Untitled record

The invention relates to a device having a pulse-gated magnetron oscillator or other pulse-controlled oscillator and a load circuit to be fed by the oscillator, which only operates with good efficiency when the load represents a well-matched impedance.

For linear electron-accelerated wave accelerators using a pulse-gated magnetron oscillator, it takes a relatively long time for the resonant system to settle in. During this sweep, power is reflected off the magnetron, thereby affecting the operating frequency and / or the efficiency can be reduced. To achieve a good efficiency and a low detuning of Magnetro-ns must after. the magnetic resonance reflected power can be reduced, for example, characterized in that an energy absorbing resistor is provided between the magnetron and the resonance system. This results in a loss of 30 to 401 / o of the high-frequency power after completion of the transient process.

For linear accelerator-based accelerators, where feedback is used, it takes a long time for the accelerating field to reach its constant value. During this time, a portion of the 1-magnetron power is absorbed in the replica impedance of a fork feedback bridge, while nearly no power is reflected toward the magnetrome. In contrast to the energy absorbing impedance of the accelerating shaft based accelerator a, this replica impedance is nearly zero in power after the transient transient is completed.

The invention aims to provide a device of the type described above with an existing from the high-frequency system of an exploiting standing waves based electron accelerator coordinated load, in which the mentioned disadvantages are largely eliminated.

According to the invention, for this purpose, the oscillator output is coupled to the tuned load circuit via two hybrid circuits, the (first) hybrid circuit connected to the oscillator comprising one with a replica impedance. and the output arm connected to the (second) fork is a distribution fork, which is used to distribute the oscillator energy to the tuned load circuit and a feedback loop, the feedback loop an output of the fork with an input of the first fork combines.

The distribution fork has expediently a Eingangsarie and two output arms, the latter on each one. Connection circuit are connected to the matched load or the feedback circuit, while the connection circuit and the feedback circuit each include a phase shifter.

With a suitable design of Verteilerga, this z. B. from a T-connection. If. However, if very short shafts are used, the distribution fork can also consist of a waveguide with a coupling slot in its side, through which slot the feedback circuit is connected.

In a preferred embodiment, the splitter fork is comprised of a replica impedance hybrid circuit having an input arm and two output arms. Such a fork assembly consisting of a fork can, for. B. be designed as a ring fork.

The invention is explained in more detail below with reference to the drawings, wherein Figure 1 is a block diagram of an embodiment of a device according to the invention with as the first fork of a ring fork and as (second) distribution fork a T-connector, while Fig. 2 also in the block diagram a preferred embodiment shows, in which the two forks are designed as ring forks.

In Fig. 1, the hybrid circuit 1 consists of a ring fork with input arms 2 and 4, a post-training impedance 3 and. an output arm 5. The input arm 2 is a waveguide and connects to a pulse-gated Magn, otronosz, illa; gate 6 ago. The replica impedance 3 consists of a hollow conductor suitably selected length. The arm 5 is a waveguide which is connected to the input arm of a T-connection 7, whose two output arms, via a. Waveguide with the matched load 8 or via a ring consisting of a waveguide feedback circuit are connected to the input arm 4 of the hybrid circuit 1. The tuned stress 8 is the high frequency system of an exploiting wave based electron accelerator.

In the connected to the output arms of the T-connection waveguide or feedback loop is ever a phase shifter 9 or 10 is inserted. However, these phase shifters may consist of the waveguide itself, wherein the length of the waveguide in each particular case, the phase difference. determined between the voltages at both ends.

In steady state operation, with proper sizing of the phases and amplitudes, the voltages supplied by the arms 2 and 4 are added to the arm 5, while canceling at the replica impedance 3, so that practically no power losses occur. During commissioning, the control can be done by controlling the Phasen.-slide 10 (or by regulating the length of the corresponding waveguide).

During the settling of the tuned load 8, power is reflected from this load 8, whereby a part of this power is absorbed in the replica, impedance 3, while another part reaches the oscillator 6 through the arm 2. The amount of power entering the arm 2 in each individual system is minimized during commissioning by controlling the phase shifter 9 (or the corresponding length). At the end of the closed-loop control, at most three-quarters of the reflected power reaches the oscillator 6. This is sufficient to enable a quick settling to the correct operating frequency; However, damping applied to the heel of the reflected power is effective only during the transient process. After completion of the transient process no losses occur.

The best power distribution ratio of the ring fork 1 depends on the power distribution ratio of the T-connector 7.

In Fig. 2, the, corresponding to Fig. 1 parts are designated by the same reference numerals.

In this embodiment, the T-connector 7 is replaced by a second ring fork 11. The second ring fork 11 has an input arm 12, two output arms 13 and 15 and a replica impedance 14. The input arm 12 is connected to the output arm 5 of the ring fork 1, while the output arms 15 and 13 each via a waveguide with the matched load 8 and are connected via the feedback circuit with the input arm 4 of the ring fork l. As in Fig. 1 The waveguide for tuned load 8 and the feedback circuit each include a phase shifter 9 and 10, respectively. If desired, this waveguide and feedback loop can also be designed such that the desired phase shift is achieved by regulating their length. Again, in a steady state, the phase shift in the feedback loop 4, 10, 13 is such that virtually no power is absorbed by the replica impedance 3.

In this device, however, the transient can be considered as two stages. The first transient stage is in circuit 1, 5, 12, 11, 13, 10, 4, and in a typical case of a 10 cm circuit may take about 1 / 2o microseconds, with power being absorbed by replica impedance 3. The second transient stage is in the tuned load 8 and may take about 1/2 microsecond under the same conditions, absorbing power from the replica impedance 14.

In this embodiment, the phase shift in the path 15, 9, 8 can also be controlled so that at most three quarters of the power reflected by the tuned load 8 reaches the oscillator 6; this number can even be a ninth.

To achieve the best possible results, the power distribution ratios of the two ring forks 1 and 11 must be equal to each other.