Note: Descriptions are shown in the official language in which they were submitted.
i6~
A control system for internal combustion engines
.... . _
The present invention relates to a control system
for an internal combustion engine, particularly of the
type used in an automobile.
Desirable attributes of an automotive engine are
a reduced fuel consumption and improved acceleration.
As a way of satisfying these demands, there has
been known in the art a system for supplying an internal
combustion engine with a lean air/fuel mixture, as
10 disclosed in Japanese Patent Laid-Open No. 59-224499 (1984).
However, -this approach is accompanied by the problem that
a sufficient acceleration cannot be attained, because of
the leanness of the mixture.
In order to eliminate this lack of good acceler-
ation, on the other hand, it is effective to prepare a
rich mixture and supply it to the engine during acceler-
ation. However, there then arises the second problem
that the engine torque will suddenly change, causing a
deterioration in drivability, when the mixture makes a
transition from rich to lean, or vice versa.
In order to eliminate such a sudden change of
the torque, it is conceivable to have a gradual trans-
ition of the richness of the mixture~ However, there
exists between the lean mixture zone and the rich mixture
zone a region, i.e. about 15 to 18 in air/fuel ratio,
: , :
.:' ' ~ '
.,: -: ~ . .. .
-- 2
where nitrogen oxides (NOX) are generated to a maximum
extent, which adversely affects the noxious components
in the exhaust gas.
It is, therefore, an object of -the present in~
vention to provide a control system for an internal
combustion engine, that can ensure a sufficient acceler-
ation without increasing the noxious components in
exhaust gas.
The present invention is characterized in that
the uniformity of the engine torque is enhanced by
controlling the operation of an actuator for driving
the throttle valve in accordance with the depression
rate of the accelerator pedal.
In the drawings:
Fig. l is a schematic diagram showing the overall
construction of a control system according to an
embodiment of the present invention;
Fig. 2 is a schematic diagram showing the
detailed construction of a throttle valve actuator;
Fig. 3 is a block diagram explaining the basic
concept of control according to the embodiment of the
present invention;
Fig. 4 is a block diagram showing an example of
a detailed construction of a control unit used in the
control system of the embodiment;
E'ig. 5 is a flow chart showing schematically
the main routine of the control;
Figs. 6 and 7 are diagrams showing the relations
for obtaining the corrected accelerator pedal position
from actual accelerator pedal position signals;
Fig. 8 is a detailed flow chart showing step 170
in the control routine of Fig. 5;
Fig. 9 is a diagram showing the relation between
the correc-ted accelerator pedal posi-tion signal and -the
amount of fuel injection Ti;
. -~ ;
. . .
Fi~. 10 is a diagram showing the relation between
the en~ine cooling water temperature and its correction
coefficient KTW;
Fig. 11 is a table showing a correction coefEicient
K~ obtained by an air/fuel ratio sensor;
Fig. 12 is a table showing a correction coefficient
KMR for set-ting the air/fuel ratio according to the
operational condition of the engine;
Fig. 13 is a diagram showing the relation between
the air temperature upstream of a throttle valve and a
correction coefficient KA;
Fig. 1~ is a diagram showing the relations for
determining the openi.ng of the throttle valve from the
ratio of Ti x KA/KMR and the number of revolutions of the
engine;
Fig. 15 is a diagram showing the relations for
determining an ignition timing BTDC from the amount of
fuel injection and the number of revolutions of the
engine;
Fig. 16 is a detailed flow chart showing step
280 in the control routine of Fig. 5;
Fig. 17 is a flow chart showing a modification of
step 280 as shown in Fig. 16;
Fig. 18 is a flow chart of the control, which is
added to the main control routine of Fig. 5, when a
suction pressure sensor is used to improve the metering
accuracy of the air flow rate;
Fig. 19 is a table of a correction coefficient
K~ of the opening of the throttle valve;
Fig. 20 is a f].ow chart showing a modifica-tion
of Fig. 18 when an air flow sensor is used in place of
the suction pressure sensor to lmprove the metering
accuracy of the air flow rate; and
Figs. 21 and 22 are diagrams respectively
explaining the effects of the present invention in
comparison with those of the prior art.
Fig. 1 is a schematic diagram showing the overall
construction of one embodiment of the present invention.
Here is shown in section one cylinder of a multi-cylinder
engine. The reciprocal movements of a piston 102 in the
cylinder 101 are converted into revolutions of a crankshaft
103 and outputted as driving power.
In accordance with the movements of the piston
102, on the other hand, an intake valve 104 and an exhaust
valve 105 are opened or closed. In synchronism with
opening of the intake valve 104, fuel is injected into an
intake pipe 107 by an injection valve or injector 109.
The fuel thus injected is mixed with the suction air to
fill the inside of the cylinder, i.e., the combustion
chamber 108 and is compressed by the piston 102. Then,
the air-fuel mixture is ignited by a spark plug 106.
Exhaust gas is discharged into an exhaust pipe 110 when
the exhaust valve 105 is opened. There is disposed at
the collector portion o~ an exhaust mani~old an air/fuel
(A/F) ratio sensor 124 for detecting the ratio of the air
to the fuel in the mixture sucked into the engine, in
terms of the concentration of excess oxygen contained in
the exhaust gas.
Downstream of an air cleaner 121, on the other
hand, there are arranged: a suction air temperature sensor
102 (such as a thermocouple or a resistance bulb) for
detecting the suction air temperature; an air flow sensor
119 for detecting the flow rate of the suction air; and
an opening sensor 118 for detecting the opening of a
throttle valve 116. There are also arranged: an
accelerator pedal position sensor 113 for detecting the
accelerator pedal position; a water temperature sensor
123 for detecting the tempera-ture of the engine cooling
water or cylinder wall; and a crank angle sensor 111 for
detecting the angle of the crankshaft 103.
: ' .
: ` ..
":"
. .
- 5 - ~ 3i 3
All the signals detected by these sensors are
inputted to and processed by a control unit 112 having a
built-in computer, so that signals representing the
in~ection valve opening time, the ignition timing and the
throttle valve opening are produced and fed to the in-
jector 109, the plug 106 and a throttle valve actuator 114.
The amount Qa of the air sucked into the engine
can be calculated by not only the output signal of the
air flow sensor 119 bu-t also the output signal of a
pressure sensor 115 disposed midway of the intake pipe
107 and the number of revolutions of the engine, i.e.,
the output signal of the crank angle sensor 111.
In the vicinity of the intake valve 104 of the
intake pipe 107, on the other hand, there i5 buried into
an inner wall of the intake pipe 107 a flush-type heating
resistor 132 which can have its calorific value controlled
from the outside. The current to be applied to the heating
resistor 132 is controlled by a heater driver 131 connected
to the control unit 112, by which it is controlled in
accordance with the respective output signals of the above-
specified sensors, such that it is fed with a larger
current when the engine is started, but has its current
flow decreased gradually after the engine has been warme~
up. Reference numeral 122 denotes a battery.
Fuel is supplied to the injector 109 from a fuel
reservoir 125 by way of a strainer 126, a pump 127, a
regulator 128 and a fuel pipe 129 with its pressure
controlled at a predetermined value.
Fig. 2 is a diagram showing the detailed con-
struction of the throttle valve actuator 114. The
necessary opening of the throttle valve 116 is determined
by the arithmetic operation (which will be described
hereinafter) of the control unit 112. In accordance with
the throttle valve opening determined, a step motor
driver 142 generates a signal for determining the direction,
angle and velocity of the rotation of a step motor 143.
-
- 6 - ~ 9~?~
In response to this signal, the step motor 1~3 is rotated
-to t~lrn the throttle valve llG to a predetermined openiny
through a reduction gear 1~4.
A potentiometer 145 is provided to measure the
actual opening of the throttle valve 116 and is used to
provide a feedback loop to ensure that the opening is
the one de-termined by the control unit 112. More specific-
ally, the voltage level of the potentiometer 145 is intro-
duced to the control unit 112 when the current to be fed
to the step motor 143 is at zero, i.e., when the throttle
valve 116 is fully closed, and the throttle valve opening
is measured by using that voltage level as a reference.
Thus, the dispersion of the resistances and adjustments
of the individual potentiometers can be automatically
corrected by the control unit 112.
Since, rnoreover, the step motor 143 is a motor that
will turn one step when it receives one pulse, as will be
described hereinafter, the throttle valve opening can be
determined without the potentiometer 145, if the number
of pulses applied to the step motor 143 is integrated from
the instant when the current fed to the step motor 143 is
at the zero level.
Although not shown, moreover, there is provided a
tension return spring that will urge the throttle valve
116 in the closing direction, so that the throttle valve
116 is closed by such spring if the current supply to the
step motor 143 is interrupted in an abnormal operation.
The concept of a control system will next be
described with reference to the block diagram of Fig. 3.
In accordance with an accelerator pedal position signal
~A, a changing rate signal d~A/dt thereof with respect to
time, an engine number-of-revolution (per unit time)
signal N and a transmission position signal S, the opening
t~ Ti of the fuel injector 109 is determined by the control
unit 112 and is set in an output circuit 117. That
opening time Ti is determined by the following formula:
Ti = f(OA, d~A/dt, N, Sj.
. :
. , , :
: .... :
.
... .
-- 7
Th~ opening time Ti is subjected to a feedback
control by the signal of the A/F ratio sensor 124, which
is denoted by K~.
On the other hand, the -throttle valve opening ~T
is determined by the control unit 112 in accordance with
the opening time Ti of the injector ].09, the signal N
and an air temperature signal TA, and is set in an output
circuit 133.
The throttle valve opening ~T is determined by
~he following formula:
~T = f (Ti, N, Ta).
This throttle valve opening ~T is subjected to a
feedback control with the signal coming from the potentio-
meter 145. This feedback signal is denoted by K9.
Next, an example of a detailed construction of
the control system will be described with reference to
Fig. 4.
With a microprocessor (CPU) 134, there is connected
through a bus a timer 135, an interruption controller 136,
a number-of-revolution counter 137, a digital input port
138, an analog input port 139, a RAM 140, a ROM 141, and
the output circuits 117 and 133. The signals of the
A/F ratio sensor 124 and the accelerator pedal position
sensor 113 are introduced in-to the analog input port 139
If necessary, the signals of the air flow meter 119, the
water temperature sensor 123 and the throttle valve
opening sensor 118 are also introduced .into the analog
input por-t 139.
The signal of a transmission position sensor 151
is inputted i.nto the digital input port 138. If an
ignition switch IG is turned on, the electric power is
supplied rom the battery 122 to the control unit 112.
Incidenta].ly, the RAM 140 is always supplied with power.
When the ignition switch IG is turned on, the
control of the main routine shown in Fig. 5 is s-tarted
.. ~
.. : . ;. ' ~., '.. .
- 8 ~ 6C3~
by executing a program that is stored in advance in the
ROM 141. The main output signals of the control unit
112 are the signals of the fuel injection valve opening
time, the ignition timing, the throttle valve opening
and so on.
Next, the content of the main routine of Fig. 5
will be described with reference to Figs. 6 to 20. If
the main routine is started, the initialization is
conducted. Then, at step 160, the accelerator pedal
position ~A, the transmission position S and the engine
number-of-revolution N are read. At step 170, as a pre-
processing for determining the basic amoun~ of fuel
injection Ti, the corrected accelerator pedal position
~C2 is obtained by retrieving the relation provided
therefor in advance on the basis of 0A.
When a driver desires to accelerate quickly or
to decelera~e an automobile, he rapidly depresses or
releases the accelerator pedal. The accelerator pedal
position signal ~A is read into the control unit 112,
in which the changing rate Q~A (i.e., d~A/dt) thereof
for a predetermined time (e.g., 40 to 60 ms) is obtained.
In accordance with the absolute value ¦~A¦ of the
changing rate Q~A, one of curves MODl, MOD2 and MOD3
shown in Fig. 6 is selected. For example:
MODl for 0 < ¦Q~A¦ < C1;
MOD2 for Cl < ¦Q~A¦ < C2; and
MOD3 for C2 < ¦QQA¦ .
Cl and C2 are constants arbitrarily set in
accordance with the required style of driving, i.e.
sporty driving or an economic one. Cl and C2 for
sporty driving are selected with smaller values than
those for economic driving.
Subsequently, a primary corrected accelerator
_ 9 _ ~ ~ 6~g3(~
pedal position ~Cl is obtained by retrieving the
selected one of the three curves on the basis of the
actual accelerator pedal position ~A. By the use of
Fig. 7, moreover, the primary corrected position ~Cl is
converted into a secondary one ~C2 in accordance with
the transmission gear position S.
This is to change the fuel increasing rate with
respect to the amount of depression of the accelerator
pedal in accordance with the transmission position, so
that the changing rate (or acceleration) of the auto-
mobile speed with respect to the amount of depression
of the accelerator pedal may be substantially identical
over the low to high gear positions of the transmission.
More specifically, when the transmission is in the 4th
position, the torque to be transmitted to the wheels is
lower than that in the 1st position, so the acceleration
becomes lower. In the case of the ~th position, there-
fore, the increasing rate of the fuel with respect to
the accelerator pedal position is enlarged.
Details of the step 170 described above are
shown in Fig. 8~ More specifically, the changing rate
of the accelerator pedal position is calculated at step
300. At steps 301 and 303, it is judged which one of
the modes MODl, MOD2 and MOD3 the changing rate is
25 located in. At steps 302, 30~ and 305, the primary
corrected accelerator pedal position 0Cl is retrieved
in accordance with each of the modes MODl to MOD3. Next,
at steps 306, 308 and 310, it is judged what the trans-
mission gear position is. At steps 307, 309, 311 and 312,
the secondary corrected accelerator pedal position ~C2
is retrieved.
According to this processing of Fig. 8, the
accelerator pedal position is corrected on the basis of
the mode MODl to MOD3, which are different depending
upon the changing rate of the accelerator pedal, so that
..~ .
~:
30~
-- 10 --
the acceleration can be improved whatever the trarlsmisslon
gear position.
Reverting to Fig. 5, at step 180, the basic amoun-t
of fuel injection Ti is retrieved. Here, the amount Ti
of the fuel to be injected auring one suction stroke
is retrieve~ on the basis of the secondary corrected
accelerator pedal position ~C2, by the use of Fig. 9.
Next, at step 190, a correction coefficient is retrieved
from the relation between the cooling water temperature
and the correction coefficient characteristics shown in
~ig. 10.
The correction of the output of the A/F ratio
sensor 124 at and after the step 200 will now be described.
At step 200, an A/F ratio signal V is read from
the A/F ratio sensor 124 disposed in the exhaust pipe
110. Then, at step 210, an A/F ratio reference VR
determined in advance is selected in accordance with the
running state. At step 220, the set A/F ratio reference
VR is compared with the A/F ratio signal V. Next, the
correction coefficient K~ is calculated at step 230 in
accordance with a formula described thereat and is
stored in the RAM 140 at step 240. Here, the coefficient
K~ is one for integration control.
Incidentally, the RAM 140 has a table of the
coefficient K~, in which the number of revolutions of
the engine N and the basic amount of fuel injection Ti
are respectively shown on the abscissa and ordinate in
Fig. 11. This table is renewed each time a new value
of the coefficient K~ is obtained at step 230. As a
result, the content of the table is gradually made
appropriate by the learning effect. The table thus
renewed will not have its content erased, even if the key
switch IG of the engine is -turned oEf, because the RAM
140 is always supplied with electric power.
Next, at step 250, the correction coefficient
. .
.
.
. .
.
6~
K~ is retrieved on the basis of the basic amount of fuel
injection Ti and the number of revolutions ~ in the
table of Fi~. 11, and the amoun-t of fuel injection Tinj
is obtained by the following formula:
Tinj = Ti x COEFF x K~ ...................... (1),
wherein the following coefficients are used solely, or in
combination, as the "COEFF":
KAS: coefficient for increment of fuel after
start;
KAI: coefficient for increment of fuel after
idle;
TADD: coefficient for acceleration increment;
and
KDEC: correction coefficient of deceleration.
Next, at step 260, an A/F ratio correction
coefficient KMR is retrieved. Fig. 12 is a table of the
correction coefficient KMR for setting the optimum A/F
ratio in each operational state of the engine.
The region where the value of the basic amount of
fuel injection Ti is high corresponds to the so called
"power region", in which the depression of the accelerator
pedal is large to provide a rich mixture to increase the
engine output. For a high speed region, a rich mixture
i5 prepared to prevent seizure of the engine. In
another running region, i.e. a partial load region, on
the other hand, a lean mixture is prepared to reduce the
fuel consumption rate. The correction the coefficient
KMR is retrieved from -the basic amount of fuel injection
Ti and the number of revolutions N by use of the table
above.
At step 270, the signal of the suction air
temperature sensor 120 disposed upstream of the throttle
valve 116 is read to retrieve a correc-tion coefficien-t
KA by the use of the relation shown in Fig. 13 and to
- 12 -
calculate the value of Ti x KA/KMR.
At step 280, the opening of the throttle valve
lL6 and the iynition timing are retrieved. At first,
the opening ~T of the throttle valve 116 is retrieved
on the basis of the value of Ti x KA/KMR and the number
of revolutions N of the engine, by using the relation of
Fig. 14. This relation is so set that ~he amount of
suction air Qa/N per one suction stroke of the engine
satisfies the following formula (2):
10 Qa/N = Kq x Ti/KMR ...................... .(2)
wherein Kq: a constant.
On the other hand, the amount of fuel injection
Qf/N during the engine suction stroke is expressed by
the following formula (3):
15 Qf/N = K10 x Tinj ....................... (3)
wherein K10: a constant.
As a result, the set A/F ratio is expressed
from formulas (2) and (3) by the following formula (4):
(A/F)A = Kll x Ti/(KMR x Tinj) .......... (4)
Incidentally, for an A/F ratio range of 15 to
18, the concentration of NOX as noxious exhaust
components rises to a high value, as is well known in
the art. In order to reduce the emission of NOX, there-
fore, the set A/F ratio has to avoid the range of 15
to 18.
The detail of step 280 is shown in Fig. 16. At
step 400, the set A/F ratio (A/F)A is calculated by
use of formula (4). If this (A/F)A is within the range
of 15 to 18, namely, if it is judyed at step 401 that
15 < (A/F)A < 18, the emission of NOX increases, as
has been described above. At steps 405 and 406, the
value KMR is detected for (A/F)A = 15 or (A~F)A = 18
by use of formula (4) to prepare a rich mixture. More
,
:. .
.;
:~ .
- 13 ~ t ~ 3 ~ ~
specifically, the amount of air is reducecl while the
amount of fuel is left as it is, so that the A/F ratio
of high NO~ emission may be avoided.
A new throttle valve opening ~T is then obtained
by use of the newly retrieved KMR from the value of
Ti x KA/KMR and the revolutions N in view of the relations
of Fig. 14. Thus, the emission of NOX can be reduced.
Returning to Fig. 5, at step 290, the opening
signal aT of the throttle valve 116 thus obtained is
sent to the throttle valve actuator 114. The difference
A~T between the set throttle valve opening ~T and the
present throttle valve opening detected by the po-tentio-
meter 145 is obtained to control the throttle valve 116.
Incidentally, if the step motor 143 is used as the throttle
valve actuator 114, the number of pulses corresponding to
the difference ~T is given to the step motor 143. If
it is necessary to set the throttle valve opening highly
accurately, moreover, the actual throttle valve opening
is measured by the potentiometer 145 to provide a feedback
loop to set the throttle valve opening ~T.
The ignition timing is obtained by an inter-
polation from the relation between the number of revolu-
tions N of the engine and the basic amoun-t offuel injection
Ti, which is shown by using the ignition timing BTDC as
a parameter.
Incidentally, the aforementioned relations~and
tables of Figs. 6 to 10 and Figs. 12 to 15 are stored in
advance in the ROM 141 of the control unit 112.
In the lean mixture combustion system of the
prior art, the accelerator and the -throttle valve are
mechanically connected through a link or the like, so
that the throttle valve has its opening increased to
increase the amount of suction air with the increase in
the depression amount OA, as shown in Fig. 22c. In
the vicinity of the maximum depression aA of the accelerator
3~
- 14 -
pedal, on the other hand, a high engine output is required,
so that: there is nothing for it to do but enrich the
mixture. As a result, the characteristics of the A/F
ratio for the depression amount ~A of the accelerator
pedal are expressed by the curve shown in Fig. 22d.
Accordingly, the characteristics of the opening
time of the fuel injector 109, i.e., the amount of fuel
for the depression ~A of the accelerator pedal are
expressed by the curve shown in Flg. 22a, and the engine
torque is expressed by the curve shown in Fig. 22b,
because it is in proportion to the amount of fuel. More
specifically, the torque is characterized in that its
increment is low for the range of a smaller depression
~A of the accelerator pedal, but suddenly becomes high
in the vicinity of the maximum depression. With these
characteristics, however, the torque is so small as to
make the driver experience a lack of acceleration in the
lower range of the depression ~A of the accelerator pedal.
On the other hand, the characteristics of the
20 A/F ratio are expressed by the curve shown in Fig. 22d,
such that the A/F ratio continuously changes from a lean
mixture to a rich mixture in the range oE a large
depression ~A of the accelerator pedal. This raises the
defect that much NOX is emitted, because the A/F ratio
25 range of 15 to 18 is passed with the increase in depression
~A of the accelerator pedal.
According to the control shown in the flow chart
of Fig. 5, on the contrary, the amount of fuel can be
characterized to be generally proportional to the
30 depression ~A of the accelerator pedal, as shown in Fig.
21a. As a result, the control of Fig. 5 has the advantage
that the torque is uniformly increased with the depression
~A of the accelerator pedal, as shown in Fig. 21b, so that
the acceleration is smooth from a small depression to a
35 larger one.
': ,
'
- 15 -
Since, moreover, the amount of suction air to the
englne can be freely set by the throttle valve actuator
114, the A/F ratio can be characterized, as shown ln Fig.
21d, if the amount of suction air is set as shown in Fig.
21c. As a result, the operation can be conducted without
any increase in NOX emissions from a small depression
to a large one, by skipping over the A/F ratio range of
15 to 1~.
In other words, the amount of suction air can be
set freely in accordance with the command of the control
unit 112 by adopting the throttle valve actuator 114, so
that the torque to be produced ~y the engine and the A/F
ratio of the fuel mixture can be controlled independently
of each other. As a result, there can be attained the
effect that the countermeasures against noxious exhaust
emission can be simplified, the acceleration performance
and the fuel economy being caused to be compatible by
making the amount of fuel injection proportional to the
depression ~A of the accelerator pedal.
Incidentally, if the A/F ratio is controlled to
vary at steps 405, 406 and 403 of Fig. 16, the torque
fluctuates a little. In this case, however, these torque
fluctuations can be suppressed, if the ignition timing is
corrected at steps 407 and 408, as shown in Fig. 17. More
specifically, the torque fluctuations are suppressed by
retarding the ignition timing for the lower A/F ratio
and advancing it for the higher A/F ratio.
Fig. 1~ shows another example in which a correction
is made by the intake pressure.
Generally speaking, the amount of air Qa/N to be
sucked into the cylinder for one revolution of the engine
is expressed by the following formula (5) if a suc-tion
pressure Pm is used:
Qa/N = K2 x Pm x r~ x KAIR ................. (5)
wherein:
- 16 ~ 6~3Ct~
K2 : a constan-t;
~ : a suction efficiency; and
KAIR: a correction coefficient of suction air
temperature.
The pressure of suc~ion air Pm for giving the
amount of suction air for a set A/F' ratio is given by
the following formula (6) with the basic amount of fuel
injection Ti and the correction coefficient KMR of A/F
ratio depending upon the engine operatlonal condition:
10 Pm = K3 x Ti/ (n x KAIR x KMR) ............ (6)
wherein K3: a constant.
Therefore, if the opening of the throttle valve
116 is subjected to the feedback loop control, while the
aetual suetion air pressure is being measured by the
suetion air pressure sensor 115, the set suetion air
pressure of formula (6) may be attained, and a hi~hly
aeeurate control of the amount of suetion ean be realized.
New steps will now be described with referenee
to the flow ehart of Fig. 18. At step 500, the set
pressure of suetion air Pm is calculated by the use of
formula (6). At step 501, the aetual pressure of suetion
air Pmr is read by the use of the suetion air pressure
sensor 115.
At step 502, the throttle valve opening eorreetion
eoefficient K~ is ealeulated by use of the following
formula (7) from the set suetion air pressure Pm calculated
at step 500 and -the actual suction air pressure Pmr, and
is stored in the RAM 1~0 of the control unit 112:
K~ = K4(Pm ~ Pmr) ~ K5 J (Pm -Pmr)dt....... (7)
wherein:
K4: a constant of proportion; and
K5: an integration constant.
At step 504, the correction coefficient K~ of
, . .. .
' ' ~ : ' '' ':''" ',
- 17 -
the throttle valve openin~ is retrieved in the t~ble shown
in Fig. 19. Fig. 19 tabulates the correction coefficient
K~ for the basic amount of fuel injection Ti retrieved at
step 180 (Fig. 5) and -the number of revolution N of the
engine.
At step 505, a corrected opening ~T' is calculated
from the following formula (8):
9T' = K0 x ~T ..................... (8),
wherein:
K0: a correction coefficient of throttle valve
opening; and
~T: a set opening of throttle valve.
Fig. 20 shows a modification of E'ig. 18, in which
the air Elow sensor 119 is used in place of the suction
air pressure sensor 115 of Fig. 18. From the basic amount
of fuel injection Ti and the correction coefficient of
the A/F ratio KMR depending upon the engine operational
condition, the amount of suction air Qa for the set A/F
ratio is given by the following formula (9):
Qa = K6 x Ti x N/KMR .............. (9)
wherein K6: a constant.
Therefore, the opening of the throttle valve lL6
is subjected to a closed loop control, while the actual
air flow rate is being metered by the air fLow sensor 119,
so that -the suction air amount Qa may be the set one
given by formuLa (9). New steps will now be described
with reference to the flow chart of Fig. 20.
At step 600, the air fLow rate Qa is caLcuLated
by the use of formuLa (9) from the Ti of step L80 and
ICMR of step 2L0. The value thus calculated is designated
at Qa. At step 601, the ac-tual air fLow rate is read
by the use of the air flow sensor 119.
At step 602, the throttle valve correction
coefficient K~ is calcuIated by the use of -the following
'; ' ~
- 18 -
formula (10) from the set air flow rate Qar. At step 603,
the value K~ is stored in the RAM 140:
K0 = K7(Qa - Qar) + K8 ~ (Qa - Qar)dt...... (10)
wherein:
S K7: a constant of proportion; and
K8: an integration constant.
This value KO is similar to that of Fig. 19 and
steps 604 and 605 are also similar to the steps 504 and
505.
Although we have shown and described only some
forms of apparatus embodying the invention, it is under-
stood that various changes and modifications may be
made therein within the scope of the appended claims
without departing from the spirit and scope of the
invention.
'
': :
