[0001] The invention relates to a method and an apparatus for operating an injection valve.
[0002] Increasingly stringent rules concerning the admissibility of noxious emissions from
internal combustion engines which are arranged in vehicles render it necessary to
take various measures which reduce the emission. One way to reduce these emissions
is to improve the combustion process in the internal combustion engine. Injection
valves are in wide spread use, in particular for internal combustion engines where
they may be arranged in order to dose fluid into an intake manifold of the internal
combustion engine or directly into the combustion chamber of a cylinder of the internal
combustion engine.
[0003] The object of the invention is to provide a method and an apparatus for operating
an injection valve which contribute to a reliable and precise function of the injection
valve.
[0004] This object is achieved by the features of the independent claims. Advantageous embodiments
of the invention are given in the sub-claims.
[0005] The invention is distinguished by a method and a corresponding apparatus to operate
an injection valve. The injection valve comprises a valve needle preventing a fluid
flow out of an injection nozzle in a closing position and enabling the fluid flow
of the injection nozzle apart from the closing position. Furthermore the injection
valve comprises an electro-magnetic actuator unit being designed to actuate the valve
needle. The actuator unit is activated according to a predetermined activation signal
with a given activation period for effecting a fluid flow out of the injection nozzle.
An actuator unit voltage characteristic is captured at least over a period of time
during which the valve needle could reach the closing position. Depending on the actuator
unit voltage characteristic a closing time is determined representing a time when
the valve needle reaches the closing position.
[0006] In this way it is possible to determine the time, the valve needle effectively reaches
the closing position, very precisely. To control injection durations and quantise
of the injection valve the actuator unit is activated by the predetermined activation
signal. Therefore the actuator unit may be controlled by a drive unit. The amount
of injected fluid depends on the pulse or signal shape and the activation period of
the activation signal. The activation period is a preferred parameter for controlling
the amount and/or flow rate of injected fluid. The activation period characterizes
a complete control time that could be divided in different phases according given
requirements. But for a precise dosing of the fluid it is advantageous to know the
time the valve needle effectively reaches the closing position. For instance, the
injection valve is indeed calibrated when new to correlate the activation signal with
the amount of fluid injected. But such calibrations are approximate because the amount
of fluid injected varies with engine operating conditions and/or with age and/or wear
of the injection valve. So, a closing period, representing a time period between an
end of the activation period and the time the valve needle reaches the closing position,
of a particular injector valve can vary, which can affect the estimation of the flow
rate and/or the amount of fluid injected. A variation of the closing period for a
particular injector valve becomes especially relevant, if the closing period becomes
of the same order of the activation period.
[0007] The actuator unit may comprise an armature and a coil. A core may be assigned to
the actuator unit. The coil and the core form an electromagnet. For instance the core
may be formed by an inlet tube. But there may be other elements guiding and/or amplifying
a magnetic field induced by the coil, like a valve body and/or a housing and/or the
armature. The actuator unit and the other elements guiding and/or amplifying a magnetic
field induced by the coil form an electro-magnetic circuit. An electrical behaviour
of this electro-magnetic circuit can be characterized by the actuator unit voltage.
For instance, the actuator unit voltage characteristic can be measured by a voltage
sensor. There may be additional and/or other components guiding and/or amplifying
the magnetic field induced by the coil and therefore may be additionally or alternatively
be considered in the electro-magnetic circuit.
[0008] During activation the magnetic field induced by the coil is controlled by the activation
signal. In this way the armature is directly controlled by the activation signal during
the activation period. After the activation period, when the activation signal has
returned to zero or another off state this force coupling is interrupted and there
is a transient phase during which the actuator unit voltage characteristic returns
to zero. During the transient phase the armature and other with the armature mechanically
coupled components move themselves depending upon their inertia and/or a mechanic
condition and/or a hydraulic condition to close the injector. But there is still some
energy stored in the electromagnetic circuit. During this transient phase the electromagnetic
circuit is discharged and the actuator unit voltage characteristic returns to zero.
[0009] According to a preferred embodiment a second temporal derivate of the actuator unit
voltage characteristic is determined. Depending on the second temporal derivate of
the actuator unit voltage characteristic the closing time is determined.
[0010] According to a further preferred embodiment the closing time is determined in correlation
with a first zero crossing of the second temporal derivate of the actuator unit voltage
characteristic.
[0011] According to a further preferred embodiment a smoothed actuator unit voltage characteristic
is determined by at least filtering the actuator unit voltage characteristic once
with a smoothing filter. Furthermore the second temporal derivate of the smoothed
actuator unit voltage characteristic is determined and depending on this second temporal
derivate the closing time is determined. In this way noise related to the system and/or
to external noise sources can be eliminated and subsequent processing and/or calculation
steps can be simplified.
[0012] According to a further preferred embodiment the closing time is determined in correlation
with a first zero crossing of the second temporal derivate of the smoothed actuator
unit voltage characteristic.
[0013] In a further preferred embodiment the smoothing filter is designed to determine for
a respective value of the captured actuator unit voltage characteristic or a pre-processed
actuator unit voltage characteristic an average value or a weighted average value
depending on the value itself and a given second number of respective previous and
subsequent values. High frequency signal parts, typical for noise, can be sub-pressed
by such an averaging. It is possible to measure the actuator unit voltage by sampling
the actuator unit voltage by given sampling instances.
[0014] In a further preferred embodiment a first number of filter cycles is defined depending
on the actuator unit voltage characteristic and/or how it is captured. This has the
advantage that depending on a particular application the filtering procedure can be
repeated several times, for instance, until high frequency signal parts disappear
which are typical for noise.
[0015] In a further preferred embodiment the second number of respective previous and subsequent
values is defined depending on the captured actuator unit voltage characteristic and/or
how it is captured. Thus, an additional parameter is available for optimizing the
filtering procedure according to predefined requirements.
[0016] Exemplary embodiments of the invention are shown in the following with the aid of
schematic drawings. These are as follows:
- Figure 1
- an embodiment of an injection valve 100,
- Figure 2
- a diagram of an actuator unit voltage characteristic Uc and
- Figure 3a to 3c
- diagrams for evaluating a closing time t_close.
[0017] Elements of the same design and function that appear in different illustrations are
identified by the same reference character.
[0018] An injection valve 100 (figure 1), that is particular suitable for dosing fuel into
an internal combustion engine, comprises e. g. a valve assembly 11 and an inlet tube
12.
[0019] The valve assembly 11 comprises a valve body 14 with a central longitudinal axis
L and a housing 16. The housing 16 is partially arranged around the valve body 14.
Furthermore a cavity 18 is arranged in the valve body 14.
[0020] The cavity 18 takes in a valve needle 20, an armature 22 and in this particular case
a damper element, e. g. a damper spring 46. The damper spring 46 forms a soft stop
element for the armature 22. In the shown embodiment the armature 22 has an upper
guide 24 formed as a collar around the valve needle 20. The upper guide 24 is mechanically
coupled with the valve needle 22.
[0021] A calibration spring 28 is arranged in a recess 26 provided in the inlet tube 12.
[0022] The valve needle 20 comprises, for example, a valve needle body and a sealing element.
The sealing element is mechanically coupled with the valve needle body. The valve
needle body preferably has a cylindrical shape. The sealing element has for example
a spherical shape. Alternatively, the sealing element can have a conical shape. In
a closing position of the valve needle 20, the sealing element rests on a seat preventing
a fluid flow through at least one injection nozzle of the injection valve 100. The
injection nozzle may be, for example, an injection hole. However, it may also be of
some other type suitable for dosing fluid. The sealing element permits the fluid injection
into the combustion chamber in further positions, i. e. when it does not rest on the
seat. The further positions represent non-closing positions.
[0023] The valve assembly 11 is provided with an actuator unit 36 which is preferably an
electro-magnetic actuator. The actuator unit 36 comprises, for example, an armature
22. The actuator unit 36 comprises a coil 38, which is preferably arranged inside
the housing 16 and overmolded. A core may be assigned to the actuator unit 36. The
coil 38 and the core form an electromagnet. In this example the core is mainly formed
by at least a part of the inlet tube 12. But also the armature 22, the housing 16
and the valve body 14 are affected by a magnetic field induced by the activated coil
38. There may be other and/or additional elements guiding and/or amplifying a magnetic
field induced by the coil.
[0024] The actuator unit 36 and the other elements guiding and/or amplifying the magnetic
field induced by the coil 38 form an electro-magnetic circuit. An electrical behaviour
of this electro-magnetic circuit can be characterized by an actuator unit voltage.
For instance, the actuator unit voltage can be measured by a voltage sensor. There
might be additional and/or other components which are affected by the magnetic flux
and therefore may be considered in the electro-magnetic circuit.
[0025] In the case when the electro-magnetic actuator unit 36 with the coil 38 is activated
by a predefined activation signal during a given activation period Ti, the electromagnet
may effect, depending on the activation signal, an electro-magnetic force on the armature
22. The armature 22 may be attracted by the electromagnet and moves in the direction
of the longitudinal axis L away from a fluid outlet. The armature 22 pushes on the
upper guide 24, which is mechanically coupled with the valve needle 20 and therefore
the valve needle 20 moves in axial direction out of the closing position.
[0026] After the activation period Ti, when the activation signal has returned to zero or
another off state this force coupling is interrupted and there is a transient phase
T_phase because some energy is still stored in the electro-magnetic circuit. Depending
on a force balance between the force on armature 22 caused by the electromagnet and
the force on the armature 22 caused by the calibration spring 28 the valve needle
20 moves in its closing position. The motion of the armature 22, e. g. together with
the upper guide 24, has an impact on the electrical behaviour of the electro-magnetic
circuit and therefore on the actuator unit voltage during the transient phase T_phase.
Depending on an architecture of the injection valve 100 the armature 22 has a different
dynamic behaviour and so a different impact on the actuator unit voltage.
[0027] Figure 2 shows a diagram of an actuator unit voltage characteristic Uc during the
activation period Ti and the transient phase T_phase. The actuator unit 36 is activated
according to a predetermined activation signal with a given activation period Ti for
effecting a fluid flow out of the injection nozzle.The actuator unit 36, for instance,
comprises actuator unit control pins. The activation signal may be applied to these
control pins. The actuator unit voltage can, e. g., be measured on these control pins
by sampling the actuator unit voltage with given sampling instances.
[0028] A variation of the armature dynamic, which happens in a moment the valve needle 20
reaches the closing position, can be detected depending on a the actuator unit voltage
characteristic Uc.
[0029] The actuator unit voltage characteristic is captured at least over a period of time
during the valve needle could reach the closing position. For instance, the actuator
unit voltage may be captured during the activation period Ti and the transient phase
T_phase. Alternatively it is also possible that actuator unit voltage may be captured
only during the transient phase T_phase. The captured actuator unit voltage characteristic
Uc may be noisy, because of a system noise and/or noise from external sources. For
evaluating the actuator unit voltage characteristic Uc it may be advantageous to filter
the actuator unit voltage characteristic Uc by a smoothing filter to sub-press high
frequency signal parts, which may be mainly caused by the noise. The smoothing filter
may be a linear smoothing filter, wherein the smoothing filter is designed to determine
for a respective value of the captured actuator unit voltage characteristic or a pre-processed
actuator unit voltage characteristic an average value depending on the value itself
and a given second number of respective previous and subsequent values according equation
1:

i: sample time index
j: number of current filter cycle (j = 1 ... M)
N: number of previous samples and number of subsequent samples, i. e. average radius
[0030] A filter procedure, for instance according equation 1, may be applied several times
until the high frequency signal parts typically for noise disappear.
[0031] Alternatively the smoothing filter may be a linear smoothing filter which is designed
to determine for a respective value of the captured actuator unit voltage characteristic
or a pre-processed actuator unit voltage characteristic a weighted average value depending
on the value itself and a given second number of respective previous and subsequent
values. Also a non-linear smoothing filter may be applied. In the case the filter
procedure is performed several times, also different smoothing filter types may be
used.
[0032] With respect to the time line the diagrams of Figure 3b and 3c show a part of the
transient phase T_phase. The valve needle 20 reaching the closing position by contacting
the seat causes a change in the dynamic behaviour of the armature 22. This change
can be detected depending on the, in particular, smoothed injector voltage Us. A first
zero crossing of the second temporal derivative of the smoothed actuator unit voltage
characteristic Us" characterizes the time the valve needle 20 reaches the closing
position. The closing time t_close may be determined in correlation with the first
zero crossing of the second temporal derivate of the smoothed actuator unit voltage
characteristic Us".
1. Method to operate an injection valve (100) comprising
- a valve needle (20) preventing a fluid flow out of an injection nozzle in a closing
position and enabling the fluid flow of the injection nozzle apart from the closing
position, and
- an electro-magnetic actuator unit (36) being designed to actuate the valve needle
(20),
wherein
- the actuator unit (36) is activated according to a predetermined activation signal
with a given activation period (Ti) for effecting a fluid flow out of the injection
nozzle,
- an actuator unit voltage characteristic (Uc) is captured at least over a period
of time during which the valve needle (20) could reach the closing position and
- depending on the actuator unit voltage characteristic (Uc) a closing time (t_close)
is determined representing a time when the valve needle (20) reaches the closing position.
2. Method in accordance to claim 1, wherein
- a second temporal derivate of the actuator unit voltage characteristic (Uc") is
determined and
- depending on the second temporal derivate of the actuator unit voltage characteristic
(Uc") the closing time (t_close) is determined.
3. Method in accordance to claim 2, wherein
the closing time (t_close) is determined in correlation with a first zero crossing
of the second temporal derivate of the actuator unit voltage characteristic (Uc").
4. Method in accordance with one of the preceding claims, wherein
- a smoothed actuator unit voltage characteristic (Us) is determined by at least filtering
the actuator unit voltage characteristic (Uc) once with a smoothing filter,
- the second temporal derivate of the smoothed actuator unit voltage characteristic
(Us") is determined,
- depending on the second temporal derivate of the smoothed actuator unit voltage
characteristic (Us") the closing time (t_close) is determined.
5. Method in accordance to claim 4, wherein
the closing time (t_close) is determined in correlation with a first zero crossing
of the second temporal derivate of the smoothed actuator unit voltage characteristic
(Us").
6. Method in accordance with claim 4 or 5, wherein
the smoothing filter is designed to determine for a respective value of the captured
actuator unit voltage characteristic (Uc) or a pre-processed actuator unit voltage
characteristic an average value or a weighted average value depending on the value
itself and a given second number of respective previous and subsequent values.
7. Method in accordance with one of the preceding claims 4 to 6, wherein
depending on the actuator unit voltage characteristic (Uc) and/or how it is captured
a first number (M) of filter cycles is defined.
8. Method in accordance with one of the preceding claims 4 to 7, wherein
depending on the captured actuator unit voltage characteristic (Uc) and/or how it
is captured the second number (N) of respective previous and subsequent values is
defined.
9. Apparatus to operate an injection valve (100), wherein the injection valve (100) comprises
- a valve needle (20) preventing a fluid flow out of an injection nozzle in a closing
position and enabling the fluid flow of the injection nozzle apart from the closing
position, wherein in the closing position the valve needle (20) rests on a seat, and
- an electro-magnetic actuator unit (36) being designed to actuate the valve needle
(20), and
wherein the apparatus is designed
- to activate the actuator unit (36) according to a predetermined activation signal
with a given activation period (Ti) for effecting a fluid flow out of the injection
nozzle,
- to capture an actuator unit voltage characteristic (Uc) at least over a period of
time during which the valve needle (20) could reach the closing position and
- to determine a closing time (t_close) depending on the actuator unit voltage characteristic
(Uc), wherein the closing time (t_close) represents a time when the valve needle (20)
reaches the closing position.