JPH07105862A - Feeding method of rf generator, and controller thereof - Google Patents

Abstract

PURPOSE: To improve the operating reliability, and controllability of an RF generator. CONSTITUTION: A control device, which is formed of a primary current rectifier, a smoothing device, a primary current measuring element for outputting the signal proportional to the primary current iP and a transformer, improves the operation reliability of the output of magnetron and controllability thereof. This improvement is carried out by a converter, which is formed of a power switch S1 and a primary free-traveling diode D1 , and during the on-time of the power switch, an inductor L1 absorbs an energy amount E0 =1/2L1 .imax <2> , which can be regulated, and outputs the same energy amount to a secondary side of the transformer during a following off-time. Primary current proportion signal and reference signal are supplied to a comparator, and when the current proportion signal overshoots the reference signal, the power switch S1 is turned form on to off.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an RF generator, and more particularly to an RF generator feeding method and a control device for feeding a magnetron.

[0002]

BACKGROUND OF THE INVENTION Methods and devices of the type described above are used to power RF generators. In particular, a magnetron can be used as the RF generator. The magnetron is powered by a DC voltage. Its characteristics (U _A = f
(I _A )) typically has at least the threshold U _S needed to enable operation. Adjacent to the threshold value U _S is the operating range where the characteristic is flat. That is, the difference internal resistance R ₁ is low. This is shown in FIG. For the reasons stated, it turns out that low-impedance power supply units and controls for loads with such characteristics are very disadvantageous. On the one hand, the dispersion associated with the production of the properties of a particular type of magnetron is very high, and on the other hand, the properties of the various types of magnetron deviate very strongly from each other, so that in principle the individual magnetrons The output must be matched to the individual power packs to output it. In this regard, the lower voltage source's internal resistance results in high fluctuations in the power supply output, as the magnetron's properties virtually completely determine the operating point itself. The current source characteristic (which has a high internal resistance) is more convenient in this case, but also due to the dispersion of the characteristic due to the current source power supply unit, interference effects are unavoidable.
In addition, the current source characteristics are technically more complex than the voltage source characteristics.

A controller for a heat permeable constant current magnetron is known from Valvo, Technical Information 830228. Here, the magnetron has a permanent magnet and a complementary electromagnet. A control current can be applied to the electromagnet via the control device, the result being that the effective properties of the magnetron can be changed or controlled by the behaviour of the complementary electromagnet. The magnetron itself is operated with an unfiltered DC half-wave, as shown in FIG. The secondary voltage (high voltage side)
A 50Hz, 220V transformer with full-wave rectification is provided for this purpose. It consists of a half wave of a series of known sinusoidal voltages, as FIG. 2 shows. According to the described properties of the magnetron, the pulsating pulse type magnetron current i
_A is formed. In relation to the 10 msec half-wave spacing, the angle θ of the current flow, ie the relative conduction spacing, is small. The reason for this is the low internal resistance of the magnetron. That is, the maximum value of the rectified mains voltage is allowed to lie slightly above the threshold value U _S. This is especially a problem with main variations and threshold voltage tolerances. In addition, the energy output is strongly pulsating. Similarly, a uniform power output is not achieved.

[0004]

In the cited example, it is proposed to correct the permissible effects (main voltage, threshold voltage, type-dependent dispersion) by appropriate field control of the electromagnets complementary to the magnetron. In this regard, the magnetron current and voltage must be taken into account, which is complicated and difficult to implement in circuits without potential difference. For implementation, the simpler method is then preferred. In this case, the change in output power is achieved without continuous adjustment, but rather the average value of the output power is changed by block sampling, i.e. by periodic, variable interruption of the pulse train.

Therefore, it is an object of the present invention to improve the operational reliability and controllability of RF generators.

[0006]

This object is achieved in particular by the features of claim 5 and by the control device according to the preamble of claim 5. The same object is achieved by the method initially described by the combination of features specified in the characterizing part of claim 1.

Advantageous developments of the method according to the invention according to claim 1 and of the control device according to the invention according to claim 5 are specified in claims 2 to 4 and 6 to 21, respectively.

An essential concept of the present invention is that it provides a precisely measured power flow. The voltage at the load is thereby freely set. Output power P
_A corresponds in this case to the input power P _e times the device efficiency η (eta), ie P _A = η · P _e (1).

The efficiency η is known and can be calculated or, in some cases, empirically determined precisely. The concept according to the invention allows the tolerances of the mains and of the load characteristics and the various magnetron characteristics to no longer be in the actual power output.

The object of the invention of supplying a precisely allocated energy to the load, in particular the magnetron, and controlling said energy independently of the characteristics of the magnetron, is achieved by the method according to claim 1. It In this connection, the effect according to the invention is that on the primary side individual power packs can be momentarily separated and transmitted to the secondary side, and said energy packs are of primary order in their amplitude or magnitude. Can be changed freely on the side. The magnitude of the energy conversion can be, for example, a current or a voltage, and according to the embodiment, a yoke (inductor).
Energy is used for this.

Output load power (RF
Still other purposes, such as maintaining power), are likewise achieved by the method according to claim 1 or by the control device according to claim 5. As a result, the advantage of the invention resides in that the amplitude of the intermediate circuit DC voltage supplied to the converter no longer enters the power output of the magnetron. The change or control or adjustment of the amplitude value of the chioke current i _max forms the sizing of the energy packet. The energy stored in the chioke is determined only as a function of the magnitude of the inductance and as a function of the chioke current i _max that is stored in the case of comparator disconnection. The amplitude of the chioke current can be varied and, as a result, the magnitude or level of the energy packet or amount of energy can be varied according to the present invention.

Yet another object of the present invention is to provide unit-to-unit magnetron unit-to-unit dispersion or matching from the same controller to various types of magnetrons. The same is achieved by the characteristics. The effect according to the invention lies in this case independently of the properties of the magnetron. The energy supplied to the magnetron is output by the magnetron considering the efficiency η, and the magnetron current or the magnetron voltage (load current, load voltage) can be set freely.

According to a convenient development, the load power can likewise be changed by changing the instantaneous separation of the energy packets. Cycle (1->2-> 1
-> 2 ...) Phases (1) and (2) which in this case do not directly follow one another periodically, but rather further phases (3) are inserted. here,

(1) is the accumulation of a specified amount of energy in, for example, a chain yoke,

(2) A load can be connected 2
Output or conversion of said energy quantity to the secondary side, and

(3) Idling, that is, the re-accumulation of a new amount of energy is also the output of the amount of energy from the primary side to the secondary side.

The energy output is changed by changing the time interval of phase (3) to cycle (1->2->3->1->).
2>->3-> 1 ...), that is to say the amount of energy stored and the output can be kept constant thereby.

According to an advantageous development, the method according to the invention or the control device according to the invention can be further supplemented by monitoring and / or protective means, which are the subject of the dependent claims.

The invention will now be described in detail with reference to exemplary embodiments.

[0020]

EXAMPLES FIGS. 1 and 2 have already been described in the introductory part of the specification.

The main element of FIG. 3 is a circulating coil or transformer Tra having a primary winding and a secondary winding. The primary winding has a number of turns w ₁ and the secondary winding has a number of turns w ₂ . Electronic switch S ₁ is disposed between the intermediate circuit DC voltage U ₁ that can be embodied as or Batsuteri as primary winding w ₁ and the rectifying and smoothing AC voltage of the transformer Tra. The electronic power switch (S ₁ ) can be formed from any type of electronic switch such as a power transistor, power MOS-FET, gate-off (GTO) thyristor and the like. It is turned on and off periodically and the switching frequency depends or is constant on the desired converted total energy. A power diode D ₁ is arranged between the secondary winding w ₂ and the connected load (magnetron). Furthermore, smoothing capacitor C
_s can be provided in parallel with the load. When the power switch S ₁ is connected to the primary side or is triggered (connection phase) to conduct, the secondary diode D ₁ is cut off. Specified or configurable maximum current i
_{When max} is reached, the primary power switch S ₁ is turned off. The main inductor L _h of the converter Tra has absorbed a defined amount of energy E ₀ during this on-phase.
The amount is

E ₀ = ½L ₁ · i _max ² (2)

L ₁ represents the primary chain inductor or the primary main inductor of the transformer.

In the circuit according to FIG. 3, the primary chain inductor is formed by the main inductor of the transformer Tra. The accumulation process is terminated by turning off S ₁ ,
Immediately after, the stored energy has the effect of rectifying the primary current to the secondary side of the transformer, the secondary diode D ₁ becomes conductive and the stored energy is output to the secondary side. This process (accumulation-conversion, accumulation-conversion) is f
If repeated (frequency f = 1 / T), the result is the converted power,

P _A = f · ½L ₁ · i _max ² · η = f · ½L _h · i _max ² · η (3)

According to FIGS. 4 and 5, i _max represents the maximum primary current.

By appropriate selection of the maximum primary side current i _max , the output power can thus be predetermined and achieved and is free of possible variations in U ₁ and / or load variations. With the smoothing capacitor C _s , smoothing of the energy packet output on the secondary side allows continuous energy output to the load, and further smoothing can be provided in the form of an LC low pass filter if desired. .

The frequencies at which the accumulation-conversion process is repeated are described above. The frequency is about 20KH
z, especially in the range beyond the audible limit,
It also drops at lower output load powers. This suggests two possibilities for energy control. Ie 1
Constant frequency while changing the maximum current i _max on the secondary side,
And a constant maximum primary current strength paired with a variable frequency. It is also possible to combine two influencing parameters so that the maximum primary current is changed in a certain range while the frequency f is changed in the other range or the frequency of the primary current and An operating range can be provided in which the amplitude is changed.

When the switch S ₁ is turned off, the transformer Tr
The primary energy carried by the main inductor of a and the primary current i _max will be completely transferred to the secondary side if the primary current disappears when switch S ₁ is turned on again. It If the remaining current i _p ≠ 0 is present when S ₁ is turned on again, only one part of the energy is 2
It is output to the next side. Such an operation is likewise possible, the output energy being in this case the energy stored in the transformer when switch S ₁ turns off and the energy still remaining in the transformer when switch S ₁ turns on again. Calculated from the energy difference between

FIG. 6 shows a further block diagram in which _two primary power switches S ₁ and S ₂ and two primary free running diodes D ₁ and D ₂ are provided. The primary winding of the transformer Tra is connected to the power switch and the free running diode in an asymmetric bridge. Switches S ₁ and S ₂ are controlled synchronously. Secondary side is Figure 3
Not changed by comparison with. The circuit according to FIG. 6 has the advantage that degaussing of the main inductor of the transformer is newly carried out on the primary and secondary sides. For example, if the load is damaged or unloaded, the secondary overvoltage will cause the energy packet to travel through the primary side free running D ₁ and D ₂ to the primary backup capacitor or battery (see FIG. 7). Can be avoided by supplying the current characteristics (see the current characteristics). Energy is thereby not output to the secondary side and undesired voltage increases can be avoided there. The device resists opening the circuit.

A pre-requisite for the design of the chain yoke L ₁ or main inductor L _h is to establish the basic operating parameters for characterisation.

U _{1 min} ... the minimum battery voltage that can be considered

U _{1 max} : maximum possible battery voltage

U _{2 min} ... Possible minimum load voltage

U _{2 max} : maximum possible load voltage

F ··· Pulse frequency

T ... Cycle

P _{A max} ... The required maximum allowable error irrespective of the predetermined output power

K ・ ・ ・ ・ Transformation ratio

The first step is to establish w ₁ / w ₂ = K.

U _{2 max} · K is less than or substantially less than U _{1 min} (U _{2 max} <or −U _{1 min} ), then K =
U _{1 min} / U _{2 max} (4)

This means that when U _{2 max} and U _{1 min} are predominant at the same time, the feedback is not yet done to the battery / buffer capacitor via D ₁ and D ₂ .

It is the maximum voltage distortion for diode D ₁ (U _Dmax ), ie

U _Dmax = 2 · U _{1 min} / K = 2 · U _{2 max} / U _{1 min} (5) is held.

Relationships that are determined separately, that is,

L ₁ = [η / (2 · f · P _{A max} )] · [U _{1 min} / (1 + U _{2 max} / U _{2 min} )] (6) is (not described in more detail here) L ₁ Held to establish.

0.9 can be estimated for the efficiency η (eta).

Furthermore, the chioke is dimensioned such that P _{A max} can be converted without saturation effects.
Ignoring the stray magnetic field, the required effective core capacitance V is

V = 2 · μ ₀ · μ _reff · P _{A max} / (η · f · B _max ² ) (7)

Here, μ _reff = μ _r / (1 + μ _r · l ₁ / l ₂ )

Μ ₀ ... 1.26 · 10 ^{−6 Tm} / A

Μ _r ··· Permeability of core material

Μ _reff ... about 50 to 1 depending on void
Effective relative permeability of 00

L ₁ ... Effective magnetic path length of core (m)

L ₂ ... void (m)

B _max ...- Maximum transmission induction T
(0.2T-0.3T)

Η (efficiency) ... efficiency

It is advantageous to choose V first based on a relatively large μ _reff , thus obtaining a reservation for larger voids and for P _{A max} .

The inductor is L ₁ = w ₁ ² · G _m ,

Here, G _m = μ ₀ · μ _reff · F / l ₁ (8)

G _m ... Magnetic permeability

F ... Effective magnetic cross section of core

G _m is determined using (7) and its parameters taken in this case or μ _reff
Known on the basis of. The number of turns w ₁ can thus be determined on the basis of (8). In this connection, the calculation should preferably be carried out with a reduced G _m · w ₁ ² . To be precise, it is possible in that case to set L ₁ precisely by empirical selection of the air gap.

In practice, the power supply unit is expected to have short circuit compatibility and open circuit protection. If the device comprises an overcurrent circuit in addition to open circuit protection, it is a short circuit guarantee.

To be precise, the short-circuit current is
Only rise in moderation and allow the pulse to turn off in time. If the output is opened, the output voltage will rise only until the diodes D ₂ and D ₃ are conducting on the primary side and the chijo energy will flow to the battery / buffer capacitor. The device thus also resists opening the circuit idling proof.

The embodiment of FIG. 8 is summarized as follows.

Main rectification

Storage chain with load circuit

Output stage with controller

Sequence control device and protection function

Main rectification

The mains voltage of 220 V rectified by G _L1 supplies the battery voltage U ₁ . + 10% and -1
In the case of a mains voltage tolerance regarded as a maximum of 5% and a maximum residual ripple of 20 V, U _{1 min} = 245
V and U _{1 max} = 345V are obtained.

Supply chain with load circuit

Specifications: U _{1 min} = 245V, U _{1 max}
= 345 V, U _{2 min} = 2500 V, U _{2 max} = 3
500V, f = 20KHz, P _{A max} = 1.2KW,
η = 0.9 (estimation or measurement)

According to (4), the result of K = 14 is obtained.

Applying (6), L ₁ ≈ 2.10 ^-4 H
= 0.2 mH is obtained.

The core capacity is based on the following assumption: μ _reff
= 150 (small void), B _max = 0.3 T, η = 0.
Calculated by applying (7) by 9.

Therefore, V = 280.000 mm ³

Corresponding core UU93 / 152/3
0 is a parameter, V '= 280.000 mm ³ , l ₁ =
It has 345 mm (magnetic path length) and F = 826 mm ² (magnetic cross section).

By applying (8), G _m = 4.
5 · 10 ⁻⁷ H / w ² is obtained.

Since w ₁ = 20, the result of L ₁ is 1.8 ·
Calculated value of 10 ⁻⁴ H. Actually 0.2 mH
A nominal value of 2.1 mm = 2 mm is almost achieved with an air gap of 2 mm.

With respect to the diode D ₁ of FIG. 6, the result of the retention voltage U _{D 1} max required according to (5) is U
_{D 1 max} ≈10 KV, and the filter capacitor C _s must be able to support 5000 V in its obvious position. These relationships are virtually infeasible, resulting in 12 (n) splits of the secondary winding. That is, w ₂ ″ = w ₂ / n = w ₁ / (K · 12)
Is. Throughout this specification, "" is a substitute for "★".

This result is U for diode D ₁ ″.
_{D1 max} = 820V, and the filter capacitor C _s ″ is loaded only by a maximum of 410V, C _s ″.
= To hold the 12 · C _s.

An additional LC element (L
_3, C ₃ ) can be achieved. The load current is thereby virtually continuous and the output power loses its pulsating nature.

Output stage with controller

It consists of an asymmetric bridge circuit, the relevant components of which are the field effect transistors T _r1, T _r1 ′ and also the diodes D ₂ , D ₃ . The control is handled by the device consisting of the driver Dri1, the transformers Trab, Trab 'and the drivers Dri2, Dri2'. The FET is controlled via an isolation controller (optically and magnetically).

Sequence control device and protection function

The main task of the sequence controller is _ip
= I _{p max} (P _A ) = i _max (P _A ) is achieved by periodically connecting and respectively opening the switching stages, for example at 20 KHz. i _p (P _A ) is the peak value at which the desired power flow P _A is achieved. The acquisition and release of current is performed by the resistor R ₁ and the comparator K ₁ . If i _p · R ₁ > U _comp then K ₁ is triggered. In this case, U _comp is U _comp = _{ip max}
It is set according to (P _A ).

The clock pulse generator is an AND gate G ₁
, A narrow pulse at 20 KHz applied to the. If the gate G ₁ is open, the pulse is routed to the RS flip-flop F via the setting input S, which has a clock pulse each.
Set F2. As a result, the switching stage is Dri1.
To turn on. When _{ip max} · R ₁ = U _comp ,
K _{1 In} each case, FF2 is turned off via the reset input R.

The "active" state is triggered by setting flip-flop FF1 by means of a start pulse. The operation is terminated by resetting FF1. This is the case when OR gate G ₂ reports Stop = H. This is done when the prescribed protection monitoring responds, or when the "end" is identified.

The sequence controller shown reproduces a schematic / block diagram. An integrated circuit can be used for this. Power switch / diode D ₂ , D ₃ ,
T _{r1 and} T _r1 'are separately provided in the embodiment. However, if the level of integration increases and the performance of integrated circuits increases, they can likewise be provided as integral elements or in the form of power modules.

[0091]

As described above, according to the present invention,
It has the following features and effects.

1. The energy storage chain (transformer, circulation coil) has two windings w ₁ and w ₂ . The electronic switch is arranged between the primary winding and the supply voltage. The half-wave rectifier is arranged between the secondary winding w ₂ and the load. The winding direction is chosen so that when the switch is conducting, the rectifier shuts off by polarity. (This corresponds to an isolation transformer)

2. The switch is periodically triggered to make and break. During the conduction phase, an "energy packet" (energy portion, amount of energy) is passed from the battery to the yoke (main inductor of the transformer). During the shutoff phase (off phase), the packet is fully output to the load. Power flow (P) is "energy packet"
(W) corresponds to the frequency to be multiplied (f) or P = W · f. This corresponds to the triangular operation of an isolation transformer with fixed frequency and superimposed triangular current characteristics. Also,
I _p> 0 or i _p DC component I ₀ is for all of t
Can be superposed to be approximately equal to.

3. The function of the above 2 is the conversion ratio K = w ₁ /
Guaranteed by all operating conditions as taken into account by appropriate sizing of w ₂ and the primary main inductor L _h .

4. The switch according to the above 1 is, for example, 2
It is realized by an asymmetric bridge (H bridge) consisting of one transistor T _r1, T _r1 ′ or S _1, S ₂ and two diodes D ₂ , D ₃ .

5. The secondary winding according to the above 1 is divided n times. Each split winding has its own rectifier D ₁ ″ and its own smoothing capacitor C _s ″. As a result, the breaking distortion for the secondary diode D ₁ ″ is reduced.

Furthermore, the insulation of the windings is simplified and the secondary storage capacitor can have low voltage endurance.

6. "Energy packet" according to 2 above
Is a measuring shunt R ₁ and a comparator K in order to determine on the primary side that a suitable current strength i _max is reached and in the course of which the conduction stage of the switches T _r1, T _r1 'is interrupted. Dimensioned by using ₁ .

7. The output voltage is smoothed by the output voltage (U ₂ )
In the case of higher demands, i.e. in the case of the desired small residual ripple, is handled by a further secondary LC low-pass filter. L
₃ / C ₃ has a capacitor C _s ″ with a triangular secondary current i
_It can be omitted if it is already large enough to sufficiently smooth ₂ (t).

[Brief description of drawings]

FIG. 1 is a characteristic diagram showing typical characteristics of a magnetron.

2 is a characteristic diagram showing current and voltage characteristics in the case of feeding a magnetron having the characteristics according to FIG. 1 with a rectified sinusoidal voltage half-wave.

FIG. 3 is a block diagram showing a single-ended isolated converter for use in an embodiment of the present invention.

4 is a characteristic diagram showing a primary current characteristic in the case of the circuit device according to FIG.

5 is a characteristic diagram showing the current characteristic of the circuit arrangement according to FIG. 3, in which the energy storage stage of the energy quantity is not directly connected at the end of the energy output stage.

FIG. 6 is a block diagram showing another modified embodiment of the present invention which similarly holds the current characteristics of FIGS. 4 and 5;

7 is a characteristic diagram showing the current characteristic of the circuit according to FIG. 6 in the case of secondary idling.

8 is a block diagram showing yet another embodiment of the invention according to FIG. 6 with detailed control electronics and additional monitoring and protection electronics.

[Explanation of symbols]

Tra transformer w ₁ primary winding w ₂ secondary winding S ₁ switch S ₂ switch C _s capacitor D ₁ diode L _h inductor T _r1 transistor

Claims (21)

[Claims] 1. An RF generator, especially an RF generator feeding method for feeding a magnetron, wherein the energy amount (E ₀ = 1 / 2L ₁ · i _max ² ) instantaneously separated is the primary main voltage. The RF generator is connected as a load from the side 2
The nominal load output (P _nom ) desired for the load, which is transmitted to the secondary high voltage side, is defined on the primary side, and is momentarily separated and transmitted (E ₀ = 1 / 2L ₁ · i).
The level (E ₀ ) of _{m ax} ² ) is the nominal load output (P _nom ) defined above and the efficiency of the entire device (η, eta).
The actual load output (P _A , P _act ) on the secondary side, which is varied as a function of
An RF generator feeding method, characterized in that it is defined in any way. 2. The energy level (E ₀ ) varies the maximum current (i _max ) of the inductors (L ₁ , L _h ) that periodically store and output the energy level (E ₀ ). The RF generator power feeding method according to claim 1, wherein the method is provided by: 3. The effective actual load output (P _A , P
_act ) is defined or varied by the level of the energy content (E ₀ ) and by the instantaneous separation (T, n / f) of the transmitted energy content (f · E ₀ = P _A ). The RF generator power feeding method according to claim 1 or 2, characterized in that. 4. The amount of energy (E ₀ ) is equal to 1 by a circulating coil or a transformer (T) in a potential-free manner.
Converted from the secondary side (w ₁ ) to the secondary side (w ₂ , w ₃ ″) and there one or more stages (C _s , C _s ″, L)
_3, C ₃ ) and each transformed energy quantity (E ₀ ) is subdivided into a plurality of symmetrical secondary windings ((w ₂ ″), said windings being loaded by load voltages (U _2, U _2). RF generator feeding method according to any one of the preceding claims, characterized in that they are connected in series to form _A ). 5. A primary device, preferably a rectifying and smoothing device for providing an intermediate circuit buffer DC voltage (U ₁ ), preferably a primary device for carrying out the method according to any one of the preceding claims. current (i _p) 1 primary current measuring device for outputting a proportional signal (U _p) in (R _1),
And at least one RF generator consisting of a transformer (Tra) having at least one primary and secondary side,
Especially in a magnetron controller, at least 1
One power switch (S _1, S ₂ , T _rl, T _rl ') and at least one primary free running diode (D ₁ , D
₂ ) a converter, in which case said at least one power switch (S _1, S ₂ , T _r1, T _r1 ′)
During the on-state, the inductor (L ₁ , L _h ) absorbs a definable amount of energy (E ₀ = ½L ₁ · i _max ² ) and the at least one power switch (S 1
_1, S ₂ , T _r1, T _r1 ') During the next turn-off, the amount of energy (E ₀ = 1 / 2L ₁ · i _max ² ) that can be defined to at least one secondary side of the transformer (Tra). ) And is fed to a comparator (K ₁ ) with a primary current proportional signal (U _p ) and a reference signal (P _nom ), and a control circuit (Tre1, Trab, Trab ', Tre).
2, Tre2 ′, FF2), at least one power switch (S _p ) when the current proportional signal (U _p ) overshoots the reference signal (P _nom ).
_A controller for an RF generator, which converts _1, S ₂ , T _r1, T _r1 ') from conduction (on) to interruption (off). 6. Two power switches (S _1, S ₂ ) and two free running diodes (D ₁ , D ₂ ) are provided, and the primary winding (w ₁ ) of the transformer (Tra) is asymmetrical. In the half-bridge circuit (H-circuit) method, the power switches (S _1, S ₂ ) and the free running diodes (D ₁ , D ₂ ) are connected, and the two power switches (S _1, S ₂ ) are connected. 6. The RF generator control device according to claim 5, wherein the RF generator control device is turned on or off in a relatively synchronous manner. 7. RF controller according to any one of the preceding claims, characterized in that the inductor (L ₁ ) is formed by the main inductor (L _h ) of the transformer (T). . 8. The at least one power switch (S _1, S ₂ , T _r1, T _r1 ′) comprises a control circuit (Dri1,
FF2, Trab, Trab ', Dri2, DrI
Through the 2 '), and characterized in that it is converted from the cutoff state when the primary current (i _p) or a current proportional signal (U _p) is the disappearance and / or a new timing clock pulses in a conducting state An RF generator controller as claimed in any one of the preceding claims, wherein 9. The secondary side of the transformer (Tra) has a plurality of symmetrical secondary windings (w ₂ ″), these secondary windings being:
6. RF controller according to claim 5, characterized in that it is connected in series by means of a DC voltage in order to achieve the desired high secondary output voltage (U ₂ ). 10. A diode (D ₁ ″) is connected in series with said plurality of symmetrical secondary windings (w ₂ ″), the anode of each diode (D ₁ ) being the respective secondary winding (w ₁ ).
Is connected to one terminal of the ₂ "), wherein its terminal least one Pawasuitsuchi (S _1, S _2, T
_r1, T _r1 ') has a negative voltage with respect to the other terminal of each secondary winding (w ₂ ″) during the on-time, and the secondary capacitor (C _s ″) is in each case respectively 10. RF controller according to claim 9, characterized in that it can be connected between the cathode of the diode (D ₁ ″) and the other terminal of the respective secondary winding (w ₂ ″). . 11. A diode (D ₁ ″) is connected in series with each of the plurality of symmetrical secondary windings (w ₂ ″), the cathode of each diode (D ₁ ) being the respective secondary winding (W ₁ ″). w ₂ ″), which is connected to said at least one power switch (S _1, S ₂ , T).
_r1, T _r1 ') of each of the secondary winding during the ON (w
₂ "has a positive voltage to the other terminal of) and secondary capacitor (C _s ') is the anode and the respective secondary windings of each of the diodes (D _1') in each case (w
₂ )) The control device for the RF generator according to claim 9, which can be connected to another terminal. 12. Each secondary capacitor (C _s ")
Are connected in series and the capacitor terminal with the highest potential difference and the capacitor terminal with the lowest potential difference can be supplied to the load as a supply voltage (U ₂ ). Generator controller. 13. A control device for an RF generator, characterized in that a further smoothing filter, preferably an LC filter (L _3, C ₃ ) is connected between the load and the secondary side of the transformer. . 14. The smoothed DC voltage (U ₁ ) of the intermediate circuit is generated from the feeder AC power supply by the rectifying and smoothing device, and the first power switch (S 1).
_1, T _r1 ), the primary winding (w ₁ ) of the transformer (Tra)
And a second power switch (S ₂ , T _r2 ) can be connected in series and connected to the smoothed DC voltage (U ₁ ) of the intermediate circuit, and the first free running diode (D ₂ ) Intermediate circuit DC voltage (U ₂ ) positive terminal, second power switch (S ₂ , T _r2 ) and the primary winding (w
₁ ) is provided between the connection point of the second terminal and the second
The free running diode (D ₃ ) is connected to the negative terminal of the intermediate circuit DC voltage (D ₁ ) and the first power switch (S ₁ T _r1 ).
Provided with between the first terminal connection point and the primary winding (w _1), and the two free running diodes (D _2,
D ₃ ) are the power switches (S _1, S ₂ , T)
7. The RF generator control device according to claim 5 _{, wherein r1 and} T _r2 ) are connected so as to be cut off when turned on. 15. Each of said at least one power switch (S _1, S ₂ , T _r1, T _r1 ') is composed of a plurality of symmetrical individual power switches of correspondingly reduced load carrying capacity. The control device for an RF generator according to claim 5, comprising a parallel circuit. 16. A memory circuit (FF2) in the RS flip-flop method which can be set by an ON signal (S), preferably a clock pulse (f, 1 / T) lying above the audible limit. ), And a gate circuit (G ₁ ) for coupling the error signal with the above-described clock pulse signal (f, 1 / T), said gate circuit (G ₁ ) being an error. The control device for an RF generator according to any one of the preceding claims, which outputs the ON signal for setting the flip-flop (FF2) when there is no message or clock pulse. 17. The RF generator controller of claim 16 wherein the error signal indicative of the error message comprises a separate coupling of individual error sources such as overcurrent, overtemperature, overvoltage and the like. 18. R according to claim 16 or 17, characterized in that the output of the comparator (K ₁ ) controls the reset input (R) of the flip-flop (FF2).
Control device for F generator. 19. The clock pulse source is equidistant pulses, according to claim 16 in which the width of the pulse is equal to or substantially less than the time interval so as to achieve the maximum primary current (i _{m ax)} Control device for RF generator. 20. The reference signal (P _nom ) defining the nominal load output (P _nom ) is output by a potentiometer, a voltage divider or an output defining circuit. A controller or method for an RF generator according to claim 1. 21. The nominal load output (P _nom ) is defined as a non-linear function of the primary side current amplitude (i _{m ax} ) and according to the overall efficiency (η, eta) of the device. 21. The controller or method for an RF generator of claim 20.