Method of contactless ignition of a welding arc with high frequency ignition pulse packets

The invention relates to a method for contactless ignition of a welding arc, in which high-frequency ignition pulses are applied between the welding electrode and the workpiece to be processed to ionize the path between the welding electrode and the workpiece, and in which the welding current is switched on after the ignition of the welding arc, as well as a circuit for the contactless ignition of a welding arc, with a charging circuit, at least one pulse capacitor, a discharge circuit containing at least one switch, and a high-voltage transformer for coupling the high-frequency ignition pulses discharged from the pulse capacitor via the switch to the welding electrode.

The ignition process according to the invention and the ignition device are in principle suitable for a wide variety of welding processes, for example TIG (tungsten inert gas) welding processes and plasma welding processes, as well as for welding processes with a consumable electrode or a non-consumable electrode. Furthermore, the application of the ignition process or the ignition device for hand-held welding torches as well as for welding torches in automatic applications, especially in robot applications.

When igniting welding arcs, a distinction is made between contact ignition and non-contact HF ignition. During contact ignition, the welding electrode is placed on the workpiece and then the welding electrode is slightly lifted off the workpiece again while the welding current is switched on, whereupon the arc is ignited. This method is relatively simple and inexpensive and avoids interference with other electronic components of the welding systems due to the high frequency used for HF ignition. However, the contact between the welding electrode and the workpiece can result in inclusions and impurities on the workpiece and, on the other hand, the electrode is worn out by the contact.

With non-contact HF ignition, on the other hand, the welding electrode always remains at a distance from the workpiece and a high-frequency high voltage is applied between the welding electrode and the workpiece, which ionizes the gas molecules in the space between the electrode and the workpiece, so that the welding arc can be ignited without the workpiece touching the electrode . In order to be able to prevent unintentional contact of the workpiece with the electrode, distances between the electrode and the workpiece at which the ignition can take place of at least a few millimeters to a few centimeters are aimed for. After the arc has been ignited, the welding torch is moved further away from the workpiece or moved towards the workpiece and supplied with the normal welding current, and the normal welding process can begin. The electronic switching elements used to generate the high-voltage pulses must withstand the considerable currents or voltages. For this reason, thyristors or semiconductor components and spark gaps are usually used as switches.

EP 1 197 285 A2 shows a circuit arrangement for generating ignition pulses for welding arcs with a flyback converter circuit with four thyristors in a bridge circuit, which are controlled by a trigger circuit. To create an improved control for the ignition pulses, the discharge circuit is controlled at times which are dependent on the parameters of the welding process. The disadvantage of using thyristors as switching elements is that the maximum switching frequency is relatively low and, moreover, these components have to be dimensioned correspondingly large, which leads to a substantial increase in the cost. Due to the low switching frequency of the thyristors, which is 100 Hz, for example, poorer ionization of the protective gas is achieved, so that it is not always ensured that a very rapid or safe ignition of the arc takes place.

EP 947 276 B1 shows a method and a device for igniting a welding arc in which the charging capacitor is discharged via a spark gap. Spark gaps

AT 413 953 B have the advantage over thyristors that they are very robust and have high voltages and

Withstand currents. In addition, relatively high switching frequencies can be achieved. The disadvantage of using spark gaps is the size and the most often necessary

Additional design effort required for cooling. In addition, the spark gaps are subject to high wear and tear and the generation of ozone can lead to oxidation of components. Finally, the electromagnetic interference can lead to problems, especially in the case of very sensitive robot controls, which can only be prevented by appropriate additional shielding.

US Pat. No. 4,870,248 A shows an arc welding device with improved ignitability. The voltage between the electrode and the workpiece is measured in order to detect the ignition of the arc. If the arc has not been ignited, a higher-frequency ignition pulse is applied, thereby increasing the ignition probability. As long as the arc has not been ignited, the voltage on the electrode rises until the

The arc is finally ignited. After the arc is ignited, an oscillator supplies the frequency and pulse width required for the welding process. However, the energy supplied to the electrode must be below the prescribed limit values. A more precise ignition of the arc cannot be achieved with the circuit according to this document.

DE 33 42 932 A1 describes a method for bumpless ignition in MIG-MAG welding, a high-frequency voltage being applied between the welding wire and the workpiece, the ignition sequence of which depends on the wire feed speed. This means that the wire feed speed does not have to be reduced during ignition. However, this does not result in a more precise or reliable ignition of the arc.

The object of the present invention is to create an above-mentioned ignition method and a corresponding ignition device, by means of which a more precise or reliable and very rapid ignition of the arc is possible or to create a higher ignitability. The cost of the method or the device should be as low as possible. Finally, the ignition method or the device for ignition should be able to be adapted to the respective welding conditions. Furthermore, the ignition should function as independently of the respective load as possible.

In terms of the method, the object according to the invention is achieved in that several pulse packets with a predeterminable frequency or duration are applied, with several successive ignition pulses being output in one pulse packet, with a packet pause being carried out between each pulse packets. By applying the ignition pulses in several pulse packets between which appropriately long pauses are provided, the coupled energy of the ignition pulses can be selected to be correspondingly high, so that reliable and rapid ignition occurs while the maximum permissible energy supplied is below the prescribed, averaged over time Limit values ​​can be arranged. As many ignition pulses as possible are arranged per pulse packet, which requires the highest possible ignition pulse frequency. With a higher number of ignition pulses, the ionization of the gas between the welding electrode and the workpiece to be machined is made easier and thus easier ignition of the welding arc is achieved. The ignition method according to the invention is characterized by a particularly high quality and reliable or rapid ignition of the arc. The ignition process enables ignition with a greater distance between the welding electrode and workpiece or with the usual distance a significantly more precise ignition than is the case with conventional processes. The low average energy input also minimizes the safety risk, that is to say that the operator of the welding torch cannot be electrified or cannot be electrified as much. Due to this lower electrifying effect, accidents, for example the welding worker falling from a scaffold or the like, can be avoided.

The fact that the ignition pulses output in a pulse packet in number and / or

AT 413 953 Β

Frequency are changed, an optimal adaptation of the ignition process to the most varied of conditions is achieved in an advantageous manner.

The ratio of the repetition rate or the period duration of the pulse packets to the duration of the ignition pulses is advantageously high.

According to a further feature of the invention, the ignition pulses are applied to the welding electrode within a pulse packet with a period duration, for example between 25 ps and 1 ms, preferably 125 ps. Such a low period or a high repetition frequency of the ignition pulses ensures that the ionization of the gas between the welding electrode and the workpiece takes place more rapidly and more intensely, and thus the welding arc is ignited more easily. Such a low or high period

Repetition frequency was not possible with the previous ignition circuits in which thyristors were used.

According to a further feature of the invention, the period of the pulse packets is between 1 ms and 1 sec, preferably 100 ms. With a repetition frequency of this type in the range between 1 Hz to 1000 Hz with a correspondingly short duration of the pulse packets, it is achieved that the energy transmitted on average is set or can be adjusted so that, for example, the amount of energy can be specified for hand-held welding torches and thus an adaptation to certain regulations and standards can be made. In this way, reliable and rapid ignition of the arc is achieved without any loss of quality, even with a limited amount of energy at the output of the welding system.

If the welding current is applied a certain period of time after the start of the ignition pulses or the pulse packet transmission, an even better ignition can take place, since during this given period of time a pre-ionization of the air or the gas flow takes place and thus a better ignition is possible, since corresponding charge carriers between the welding electrode and the workpiece are already present.

The duration of the pulse packages or the number of ignition pulses per pulse package are advantageously set as a function of the welding parameters, such as the material of the workpiece to be machined, the material of the welding electrode and / or the protective gas used, etc. This allows the ignition process to be adapted to the respective welding conditions. For example, when using protective gases that are difficult to ionize, such as Helium, so that even then a safe and very rapid ignition is possible.

In terms of circuitry, the object of the invention is achieved in that a pulse compression circuit connected to the charging circuit is provided, comprising the pulse capacitor, the high-voltage transformer and the switch, the switch being formed by a magnetic choke so that high-frequency switching of the ignition pulses can be achieved. In contrast to thyristors or spark gaps, the use of a magnetic choke enables the ignition pulses to be switched at high levels

Frequencies in the range, for example 1-40 kHz. Due to the extremely high ignition frequencies, more precise and reliable ignition is achieved, as stronger ionization of the air or the gas flow between the welding electrode and the workpiece is achieved. Finally, the ignition circuit according to the invention achieves an ignition independent of the load and thus, for example, even with very long hose packages, which cause a high load and lower the ignition voltage accordingly, an effective ignition is achieved, since a higher voltage is set by the emission of pulse packages can, whereby the energy mean value resulting from the pulse packets is kept correspondingly low. With thyristors, given a correspondingly low voltage, no or only a particularly poor ignition would function or would have to be dimensioned correspondingly large due to long hose packages or the like.

AT 413 953 Β

The pulse compression circuit can also be constructed from two or more stages connected in series, each stage containing at least one pulse capacitor, a switch formed by a magnetic choke and a high-voltage transformer.

While the demands on the charging circuit can be reduced with multi-stage pulse compression circuits, this entails a higher circuit complexity and thus a higher space requirement.

If a device for controlling the ignition pulses is connected to the charging circuit, a corresponding control of the ignition pulses and thus an adaptation of the ignition to the respective welding parameters can be carried out.

If the control device is also connected to a welding power source to control the point in time at which the welding current is switched on after ignition has taken place, the switching on of the welding current can, for example, be delayed in relation to the ignition pulse packages, whereby a pre-ionization of the air or the gas between the welding torch and Workpiece takes place and thus an even more precise and easier ignition takes place. The coupling of the ignition pulses to the welding torch can take place capacitively via a coupling capacitor connected in series or inductively via a coupling coil.

The present invention is explained in more detail with reference to the accompanying figures.

1 shows a diagrammatic representation of a welding system or a welding device; 2 shows a block diagram of the welding device with a device for contactless ignition of the welding arc; 3 shows a basic circuit diagram of an ignition circuit with a thyristor according to the prior art; 4 shows a basic circuit diagram of an ignition circuit according to the prior art with a spark gap; Fig. 5 Figure 4 is a more detailed illustration of one embodiment of the ignition circuit according to the present invention; 6 to 10 show different time profiles of the generated ignition voltage and control voltage, in a simplified schematic form; 11 shows an embodiment of the ignition circuit according to the invention with a two-stage pulse compression circuit; Fig. 12th an embodiment for the application of the method according to the invention in an ignition circuit with thyristor known from the prior art; 13 shows a further exemplary embodiment for using the method according to the invention in an ignition circuit known from the prior art with a spark gap.

1 shows a welding device 1 or a welding system for a wide variety of processes or methods, such as MIG / MAG welding or WIG / TIG welding or electrode welding processes, double wire / tandem welding processes, plasma or soldering processes, etc. .

The welding device 1 comprises a power source 2 with a power section 3, a control device 4 and a switching element 5 assigned to the power section 3 or the control device 4 for a gas 8, in particular a protective gas such as CO₂, Helium or argon and the like., Is arranged between a gas reservoir 9 and a welding torch 10 or a burner.

In addition, a wire feeder 11, which is common for MIG / MAG welding, can be controlled via the control device 4, whereby a filler material or welding wire 13 from a supply drum 14 or a wire reel into the area of ​​the Welding torch 10 is supplied. It is of course possible that the wire feed device 11, as is known from the prior art, is integrated in the welding device 1, in particular in the base housing, and not, as shown in FIG. 1, as a

Additional device is formed.

AT 413 953 Β

It is also possible that the wire feeder 11 feeds the welding wire 13 or the filler material outside the welding torch 10 to the process point, with a non-melting electrode preferably being arranged in the welding torch 10, as is usual in TIG / TIG welding.

The current for setting up an arc 15, in particular a work arc, between the electrode and a workpiece 16 is fed via a welding line 17 from the power part 3 of the power source 2 to the torch 10, in particular the electrode, whereby the workpiece 16 to be welded, which consists of several Parts can be formed, via a further welding line 18, likewise with the welding device 1, in particular with the power source

2, is connected and thus via the arc 15 or the plasma jet formed for a

Process a circuit can be built.

To cool the welding torch 10, the welding torch 10 15 can be connected to a liquid container, in particular a water container 21, via a cooling circuit 19 with the interposition of a flow monitor 20

Welding torch 10, the cooling circuit 19, in particular a liquid pump used for the liquid arranged in the water tank 21, is started and the welding torch 10 can thus be cooled.

The welding device 1 also has an input and / or output device 22 by means of which the most varied of welding parameters, operating modes or welding programs of the welding device 1 can be set or called up. The welding parameters, operating modes or welding programs set via the input and / or output device 22 are forwarded to the control device 4, and the control device 4 then outputs the individual

Components of the welding system or of the welding device 1 controlled or corresponding

Setpoints for the regulation or control are specified.

Furthermore, in the exemplary embodiment shown, the welding torch 10 is connected to the welding device 1 or the welding system via a hose package 23. The individual lines from the welding device 1 to the welding torch 10 are arranged in the hose package. That

Hose package 23 is connected to welding torch 10 via a coupling device 24, whereas the individual lines in hose package 23 are connected to the individual contacts of welding device 1 via connection sockets or plug connections. In order to ensure a corresponding strain relief of the hose package 23, the hose package 23 is connected to a housing 26, in particular to the base housing of the welding device 1, via a strain relief device 25. It is of course possible that the coupling device 24 can also be used for the connection to the welding device 1.

Basically it should be mentioned that for the different welding processes or welding devices 1, such as, for example, TIG devices or MIG / MAG devices or plasma devices, not all of the components mentioned above have to be used or deployed. For this purpose it is possible, for example, for the welding torch 10 to be designed as an air-cooled welding torch 10.

2 shows a basic block diagram of a circuit for the contactless ignition of a welding arc with a welding power source or power source 2, which supplies the welding torch 10 with the corresponding current and voltage during the welding process with respect to the workpiece 16. For the contactless ignition of the welding arc 15 between a welding electrode 27, in this example a non-consumable electrode, and the workpiece 16 to be processed is parallel to the

Welding electrode 27 and the workpiece 16, an ignition circuit 28 is arranged, which applies high-frequency ignition pulses with a correspondingly high voltage between the welding electrode 27 and the workpiece 16, so that the between the welding torch 10 and the

Workpiece 16 arranged air or the gas 8 is ionized and the formation of the weld7

AT 413 953 Β arc 15 facilitated. The control device serves to control the ignition circuit 28

4, which can also be connected to the welding power source 2. The control device 4 has, for example, the ignition circuit when a start switch is actuated

To activate welding torch 10 and to deactivate it in turn when the arc 15 is ignited, so that the high-voltage pulses are not emitted during the welding process. Of course, it would be possible for the high-voltage pulses to be emitted over the entire welding process, but there is a risk that neighboring devices could be disturbed as a result. Usually, the high-voltage pulses are terminated after the arc has been ignited or, if the conditions are appropriate, they are activated for a short time again, that is to say that, for example, in the case of an alternating current welding, the HF ignition, i.e. the ignition circuit 28, is activated synchronously with the zero crossing, in order to achieve a better and earlier Above all, to achieve safe re-ignition of the arc 15.

3 and 4 show basic block diagrams of ignition circuits as they exist according to the prior art. 3 shows an ignition circuit 28 in which a thyristor 29 is used as a switch to charge a pulse capacitor 30, which is generated by a charging circuit 31, via a high-voltage transformer 32 to the welding electrode 27 and the workpiece 16 (not shown) transfer. It is possible that any charging circuit known from the prior art can be used. The discharge circuit can also be constructed by four thyristors 29 arranged in a bridge circuit, as is described, for example, in EP 1 197 285 A2. Thyristors 29 as switches are disadvantageous in that only relatively low switching frequencies can be achieved with them.

In the variant according to the prior art according to FIG. 4, a spark gap 33 is used as a switch, whereby higher switching frequencies can be achieved, but an increased structural effort is associated. A major disadvantage of the spark gaps 33 is that ozone is generated by them, which causes the

3o installation of spark gaps 33 in a welding device 1 can lead to destruction of the electronic components and / or the circuit boards and / or the plastics, etc. due to the increased ozone exposure. At the same time, increased electromagnetic interference is caused by the spark gaps 33, the avoidance of which requires a great deal of effort for shielding.

5 shows a basic variant of the ignition circuit 28 according to the present application, the charging circuit 31 having a connection 34 for connecting the supply voltage and a buffer capacitor 35 connected to ground Schalteie40 ment 37 connected to ground. The pulse capacitor 30 is charged via the transformer 36 by means of the supply voltage applied to the connection 34.

In addition, a current sensor 39 can be arranged which detects the charging current and sends a signal proportional to it to the controller 38. In the case of the charging circuit 31, a structure known from the prior art was shown, it being possible to use any charging circuit 31 known from the prior art. It is also possible that, instead of the additional control 38, a direct control is carried out by the control device 4 of the welding device 1.

A pulse compression circuit 40 according to the invention is connected to the charging circuit 31, this comprising a magnetic choke 41 as a switch, the pulse capacitor 30 and the high-voltage transformer 32. The charge of the pulse capacitor 30 is transferred or switched via the magnetic choke 41 to the high-voltage transformer 32 and from there to the terminals of the welding electrode 27 or the workpiece 16 (not shown).

The charge of the pulse capacitor 30 is controlled via the electronic switch

AT 413 953 Β

37, which is controlled by a corresponding controller 38 or 4, so that when the switch 37 is activated, a current flows through the primary side of the transformer 36, causing an energy transfer at the transformer 36, which resets the magnetic switch or the choke 41. When the switch 37 is deactivated, the magnetically stored energy is then transmitted via the transformer 36, as a result of which the pulse capacitor 30 is charged. If a certain voltage-time area is reached when charging the pulse capacitor 30, the magnetic choke 41 automatically switches through, so that the energy charged in the pulse capacitor 30 is discharged via the high-voltage transformer 32 and a current or voltage pulse is generated. As a result of the inventive design of the io switch as a magnetic choke 41, very high switching frequencies can be achieved and thus rapid and reliable ignition of the arc 15 can be achieved. In addition, the choke 41 is very robust with regard to the high voltages and currents that occur when the welding arc 15 is ignited. In contrast to spark gaps 33, which usually have to be cooled, the structural outlay in the case of the Dros15 according to the invention is relatively low, although it also generates considerably fewer interference signals.

In FIGS. 6 to 10, different time profiles of a formed HF output signal 42 and a control voltage 43 are shown. The HF output signal 42 was shown schematically in the form of square-wave pulses, in particular the generated current and / or

Voltage pulses shown. The square-wave pulses are usually formed by current and / or voltage pulses with a decaying curve. The HF output signal 42 is basically formed in such a way that one or more pulse packets 44 with a predefinable frequency or duration are applied to the welding electrode 27 and several successive ignition pulses 45 are output in one pulse pack 44, with a packet pause 46 being carried out between the pulse packets 44 will. The ratio of the repetition rate or period duration of the pulse packets to the duration or period duration of the ignition pulses is high. For example, the packet period, consisting of a pulse packet with a subsequent packet pause, is 1 ms to 1 s, corresponding to a repetition frequency of the pulse packets of 1-1000 Hz. The duration of the pulse packets T_w preferably corresponds to 50 ps to 300 ms, whereby this can be freely specified. The period of the ignition pulses is, for example, 25 ps to 1 ms, corresponding to a repetition frequency of 1 kHz to 40 kHz.

By sending out individual pulse packets 44, different HF output signals 42 with different temporal progressions or frequencies can be formed. In this way, the HF output signal 42 can be optimally adapted to the welding conditions, that is to say that, depending on the parameters set, the welding device 1 or the control device 4 and / or 38 corresponding RF output signals 42 are generated, so that a very rapid and reliable ignition of the arc 15 can be achieved.

For example, in FIG. 6, the RF output signal 42 is constructed in such a way that a packet period 47, which consists of a pulse packet 44 and a packet pause 46, is formed from the same clock ratio, that is to say that a time period 48 for a pulse packet 44 is equal to one Duration 49 for the packet pause 46 is. It is thus possible that, depending on the dimensioning, that is to say depending on the length of the period 48, a multiplicity of ignition pulses 45 are contained in the pulse packet 44. For example, between 1 and 300, preferably 60, ignition pulses 45 can be contained in a pulse packet. The number of possible ignition pulses 45 depends on the selected ignition period 50 or its frequency.

The main advantage of the packet-like formation of the HF output signal 42 is that the emitted energy can be reduced, but a reliable ignition of the arc 15 is ensured, that is, that by applying pulse packets to the welding electrode 27, less energy on average is transmitted to this, but with a very high amount of energy for the ignition of the arc 15 during a pulse packet 44

AT 413 953 Β is available. It is thus possible that due to a corresponding control of the

HF output signal 42, the mean energy can be set or specified so that the present ignition method can also be used with manually operated welding torches or also with automatic applications in which the maximum power or amount of energy is limited.

Another example of the RF output signal 42 is shown in FIG. In contrast to the exemplary embodiment in FIG. 6, the packet period 47 has now been doubled, for example, the pulse duty factor between the time 48 and 49 for the 10 pulse packets 44 and the packet pauses 46 being again 50%.

In this embodiment, the control takes place via the number of ignition pulses 45 in a pulse package 44, ie that the same or predetermined number of ignition pulses 45 is repeatedly formed in a pulse package 44, but the frequency of the ignition pulses 45 is changed. In order that this can be seen, in the embodiment shown in FIG. 7 the same number of ignition pulses 45 as are contained in a pulse packet 44 in FIG. 6 was used. When comparing with FIG. 6 it can now be seen that the same number of ignition pulses 45 is again contained in a pulse packet 44, but the ignition period 50 has been adjusted accordingly due to the lengthening of the duration 48 for the pulse packet 44.

Furthermore, it is also possible, as shown schematically in FIG. 8, for the frequency or ignition period duration 50 to be kept constant, as a result of which any number of ignition pulses 45 is formed in a pulse packet 44, ie that due to the duration 48 for the

Pulse packet 44 over this period 48 the ignition pulses 45 are sent out at a constant frequency. For comparison, the frequency or ignition period 50 from FIG. 6 was used in FIG. 8, so that it can now be clearly seen that, due to the longer duration 48 for the pulse packet 44, significantly more ignition pulses 45 are formed within the pulse packet 44.

In the exemplary embodiment in FIG. 9, the HF output signal 42 is formed in such a way that the frequency or ignition period 50 within the pulse packet 44 is changed. As shown schematically, a changed frequency or ignition period 50 is carried out in these time ranges 51 to 53, for example over defined time ranges 51 to 53. Furthermore, it can be seen from this embodiment that the pulse duty factor between the pulse packets 44 and the pulse pauses 46 has been changed in a packet period length 47, so that the time period 48 for a pulse packet 44 is now longer than the time period 49 for the pulse pause 46 it is possible that the time duration 48 for the pulse packet 44 can be shorter than the time duration 49 for the packet pause 46.

Furthermore, it is also possible that the voltage level 54 of the ignition pulses 45 can be changed for each pulse packet 44 or within a pulse packet 44, as is shown in FIG. 10. In this exemplary embodiment, the voltage level 54 is reduced within a pulse packet 44.

Basically, it should be mentioned that it is possible that the welding current is only switched on a certain time after the application of the first ignition pulse 45 or pulse package 44, whereby a certain pre-ionization of the air or the gas 8 between the welding electrode 27 and the workpiece 16 and thus rapid and reliable ignition of the welding arc 15 takes place. It is also possible that the pulse packet duration, i.e. the duration 48 or the number of ignition pulses 45 within a pulse packet 44 and the packet period 47 of the pulse packets 44 can preferably be adapted to the welding conditions, that is, for example, the HF output signal 42, in particular the type or form of the HF output signal 42, based on the set parameters , such as the material of the work piece 16, the material of the welding electrode 27 or the protective gas used or the like., ο

AT 413 953 Β is generated. This means that an optimal ignition process can be used for every welding process.

The coupling of the HF ignition, in particular the HF output signal 42, into the welding circuit can take place in the most varied of ways known from the prior art. Here, for example, the high-voltage ignition pulses are capacitively coupled in via a coupling capacitor. It is of course also possible to couple the high-frequency ignition pulses inductively.

11 shows a variant of the ignition circuit according to the invention in which the pulse compression circuit 40 is constructed in two stages, each stage having a pulse capacitor 30, a magnetic choke 41 and a high-voltage transformer 32. Multi-stage pulse compression circuits 40 have the advantage that the requirements placed on the charging circuit 31 are not too high, but the structural complexity is increased.

In the further exemplary embodiments in FIGS. 12 and 13, an application of the method for sending out pulse packets 44 in the systems known from the prior art, such as with a thyristor 29 or a spark gap 33, is shown.

To this end, for the sake of simplicity, a switching element 55 is integrated which interrupts the primary circuit of the high-voltage transformer 32 so that corresponding output pulses are generated. It is thus possible to generate a pulse packet 44 over a defined period of time 48 by appropriately activating the switching element 55, so that a corresponding RF output signal 42, as described in FIGS. 6 to 10, can be generated. Of course, this can be implemented by appropriate control of the thyristor 29 or a controlled spark gap 33. Furthermore, the charging circuit 31 can be controlled accordingly for the application of the method according to the invention, so that the pulse packets 44 with the intermittent pulse pauses 46 with the high-frequency ignition pulses 45 contained therein are formed at the output or at the electrode 27.

It is essential that the generated HF output signal 42 is formed by individual pulse packets 44 with packet pauses 46 arranged in between. The mean energy or power can thus be set via the pulse duty factor, so that an adaptation of the power or mean amount of energy to the national regulations or standards can be carried out.