and its behavior.

(2) The influence of specifications and shapes of fuel system parts on fuel response delay can be evaluated.

(3) Performance can be predicted during component design.

(4) The amount of transient correction and the correction method can be predicted.

(5) Development man-hours can be reduced.

Furthermore, according to the above equation 0, when the pulse width fluctuation is large or when an overshoot of N occurs after the pulse width step change, the measurement accuracy inevitably decreases.
According to Equation 0 or Equation (■) above, even when the pulse width fluctuation is large or when an overshoot of N occurs after the pulse width step change, the void ratio change IF at the time of the step change is
Z can be measured with high precision.

Note that a method for evaluating or determining the void ratio change rate Z will be described below.

(,) For example, in FIG. 9, it can be predicted that the more characteristic the void ratio change rate Z is (the lower the line is), the smaller the wall flow will be, and the better the transient performance will be.

(b) When the characteristics shown in FIG. 10 are obtained, it can be determined that transient correction is easy to perform. For example, in the case of actual #X (a straight line with a constant slope), a substantially flat fuel chamber ratio can be obtained by increasing or decreasing the injection amount by 1/Z times the amount of change in injection amount. In addition, in the case of the broken line (a straight line whose slope is constant after the first intake), if the transient correction amount is given in the form of a first-order lag waveform, the fuel chamber ratio can be set to 7 lats. An example of the characteristics shown by the solid line is when early injection is performed in a partial region after warm-up in a two-valve system.

(c) Figure 11 shows the case where the injection timing is different (the broken line shows the case where the injection is performed much earlier than the opening timing of the intake valve,
The solid line shows the case where the injection is delayed until near the opening timing of the intake valve), and Fig. 12 shows the case where the length 2 from the injector to the intake valve is different (the solid line shows the case where the length e is short, and the broken line shows the case where the length e is short). (showing a long case in terms of length), it can be determined that the larger the void ratio change rate Z is, the better, if the injection timing is the same.

(d) As shown in Figure 1113, the slope of the first intake (
8 part) may be smaller than the later slope (L part).

In the embodiment, the air-fuel ratio change T$Z is used, but it goes without saying that the air-fuel ratio change ratio may also be used.

(Effects of the Invention) In this invention, the delay in the exhaust air-fuel ratio when the supply air-fuel ratio is changed in steps is measured for each cycle, so the wall flow characteristics of the fuel injection i-system can be evaluated. At the same time, the number of man-hours required for developing the fuel injection system can be reduced.

[Brief explanation of the drawing]

Fig. 1 is a diagram corresponding to the claims of this invention, Fig. 2 is a block diagram of one embodiment, Fig. 3 is a waveform diagram for explaining the timing of reading data in this embodiment, and Fig. 4 is a diagram of this embodiment. Waveform diagrams after pulse width step change correction, Figures 5 to 7 are waveform diagrams for explaining the method of obtaining steady state data in this embodiment, and Figure 8 is the fuel chamber ratio change rate Z in this embodiment. Figures 9 to 13 are plots of the evaluation of the fuel chamber ratio change rate Z. It is a characteristic diagram for showing a judgment method. 11.12...Amplifier, 13...A/D converter, 1
5... Pulse width calculation circuit, 17... Crank angle sensor selection circuit, 18... Engine data input device, 19
... Pulse width step circuit, 21... Control unit, 25... Output display device, 31... Intake negative pressure sensor, 32... Air-fuel ratio sensor, 33... Crank angle sensor, 34... ...timing setting means, 35.3
6...Sampling means, 37.38...Storage means, 39...Fuel injection pulse width calculation means, 40...Storage means, 41...Pulse width step variable device, 42.
. . . Fuel response calculation means, 43 . . . Output device. Figure 4 Number of cycles Figure average value is MR○. Figure Intake port k (times) Figure 9 Number of intakes (times) Figure 10 Figure 11 Figure 12 Figure III Drama tie base 7 Figure 13 Number of intakes

Claims (1)

[Claims] A sensor that outputs an output according to the intake negative pressure, a sensor that outputs an output according to the air-fuel ratio in the exhaust, a sensor that outputs a crank angle signal, and data for each cycle is obtained by receiving this crank angle signal. means for setting the sampling timing so as to set the sampling timing, means for sampling each of the sensor outputs of the suction negative pressure and air-fuel ratio in synchronization with the timing signal, and means for setting each of the sampled data of the suction negative pressure and the air-fuel ratio at a predetermined value. a means for storing the number of cycles;
Similarly, means for calculating the fuel injection pulse width per cycle in synchronization with the timing signal, means for storing data of the calculated fuel injection pulse width for a predetermined number of cycles, and means for calculating the fuel injection pulse width of the injector. the suction negative pressure for a predetermined number of cycles before and after the step change in fuel injection pulse width;
Using each data of air-fuel ratio and fuel injection pulse width, 1
A wall flow measuring device for an engine, comprising means for calculating a fuel response for each cycle, and a device for outputting the calculated fuel response.