The present invention relates to a measurement device in an electronically-controlled radio frequency ignition system for an internal combustion engine, suitable for measuring of the ionization current of the gases in the cylinders of the engine upon combustion.
The ionization current of the gases in the cylinders of the engine is typically measured after the end of the ignition and is particularly advantageously applicable, for example, for detecting the crankshaft angle corresponding to the pressure peak of the combustion chamber, for detecting pinking or even for identifying misfiring.
Circuits for measuring the ionization current for a conventional ignition system are known, wherein the operation consists in polarizing the mixture in the combustion chamber after the generation of the spark between the electrodes of the sparkplug, in order to measure the current resulting from the propagation of the spark.
Such circuits are conventionally positioned at the foot of the secondary of an ignition coil connected to the sparkplug.
However, these circuits need to be dedicated to the characteristics of conventional ignition and are therefore not suitable as such to plasma generation ignition systems which implement sparkplugs of the radio frequency plug-coil type, as described in detail in the following patent applications filed in the name of the Applicant: FR 03-10766, FR 03-10767 and FR 03-10768.
The specific features of radio frequency ignition engender a number of constraints in measuring the ionization current. In practice, the ionization current is measured after the end of ignition. Its amplitude depends on the DC voltage or “polarization voltage” applied between the high-voltage electrode of the plug and the engine ground. The polarization voltage typically lies between the battery voltage and a few hundred volts. Experience shows that the signal representative of the ionization current has an amplitude of between 0.1 μA and 1 mA depending on the conditions of the combustion chamber (temperature, pressure, composition of the mixture, etc.). Now, the ignition control signal induces significant currents which have an amplitude deviation of almost 120 dB with the ionization current that is to be measured. The measurement circuit therefore undergoes a glare time during which it cannot acquire a low current.
Furthermore, this type of ignition makes it possible to develop 2 types of discharge (a multi-filament spark or a mono-filament arc), which have a different influence on the measurement system. There is therefore difficulty in guaranteeing independence of the measurement of the ionization current relative to the type of discharge generated.
The present invention therefore aims at least partly to resolve these drawbacks by proposing a device for measuring the ionization current that is suited to a radio frequency ignition system.
With this objective in mind, the invention therefore relates to a radio frequency ignition device for an internal combustion engine, characterized in that it comprises:
Advantageously, the measurement capacitor is connected in series between the secondary winding of the transformer and the resonator, at the level of a ground return wire of the transformer and of the resonator.
According to one embodiment, the measurement circuit comprises a transistor mounted in common base configuration, a first electrode of which is connected to a terminal of the measurement capacitor and a second electrode of which is connected to the polarization voltage via the first amplification means.
According to another embodiment, the first electrode of the transistor mounted in common base configuration is also connected to an input resistor of the measurement circuit.
Advantageously, the means of measuring the voltage representative of the amplified ionization current are also connected to the input resistor via second amplification means, which are able to amplify the current circulating in the input resistor and having an amplification gain identical to the first amplification means.
According to one embodiment, the amplification means comprise a current mirror.
According to one embodiment, the means of measuring the voltage representative of the amplified ionization current comprise a measurement resistor.
According to another embodiment, a primary winding of the transformer is connected on one side to a power supply voltage and on the other side to the drain of at least one switching transistor controlled by a control signal, the switching transistor applying the power supply voltage to the terminals of the primary winding at a frequency defined by the control signal.
Preferably, the transformer has a variable turns ratio.
Other features and benefits of the present invention will become more clearly apparent from reading the following description given by way of illustrative and nonlimiting example and with reference to the appended figures in which:
The plug coil implemented in the context of controlled radio frequency ignition is electrically equivalent to a resonator 1 (see
In practice, when the resonator is powered by a high voltage at its resonance frequency Fc (1/(2π√{square root over (Ls*Cs)})), the amplitude at the terminals of the capacitor Cs is amplified so that multi-filament discharges develop between the electrodes, over distances of the order of a centimeter, at high pressure and for peak voltages of less than 20 kV.
The sparks are then said to be “branched”, inasmuch as they involve the simultaneous generation of at least a number of ionization lines or paths in a given volume, their branches also being omnidirectional.
This application to radio frequency ignition then requires the use of a power supply circuit, capable of generating voltage pulses, typically of the order of 100 ns, possibly being able to reach amplitudes of the order of 1 kV, at a frequency very close to the resonance frequency of the plasma generation resonator of the radio frequency plug coil.
This circuit consists of an intermediate DC power supply Vinter that can vary from 0 to 250 V, a power MOSFET transistor M and a parallel resonant circuit 4, comprising a coil Lp in parallel with a capacitor Cp, also with a resonance frequency close to 5 MHz. The transistor M is used as a switch to control the switchings at the terminals of the parallel resonant circuit and of the plasma generation resonator 1 intended for connection to an output interface OUT of the power supply circuit.
The transistor M is driven on its gate by a control logic signal V1, supplied by a control stage 3, at a frequency that should be roughly aligned on the resonance frequency of the resonator 1.
The intermediate DC power supply voltage Vinter, which can vary between 0 and 250 V, can advantageously be supplied by a high-voltage power supply, typically a DC/DC converter.
Thus, in proximity to its resonance frequency, the parallel resonator 4 transforms the DC power supply voltage Vinter into an amplified periodic voltage, corresponding to the power supply voltage multiplied by the quality factor of the parallel resonator and applied to an output interface of the power supply circuit at the level of the drain of the switching transistor M.
The switching transistor M then applies the amplified power supply voltage to the output of the power supply, at the frequency defined by the control signal V1, that is to be made as close as possible to the resonance frequency of the plug coil, so as to generate the high voltage at the terminals of the electrodes of the plug coil that are necessary to the development and sustaining of the multi-filament discharge.
The transistor thus switches high currents (Ipeak≈20 A) at a frequency of approximately 5 MHz, and with a drain-source voltage that can reach 1 kV. The choice of the transistor is therefore critical and necessitates a trade-off between voltage and current.
Also, according to the embodiment illustrated in
The secondary winding LN of the transformer, one side of which is linked to ground by a ground return wire 6, is designed to be connected to the plug coil. In this way, the resonator 1 of the plug coil, connected to the terminals of the secondary winding by link wires 5 and 6, including the ground return wire 6, is therefore powered by the secondary of the transformer, as illustrated in
Adapting the turns ratio then makes it possible to reduce the drain-source voltage of the transistor. Reducing the voltage on the primary however induces an increase in the current passing through the transistor. It is then possible to compensate this constraint by, for example, placing two transistors in parallel controlled by the same control stage 3.
Upon ignition, it is essential for the branched spark to develop in volume in order to ensure combustion and optimum engine operation. Measuring of the ionization current therefore entails using a component which does not degrade the energy efficiency of the ignition.
The solution retained for this purpose consists in connecting a measurement capacitor CMES in series between the secondary winding of the transformer T and the resonator 1, to the ground return wire 6. The measurement capacitor is thus advantageously placed in the circuit at a point where the potential differences relative to ground are as low as possible.
A capacitor of reduced capacitance, typically around ten nanofarads, makes it possible not to disturb the ignition system while having the possibility of performing low-frequency measurements of the ionization current.
Thus, the main benefit from the choice of this measurement component over other passive components lies in its radio frequency behavior. In practice, at high frequencies, those skilled in the art know that the high-frequency equivalent circuit of a capacitor consists of a series resonator. Now, a resonator has an impedance that changes according to the frequency of the signal applied to its input, and is minimal at the resonance frequency of the resonator. This characteristic of the change in impedance of a resonator as a function of the frequency then enables the capacitor to present a very low impedance in the vicinity of the ignition resonance frequency and a high impedance in the frequency band used for the ionization signal (FION<15 kHz). The measurement capacitor is therefore judiciously chosen so as to present its lowest impedance in the range of frequencies used for the ignition control signal. This makes it possible to minimize the voltage at the terminals of the measurement capacitor to protect the measurement circuit, which will now be described with reference to
Useful combustion information can be extracted from the ionic signal soon after the end of ignition. With the combustion lasting on average 40° crankshaft, it is tolerable for the information to be masked for up to 200 μs after the end of the spark (or approximately 8° crankshaft for an engine speed of 6500 rpm). It is then necessary to provide a measurement circuit that is able to be available very rapidly to perform the measurement. Since the measurement circuit is saturated throughout the ignition phase, because of the major currents induced by the ignition control signal, it is then necessary for the circuit desaturation time to be no more than 200 μs in order to be able to acquire the measurement signal in linear mode.
Also, the measurement circuit 10, connected to the terminals of the measurement capacitor CMES, as illustrated in
Thus, the low input impedance of the generator makes it possible to keep the voltage constant at the terminals of the capacitor and/or rapidly bring its voltage to VPOLAR after the spark. This impedance is sufficiently low for the current IION representative of the trend of the combustion of the gases in the combustion chamber to be supplied by the transistor TB, the operation of which will be described in more detail hereinbelow, and not by the capacitor C. It is this discharge current IION that is to be measured via the measurement circuit 10 of
Thus, the polarization voltage VPOLAR is applied to the circuit via a polarization stage 12, comprising a bipolar transistor TB mounted in common base configuration with an output on the emitter of the transistor, connected to a terminal of the measurement capacitor CMES. The mounting of the transistor TB in common base configuration is characterized notably by its low input impedance, advantageously making it possible to obtain the desired reactivity on the measurement circuit.
For example, by connecting this circuit to the measurement capacitor, an input impedance ZE is obtained that is equivalent to:
In which: RIN is the resistor placed at the input of the measurement circuit,
Rbe indicates the intrinsic resistance of the transistor TB, and
β corresponds to the gain of the transistor TB.
Typically, by choosing: RIN=8 kΩ, Rbe=1 kΩ and β=100, we obtain ZE≈10Ω.
The output current IS of the circuit 12, representative of the ionization current IION, is measured via the output resistor RS of the measurement circuit which, as will be seen in more detail hereinbelow, is in fact passed through by a current which confers on it a voltage VS at its terminals, the measurement of which will then provide an image in voltage of the ionization current.
This current IS is roughly equal to the current difference between the current Ic entering into the transistor TB and the current Ip circulating in the circuit's input resistor RIN.
Now, since the amplitude of the ionization current that is to be measured is low and mostly less than 1 mA, the measurement circuit advantageously comprises current amplification means. To this end, the measurement circuit comprises a first current mirror M1, connected between the polarization voltage source VPOLAR and the input of the transistor TB, and having an amplification gain Gm=RA/RB, defined by the values of the resistors RA and RB respectively present in each branch of the current mirror M1. The current mirror M1 therefore makes it possible to amplify the current of the signal Ic entering into the transistor TB, to copy this amplified signal intended for the resistor RS, connected to the output of the current mirror M1.
As has been seen, the current Ic is the sum of the ionization current IION and the current IR circulating in the input resistor RIN. Also, in order to measure a voltage VS at the terminals of RS that is representative of the only ionization current, it is necessary to subtract the unwanted component corresponding to the current circulating in the input resistor RIN from the amplified signal obtained at the output of the current mirror M1.
To do this, the measurement circuit comprises a second current mirror M2, connected between the input resistor RIN of the circuit and ground, and having an amplification gain Gm that is identical to the first current mirror M1, defined by the values of the resistors R′A and R′B respectively present in each branch of the current mirror M2.
The output resistor RS, connected to the output of the second amplification means M2, is therefore passed through by the current difference Ic−IR, roughly equal to the ionization current IION, amplified by the ratio Gm=RA/RB=R′A/R′B. In other words, the output resistor RS is passed through by an amplified image of the ionization current, so as to obtain the output voltage VS at its terminals according to the relation:
V
S
=G
m
×R
S
×I
ION
In order to obtain a high ionization current, the circuit must be polarized with a DC voltage that is as high as possible, but limited by the maximum voltages and currents supported by the transistors of the circuit. Also, the maximum voltage accepted by the transistors of the measurement circuit determines the polarization voltage of the circuit. Similarly, the input current must remain sufficiently low to guarantee linear-mode operation. This constraint conditions the gain applied to the current mirrors. Thus, in the case of a short circuit on the input (on the terminals of the measurement capacitor), the current increases in the resistor RA of the current mirror M1. By amplification, the current in the resistor RB is increased. To protect the circuit, it is possible to add a diode D2 from the collector to the base of the second transistor of the current mirror M1.
It will also be observed that the voltage at the terminals of the measurement capacitor is also a function of the type of spark generated. A mono-filament spark between the electrode of the plug and a ground plan leads to an abrupt increase in the current circulating in the measurement capacitor and consequently a high variation of the voltage at its terminals, potentially damaging for the measurement circuit. The measurement circuit can therefore provide a protection diode D1, making it possible to transfer the excess energy into a buffer capacitor CT and ensure that the voltage at the terminals of the measurement capacitor does not exceed the polarization voltage VPOLAR.
Number | Date | Country | Kind |
---|---|---|---|
0758795 | Nov 2007 | FR | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/FR08/51776 | 10/1/2008 | WO | 00 | 6/23/2010 |