[0001] The invention relates to method and apparatus for detecting faults in glow plugs
provided in diesel engines.
[0002] In order to improve the engine starting performance during start-up of a diesel engine,
it has been known to pre-heat the diesel engine by applying current to glow plugs
provided in the diesel engine. Such a glow plug includes a heating element that is
energized so as to generate heat. When the heating element itself suffers from disconnection
or an electric power supply line that leads to the heating element suffers from disconnection
or cut-off, a problem arises in the engine starting performance, in particular, during
a cold start of the engine. To prevent this problem, it has been proposed to provide
a glow-plug disconnection detecting device for detecting faults or abnormalities in
the glow plugs. Examples of such detecting devices are disclosed in, for example,
Japanese laid-open Patent Publications No. 11-182400, No. 57-26275 and No. 58-113581.
[0003] In the known devices, the presence of a fault, such as disconnection of a glow plug,
is determined based on voltage changes resulting from energization of the glow plug
during start-up of the engine. However, a fault in the glow plug may not be determined
with high accuracy if the fault detection utilizes normal energization control performed
on the glow plug, since in some cases the fault determination must be made under a
situation in which highly accurate detection is impossible or difficult.
[0004] It is therefore proposed to implement forced energization and deenergization of the
glow plug for the purpose of diagnosis, separately from or independently of the normal
glow-plug energization control intended for warm-up of the engine, so as to increase
a degree of freedom in the timing of implementation of glow-plug fault detection and
thus permit highly accurate fault detection or determination under a situation suitable
for fault detection.
[0005] Although a fault in the glow plug may be determined with high accuracy if the glow
plug is forced to be energized and deenergized for the purpose of diagnosis independently
of the normal energization control, a circuit switching mechanism, such as a glow
relay, is more frequently switched between ON and OFF because of the diagnostic energization
and deenergization. With the number of times of switching thus increased, the durability
of the circuit switching mechanism may be reduced.
[0006] It is therefore an object of the invention to provide glow-plug fault detecting method
and apparatus which are able to detect or determine a fault in a glow plug with sufficiently
high accuracy, without suffering from a reduction in the durability of a circuit switching
mechanism for the glow plug.
[0007] To accomplish the above object, there is provided according to a first aspect of
the invention a method of detecting a fault in a glow plug provided in a diesel engine,
including a process of determining the presence of a fault in the glow plug by utilizing
a forced change in a state of energization of the glow plug, characterized in that
a process of determining a possibility of a fault in the glow plug without utilizing
the forced change in the state of energization of the glow plug is executed prior
to the process of determining the presence of the fault, and the process of determining
the presence of the fault is not executed if it is determined in the process of determining
the possibility of the fault that there is no possibility of a fault in the glow plug.
[0008] A fault in the glow plug is detected with relatively low reliability if the fault
detection does not involve the forced change in the state of energization of the glow
plug for the purpose of diagnosis, as compared with the case where the fault detection
utilizes the forced change in the state of energization of the glow plug. However,
a possibility of a fault in the glow plug can be determined with sufficiently high
reliability even without using the forced change in the state of energization. According
to the above aspect of the invention, the process of determining the possibility of
the fault is executed before the presence of the fault is determined by utilizing
the forced change in the state of energization of the glow plug, and, if it is determined
in the process of determining the possibility of the fault that there is no possibility
of a fault in the glow plug, there is no need to execute the process of determining
the presence of the fault by utilizing the forced change in the state of energization
of the glow plug.
[0009] Accordingly, the process of determining the presence of the fault that assures high
reliability in fault detection may be executed only if the fault possibility determination
reveals that there is a possibility of a fault in the glow plug, namely, if a determination
is made which does not necessarily negate a possibility of the fault. Thus, the number
of times the glow plug is forced to be energized or deenergized for the purpose of
diagnosis can be reduced. Consequently, otherwise possible reduction of the durability
of the circuit switching mechanism can be suppressed while assuring accurate determination
of the fault in the glow plug.
[0010] The "forced change" mentioned herein means a process of changing the state of energization
of the glow plug for the purpose of detecting a fault in the glow plug. This definition
applies in the other portions of the description and claims.
[0011] The above-indicated forced change in the state of energization of the glow plug may
be in the form of switching of the glow plug between an energized state and a deenergized
state.
[0012] As described above, the forced change in the state of energization may be, in particular,
caused by switching of the glow plug between the energized state and the deenergized
state. With the forced change thus made, a phenomenon associated with a fault in the
glow plug is likely to occur, but at the same time the durability of the circuit switching
mechanism tends to be lowered. By employing the fault possibility determination that
there is a possibility of a fault as a precondition for execution of the process of
determining the presence of the fault, the number of changes in the state of energization
of the glow plug can be reduced, and otherwise possible reduction in the durability
of the circuit switching mechanism can be suppressed while assuring accurate fault
detection.
[0013] In one embodiment of the first aspect of the invention, the possibility of the fault
in the glow plug is determined based on an operating state of the diesel engine detected
before completion of a starting cycle of the engine while the glow plug is in an energized
state.
[0014] If current is applied to the glow plug and the glow plug normally generates heat,
electric energy is consumed for generating heat, and the diesel engine can be smoothly
started. However, if current is not actually applied to the glow plug even if a command
to energize the glow plug is generated, or the amount of current applied to the glow
plug is extremely small, no consumption of electric energy for heating of the glow
plug takes place, and the engine cannot be started smoothly. In this case, the fault
in energization of the glow plug is reflected by the operating state of the engine
detected before completion of the engine starting cycle. Thus, the possibility of
the fault in the glow plug can be determined based on the engine operating state detected
before completion of the starting cycle of the engine.
[0015] In another embodiment of the first aspect of the invention, when it is determined
in the process of determining the possibility of the fault that there is a possibility
of a fault in the glow plug, and the process of determining the presence of the fault
cannot be completed within a period of time set for energization of the glow plug
started for start-up of the diesel engine, the process of determining the presence
of the fault is not executed.
[0016] With the above arrangement, the process of determining the presence of the fault
can be completed within the set time for energization of the glow plug that is started
for engine start-up. Thus, normal energization of the glow plug is not prolonged in
vain because of the forced change of the state of energization for the purpose of
determining the presence of the fault. Accordingly, energy consumption due to the
process of determining the presence of the fault in the glow plug can be reduced.
[0017] In a further embodiment of the first aspect of the invention, when it is determined
in the process of determining the possibility of the fault that there is a possibility
of a fault in the glow plug, the process of determining the presence of the fault
is executed provided that the diesel engine is under idle speed control.
[0018] Since the presence of the fault is determined on limited occasions, namely, only
during idling of the engine, the determination is less likely to be influenced by
other factors, thus assuring further improved accuracy in the determination of the
presence of the fault.
[0019] In a still further embodiment of the above aspect of the invention, when it is determined
in the process of determining the possibility of the fault that there is a possibility
of a fault in the glow plug, the process of determining the presence of the fault
is executed provided that a battery voltage is within a reference voltage range.
[0020] If the battery voltage is too low, the load on a power generating mechanism, such
as an alternator, may reach 100%. Even if the state of energization of the glow plug
is forced to be changed for the purpose of diagnosis in this condition, substantially
no change may occur in the load on the diesel engine, and the presence of the fault
may not be determined with high accuracy. If the battery voltage is too high, on the
other hand, the load on the power generating mechanism is reduced down to 0%. Even
if the state of energization of the glow plug is forced to be changed in this condition,
substantially no change may occur in the load on the diesel engine, and therefore
the presence of the fault may not be determined with high accuracy. In view of these
situations, the reference voltage range is provided, and the process of determining
the presence of the fault is not executed if the battery voltage is outside the reference
range, so as to prevent a reduction of the accuracy in the determination of the presence
of the fault.
[0021] In another embodiment of the first aspect of the invention, the presence of the fault
in the glow plug is determined based on a change in an operating state of the diesel
engine which is caused by the forced change in the state of energization of the glow
plug.
[0022] More specifically, the presence of a fault in the glow plug can be determined with
high reliability by measuring a change in the operating state of the diesel engine
that occurs in response to the forced change in the state of energization of the glow
plug. Furthermore, since the process of determining the presence of the fault is executed
under a precondition that there is a possibility of a fault in the glow plug, the
state of energization of the glow plug is prevented from being changed for the purpose
of diagnosis when there is no possibility of the fault. While the forced change in
the state of energization of the glow plug may be effected once or may be repeated
twice or more, it is preferable to change the state of energization of the glow plug
only once or twice.
[0023] In a further embodiment of the first aspect of the invention, the process of determining
the presence of the fault comprises (a) a step of causing a first forced change in
the state of energization of the glow plug, (b) a step of making a first determination
based on a change in the operating state of the diesel engine which is caused by the
first forced change, (c) a step of finishing the process of determining the presence
of the fault and resuming the state of energization of the glow plug established before
the first forced change if the first determination indicates that the glow plug is
normal, (d) a step of causing a second forced change in the state of energization
of the glow plug so as to bring the glow plug into the state of energization established
before the first forced change if the first determination does not indicate that the
glow plug is normal, (e) a step of making a second determination based on a change
in the operating state of the diesel engine which is caused by the second forced change,
(f) a step of finishing the process of determining the presence of the fault if the
second determination indicates that the glow plug is faulty, and (g) a step of finishing
the process of determining the presence of the fault without making a determination
on the fault in the glow plug when the second determination does not indicate that
the glow plug is faulty.
[0024] With the above method, the presence of a fault in the glow plug can be determined
with high reliability by checking the engine operating state in two steps upon the
first forced change and second forced change in the state of energization of the glow
plug. Furthermore, the first forced change is cancelled by the second forced change,
so that the glow plug can resume the original state of energization when the process
of determining the presence of the fault is finished.
[0025] The above-indicated change in the operating state of the diesel engine may be in
the form of a change in an engine load.
[0026] When current starts being applied to the glow plug or the amount of current applied
to the glow plug is increased in response to the first forced change or the second
forced change in the state of energization of the glow plug, the quantity of power
generated is increased in accordance with an increase of the speed of battery exhaustion,
and the load on the diesel engine is accordingly increased. On the contrary, if current
applied to the glow plug is cut off or the amount of current applied to the glow plug
is reduced in response to the first forced change or the second forced change, the
quantity of power generated is reduced in accordance with a decrease of the speed
of battery exhaustion, and the load on the diesel engine is accordingly reduced. When
current is not normally applied to the glow plug, therefore, no change in the engine
load appears in response to the first forced change or the second forced change in
the state of energization of the glow plug. Thus, the presence of a fault in the glow
plug can be determined based on a change in the engine load.
[0027] In another example, the above-indicated change in the operating state of the engine
may be in the form of a change in a battery voltage.
[0028] When current starts being applied to the glow plug or the amount of current applied
to the glow plug is increased in response to the first forced change or the second
forced change in the state of energization of the glow plug, the voltage of the battery
used for energization of the glow plug drops. On the contrary, if current applied
to the glow plug is cut off or the amount of current applied to the glow plug is reduced
in response to the first forced change or the second forced change, the battery voltage
rises. When current is not normally applied to the glow plug, therefore, no change
in the battery voltage appears in response to the first forced change or the second
forced change in the state of energization of the glow plug. Thus, the presence of
a fault in the glow plug can be determined based on a change in the battery voltage.
[0029] In a further example, the above-indicated change in the operating state of the engine
may include a change in a fuel injection quantity and a change in a battery voltage.
[0030] Since both the fuel injection quantity and the battery voltage change in response
to the first forced change or the second forced change in the state of energization
of the glow plug, the presence of a fault in the glow plug can be determined with
improved accuracy by measuring the changes in both of the injection quantity and the
battery voltage.
[0031] According to a second aspect of the invention, there is provided an apparatus for
detecting a fault in a glow plug provided in a diesel engine, characterized by comprising
(a) fault possibility determining means for determining a possibility of a fault in
the glow plug based on a phenomenon that occurs without involving a forced change
in a state of energization of the glow plug, and (b) fault presence determining means
for determining the presence of the fault in the glow plug based on a phenomenon that
occurs in response to the forced change in the state of energization of the glow plug
when the fault possibility determining means determines that there is a possibility
of a fault in the glow plug, wherein the fault presence determining means does not
determine the presence of the fault by causing the forced change in the state of energization
of the glow plug when the fault possibility determining means determines that there
is no possibility of a fault in the glow plug.
[0032] The fault detecting apparatus according to the second aspect of the invention provides
similar advantage effects to those provided by the fault detecting method according
to the first aspect of the invention.
[0033] The foregoing and/or further objects, features and advantages of the invention will
become more apparent from the following description of exemplary embodiments with
reference to the accompanying drawings, in which like numerals are used to represent
like elements and wherein:
Fig. 1 is a view schematically showing an accumulator injection type diesel engine
and its control system that employ glow-plug fault detecting method and apparatus
according to a first embodiment of the invention;
Fig. 2 is a view showing an electric power supply system for supplying electric power
to a glow plug of the engine of Fig. 1;
Fig. 3 is a flowchart illustrating a part of a fuel injection quantity control routine
of the first embodiment:
Fig. 4 is a flowchart illustrating a part of the fuel injection quantity control routine
following that of Fig. 3;
Fig. 5 is a flowchart illustrating a starting precondition checking process of the
first embodiment;
Fig. 6 is a view explaining a map used for obtaining a normal starting period Tstanorm
from a coolant temperature THWs measured upon a start of the engine starting cycle
in the starting precondition checking process;
Fig. 7 is a view explaining a map used for obtaining a starting-time normal minimum
voltage Vstanorm from the coolant temperature THWs measured upon a start of the engine
starting cycle in the starting precondition checking process;
Fig. 8 is a flowchart illustrating a part of a process of checking conditions for
execution of glow-plug fault determination according to the first embodiment;
Fig. 9 is flowchart illustrating a part of the process of checking conditions for
execution of glow-plug fault determination, which follows that of Fig. 8, according
to the first embodiment;
Fig. 10 is a flowchart illustrating a provisional fault determination process according
to the first embodiment;
Fig. 11 is a flowchart illustrating a main fault determination process according to
the first embodiment;
Fig. 12 is a flowchart illustrating a part of a process of checking conditions for
stopping glow-plug fault determination according to the first embodiment;
Fig. 13 is a flowchart illustrating a part of the process of checking conditions for
stopping glow-plug fault determination, which follows that of Fig. 12, according to
the first embodiment;
Fig. 14 is a time chart showing one example of control according to the first embodiment;
Fig. 15 is a time chart showing another example of control according to the first
embodiment;
Fig. 16 is a time chart showing a further example of control according to the first
embodiment;
Fig. 17 is a time chart showing a still further example of control according to the
first embodiment;
Fig. 18 is a time chart showing another embodiment of control according to the first
embodiment;
Fig. 19 is a flowchart illustrating a part of a process of checking conditions for
execution of glow-plug fault determination according to a second embodiment of the
invention;
Fig. 20 is a flowchart illustrating a provisional fault determination process according
to the second embodiment;
Fig. 21 is a flowchart illustrating a main fault determination process according to
the second embodiment;
Fig. 22 is a flowchart illustrating a part of a process of checking conditions for
stopping glow-plug fault determination according to the second embodiment;
Fig. 23 is a time chart showing one example of control according to the second embodiment;
Fig. 24 is a time chart showing another example of control according to the second
embodiment;
Fig. 25 is a time chart showing a further example of control according to the second
embodiment; and
Fig. 26 is a time chart showing a still further example of control according to the
second embodiment.
[0034] Fig. 1 schematically shows an accumulator injection type diesel engine (or common
rail type diesel engine) 2 and a control system for the diesel engine 2, for which
glow-plug fault detecting method and apparatus according to a first embodiment of
the invention are employed. The accumulator injection type diesel engine 2 as one
type of automobile engines is installed on a vehicle.
[0035] The diesel engine 2 includes a plurality of cylinders (i.e., four cylinders in this
embodiment, of which only one is shown in Fig. 1) #1, #2, #3, #4, and an injector
4 is provided with respect to each of combustion chambers of the cylinders #1 through
#4. Fuel injection from the injector 4 to a corresponding one of the cylinders #1
to #4 of the diesel engine 2 is controlled by controlling the ON/OFF state of a solenoid-operated
valve 4a used for fuel injection control with respect to the injector 4.
[0036] The injector 4 is connected to a common rail 6 serving as an accumulator pipe used
in common to the four cylinders. While the solenoid-operated valve 4a used for fuel
injection control is open, fuel in the common rail is injected through the injector
4 into a corresponding one of the cylinders #1 to #4. A relatively high pressure that
is equivalent to a fuel injection pressure is accumulated in the common rail 6. To
build up the fuel injection pressure, the common rail 6 is connected to a delivery
port 10a of a supply pump 10 via a supply pipe 8. A check valve 8a, which is provided
in the supply pipe 8, serves to permit supply of the fuel from the supply pump 10
to the common rail 6, and inhibit reverse flow of the fuel from the common rail 6
to the supply pump 10.
[0037] A fuel tank 12 is connected to the supply pump 10 via a suction port 10b, and a filter
14 is provided in a pipe connecting the fuel tank 12 with the supply pump 10. In operation,
the supply pump 10 takes in the fuel from the fuel tank 12 through the filter 14.
At the same time, the supply pump 10 raises a fuel pressure to a required high pressure
by using a cam (not shown) that moves in synchronism with the rotating diesel engine
2 for reciprocating a plunger, and supplies the highly pressured fuel to the common
rail 6.
[0038] A pressure control valve 10c is provided in the vicinity of the delivery port 10a
of the supply pump 10. The pressure control valve 10c serves to control the pressure
of fuel delivered from the delivery port 10a toward the common rail 6. When the pressure
control valve 10c is opened, a redundant portion of the fuel that is not delivered
from the delivery port 10a is returned to the fuel tank 12 via a return port 10d provided
in the supply pump 10 and a return pipe 16.
[0039] An intake passage 18 and an exhaust passage 20 are connected to a combustion chamber
of each of the cylinders #1 to #4 of the diesel engine 2. A throttle valve (not shown)
is disposed in the intake passage 18. In operation, the flow rate of the intake air
to be introduced into the combustion chamber is controlled by adjusting the opening
angle of the throttle valve according to the operating state of the diesel engine
2.
[0040] In addition, a glow plug 22 is disposed in the combustion chamber of each of the
cylinders #1 to #4 of the diesel engine 2. The glow plug 22 glows (i.e., heats and
turns red) when current is applied to the plug 22 via a glow relay 22a immediately
before start-up of the diesel engine 2, and a part of fuel spray is directed at the
hot glow plug 22, thereby to promote or aid firing and combustion in the combustion
chamber. Thus, the glow plug 22 functions as a start assist device for aiding combustion
upon start-up of the engine 2. A fault, such as disconnection, of the glow plug 22
is determined through a process as described later.
[0041] The diesel engine 2 is equipped with various sensors as follows, for detecting operating
conditions of the diesel engine 2 in the first embodiment. As shown in Fig. 1, an
accelerator stroke sensor 26 for detecting the accelerator pedal position or degree
of depression of the accelerator pedal ACCPF is provided in the vicinity of an accelerator
pedal 24 as shown in Fig. 1. Also, the diesel engine 2 is provided with a starter
30 for starting the diesel engine 2. The starter 30 includes a starter switch 30a
for detecting an operating state of the starter 30. A water temperature sensor 32
for detecting the temperature of a coolant (coolant temperature THW) is mounted in
the cylinder block of the diesel engine 2. In addition, an oil temperature sensor
34 for detecting the temperature THO of engine oil is provided in an oil pan (not
shown). Also, a fuel temperature sensor 36 for detecting a fuel temperature THF is
provided in the return pipe 16, and a fuel pressure sensor 38 for detecting the pressure
of fuel in the common rail 6 is provided in the common rail 6. A NE sensor 40 for
detecting the engine speed is disposed in the vicinity of a pulser (not shown) provided
at a crankshaft (not shown) of the diesel engine 2. In operation, rotation of the
crankshaft is transmitted to a camshaft (not shown) for opening and closing an intake
valve 18a and an exhaust valve 20a via a timing belt, and the like. The speed of rotation
of the camshaft is set to be a half of the speed of rotation of the crankshaft. A
cylinder discrimination sensor 42 is disposed in the vicinity of a pulser (not shown)
provided at the camshaft. In the first embodiment, the engine speed NE, crank angle
CA, and the top dead center (TDC) of the intake stroke of the first cylinder #1 are
calculated based on pulse signals transmitted from these sensors 40, 42. A transmission
44 is provided with a shift position sensor 46 for detecting the gear stage or position
of the transmission 44. A vehicle speed sensor 48 is provided on the side of the output
shaft of the transmission 44, for detecting the vehicle speed SPD based on the speed
of rotation of the output shaft. In addition, an air conditioner (not shown) to be
driven with power of the diesel engine 2 is provided, and an air-conditioner switch
40 for generating a command to drive the air conditioner is provided.
[0042] In the first embodiment, an electronic control unit (ECU) 52 is provided for governing
various controls of the diesel engine 2. The ECU 52 performs control processes, such
as fuel injection quantity control and glow-plug energization control as described
later, for controlling the diesel engine 2, a process of determining a fault or abnormality
in glow-plug energization as described later, and so forth. The ECU includes, as a
main component, a microcomputer including a central processing unit (CPU), a read-only
memory (ROM) that stores in advance various programs, maps, and so forth, a random
access memory (RAM) that temporarily stores operation results of the CPU, and a backup
RAM that stores operation results, pre-stored data, and so forth. The microcomputer
further includes a timer or a counter, an input interface, an output interface, and
other components. The accelerator stroke sensor 26, water temperature sensor 32, oil
temperature sensor 34, fuel temperature sensor 36, fuel pressure sensor 38 and others
are connected to the input interface of the ECU 52 via respective buffers, multiplexers,
and A/D converters (not shown). Also, the NE sensor 40, cylinder discrimination sensor
42, vehicle speed sensor 48 and others are connected to the input interface of the
ECU 52 via waveform shaping circuits (not shown). Furthermore, the starter switch
30a, shift position sensor 46, air-conditioner switch 50 and others are directly connected
to the input interface of the ECU 52. Other than signals from these sensors, the ECU
52 also receives battery voltage VB, control duty DF of an alternator 54, and other
parameters, and read values thereof. The CPU reads signals of the above-indicated
sensors and switches via the input interface. On the other hand, the solenoid-operated
valves 4a, pressure control valve 10c, glow relays 22a and others are connected to
the output interface of the ECU 52 via respective drive circuits. The CPU performs
calculations based on the input values read via the input interface, and favorably
controls the solenoid-operated valves 4a, pressure control valve 10a, glow relays
22a and others via the output interface.
[0043] As shown in Fig. 2 illustrating an electric power supply system of the diesel engine
2, the alternator 54 and an air-conditioner compressor 56 are driven or rotated by
the crankshaft 2a of the diesel engine 2 via a belt 2b. A voltage regulator 54a is
provided in the alternator 54. The voltage regulator 54a causes the alternator 54
to output a voltage corresponding to a duty signal received from a controller 58 for
the alternator 54. The controller 58 detects voltage VB of the battery 60, and performs
duty control on the voltage regulator 54a so that the battery 60 is kept in an appropriate
charged condition. When the ECU 52 turns on the glow relay 22a, electric power is
supplied from the battery 60 and the alternator 54 to the glow plug 22 so that the
glow plug 22 can generate heat.
[0044] Next, the fuel injection quantity control routine and the glow-plug energization
control routine, which are executed by the ECU 52 in the present embodiment, will
be described.
[0045] Fig. 3 and Fig. 4 illustrate the fuel injection quantity control routine. This routine
is an interrupt routine that is executed at regular intervals of a fixed crank angle
(i.e., executed for each explosion stroke).
[0046] Upon start of the fuel injection quantity control routine, data necessary for control
are read into a work region within the RAM of the ECU 52 in step S110. The data may
include the engine speed NE detected by the NE sensor 40, the degree of depression
of the accelerator pedal ACCPF detected by the accelerator stroke sensor 26, the shift
position SFT detected by the shift position sensor 46, and the vehicle speed SPD detected
by the vehicle speed sensor 48.
[0047] In the next step S120, a governor injection quantity command value QGOV1 for an engine
idling state is calculated from an idling governor injection quantity command value
map in which the engine speed NE and the degree of depression of the accelerator pedal
ACCPF are used as parameters, based on the engine speed NE and the degree of depression
of the accelerator pedal ACCPF. This map, which is empirically determined in advance
for the idling state, is stored in the ROM of the ECU 52. Since discrete values are
arranged in this map, the governor injection quantity command value QGOV1 is determined
through interpolation when no value coincides with map values as parameters. It is
to be understood that setting of a map and calculation through interpolation are similarly
performed with respect to other maps.
[0048] In the following step S130, a governor injection quantity command value QGOV2 for
an engine operating state other than the idling state is calculated from a non-idling
governor injection quantity command value map in which the engine speed NE and the
degree of depression of the accelerator pedal ACCPF are used as parameters, based
on the engine speed NE and the degree of depression of the accelerator pedal ACCPF.
Furthermore, an auxiliary governor injection quantity command value QGOV3 that gives
an auxiliary characteristic to the non-idling governor injection quantity command
value QGOV2 is calculated in step S140 from an auxiliary governor injection quantity
command value map using the engine speed NE and the degree of depression of the accelerator
pedal ACCPF as parameters, based on the engine speed NE and the degree of depression
of the accelerator pedal ACCPF.
[0049] Next, it is determined in step S150 whether the diesel engine 2 is in an operating
state other than the idling state. For example, the engine 2 is determined to be in
the idling state when the vehicle speed SPD is equal to 0 km/h and the accelerator
stroke sensor 26 indicates that the degree of depression of the accelerator pedal
ACCPF is equal to 0% after completion of warm-up. If the engine 2 is in the idling
state (i.e., when a negative determination is made in step S150), an engine speed
deviation NEDL of the actual engine speed NE from a target engine speed NTRG for the
idling state is calculated in step S160 as indicated in the following expression (1).

[0050] Next, an injection quantity correction value QIIDL corresponding to the engine speed
deviation NEDL is determined in step S170 from a map using the engine speed deviation
NEDL as a parameter. Alternatively, the injection quantity correction value QIIDL
may be determined from a function using the engine speed difference NEDL as a parameter.
[0051] Then, an idling injection quantity correction value QII is calculated in step S180
based on the injection quantity correction value QIIDL, according to the following
expression (2).

[0052] In the above expression (2), QII in the right side represents an idling injection
quantity correction value QII determined in the last control cycle, and "± QIIDL"
means "+ QIIDL" when NTRG is equal to or greater than NE and "- QIIDL" when NTRG is
smaller than NE.
[0053] In step S190 following step S180, a governor injection quantity command value QGOV
is calculated according to the following expression (3). If it is determined in step
S150 that the engine 2 is in an operating state other than idling (i.e., if an affirmative
determination is made in step S150), the control directly proceeds to step S190 for
calculating the governor injection quantity command value QGOV.

where QIP is an offset value in the case where the air conditioner, or the like, causes
a load during idling, and QIPB is an offset value in the case where the air conditioner,
or the like, causes a load when the engine 2 is in an operating state other than idling.
MAX ( ) is an operator that selects the maximum value from values in the parentheses.
[0054] Next, it is determined in step S200 whether the vehicle is accelerating or decelerating.
This determination is made by, for example, determining whether the governor injection
quantity command value QGOV is larger or smaller than a basic injection quantity command
value QBASEOL calculated in the last control cycle.
[0055] If the vehicle is accelerating or decelerating (i.e., if an affirmative determination
is made in step S200), an operation to restrict an increase or a decrease of the governor
injection quantity command value QGOV is performed. This operation aims at preventing
shocks that would occur when the governor injection quantity command value QGOV rapidly
changes. If the calculation result of step S190 indicates that the governor injection
quantity command value QGOV has largely changed from the basic injection quantity
command value QBASEOL, the command value QGOV is corrected so as not to cause shocks.
[0056] In the next step S220 (Fig. 4), the governor injection quantity command value QGOV
is set as the basic injection quantity command value QBASE. If step S200 determines
that the vehicle is not accelerating nor decelerating (namely, if a negative determination
is made in step S200), the control directly proceeds to step S220.
[0057] In step S230 following step S220, the final basic injection quantity command value
QFINC is calculated by subjecting the basic injection quantity command value QBASE
to a guard process to limit the command value QBASE to the maximum injection quantity
command value QFULL according to the following expression (4).

where MIN( ) is an operator that selects the minimum value from values in the parentheses.
[0058] In the next step S240, a main injection quantity command value QFPL is calculated
by subtracting a pilot injection quantity command value QPL from the final basic injection
quantity command value QFINC, as indicated in the following expression (5).

[0059] Next, a main injection duration TQFPL is calculated in step S250 from a map or a
function fq, based on the main injection quantity command value QFPL. Furthermore,
a pilot injection duration TQPL is calculated in step S260 from a map or a function
fq, based on the pilot injection quantity command value QPL. Then, the basic injection
quantity command value QBASE calculated in the current control cycle is set as the
last basic injection quantity command value QBASEOL in step S270. In this manner,
the fuel injection quantity control routine is once finished.
[0060] In the meantime, the ECU 52 performs the above-mentioned glow-plug energization control
process in the following manner. When turn-on of the ignition switch is detected,
a command to energize the glow plug 22 is generated, and current is applied from the
battery 60 to the glow plug 22, so that the glow plug 22 starts generating heat. Then,
a starting cycle of the diesel engine 2 is started upon turn-on of the starter 30.
When the starting cycle of the diesel engine 2 is completed, and a certain delay time
elapses, energization of the glow plug 22 is stopped. The delay time is set to be
shorter as the coolant temperature THW increases.
[0061] Next, operations to detect a fault, such as disconnection, of the glow plug 22 will
be described, referring to Fig. 5 and Fig. 8 through Fig. 13.
[0062] Initially, a starting precondition checking process (Fig. 5) will be described. This
process is repeatedly executed at certain time intervals after a power supply of the
ECU 52 is turned ON. Upon a start of this process, it is determined in step S300 whether
starting preconditions as described below have been determined. Since the starting
preconditions have not been determined in the initialized state at the time when the
power supply is turned ON, a negative determination is made in step S300, and it is
determined in step S301 whether the coolant temperature THW detected by the water
temperature sensor 32 is equal to or lower than a predetermined water-temperature
threshold value. If the coolant temperature THW is larger than the threshold value,
a noticeable difference in the engine starting performance does not appear between
the case where the glow plug 22 normally generates heat and the case where the glow
plug 22 does not generate heat because of, for example, disconnection. Therefore,
the process of step S304 through step S314 for checking the starting preconditions
cannot be effected with sufficiently high accuracy. In this case, the starting preconditions
are determined to be satisfied in step S318, and this process is once finished. Since
the starting preconditions are determined in this manner, an affirmative determination
is made in step S300 in the next control cycle, and a substantial portion of the starting
precondition checking process (Fig. 5) is terminated. With the starting preconditions
thus determined to be satisfied, a glow-plug fault determination is executed by effecting
forced energization and deenergization of the glow plug 22, namely, applying current
to the glow plug 22 and cut off current applied to the glow plug 22 for the purpose
of diagnosis.
[0063] When the coolant temperature THW is equal to or lower than the threshold value (i.e.,
if an affirmative determination is made in step S301), on the other hand, it is determined
in step S302 whether the starting cycle of the diesel engine 2 has been completed.
If the starting cycle has not been completed (i.e., if a negative determination is
made in step S302), the process of Fig. 5 is once finished. Thus, a negative determination
is repeatedly made in step S302 until the starting cycle is completed. During this
period, the diesel engine 2 is started by cranking, and the engine speed NE is increased
by combustion of fuel injected from the injectors 4 until it reaches a speed level
that indicates completion of the starting cycle.
[0064] When the engine starting cycle is completed (i.e., when an affirmative determination
is made in step S302), a starting period Tsta is read in step S304. The starting period
Tsta is a count value obtained by the ECU 52 that counts the period of time from the
start of the starting cycle to completion thereof by a process that is separately
executed.
[0065] Next, a starting coolant temperature THWs measured at the time of a start of the
engine starting cycle is read in step S306. The starting coolant temperature THWs
is a coolant temperature THW that was detected at the time of the start of the engine
starting cycle and stored in a memory by a process separately executed by the ECU
52.
[0066] Next, a normal starting period Tstanorm (corresponding to the reference starting
period) is calculated in step S308. The normal starting period Tstanorm is the upper
limit of the starting period that is expected when the glow plug 22 normally generates
heat, or represents an allowable range over the upper limit. Since the normal starting
period Tstanorm changes depending upon the coolant temperature THWs measured upon
engine start, the normal starting period Tstanorm is calculated based on the coolant
temperature THWs with reference to, for example, a map of Fig. 6 that is stored in
the ROM of the ECU 52.
[0067] Subsequently, it is determined in step S310 whether the actual starting period Tsta
is longer than the normal starting period Tstanorm. If Tsta is equal to or shorter
than Tstanorm (i.e., if a negative determination is made in step S310), it means that
the glow plug 22 normally generates heat so as to smoothly start the diesel engine
2. In this case, the starting preconditions are determined to be not satisfied in
step S316 so that the fault determination process including forced energization and
deenergization of the glow plug 22 as described later is not carried out. Then, the
process of Fig. 5 is once finished. With the starting preconditions thus determined
to be not satisfied, an affirmative determination is made in step S300 in the next
control cycle, and the substantial starting precondition checking process (Fig. 5)
is finished. Thus, the engine operating state in which the starting preconditions
are set to be not satisfied means that there is no possibility of a fault in glow-plug
energization.
[0068] If the actual starting period Tsta is longer than the normal starting period Tstanorm
(i.e., if an affirmative determination is made in step S310), the history of the battery
voltage VB during starting is read in step S312. During starting of the engine 2,
namely, during a period from start of the engine starting cycle to completion thereof,
the battery voltage VB is detected by a process that is separately executed by the
ECU 52. Through this process, the history of the battery voltage VB is stored in a
memory, which history indicates whether the battery voltage VB has ever been lower
than the starting-time normal minimum voltage Vstanorm (corresponding to the reference
voltage) calculated based on the starting-time coolant temperature THWs with reference
to, for example, a map shown in Fig. 7. The starting-time normal minimum voltage Vstanorm
indicates the smallest degree of a voltage drop of the battery that is expected when
the glow plug 22 is normally energized to generate heat, or indicates an allowable
range over the minimum voltage drop.
[0069] On the basis of the history of the battery voltage VB, it is determined in step S314
whether the battery voltage VB has ever dropped below the starting-time normal minimum
voltage Vstanorm during starting of the engine 2. If the battery voltage VB dropped
below the normal minimum voltage Vstanorm (i.e., if an affirmative determination is
made in step S314), it means that the glow plug 22 normally generates heat, and the
starting period Tsta, which is longer than Tstanorm, is considered to be caused by
another factor that is different from a fault in glow-plug energization.
[0070] If an affirmative determination is made in step S314, therefore, the starting preconditions
are determined to be not satisfied in step S316 as described above, and the process
of Fig. 5 is once finished. With the starting preconditions thus determined, an affirmative
determination is made in step S300 in the next control cycle, and the substantial
starting precondition checking process of Fig. 5 is finished. The engine operating
state in which the starting preconditions are determined to be not satisfied means
that there is no possibility of a fault in glow-plug energization.
[0071] If the battery voltage VB has never dropped below the starting-time normal minimum
voltage Vstanorm during starting (i.e., if a negative determination is made in step
S314), the starting preconditions are determined to be satisfied in step S318, and
the process of Fig. 5 is once finished. With the starting preconditions thus determined,
an affirmative determination is made in step S300 in the next control cycle, and the
substantial starting precondition checking process of Fig. 5 is finished.
[0072] The engine operating state in which the starting preconditions are determined to
be satisfied means that there is a possibility of a fault in glow-plug energization
though an affirmative determination may be made in step S310 and a negative determination
may be made in step S314 for a reason other than a fault in the glow plug 22.
[0073] Next, a process of checking conditions for execution of glow-plug fault determination
will be described with reference to Fig. 8 and Fig. 9. Namely, the process of Figs.
8 and 9 is executed to check conditions for execution of a provisional fault determination
process (Fig. 10) and a main fault determination process (Fig. 11) as described later.
The process of Fig. 8 and Fig. 9 is repeatedly executed at certain time intervals.
Upon a start of this process, it is determined in step S400 whether the starting preconditions
have been determined by the starting precondition checking process (Fig. 5). If the
starting preconditions are not determined to be satisfied or not satisfied (namely,
if a negative determination is made in step S400), the process of Figs. 8 and 9 is
once finished.
[0074] If the starting preconditions are determined to be satisfied or not satisfied in
the starting precondition checking process of Fig. 5 (i.e., if an affirmative determination
is made in step S400), it is determined in step S402 whether execution of glow-plug
fault determination has been permitted. Since execution of the glow-plug fault determination
is not permitted in the initialized state at the time of turn-on of the power supply,
a negative determination is made in step S402. In this case, the control proceeds
to step S404.
[0075] In step S404, it is determined whether sensors used for determination are normally
operating. More specifically, it is determined whether any fault or abnormality is
observed in any of the water temperature sensor 32, vehicle speed sensor 48 and other
sensors from which information is needed so as to determine whether the glow-plug
fault determination process is permitted to be executed. The determination whether
the sensors are normally operating may be made by using data obtained, for example,
by a sensor failure detecting process that is separately executed by the ECU 52.
[0076] If any fault is detected in one or more of the above sensors (i.e., if a negative
determination is made in step S404), the process of Figs. 8 and 9 is once finished.
Since accurate glow-plug fault determination cannot be effected as long as a fault
is found in any one or more of the above sensors, the control does not proceed to
step S426 to permit execution of the glow-plug fault determination process.
[0077] If all of the sensors are normal (i.e., if an affirmative determination is made in
step S404), it is then determined in step S406 whether a coolant temperature THW detected
by the water temperature sensor 32 is within a predetermined reference temperature
range (e.g., a range of 0 °C to 20°C). The reference temperature range is set in accordance
with the type of the diesel engine. More specifically, the reference temperature range
is set to a range suitable for fault detection, in which the engine operating state
apparently differs between the case where the glow plug 22 normally generates heat
during idling and the case where the glow plug 22 is not able to generate sufficient
heat. Namely, when the coolant temperature THW is below the lowest temperature in
the reference temperature range, the vehicle operator is able to detect a fault in
glow plug energization by himself/herself since faulty glow-plug energization and
insufficient heat generated by the glow plug 22 make it difficult for the engine 2
to be started. In addition, when the coolant temperature THW is excessively low, combustion
in the idling state after the start of the engine is unstable, and a fault in the
glow-plug energization cannot be detected with high accuracy, even if the engine 2
can be started due to normal heat generation by the glow plug 22. When the coolant
temperature THW is higher than the highest temperature in the reference temperature
range, the engine is smoothly operated even if the glow plug 2 fails to heat sufficient
heat. In this case, therefore, a fault in glow-plug energization cannot be detected
with high accuracy.
[0078] If the coolant temperature THW is not within the reference temperature range (i.e.,
if a negative determination is made in step S406), the process of Figs. 8 and 9 is
once finished.
[0079] If the coolant temperature THW is within the reference temperature range (i.e., if
an affirmative determination is made in step S406), it is then determined in step
S408 whether the degree of depression of the accelerator pedal ACCPF detected by the
accelerator stroke sensor 26 is equal to 0%, namely, whether the accelerator pedal
24 is completely released. If the accelerator pedal 24 is not released, namely, is
depressed by some degree (i.e., if a negative determination is made in step S408),
which means that the vehicle is not in the idling state, the fuel injection quantity
is not stable, and a fault in glow-plug energization cannot be detected with sufficiently
high accuracy in the fault determination process as described later. In this case,
therefore, the process of Figs. 8 and 9 is finished.
[0080] If the accelerator pedal 24 is completely released (i.e., if an affirmative determination
is made in step S408), it is determined in step S410 whether the vehicle speed detected
by the vehicle speed sensor 48 is equal to 0 km/h. If the vehicle speed is not equal
to 0 km/h (i.e., if a negative determination is made in step S410), namely, if the
vehicle is running, the vehicle is not in the idling state, and the fuel injection
quantity is not stable. Therefore, a fault in glow-plug energization cannot be detected
with sufficiently high accuracy in the fault determination process as described later.
In this case, therefore, the process of Figs. 8 and 9 is finished.
[0081] If the vehicle speed is equal to 0 km/h (i.e., if an affirmative determination is
made in step S410), it is determined in step S412 whether the battery voltage VB is
within a predetermined reference voltage range. The reference voltage range is set
in accordance with the type of the diesel engine. More specifically, the reference
voltage range is set to a range in which the engine operating state, in particular,
the fuel injection quantity, apparently differs between the case where current is
normally applied to the glow plug 22 and the case where no current is applied to the
glow plug 22. If the battery voltage VB is lower than the lowest voltage in the reference
voltage range, the load on the alternator 54 reaches 100%. In this condition, substantially
no change in the engine load may occur even if the glow plug 22 is switched between
the energized state and the deenergized state, and therefore any fault in glow-plug
energization cannot be detected with high accuracy. If the battery voltage VB is higher
than the highest voltage in the reference voltage range, the load on the alternator
54 is almost equal to 0%. In this condition, substantially no change in the engine
load may occur even if the glow plug 22 is switched between the energized state and
the deenergized state, and therefore any fault in glow-plug energization cannot be
detected with high accuracy.
[0082] If the battery voltage VB is not within the reference voltage range (i.e., if a negative
determination is made in step S412), the process of Figs. 8 and 9 is once finished.
[0083] If the battery voltage VB is within the reference voltage range (i.e., if an affirmative
determination is made in step S412), it is determined in step S414 whether a deviation
(|NE - NTRG|) of the engine speed NE detected by the NE sensor 40 from the target
idle speed NTRG set by the ECU 52 is within a predetermined reference deviation range.
The reference deviation range is set in accordance with the type of the diesel engine,
and is defined as a range in which the fuel injection quantity does not largely change
under the idle speed control. This step makes it possible to clearly determine a difference
in the fuel injection quantity by switching the glow plug 22 between the energized
state and the deenergized state in the fault determination process as described later.
[0084] If the deviation (|NE - NTRG|) is not within the reference deviation range (i.e.,
if a negative determination is made in step S414), the process of Figs. 8 and 9 is
once terminated.
[0085] If the deviation (|NE - NTRG|) is within the reference deviation range (i.e., if
an affirmative determination is made in step S414), it is determined in step S416
(Fig. 9) whether an engine stabilization time has passed after completion of the engine
starting cycle. In this connection, the fuel injection quantity is likely to be unstable
immediately after completion of the engine starting cycle. Therefore, the engine stabilization
time, which is the time it takes until the fuel injection quantity is supposed to
be stabilized, is set so that a difference in the fuel injection quantity due to switching
of the glow plug 22 between the energized state and the deenergized state can be determined
with high certainty. Since it takes a longer time to stabilize the engine with a reduction
in the coolant temperature, the engine stabilization time may be set with reference
to a map, or the like so that the engine stabilization time increases as the coolant
temperature THW decreases.
[0086] If the engine stabilization time has not passed after completion of the starting
cycle (i.e., if a negative determination (NO) is made in step S416), the process of
Figs. 8 and 9 is once finished to ensure a stand-by time until the fuel injection
quantity becomes stable.
[0087] When the engine stabilization time has passed (i.e., when an affirmative determination
is made in step S416), it is determined in step S418 whether the remaining glow-plug
energization time in which the glow-plug energization control process as described
above is executed is equal to or longer than a predetermined period of time (which
will be called "required time for fault determination") that permits execution of
the fault determination process. The required time for fault determination means time
required for performing two fault determination processes (of Fig. 10 and Fig. 11)
as described later, in which the glow plug 22 is forced to deenergized for detection
of a change in the engine operating state, and is forced to be energized for detection
of a change in the engine operating state.
[0088] If the remaining glow-plug energization time is shorter than the required time for
fault determination at the time of execution of step S418 (i.e., if a negative determination
is made in step S418), the process of Fig. 8 and Fig. 9 is finished. In this case,
the fault determination processes of Fig. 10 and Fig. 11 as described later will not
be carried out in the current trip of the vehicle.
[0089] If the remaining glow-plug energization time is equal to or longer than the required
time for fault determination (i.e., if an affirmative determination is made in step
S418), on the other hand, it is determined in step S420 whether glow-plug energization
switching control for switching the glow plug 22 between the energized state and the
deenergized state for the purpose of diagnosis has ever been executed in the fault
determination processes of Fig. 10 and Fig. 11 during the current trip. Thus, the
fault determination processes of Figs. 10 and 11 are permitted only once during one
trip, so that switching of the glow relay 22a is minimized, thereby to avoid or suppress
a reduction in the durability of the glow relay 22a.
[0090] If the fault determination processes of Figs. 10 and 11 have been executed in the
current trip (i.e., if an affirmative determination is made in step S420), the fault
determination processes of Figs. 10 and 11 are not executed again in the current trip,
and the process of Fig. 8 and Fig. 9 is terminated.
[0091] Next, it is determined in step S422 whether the starting preconditions are satisfied
in the starting precondition checking process (Fig. 5). If the starting preconditions
are not satisfied (i.e., if a negative determination is made in step S422), it is
determined in step S424 whether the main fault determination stored by the ECU 52
in the last or other previous trip has already been made affirmative. If the main
fault determination was not made affirmative in the last trip (i.e., if a negative
determination is made in step S424), the present process is terminated.
[0092] If the main fault determination was affirmative in the last or other previous trip
(i.e., if an affirmative determination is made in step S424), the main fault determination
is returned to be negative in step S425, and execution of glow-plug fault determination
is permitted in step S426.
[0093] If the starting preconditions are satisfied (i.e., if an affirmative determination
is made in step S422), the normality determination is made negative in step S423,
and execution of the glow-plug fault determination is permitted in step S426. Namely,
when an affirmative determination is made in step S422, there is a possibility of
a fault in glow-plug energization in the starting precondition checking process (Fig.
5) as described above ("NO" in step S314), or determination on a possibility of a
fault was impossible ("NO" is step S301). In these cases, therefore, the presence
of a fault is determined by executing the fault determination processes of Fig. 10,
Fig. 11 as described later.
[0094] When an affirmative determination is made in step S424, there is a possibility that
repair has already been completed after it was determined in the last or other previous
trip that a fault is present in glow-plug energization. Thus, even if it is determined
in step S316 that there is no possibility of a fault in glow-plug energization in
the starting precondition checking process (Fig. 5), the fault determination processes
of Fig. 10 and Fig. 11 as described later are executed so as to provide information
as to whether the glow plug 22 has been brought into a normal state by repair.
[0095] Next, the fault determination processes of Fig. 10 and Fig. 11 in which the glow
plug 22 is switched between the energized state and the deenergized state will be
described. Fig. 10 illustrates a provisional fault determination process, and Fig.
11 illustrates a main fault determination process. These processes are repeatedly
executed at regular time intervals.
[0096] Once the provisional fault determination process (Fig. 10) is started, it is determined
in step S500 whether execution of glow-plug fault determination is permitted in step
S426 of the process of Fig. 8 and Fig. 9 for checking conditions for executing glow-plug
fault determination as described above. If the glow-plug fault determination is not
permitted (i.e., if a negative determination is made in step S500), the process of
Fig. 10 is finished, and no substantial fault determination process is performed.
[0097] If execution of glow-plug fault determination is permitted (i.e., if an affirmative
determination is made in step S500), it is then determined in step S502 whether this
is the first cycle since execution of glow-plug fault determination was permitted.
If this is the first cycle (i.e., if an affirmative determination is made in step
S502), the target idle speed NTRG obtained at this time is stored in the memory as
a target idle speed NTold immediately before deenergization of the glow plug 22 in
step S504. This target idle speed NTold is used for determining the presence of a
change in the target idle speed NTRG in a process of Fig. 12 for checking conditions
for stopping glow-plug fault determination as described later.
[0098] In the next step S506, the current final basic injection quantity command value QFINC
(obtained in step S230 of Fig. 4) is stored in the memory as a final basic injection
quantity command value Qold1 immediately before deenergization of the glow plug 22.
Then, the current battery voltage VB is stored in the memory as a battery voltage
VBold immediately before deenergization of the glow plug 22 in step S508.
[0099] In the next step S510, energization of the glow plug 22 that has continued from turn-on
of the ignition switch to the current point of time is forced to be stopped. Namely,
the glow plug 22 is forced to be deenergized for the purpose of diagnosis. As a result,
electric energy that has been supplied to the glow plug 22 disappears, and the amount
of consumption of electric energy as a whole is significantly reduced. Therefore,
the load on the diesel engine 2 is reduced in accordance with the reduction of the
electric energy consumption. In the idle speed control executed as a part of the fuel
injection quantity control process (of Fig. 3 and Fig. 4), therefore, the final basic
injection quantity command value QFINC is reduced so as to maintain the same target
idle speed NTRG. Also, the battery voltage VB rises in response to the stop of energization
of the glow plug 22, namely, cut-off of current that has been applied to the glow
plug 22
[0100] Subsequently, a change in the final basic injection quantity command value QFINC
under idle speed control and a change in the battery voltage VB are determined so
as to determine whether the load on the diesel engine 2 has been normally reduced.
[0101] Initially, it is determined in step S512 whether the condition that satisfies the
following expression (6) has continued for a period of time required for determining
a reduction in the engine load.

where dQ1 is reduction judgment value, which is set in advance through experiments,
and represents the minimum value of the amount of reduction of the fuel injection
quantity that corresponds to the above-mentioned reduction of the electric energy
consumption.
[0102] If the above expression (6) is not satisfied or if the condition that satisfies the
expression (6) has not continued for the required time (i.e., if a negative determination
is made in step S512), it is determined in step S514 whether the condition that satisfies
the following expression (7) has continued for a period of time required for determining
an increase in the battery voltage VB.

where dV1 is increase judgment value, which is set in advance through experiments,
and represents the minimum value of the amount of increase of the battery voltage
VB that would occur at the time of deenergization of the glow plug 22.
[0103] If the above expression (7) is not satisfied or if the condition that satisfies the
expression (7) has not continued for the required time (i.e., if a negative determination
is made in step S514), it is determined in step S516 whether the condition that satisfies
the following expression (8) has continued for a period of time required for determining
no change in the engine load.

This expression (8) represents a condition that the above-indicated expression (6)
is not satisfied.
[0104] If the above expression (8) is not satisfied or if the condition that satisfies the
expression (8) has not continued for the required time (i.e., if a negative determination
is made in step S516), the process of Fig. 10 is once finished.
[0105] If the condition that satisfies the above expression (6) has continued for the time
required for determining a reduction in the engine load (i.e., if an affirmative determination
is made in step S512), or if the condition that satisfies the above expression (7)
has continued for the time required for determining an increase in the battery voltage
VB (i.e., if an affirmative determination is made in step S514), normality determination
is made affirmative in step S518. Namely, glow-plug energization is determined to
be normal. When an affirmative determination is made in step S512 or step S514, it
indicates that the glow plug 22 had been normally energized before the glow-plug energization
was forced to be stopped in step S510, and energization of the glow plug 22 was stopped
(i.e., the glow plug 22 was deenergized) as commanded by the ECU 52 in step S510.
In this case, information that normality determination is affirmative is stored in
the backup RAM of the ECU 52. Thus, the process of Fig. 10 is finished. If normality
determination is affirmative, energization of the glow plug 22 is resumed (i.e., current
is applied to the glow plug 22 again) and the main fault determination process (Fig.
11) is not executed, in a process (Fig. 12 and Fig. 13) of checking conditions for
stopping glow-plug fault determination.
[0106] If negative determinations are made in both step S512 and step S514, and the condition
that satisfies the above expression (8) has continued for the time required for determining
no change in the engine load (i.e., if an affirmative determination is made in step
S516), provisional fault determination is made affirmative in step S520. This situation
indicates that the glow plug 22 had not been normally energized before energization
of the glow plug 22 was forced to be stopped in step S510, or glow-plug energization
was not stopped as commanded by the ECU 52 in step S510. In this case, therefore,
information that provisional fault determination is affirmative (i.e., glow-plug energization
is provisionally determined to be faulty) is stored in the backup RAM of the ECU 52.
Thus, the process of Fig. 10 is finished. With the provisional fault determination
being affirmative, the provisional fault determination process (Fig. 11) is stopped
and a substantial process of the main fault determination process (Fig. 11) is executed
in the process (Fig. 12 and Fig. 13) of checking conditions for stopping glow-plug
fault determination.
[0107] The main fault determination process (Fig. 11) will be now explained. This process
is repeatedly executed at regular time intervals. Once the process of Fig. 11 is started,
it is determined in step S600 whether the provisional fault determination is affirmative
(namely, the glow-plug energization is provisionally determined to be faulty). Since
the provisional fault determination is kept negative as set by initialization until
the provisional fault determination is made affirmative in the provisional fault determination
process (Fig. 10), a negative determination is made in step S600, and the process
of Fig. 11 is once finished without performing any substantial portion of the process.
[0108] On the contrary, if the provisional fault determination is made affirmative in the
provisional fault determination process (Fig. 10) as described above (i.e., if an
affirmative determination is made in step S600), it is determined in step S601 whether
a predetermined stand-by time has passed since the forced deenergization of the glow
plug 22 in step S510 of the provisional fault determination process (Fig. 10). The
stand-by time is set in advance for preventing switching of the glow relay 22a between
ON and OFF in a short time, thereby preventing a reduction of the durability of the
glow relay 22a. The stand-by time may be set to, for example, 100 msec though it depends
on the type of the glow relay 22a and the amount of current that flows through the
glow relay 22a.
[0109] If the stand-by time has not passed (i.e., if a negative determination is made in
step S601), the process of Fig. 11 is once finished, and the substantial portion of
the main fault determination process is not started.
[0110] If the stand-by time has passed (i.e., if an affirmative determination is made in
step S601), it is determined in step S602 whether this step is executed for the first
time after an affirmative determination is made in step S601. If step S602 is executed
for the first time (i.e., if an affirmative determination is made in step S602), the
final basic injection quantity command value QFINC obtained at this time is stored
in the memory as a final basic injection quantity command value Qold2 immediately
before energization of the glow plug 22 in step S604. Then, the current battery voltage
VB is stored in the memory as a battery voltage VBold2 immediately before energization
of the glow plug 22 in step S606.
[0111] In step S608 following step S606, the glow plug 22 that has been deenergized is forced
to be energized for the purpose of diagnosis. Namely, current starts being applied
to the glow plug 22 again, and the amount of consumption of electrical energy is greatly
increased. As a result, the load on the diesel engine 2 is increased in accordance
with the increase in the electrical energy consumption. In the idle speed control
executed as a part of the fuel injection quantity control process (Fig. 3 and Fig.
4), therefore, the final basic fuel quantity command value QFINC is increased so as
to maintain the same target idle speed NTRG. Furthermore, the battery voltage VB drops
due to the start of energization of the glow plug 22.
[0112] Subsequently, a change in the final basic injection quantity command value QFINC
under idle speed control and a change in the battery voltage VB are determined so
as to determine whether the load on the diesel engine 2 has been normally increased.
[0113] Initially, it is determined in step S610 whether the condition that satisfies the
following expression (9) has continued for a period of time required for determining
an increase in the engine load.

where dQ2 is increase judgment value, which is set in advance through experiments,
and represents the minimum value of the amount of increase of the fuel injection quantity
that corresponds to the above-mentioned increase of the electric energy consumption.
[0114] If the above expression (9) is not satisfied or if the condition that satisfies the
expression (9) has not continued for the required time (i.e., if a negative determination
is made in step S610), it is determined in step S612 whether the condition that satisfies
the following expression (10) has continued for a period of time required for determining
a drop of the battery voltage VB.

where dV2 is drop judgment value, which is set in advance through experiments, and
represents the minimum value of the amount of drop of the battery voltage VB that
would occurs at the time of energization of the glow plug 22.
[0115] If the above expression (10) is not satisfied or if the condition that satisfies
the expression (10) has not continued for the required time (i.e., if a negative determination
is made in step S612), it is determined in step S614 whether the condition that satisfies
the following expression (11) has continued for a period of time required for determining
no change in the engine load.

This expression (11) represents a condition that the above-indicated expression (9)
is not satisfied.
[0116] If the above expression (11) is not satisfied or if the condition that satisfies
the expression (11) has not continued for the required time (i.e., if a negative determination
is made in step S614), the process of Fig. 11 is once finished.
[0117] If the condition that satisfies the above expression (9) has continued for the time
required for determining an increase in the engine load (i.e., if an affirmative determination
is made in step S610) or if the condition that satisfies the above expression (10)
has continued for the time required for determining a drop of the battery voltage
VB (i.e., if an affirmative determination is made in step S612), provisional fault
determination is made negative in step S616. Namely, glow-plug energization is provisionally
determined to be normal. When an affirmative determination is made in step S610 or
step S612, it indicates that the glow plug 22 had certainly been in the deenergized
state before the forced energization of the glow plug 22 in step S608, and the glow
plug 22 was energized again as commanded by the ECU 52 in step S608. This situation
indicates that the forced deenergization of the glow plug 22 (in step S510 of Fig.
10) was not clearly detected for some reason in the provisional fault determination
process (Fig. 10) even though the glow plug 22 was actually deenergized for the purpose
of diagnosis. In the main fault determination process (Fig. 11), however, it was confirmed
that the glow plug 22 functions normally.
[0118] In the above case, therefore, the provisional fault determination is made negative,
and this information is stored in the backup ROM of the ECU 52. Then, the process
of Fig. 11 is once finished. In the following control cycle, a negative determination
is made in step S600 since the provisional fault determination is negative, and therefore
the substantial portion of the main fault determination process (Fig. 11) is not executed.
Since energization of the glow plug 22 is resumed in this case, current applied to
the glow plug 22 is cut off after the glow-plug energization continues for the remaining
part of the glow-plug energization time set in the glow-plug energization control
process started upon turn-on of the ignition switch.
[0119] If negative determinations are made in both step S610 and step S612, and the condition
that satisfies the above expression (11) has continued for the time required for determining
no change in the engine load (i.e., if an affirmative determination is made in step
S614), main fault determination is made affirmative in step S618. This situation indicates
that glow-plug energization is determined to be faulty in the provisional fault determination
process (Fig. 10), and is also determined to be faulty in the main fault determination
process (Fig. 11). In this case, glow-plug energization is determined with high certainty
to be faulty, and information that main fault determination is affirmative is stored
in the backup RAM of the ECU 52. Thus, the process of Fig. 11 is finished. With the
main fault determination being affirmative, the main fault determination process (Fig.
11) is stopped in the process (Fig. 12 and Fig. 13) of checking conditions for stopping
glow-plug fault determination.
[0120] The process (Fig. 12 and Fig. 13) of checking conditions for stopping glow-plug fault
determination will be now explained. This process is repeatedly executed at regular
intervals, for stopping the provisional fault determination process (Fig. 10) and
the main fault determination process (Fig. 11).
[0121] Upon start of the process of Figs. 12 and 13, it is determined in step S700 whether
execution of glow-plug fault determination is permitted in step S426 of the process
(Figs. 8 and 9) of checking conditions for executing glow-plug fault determination.
If execution of glow-plug fault determination is not permitted (i.e., if a negative
determination is made in step S700), the present process is finished, and a substantial
portion of this process is not performed.
[0122] If execution of glow-plug fault determination is permitted (i.e., if an affirmative
determination is made in step S700), the same operations as those of step S404 through
step S412 of the process (Figs. 8 and 9) of checking conditions for execution of glow-plug
fault determination are performed in step S702 through step S710. More specifically,
it is determined in step S702 whether the sensors used for determination are normal.
If the sensors are normally operating (i.e., if an affirmative determination is made
in step S702), it is determined in step S704 whether the coolant temperature THW is
within the reference temperature range. If the coolant temperature THW is within the
reference temperature range (i.e., if an affirmative determination is made in step
S704), it is determined in step S706 whether the degree of depression of the accelerator
pedal ACCPF is equal to zero (namely, the accelerator pedal 24 is released). If the
accelerator pedal depression ACCPF is equal to zero (i.e., if an affirmative determination
is made in step S706), it is determined in step S708 whether the vehicle speed is
equal to 0 km/h. If the vehicle speed is equal to 0 km/h (i.e., if an affirmative
determination is made in step S708), it is determined in step S710 whether the battery
voltage VB is within the reference voltage range.
[0123] If the battery voltage VB is within the reference voltage range (i.e., if an affirmative
determination is made in step S710), it is determined in step S712 whether there is
any change in the target idle speed NTRG under idle speed control during execution
of the provisional fault determination process (Fig. 10) and the main fault determination
process (Fig. 11). This determination is made by comparing the current target idle
speed NTRG with the target idle speed NTold immediately before deenergization of the
glow plug 22, which speed NTold is stored in step S504 of the provisional fault determination
process (Fig. 10).
[0124] While a variation arises in the final basic injection quantity command value QFINC
if the target idle speed NTRG changes, it is difficult to distinguish this variation
from a variation in the injection quantity due to switching of the glow plug 22 between
the energized state and the deenergized state. Accordingly, the determination in step
S712 is made so as to avoid erroneous determinations in the provisional fault determination
process (Fig. 10) and the main fault determination process (Fig. 11).
[0125] If there is no change in the target idle speed NTRG (i.e., if a negative determination
is made in step S712), it is determined in step S714 whether both the normality determination
and the main fault determination are negative. If the normality determination and
the main fault determination are both negative (if an affirmative determination is
made in step S714), it is then determined in step S716 whether there is a change in
any of switches of various devices that use the battery 60 as a power supply. These
switches may include the air-conditioner switch 50, an electric heater switch, a tail
lamp switch, a defogger switch, a brake signal switch, and so forth. If there is a
change in any of these switches, variations arise in the battery voltage VB and the
final basic injection quantity command value QFINC. It is difficult to distinguish
these variations from variations in the battery voltage VB and the command value QFINC
due to switching of the glow plug 22 between the energized state and the deenergized
state. Thus, the determination in step S716 is made so as to avoid erroneous determinations
in the provisional fault determination process (Fig. 10) and the main fault determination
process (Fig. 11).
[0126] If there is no change in the switches (i.e., if a negative determination is made
in step S716), it is then determined in step S718 whether the provisional fault determination
is negative (i.e., glow-plug energization is provisionally determined to be faulty).
Here, if the normality determination is negative and the provisional fault determination
is negative during execution of the provisional fault determination process (Fig.
10) (i.e., if an affirmative determination is made in step S718, the process of Figs.
12 and 13 is once terminated.
[0127] If a negative determination is made in any of steps S702 - S710 and step S714 or
an affirmative determination is made in step S712 or step S716, the provisional fault
determination is made negative in step S720. In this case, a negative determination
is made in step S600 in the main fault determination process (Fig. 11), and a substantial
portion of this process is not performed.
[0128] In step S722 following step S720, forced energization process is executed, namely,
the glow plug 22 is forced to be energized, after a stand-by time has passed since
the glow plug 22 was switched from the energized state to the deenergized state in
step S510 of the provisional fault determination process (Fig. 10). The stand-by time
is provided for preventing or suppressing a reduction in the durability of the glow
relay 22a, and is set to be the same as the stand-by time as explained above with
respect to step S601 of the main fault determination process (Fig. 11).
[0129] In the forced energization process as described above, the glow plug 22 is immediately
energized if the stand-by time has already passed at a point of time when step S722
is executed. If the stand-by time has not passed at the time of execution of step
S722, on the other hand, the glow plug 22 is energized after a lapse of the stand-by
time. If the glow plug 22 was already forced to be energized for the purpose of diagnosis
in step S608 of the main fault determination process (Fig. 11), the glow plug 22 is
kept in the energized state.
[0130] In step S724 following step S722, execution of glow fault determination is not permitted,
and the process of Fig. 12 and Fig. 13 is once finished.
[0131] In the next control cycle, since permission of execution of glow-plug fault determination
is cancelled, a negative determination is made in step S700 in the process (Figs.
12 and 13) of checking conditions for stopping glow-plug fault determination, and
a substantial portion of this process is not performed. Similarly, a negative determination
is made in step S500 during execution of the provisional fault determination process
(Fig. 10), and a substantial portion of this process is not performed.
[0132] It is to be noted that a negative determination is made in step S714 when the normality
determination is made affirmative in step S518 of the provisional fault determination
process (Fig. 10), or the main fault determination is made affirmative in step S618
of the main fault determination process (Fig. 11).
[0133] If the provisional fault determination is made affirmative in step S520 of the provisional
fault determination process (Fig. 10), and a negative determination is made in step
S718, execution of the provisional fault determination process (Fig. 10) is stopped
in step S726.
[0134] Examples of the control process according to the present embodiment as described
above are illustrated in Fig. 14 through Fig. 18.
[0135] Fig. 14 illustrates the case where the starting preconditions are determined to be
not satisfied. When the ignition switch is turned ON at time t0, current is immediately
applied to the glow plug 22. At time t1, an indicator lamp indicates permission of
starting of the diesel engine 2 to the vehicle operator, and the operator immediately
turns on the starter 30, whereby the engine starting cycle is initiated. After the
engine speed NE increases up to a speed level that indicates completion of the engine
starting cycle, the ECU 52 determines completion of the engine starting cycle at time
t2. The actual starting period Tsta (period between t1 and t2) obtained at this time
is equal to or shorter than the normal starting period Tstanorm. In the example of
Fig. 14, the battery voltage VB drops to be lower than the normal minimum voltage
Vstanorm during starting.
[0136] After completion of the engine starting cycle, therefore, a negative determination
is made in step S310 of the starting precondition checking process (Fig. 5), and the
preconditions are determined to be not satisfied in step S316. Accordingly, neither
the provisional fault determination process (Fig. 10) nor the main fault determination
process (Fig. 11) is executed, and current is kept applied to the glow plug 22 for
normal glow-plug energization. If the energization time set under normal glow-plug
energization control expires at time t3, energization of the glow plug 22 is stopped,
namely, current that has been applied to the glow plug 22 is cut off. With the process
as described above, the information that the starting preconditions are not satisfied,
the main fault determination is negative, and the normality determination is negative
is recorded in the memory of the ECU 52 as internal diagnosis information.
[0137] Fig. 15 illustrates the case where no current flows through the glow plug 22 because
of disconnection even if a command to energize the glow plug 22 is generated. Initially,
the ignition switch is turned ON at time t10, and the starter 30 starts driving of
the engine 2 at time t11 in response to a signal to permit starting, which is followed
by completion of the engine starting cycle at time t12. In this case, the actual starting
period Tsta is longer than the normal starting period Tstanorm, and the battery voltage
VB has never dropped to be lower than the starting-time normal minimum voltage Vstanorm.
Therefore, an affirmative determination is made in step S310 and a negative determination
is made in step S314 in the starting precondition checking process (Fig. 5), and the
starting preconditions are determined to be satisfied in step S318. As a result, the
process (Fig. 8 and Fig. 9) of checking conditions for execution of glow-plug fault
determination is executed. If execution of glow-plug fault determination is permitted
in step S426, the provisional fault determination process (Fig. 10) is executed at
time t13 so that the glow plug 22 is forced to be deenergized. In the example of Fig.
15, neither the fuel injection quantity nor the battery voltage changes immediately
after cut-off of current to the glow plug 22, even after a stand-by time for determination.
Therefore, an affirmative determination is made in step S516, and the provisional
fault determination is made affirmative in step S520 at time t14 in Fig. 15. As a
result, the provisional fault determination process (Fig. 10) is stopped in step S726
of the process (Fig. 12 and Fig. 13) of checking conditions for stopping glow-plug
fault determination, and instead the substantive main fault determination process
(Fig. 11) is started so that the glow plug 22 is forced to be energized in step S608.
In the example of Fig. 15, neither the fuel injection quantity nor the battery voltage
changes immediately after forced energization of the glow plug 22, even after a stand-by
time for determination. Therefore, an affirmative determination is obtained in step
S614, and the main fault determination is made affirmative in step S618 at time t15
in Fig. 15. Then, a negative determination is made in step S714 in the process (Figs.
12 and 13) of checking conditions for stopping glow-plug fault determination, and
the provisional fault determination is made negative in step S720, whereby execution
of glow-plug fault determination is not permitted in step S724. Since the glow plug
22 has been forced to be energized, the energized state of the glow plug 22 is maintained
in step S722.
[0138] Subsequently, when the energization time set in normal glow-plug energization control
expires (t16), current applied to the glow plug 22 is cut off. With the process as
described above, the information that the starting preconditions are satisfied, the
main fault determination is affirmative, and the normality determination is negative
is recorded in the memory of the ECU 52.
[0139] Fig. 16 illustrates the case where energization of the glow plug 22 is determined
to be normal in the provisional fault determination process (Fig. 10). Initially,
the ignition switch is turned ON at time t20, and the starter 30 starts driving of
the engine 2 at time t21 in response to a signal to permit starting, which is followed
by completion of the engine starting cycle at time t22. In this case, the actual starting
period Tsta is longer than the normal starting period Tstanorm, and the battery voltage
VB has never dropped to be lower than the starting-time normal minimum voltage Vstanorm.
Therefore, an affirmative determination is made in step S310 and a negative determination
is made in step S314 in the starting precondition checking process (Fig. 5), and the
starting preconditions are determined to be satisfied in step S318. As a result, the
process (Fig. 8 and Fig. 9) of checking conditions for execution of glow-plug fault
determination is executed. If execution of glow-plug fault determination is permitted
in step S426, the provisional fault determination process (Fig. 10) is executed at
time t23, and the glow plug 22 is forced to be deenergized. In the example of Fig.
16, the battery voltage VB is immediately increased by an amount larger than the increase
judgment value dV1, and an affirmative determination is made in step S514. Also, the
fuel injection quantity is reduced by an amount larger than the reduction judgment
value dQ1.
[0140] In the above case, the normality determination is made affirmative in step S518 at
time t24 in Fig. 16. Accordingly, the provisional fault determination is kept being
negative in step S720, and execution of glow-plug fault determination is not permitted
in step S724, so that the provisional fault determination process (Fig. 10) is stopped.
Furthermore, in step S722, energization of the glow plug 22 is resumed at time t25
after a lapse of the stand-by time measured from the forced deenergization of the
glow plug 22 in step S510. Thus, the main fault determination process (Fig. 11) is
not executed. Subsequently, glow-plug energization is terminated when the energization
time set in normal glow-plug energization control expires at time t26. With the process
as described above, information that the starting preconditions are satisfied, the
main fault determination is negative, and the normality determination is affirmative
is recorded in the memory of the ECU 52.
[0141] Fig. 17 illustrates the case where sufficiently large changes do not occur in the
fuel injection quantity and the battery voltage when the glow plug 22 is forced to
be deenergized in the provisional fault determination process (Fig. 10), and sufficiently
large changes occur in the fuel injection quantity and the battery voltage for the
first time when the glow plug 22 is forced to be energized in the main fault determination
process (Fig. 11). The control proceeds from time t30 to t34 in the same manner in
which the control proceeds from time t10 to t14 as explained above with reference
to Fig. 15. In the main fault determination process (Fig. 11) started at time t34,
the glow plug 22 is forced to be energized in step S608. In the example of Fig. 17,
the fuel injection quantity is first increased by an amount larger than the increase
judgment value dQ2 (i.e., an affirmative determination is made in step S610), and
therefore the provisional fault determination is made negative in step S616 at time
t35. After the energization time set in normal glow-plug energization control expires
at time t36, the glow plug 22 is deenergized, namely, current applied to the glow
plug 22 is cut off. With the process as described above, information that the starting
preconditions are satisfied, the main fault determination is negative, and the normality
determination is negative is recorded in the memory of the ECU 52. It is to be noted
that execution of glow-plug fault determination is not permitted in step S724 at an
appropriate time, for example, when the accelerator pedal 24 is depressed (i.e., when
a negative determination is made in step S706).
[0142] Fig. 18 illustrates the case where a fault in the glow-plug energization was found
in the last or other previous trip, and the engine 2 is started immediately after
repair is completed. The control proceeds from time t40 to t42 in the same manner
in which the control proceeds from time t0 to t2 as explained above with respect to
Fig. 14. Since normal energization of the glow plug 22 is performed, a negative determination
is made in step S310 or an affirmative determination is made in step S314, so that
the starting preconditions are determined to be not satisfied in step S316.
[0143] If affirmative determinations are made in all of steps S400 through S420 except for
steps S402 and S420 and negative determinations are made in steps S402 and S420 in
the process (Figs. 8 and 9) of checking conditions for executing glow-plug fault determination,
it is then determined in step S422 whether the starting preconditions are satisfied
in step S422. At this time, since the starting preconditions are not satisfied assuming
that there is no possibility of a fault in the glow plug 22, a negative determination
is made in step S422, and it is determined in step S424 whether the main fault determination
has already been affirmative. Here, since the main fault determination has already
been affirmative (i.e., an affirmative determination is made in step S424), the main
fault determination is returned to be negative in step S425, and execution of glow-plug
fault determination is permitted in step S426. Thus, the substantial provisional fault
determination process (Fig. 10) is started.
[0144] In the provisional fault determination process (Fig. 10), the glow plug 22 is forced
to be deenergized. At this time, since the glow plug 22 is normally deenergized, the
battery voltage VB is immediately increased by an amount larger than the increase
judgment value dV1, and an affirmative determination is made in step S514 at time
t44. Also, the fuel injection quantity is reduced by an amount larger than the reduction
judgment value dQ1.
[0145] In the above case, the normality determination is made affirmative in step S518.
Accordingly, a negative determination is made in step S714, and the provisional fault
determination is kept being negative in step S720. Also, execution of glow-plug fault
determination is not permitted in step S724, and the provisional fault determination
process (Fig. 10) is stopped. Furthermore, the glow plug 22 is forced to be energized
again at time t45 through the process of step S722, after a lapse of the stand-by
time as measured from the forced deenergization of the glow plug 22 in step S510.
[0146] Subsequently, when the energization time set in normal glow-plug energization control
expires at time t46, the glow-plug energization is terminated. With the process as
described above, information that the starting preconditions are not satisfied, the
main fault determination is negative, and the normality determination is positive
is stored in the memory of the ECU 52.
[0147] In the glow-plug fault detecting apparatus and method as described above, the starting
precondition checking process (Fig. 5) corresponds to fault possibility determining
means, and the process (Figs. 8 and 9) of checking conditions for execution of glow-plug
fault determination, the provisional fault determination process (Fig. 10), the main
fault determination process (Fig. 11) and the process (Figs. 12 and 13) of checking
conditions for stopping glow-plug fault determination correspond to fault presence
determining means. Also, the process of calculating the starting period Tsta by counting
a period of time from the start of the engine starting cycle to completion thereof
and the process of detecting the history of the battery voltage VB for a period from
the start of the engine starting cycle to completion thereof correspond to means for
detecting an engine state before completion of the engine starting cycle. The process
(Figs. 8 and 9) of checking conditions for executing glow-plug fault determination
and the provisional fault determination process (Fig. 10) correspond to the first
fault determining means, and the main fault determination process (Fig. 11) and the
process (Figs. 12 and 13) of checking conditions for stopping glow-plug fault determination
correspond to the second fault determining means.
[0148] The first embodiment of the invention as described above yields the following effects.
(1) In the starting precondition checking process (Fig. 5), a fault of the glow plug
22 is detected based on the length of the starting period Tsta or the history of the
battery voltage VB during starting without relaying on forced energization and deenergization
of the glow plug 22. The fault detection through this process exhibits lower certainty
or reliability as compared with the case where a fault is detected through forced
energization and deenergization of the glow plug 22 for the purpose of diagnosis.
This is because the starting period Tsta may be prolonged or the battery voltage VB
may not be lowered for other reasons even if current is normally applied to the glow
plug 22.
However, if the starting period Tsta is sufficiently short or the battery voltage
VB is sufficiently lowered during energization of the glow plug 22 for warm-up of
the engine 2, no possibility of a fault in glow-plug energization can be determined
with high reliability even if the energized state of the glow plug 22 is not forced
to be changed for diagnostic purposes. Accordingly, if it is determined that there
is no possibility of a fault through the starting precondition checking process (Fig.
5) executed before the provisional fault determination process (Fig. 10) and the main
fault determination process (Fig. 11), there is no need to execute the provisional
fault determination process (Fig. 10) and the main fault determination process (Fig.
11). If it is not determined in the starting precondition checking process (Fig. 5)
that there is no possibility of a fault in the glow plug 22, namely, if it is determined
that there is a possibility of a fault, the provisional fault determination process
(Fig. 10) and the main fault determination process (Fig. 11) in which the glow plug
22 is forced to be energized or deenergized for diagnostic purposes are executed.
Thus, the number of times the energized state of the glow plug 22 is changed by driving
the glow relay 22a can be reduced.
Thus, a fault in glow-plug energization can be detected with sufficiently high accuracy
by monitoring changes in the fuel injection quantity and changes in the battery voltage
immediately after forced energization or deenergization of the glow plug 22 in the
case where it is determined that there is a possibility of a fault in glow-plug energization.
Furthermore, the number of switching of the glow plug 22 between the energized state
and the deenergized state can be reduced, and the therefore otherwise possible deterioration
of the durability of the glow relay 22a can be suppressed.
(2) If glow-plug energization is normally performed, the engine starting cycle is
completed early, and the starting period Tsta turns out to be short. If glow-plug
energization is not normally performed, to the contrary, the starting period Tsta
is increased. In the starting precondition checking process (Fig. 5), therefore, it
is determined that there is no possibility of a fault in the glow plug 22 if the starting
period Tsta is equal to or shorter than the normal starting period Tstanorm. On the
contrary, if the starting period Tsta is longer than the normal starting period Tstanorm,
a possibility of a fault in the glow plug 22 cannot be negated though the long starting
period may be caused by another reason.
If glow-plug energization is normally performed, the battery voltage VB drops off
at a time prior to completion of the engine starting cycle. On the other hand, if
no current is applied to the glow plug 22 or the amount of current applied to the
glow plug 22 is excessively small, the battery voltage VB does not drop or hardly
drops. In the starting precondition checking process (Fig. 5), therefore, it is determined
that there is no possibility of a fault in the glow plug 22 if the battery voltage
VB has ever dropped below the starting-time normal minimum voltage Vstanorm before
completion of the engine starting cycle. On the contrary, if there is no history of
a drop of the battery voltage VB below the starting-time normal minimum voltage Vstanorm
before completion of the starting cycle, a possibility of a fault in the glow plug
22 cannot be negated or denied though the absence of the voltage drop may be caused
by another reason.
Accordingly, if the starting period Tsta is longer than the normal starting period
Tstanorm and there is no history of a drop of the battery voltage VB below the starting-time
normal minimum voltage Vstanorm before completion of the engine starting cycle, it
is determined that there is a possibility of a fault in the glow plug 22.
Thus, in the starting precondition checking process (Fig. 5), the possibility of a
fault in the glow plug 22 can be easily determined even if the glow plug 22 is not
forced to be energized or deenergized for the purpose of diagnosis. Furthermore, the
normal starting period Tstanorm and the starting-time normal minimum voltage Vstanorm
are set based on the coolant temperature THWs measured at the beginning of the engine
starting cycle, and therefore the possibility of a fault can be determined with relatively
high certainty or reliability.
(3) In the process (Figs. 8 and 9) of checking conditions for executing glow-plug
fault determination, execution of glow-plug fault determination is permitted only
if the remaining portion of the glow-plug energization time set in the normal glow-plug
energization control is long enough to execute the provisional fault determination
process (Fig. 10) and the main fault determination process (Fig. 11) (i.e., if an
affirmative determination is made in step S418).
In the provisional fault determination process (Fig. 10) and the main fault determination
process (Fig. 11), therefore, determination of the presence of a fault accompanied
by forced changes of the energized state of the glow plug 22 can be completed within
the time set for normal glow-plug energization. Thus, the glow-plug energization control
is not prolonged in vain due to the forced changes of the energized state of the glow
plug 22 for the purpose of diagnosis, and energy consumption for determination of
the presence of a fault can be reduced.
The forced changes of the energized state of the glow plug 22 are effected during
normal glow-plug energization in the provisional fault determination process (Fig.
10) and the main fault determination process (Fig. 11). Since the process (Figs. 8
and 9) of checking conditions for executing glow-plug fault determination allows the
energized state of the glow plug 22 to be forced to be changed only after the engine
operating state becomes stable, no problem arises in the stability of the engine operation
after starting thereof.
(4) With the provisional fault determination process (Fig. 10) and the main fault
determination process (Fig. 11) as described above, a fault in glow-plug energization
can be detected with high reliability in two steps or stages, namely, through energization
and deenergization of the glow plug 22 for diagnostic purposes. Furthermore, by switching
the glow plug 22 from the energized state to the deenergized state and from the deenergized
state to the energized state in the two steps, the glow plug 22 can be returned to
the energized state, thus causing no problem in the stability of the operation of
the diesel engine 2.
(5) If energization and deenergization of the glow plug 22 are carried out at short
time intervals, namely, if the glow relay 22a is switched off and on at short intervals,
the durability of the glow relay 22a may be reduced. In step S601 of the main fault
determination process (Fig. 11), therefore, the forced energization of the glow plug
22 is not executed until the stand-by time set for protecting the glow relay 22a passes
from a point of time when the glow plug 22 is forced to be deenergized in step S510
of the provisional fault determination process (Fig. 10). Thus, otherwise possible
reduction of the durability of the glow relay 22a can be prevented.
(6) In the process (Figs. 8 and 9) of checking conditions for executing glow-plug
fault determination, even if it is apparently determined that there is no possibility
of a fault in glow-plug energization (i.e., a negative determination is made in step
S422), execution of glow-plug fault determination is permitted in step S426 provided
that the main fault determination was already made affirmative in the last or other
previous trip (i.e., an affirmative determination is made in step S424).
[0149] With the above arrangement, the provisional fault determination process (Fig. 10)
can be executed and the main fault determination process (Fig. 11) can be further
executed in some cases after repair is done with respect to the previously detected
fault, so that internal information relating to diagnosis, for example, can be updated
to reflect normal glow-plug energization after the repair.
[0150] Next, a second embodiment of the invention will be described. In this embodiment,
while the glow plug 22 is energized under normal glow-plug energization control performed
upon a start of the engine, only the starting precondition checking process (Fig.
5) is executed, and the starting preconditions are determined to be satisfied or not
satisfied. Then, after completion of the energization of the glow plug 22 for warm-up
of the engine 2, the glow plug 22 is energized or deenergized for the purpose of determining
the presence of a fault, only in the case where the starting preconditions are satisfied
or in the case where the main fault determination has already been affirmative.
[0151] In the second embodiment, a process as shown in Fig. 19 is executed which replaces
a part of the process of checking conditions for executing glow-plug fault determination
as shown in Fig. 9 of the first embodiment. Also, a provisional fault determination
process as shown in Fig. 20 and a main fault determination process as shown in Fig.
21 are executed in place of the provisional fault determination process (Fig. 10)
and the main fault determination process (Fig. 11). Furthermore, a process as shown
in Fig. 22 is executed which replaces a part of the process of checking conditions
for stopping glow-plug fault determination as shown in Fig. 13. With regard to the
other construction and processes, the second embodiment is substantially identical
with the first embodiment.
[0152] Referring first to Fig. 19, it is determined in step S417, as a condition for permitting
execution of glow-plug fault determination, whether a stand-by time has passed after
the normal glow-plug energization period expires. This step S417 replaces step S416
and step S418 of Fig. 9.
[0153] Thus, execution of glow-plug fault determination is not permitted (in step S426)
unless the normal glow-plug energization period expires. Thus, neither deenergization
nor energization of the glow plug 22 for a diagnostic purpose is carried out during
normal glow-plug energization for warm-up of the engine.
[0154] If affirmative determinations are made in step S400 and steps S404 through S417 and
negative determinations are made in step S402 and step S420 after the normal glow-plug
energization period expires, it is determined in step S422 whether the starting preconditions
are satisfied. If the starting preconditions set in the starting precondition checking
process (Fig. 5) are not satisfied (i.e., if a negative determination is made in step
S422), it is determined in step S424 whether the main fault determination has already
been affirmative. If the main fault determination is negative (i.e., if a negative
determination is made in step S424), the present process is terminated. Thus, the
forced energization and deenergization of the glow plug 22 for the purpose of diagnosis
are not carried out in the same trip. In this case, the control proceeds in the same
manner as shown in the time chart of Fig. 14 as explained above with respect to the
first embodiment.
[0155] The case where the starting preconditions are determined to be satisfied in the starting
precondition checking process (Fig. 5) will be now described. In this case, the forced
deenergization and energization of the glow plug 22 are not carried out within the
normal glow-plug energization period. For example, as shown in the time chart of Fig.
23, the ignition switch is turned on at time t50, and the starter 30 is allowed to
start driving the diesel engine 2 at time t51, followed by completion of the engine
starting cycle at time t52. In this particular case, the actual starting period Tsta
is longer than the normal starting period Tstanorm, and the battery voltage VB has
never dropped below the starting-time normal minimum voltage Vstanorm. Accordingly,
an affirmative determination is made in step S310 and a negative determination is
made in step S314 in the starting precondition checking process (Fig. 5), and the
preconditions are determined to be satisfied in step S318. However, the provisional
fault determination process (Fig. 20) is not immediately started but waits until the
normal glow-plug energization period expires.
[0156] If affirmative determinations are made in step S400 and steps S404 through S417 and
negative determinations are made in step S402 and step S420 in the process (Fig. 8
and Fig. 19) after the normal glow-plug energization period expires, it is determined
in step S422 whether the starting preconditions are satisfied. In this particular
case, since the starting preconditions are satisfied (i.e., an affirmative determination
is made in step S422), the normality determination is made negative in step S423,
and execution of glow-plug fault determination is permitted in step S426 at time t54.
As a result, the provisional fault determination process (Fig. 20) is started, and
affirmative determinations are made in step S800 and step S802. After execution of
steps S804 through step S808, the glow plug 22 is forced to be energized in step S810.
It will be understood that step S800 through step S808 are substantially identical
with step S500 through step S508 of the provisional fault determination process (Fig.
10) of the first embodiment.
[0157] Subsequently, it is determined in step S812 whether the condition that satisfies
the following expression (12) has continued for a period of time required for determining
an increase of the engine load.

where dQ3 is increase judgment value, which is set in advance through experiments,
and is defined as the minimum value of the amount of increase of the fuel injection
quantity that corresponds to an increase of the consumption of electric energy caused
by the energization of the glow plug 22.
[0158] If the above expression (12) is not satisfied or if the condition that satisfies
the above expression (12) has not continued for the time required for determining
the increase of the engine load (i.e., if a negative determination is made in step
S812), it is then determined in step S814 whether the condition that satisfies the
following expression (13) has continued for a period of time required for determining
a drop of the battery voltage VB.

where dV3 is drop judgment value, which is set in advance through experiments, and
represents the minimum value of the drop of the battery voltage VB that occurs in
response to the energization of the glow plug 22.
[0159] If the above expression (13) is not satisfied or if the condition that satisfies
the above expression (13) has not continued for the time required for determining
the drop of the battery drop (i.e., if a negative determination is made in step S814),
it is determined in step S816 whether the condition that satisfies the following expression
(14) has continued for a period of time required for determining no change of the
engine load.

This expression (14) represents a state in which the above-indicated expression (12)
is not satisfied.
[0160] If the above expression (14) is not satisfied or if the condition that satisfies
the above expression (14) for the time required for determining no change of the engine
load (i.e., if a negative determination is made in step S816), the process of Fig.
20 is once terminated.
[0161] As shown in the time chart of Fig. 23, if the condition that satisfies the above
expression (14) has continued for the time required for determining no change of the
engine load (if an affirmative determination is made in step S816) at time t55 before
affirmative determinations are made in step S812 and step S814, the provisional fault
determination is made affirmative in step S820. This indicates that the glow plug
22 had been in the energized state though it should have been in the deenergized state
before the glow plug 22 was energized for the purpose of diagnosis in step S810, or
the glow plug 22 was not energized as commanded by the ECU 52 in step S810. In this
case, information that the provisional fault determination is positive is stored in
the backup RAM of the ECU 52, and the process of Fig. 20 is once terminated).
[0162] With the provisional fault determination thus made positive, the provisional fault
determination process (Fig. 20) is stopped in step S726 in the process (Figs. 12,
22) of checking conditions for stopping glow-plug fault determination, and the substantial
main fault determination process (Fig. 21) starts being executed (i.e., an affirmative
determination is made in step S900).
[0163] After execution of step S900 through step S906 in the main fault determination process
(Fig. 21), the glow plug 22 is forced to be deenergized for the purpose of diagnosis
in step S908. It will be understood that the contents of step S900 through step S906
are substantially identical with those of step S600 through step S606 of the main
fault determination process (Fig. 11) through the glow plug 22 is energized rather
than deenergized in the process of Fig. 11.
[0164] Next, a change in the final basic injection quantity command value QFINC under idle
speed control and a change in the battery voltage VB are determined in order to determine
whether the load on the diesel engine 2 has normally reduced due to deenergization
of the glow plug 22.
[0165] Initially, it is determined in step S910 whether the condition that satisfies the
following expression (15) has continued for a period of time required for determining
the reduction of the engine load.

where dQ4 is reduction judgment value, which is set in advance through experiments,
and represents the minimum value of the amount of reduction of the fuel injection
quantity that corresponds to a reduction of the consumption of electrical energy due
to the deenergization of the glow plug 22.
[0166] If the above expression (15) is not satisfied or if the condition that satisfies
the above expression (15) has not continued for the time required for determining
the reduction of the load (i.e., if a negative determination is made in step S910),
it is determined in step S912 whether the condition that satisfies the following expression
(16) has continued for a period of time required for determining an increase of the
battery voltage VB.

where dV4 is increase judgment value, which is set in advance through experiments,
and represents the minimum value of the increase of the battery voltage VB that occurs
in response to deenergization of the glow plug 22.
[0167] If the above expression (16) is not satisfied or the condition that satisfies the
above expression (16) has not continued for the time required for determining the
increase of the battery voltage (i.e., if a negative determination is made in step
S912), it is determined in step S914 whether the condition that satisfies the following
expression (17) has continued for a period of time required for determining no change
of the engine load.

This expression (17) represents a state in which the above-indicated expression (15)
is not satisfied.
[0168] If the above expression (17) is not satisfied or if the condition that satisfies
the above expression (17) has not continued for the time required for determining
no change of the engine load (i.e., if a negative determination is made in step S914),
the process of Fig. 21 is once terminated.
[0169] As shown in the time chart of Fig. 23, if the condition that satisfies the above
expression (17) has continued for the time required for determining no change of the
engine load (i.e., if an affirmative determination is made in step S914) at time t56
before affirmative determinations are made in step S910 and step S912, the main fault
determination is made affirmative in step S918. This indicates that the glow plug
22 had not been in the energized state before the glow plug 22 was forced to be deenergized
in step S908, or the glow plug 22 was not deenergized as commanded by the ECU 52 in
step S908.
[0170] In the above case, the glow plug 22 is determined to be faulty in the provisional
fault determination process (Fig. 20) and is also determined to be faulty in the main
fault determination process (Fig. 21). Thus, the glow-plug energization is determined
to be faulty with high certainty, and information that the main fault determination
is affirmative is stored in the backup RAM of the ECU 52. Then, the process of Fig.
21 is once terminated.
[0171] With the main fault determination thus made affirmative, a negative determination
is made in step S714 of the process (Figs. 12 and 22) of checking conditions for stopping
glow-plug fault determination, and execution of glow-plug fault determination is not
permitted in step S724, whereby the main fault determination process (Fig. 21) is
stopped. Through the process as described above, information that the starting preconditions
are not satisfied, the main fault determination is affirmative and the normality determination
is negative is recorded in the memory of the ECU 52 as internal information relating
to diagnoses.
[0172] Referring next to the time chart of Fig. 24, the case where the normality determination
is made affirmative will be described. The control proceeds from time t60 to t64 in
the same manner in which the control proceeds from time t50 to t54 in the case of
Fig. 23. In the provisional fault determination process (Fig. 20) that is started
at time t64, the glow plug 22 is forced to be energized for a diagnostic purpose in
step S810. As current actually starts flowing through the glow plug 22 in a normal
way upon energization of the glow plug 22, the fuel injection quantity starts increasing,
and the battery voltage VB starts dropping. Then, the relationship of the expression
(13) is first satisfied at time t65 and is kept satisfied for the required time (an
affirmative determination is made in step S814), whereby the normality determination
is made affirmative in step 818. With the normality determination thus made affirmative,
a negative determination is made in step S714 in the process (Figs. 12 and 22) of
checking conditions for stopping glow-plug fault determination, and the provisional
fault determination is made negative in step S720, whereby execution of glow-plug
fault determination is not permitted in step S724. Consequently, the provisional fault
determination process (Fig. 20) and the main fault determination process (Fig. 21)
are stopped. Also, the glow plug 22 is returned to the deenergized state after a lapse
of a stand-by time in step S723 at time t66.
[0173] With the process as described above, information that the starting preconditions
are satisfied, the main fault determination is negative, and the normality determination
is affirmative is recorded in the memory of the ECU 52.
[0174] Referring next to the time chart of Fig. 25, the case where the energized state of
the glow plug 22 is normally changed for the first time in the main fault determination
process (Fig. 21) will be described. The control proceeds from time t70 to t74 in
the same manner in which the control proceeds from time t50 to t54 in the case of
Fig. 23. In the provisional fault determination process (Fig. 20) started at time
t74, a command to energize the glow plug 22 for the purpose of diagnosis is generated
in step S810. Since no current flows through the glow plug 22 in response to this
command, the above-indicated expressions (12) and (13) are not satisfied, and the
above-indicated expression (14) is kept satisfied (i.e., an affirmative determination
is made in step S816 at time t75). As a result, the provisional fault determination
is made affirmative in step S820. With the provisional fault determination thus made
affirmative, the provisional fault determination process (Fig. 20) is stopped in step
S726. In the main fault determination process (Fig. 21), a command to deenergize the
glow plug 22 for the purpose of diagnosis is generated in step 908 at time t76 after
the stand-by time passes from the forced energization of the glow plug 22 in step
S810.
[0175] If current that has flowed through the glow plug 22 is normally stopped in response
to the command to deenergize the glow plug 22, the fuel injection quantity starts
decreasing, and the battery voltage VB starts rising. Then, the relationship of the
above expression (16) is satisfied for the first time at time t77, and is kept satisfied
so that an affirmative determination is made in step S912, whereby the provisional
fault determination is returned to be negative in step S916. With the provisional
fault determination being negative, the substantial main fault determination process
(Fig. 21) is stopped (i.e., a negative determination is made in step S900). With the
process as described above, information that the starting preconditions are satisfied,
the main fault determination is negative, and the normality determination is negative
is recorded in the memory of the ECU 52. It is to be noted that execution of glow-plug
fault determination is not permitted in step S724 at an appropriate time, for example,
when the accelerator pedal 24 is depressed (i.e., when a negative determination is
made in step S706).
[0176] Fig. 26 illustrates the case where a fault in glow-plug energization was found in
the last or other previous trip, and the diesel engine 2 is started immediately after
completion of repair. The control proceeds from time t80 to t83 in the same manner
in which the control proceeds from time t0 to t3 as explained above with reference
to Fig. 14. During the period, since glow-plug energization is normally effected,
a negative determination is made in step S310 or an affirmative determination is made
in step S314 in the starting precondition checking process (Fig. 5), and the starting
preconditions are determined to be not satisfied in step S316 at time t82.
[0177] After completion of the normal glow-plug energization process at time t83, if affirmative
determinations are made in steps S400 and S404 through S417 and negative determinations
are made in step S402 and step S420 in the process (Figs. 8 and 19) of checking conditions
for executing glow-plug fault determination, it is determined in step S422 whether
the starting preconditions are satisfied. Since the starting preconditions are not
satisfied (i.e., a negative determination is made in step S422), it is then determined
in step S424 whether the main fault determination has already been affirmative in
the last or other previous trip. Since the main fault determination has been affirmative
in this example (i.e., an affirmative determination is made in step S424), the main
fault determination is made negative in step S425, and execution of glow-plug fault
determination is permitted in step S426 at time t84. As a result, the provisional
fault determination process (Fig. 20) is executed, and the relationship of the above-indicated
expression (13) is satisfied at time t85 and kept satisfied (i.e., an affirmative
determination is made in step S814), whereby the normality determination is made affirmative
in step S818.
[0178] With the normality determination thus made negative, a negative determination is
made in step S714 of Fig. 22, and the glow plug 22 is forced to be deenergized at
time t86 in step S723 after the stand-by time passes from the forced energization
of the glow plug 22 in step S810 of Fig. 20. Also, the provisional fault determination
is made negative in step S720 and execution of glow-plug fault determination is not
permitted in step S724, whereby both the provisional fault determination process (Fig.
20) and the main fault determination process (Fig. 21) are stopped. With the process
as described above, information that the starting preconditions are not satisfied,
the main fault determination is negative, and the normality determination is affirmative
is recorded in the memory of the ECU 52.
[0179] In the glow-plug fault detecting apparatus and method as described above, the process
(Figs. 8 and 19) of checking conditions for execution of glow-plug fault determination,
the provisional fault determination process (Fig. 20), the main fault determination
process (Fig. 21) and the process (of Figs. 12 and 20) of checking conditions for
stopping glow-plug fault determination correspond to fault presence determining means.
Also, the process (Figs. 8 and 19) of checking conditions for execution of glow-plug
fault determination and the provisional fault determination process (Fig. 20) correspond
to the first fault determining means, and the main fault determination process (Fig.
21) and the process (Figs. 12 and 22) of checking conditions for stopping glow-plug
fault determination correspond to the second fault determining means. The fault possibility
determining means and the means for detecting an engine state before completion of
the engine starting cycle are the same as those of the first embodiment.
[0180] The second embodiment of the invention as described above yields the following effects.
(1) This embodiment provides the same effects as those as described above at (1),
(2) and (4) through (6) with respect to the first embodiment.
(2) Since the provisional fault determination process (Fig. 20) and the main fault
determination process (Fig. 21) are executed after expiration of the normal glow-plug
energization period, these processes have no influence on the normal glow-plug energization
control. Accordingly, the diesel engine 2 operates with improved stability during
start-up, in particular, during a cold start.
[0181] In addition, since a fault in glow-plug energization is not detected within the normal
glow-plug energization period, chances to detect a fault in glow-plug energization
are sufficiently provided even if the diesel engine 2 has been warned up and the normal
glow-plug energization period is short,.
[0182] Other embodiments of the invention will be now described.
(a) While the history of the battery voltage VB used in the starting precondition
checking process (Fig. 5) means the history during the engine starting cycle (namely,
from start to completion of the starting cycle) in the illustrated embodiments, the
history of the battery voltage VB may further cover a period from turn-on of the ignition
switch to the start of the engine starting cycle, or may include the history of the
battery voltage VB immediately after completion of the starting cycle of the diesel
engine 2. Also, the history of the battery voltage VB may be limited to that during
the period from turn-on of the ignition switch to the start of the engine starting
cycle, or may be limited to that after completion of the starting cycle of the diesel
engine 2.
(b) In the illustrated embodiments, the starting period Tsta and the history of the
battery voltage VB are used for determining the starting preconditions in the starting
precondition checking process (Fig. 5). However, the starting preconditions may be
determined to be satisfied or not satisfied by using only the starting period Tsta
or using only the history of the battery voltage VB.
If one of the condition that the starting period Tsta is equal to or shorter than
the normal starting period Tstanorm and the condition that the battery voltage VB
has ever dropped below the starting-time normal minimum voltage Vstanorm is satisfied,
the starting preconditions are determined to be not satisfied. In another embodiment,
the starting preconditions are not satisfied when both of the above two conditions
are satisfied, and are satisfied when both of the two conditions are not satisfied.
(c) In the illustrated embodiments, execution of glow-plug fault determination is
permitted (in step S426) when the diesel engine 2 is in an idle state in the process
(Figs. 8, 9 and 19) of checking conditions for executing glow-plug fault determination,
and the provisional fault determination process (Fig. 10 or Fig. 20) and the main
fault determination process (Fig. 11 or Fig. 21) are executed during idling. In another
embodiment, the provisional fault determination process (Fig. 10 or Fig. 20) and the
main fault determination process (Fig. 11 or Fig. 21) may be allowed to be executed
if the diesel engine 2 is in a quasi-idle state, even if the engine 2 is not actually
idling. For example, when no power is transmitted from the diesel engine to the wheels
with a clutch being disengaged during downhill running or inertia running, the provisional
fault determination process (Fig. 10 or Fig. 20) and the main fault determination
process (Fig. 11 or Fig. 21) may be executed provided that the engine 2 can be stably
rotated.
(d) In the provisional fault determination process (Fig. 10, 20) and the main fault
determination process (Fig. 11, 21), when switching of the glow plug 22 between the
energized state and the deenergized state results in sufficiently large changes in
one of the fuel injection quantity and the battery voltage, the normality determination
is made affirmative or the provisional fault determination is made negative. In another
embodiment, a fault may be determined based solely on changes in the fuel injection
quantity or based solely on changes in the battery voltage.
In the fault determination processes as described above, when it is determined that
a sufficiently large change does not appear in the fuel injection quantity, the provisional
fault determination is made affirmative or the main fault determination is made affirmative.
In another embodiment, however, the determination may be made when a sufficiently
large change does not appear in the battery voltage, or when sufficiently large changes
do not appear in both the fuel injection quantity and the battery voltage.
In addition to the fuel injection quantity and the battery voltage, or instead of
these parameters, a change in the control duty DF associated with the alternator 54
may be regarded as a change in the operating state of the diesel engine 2. More specifically,
the normality determination may be made affirmative or the provisional fault determination
may be made negative when a sufficiently large change in the control duty DF occurs
upon forced switching between the energized state and the deenergized state of the
glow plug 22, and the provisional fault determination may be made affirmative or the
main fault determination may be made affirmative when a sufficiently large change
does not appear in the control duty DF.
While the final basic injection quantity command value QFINC is used as the fuel injection
quantity based on which the provisional fault determination and the main fault determination
are made in the illustrated embodiments, the basic injection quantity command value
QBASE, governor injection quantity command value QGOV or the idle injection quantity
correction value QII may be used as the fuel injection quantity.
Also, the parameters based on which the provisional fault determination and the main
fault determination are made are not limited to the fuel injection quantity, battery
voltage and the control duty, but may be selected from any other parameters, such
as the engine load, which are indicative of changes in the operating state of the
diesel engine.
(e) In the illustrated embodiments, the main fault determination is made affirmative
when sufficiently large changes in the fuel injection quantity or the battery voltage
do not occur upon switching of the glow plug 22 between the energized state and the
dieenergized state in the provisional fault determination process (Fig. 10 or 20)
and the main fault determination process (Fig. 11 or 21). In another embodiment, only
the provisional fault determination process (Fig. 10 or Fig. 20) is executed, and
the normality determination is made affirmative when a sufficiently large change in
one of the fuel injection quantity and the battery voltage occurs upon switching of
the glow plug 22 between the energized state and the dienergized state, and the main
fault determination is made affirmative when a sufficiently large change does not
occur in the fuel injection quantity.
(f) In the first embodiment, forced switching of the glow plug 22 between the energized
state and the deenergized state for fault detection is executed only within the normal
glow-plug energization period in which the glow plug 22 is energized for warm-up of
the engine 2. However, the first embodiment may be combined with the second embodiment
in this respect.
For example, when the forced switching of the glow plug 22 for fault detection cannot
be effected within the normal glow-plug energization period, forced energization and
denergization of the glow plug 22 for fault detection may be carried out in the same
manner as in the second embodiment when the engine is brought into an idle state after
the normal glow-plug energization period expires. With this arrangement, fault determination
utilizing forced glow-plug energization can be implemented with high certainty or
reliability at the earliest opportunity, as compared with the first and second embodiments.
(g) While the coolant temperature is used as a temperature factor in each of the illustrated
embodiments, the temperature of the engine oil may be used instead. Alternatively,
the engine temperature may be estimated based on a balance between heat generated
and heat radiated as a result of combustion in the diesel engine, and may be used
as a temperature factor.
[0183] In method and apparatus for detecting a fault in a glow plug provided in a diesel
engine, a possibility of a fault in the glow plug is determined without utilizing
a forced change in the state of energization of the glow plug for the purpose of diagnosis,
before the presence of a fault in the glow plug is determined by utilizing the forced
change in the state of energization of the glow plug. The process of determining the
presence of the fault is not executed if it is determined in the process of determining
the possibility of the fault that there is no possibility of a fault in the glow plug.