On Some Directions for Further Automation of Betatrons

V. M. Razin, N. Ya. Makarov

(Presented by the scientific seminar of the Faculty of Automation and Computer Engineering)

The wide introduction of betatrons into industry has posed a number of new tasks for their further improvement, for the development of new operating modes of accelerators and, in particular, for a more complete automation of the processes of electron acceleration.

The principal process that to a considerable degree determines the efficiency of betatron operation—the magnitude of the obtained current of high-energy electrons and the radiation intensity associated with it—is the process of capturing electrons into acceleration.

Below we consider certain questions of the further automation of this part of the electron-acceleration process.

The main control parameter is the injection phase, i.e., the moment at which electrons are supplied to the betatron chamber, while the controlled parameter is the intensity of gamma radiation [1]. In the nature of its operation, the betatron is a discrete automatic system with time-pulse (phase-pulse) modulation of the input parameter. With the above-mentioned choice of input and output parameters, the betatron behaves as an inertia-free controlled object.

One of the principal operating modes of the accelerator is the mode of maintaining, with high accuracy, the radiation intensity at a certain specified level.

Analysis of the operation of the intensity-stabilization systems used on betatrons shows that, to some extent, they do not satisfy the requirements of production. The accuracy of maintaining a constant value of intensity at a level of 0.7 of the maximum fluctuates around 2%. These parameters of the stabilization system are not optimal.

The principal destabilizing factor (external disturbing influence) is the fluctuation of the supply voltage [2], even slight changes of which lead to a sharp change in the optimal conditions for capturing electrons into acceleration. Under these conditions, combined systems may be the most effective, using as operating information both the deviation signals of the controlled quantity and the signals of disturbing influences. Such control systems for betatrons can be developed according to one of the principles considered in [2].

Owing to the increased amount of operating information in combined stabilization systems, under corresponding conditions a higher control accuracy is obtained [3].

In the course of operation of a betatron it is often necessary to obtain the maximum radiation intensity. This can be done most effectively by automatic means. The injection phase is also chosen as the regulating parameter. The dependence of the betatron radiation intensity on the injection phase has a sharply expressed extremal character and, in the course of operation, changes according to a random law.

Under these conditions, the problem of attaining the maximum productivity of the accelerator can be solved with the aid of self-adjusting automatic control systems (ACS).

The necessity of implementing an extremal-control mode follows from the very nature of the operation of the object. The radiation intensity depends on many parameters [2], which are continuously subjected to external influences having a completely irregular character of variation in time. Ensuring the necessary noise immunity of the control system is the principal factor determining the speed of response and the effectiveness of the system.

An amplitude statistical analysis of pulses from the radiation sensor (photoelectron multiplier, PEM), carried out for a 30-MeV betatron, shows a change in intensity from cycle to cycle of about 14% of the mean value of the pulse amplitude.

The random character of such fluctuations makes it possible to regard them as interference arriving, together with the useful signal, at the input of the ACS.

Under these conditions, more effective regulation, as in the intensity-stabilization mode, can be provided by combined ACSs, which make it possible in the best way to achieve independence of the regulated quantity from external conditions [3].

Figure 1 shows a functional diagram of one of the possible variants of a combined ACS for the radiation intensity of a betatron.

Fig. 1. Functional diagram of a combined system of extremal regulation of betatron radiation intensity.
1—controlled object—betatron; 2—measurement sensor—PEM; 3—switch; 4—command device; 5—search circuit; 6—tracking circuit; 7—sawtooth-voltage generator; 8—functional converter; 9—disturbance-measurement sensor; 10—compensation control device; \(I\)—radiation intensity; \(\varphi\)—injection phase.

Fig. 2. Structural diagram.

Command device 4 starts the sawtooth-voltage generator 7. A signal proportional to the magnitude of the radiation intensity \(I\), obtained from the FEU sensor 2, is fed through switch 3 to the search circuit 5. The maximum value \(I\) is recorded by a special “memory” device. In the tracking mode, after passing through the maximum, the current value of \(I\) is compared with its extremal value, and, when the difference \(\delta = I_{\max} - I\) reaches a certain value—\(\delta_{\max}\)—the direction of change of \(I\) is reversed and the memorization is removed. Thereafter the process is repeated. Sensor 9 and control device 10, through functional converter 8, make it possible to carry out effective compensation for changes in the external action.

Analytical calculation of extremal-control systems, especially with allowance for the influence of noise, presents considerable difficulties. However, the use of computing devices for the calculations makes it possible to investigate, comparatively fully and in a short time, problems of optimizing the system parameters already at the first stage of design [4].

Figure 2 presents the block diagram of a model of one of the possible variants of extremal control of intensity, developed on the basis of the MN-7 analog electronic model.

The averaged characteristic of the object, which reflects its dynamic properties in the model, is represented by individual linear segments. The approximation error does not exceed 5%; in the region of the maximum it is no more than 1%.

The channels being compared are modeled by two functional amplifiers 4 and 12. The electrical pulses in this circuit are obtained by means of relay \(P_0\) at the output of amplifier 17, controlled by a voltage of 100 V through contact 1 “\(e u\)” of the model.

The comparison detector is identical in form to the real detector, but has constant charge and discharge times for capacitors \(C_1\) and \(C_2\), matched to the time scale.

The difference voltage from the detector output is integrated by amplifier 5, which directly models the regulator integrator.

Amplifiers 9, 11, 2, and 3 are necessary for providing feedback to the controlled object.

At the input of the summing amplifier 9, along with the useful signal from amplifier 5, the noise voltage from the normal-noise generator is also applied.

The use of the described electronic model makes it possible, in the presence of random noise, to investigate the influence of individual regulator parameters on the search time, and also on the accuracy with which the maximum radiation intensity of the betatron is maintained.

REFERENCES

1. L. M. Anan’ev, A. A. Vorob’ev, V. I. Gorbunov. Induction Electron Accelerator—Betatron, Atomizdat, 1961.

2. V. M. Razin. Stabilization of betatron radiation, Proceedings of the Tomsk Polytechnic Institute, vol. 87, 206–215, 1957.

3. A. G. Ivakhnenko. Elements of the general theory of combined cybernetic systems, Avtomatika, No. 1, 1958.

4. B. Ya. Kogan. Electronic Modeling Devices and Their Application to the Study of Automatic-Control Systems, Fizmatgiz, 1959.