Equipment for Simultaneous Recording of Betatron Pulse Parameters

V. A. Moskalev, Yu. A. Gromov

The intensity of betatron radiation has a fast and a slow component of amplitude variation in time [1, 3]. The slow component is caused mainly by fluctuations of the line voltage supplying the betatron electromagnet, and hence by the departure from the optimum conditions for injection and capture of electrons into acceleration. The fast component (variation from cycle to cycle) makes a substantial contribution to the stability of the radiation, but the reasons for its occurrence are unclear. It is unknown whether it is determined entirely by the operating input pulse and nonpulse parameters of the betatron (injection voltage, inflection, injection phase, current in the bias winding, voltage on the magnet, filament current, etc.) or by its very design (vibration of the piker [3], inflector, interaction of the electron beam with the chamber walls [2], etc.). It is therefore of interest to establish the interrelation of the parameters with one another over shorter time intervals, i.e., during the acceleration cycle. However, because of the large duty factor of the betatron’s operating pulse quantities, it is not possible to use any standard recording instruments directly.

This problem has been solved with the aid of the equipment proposed by us, which makes it possible to record simultaneously, from cycle to cycle, the amplitudes and durations of pulses, as well as short time intervals between various pulse quantities. The number of channels is unlimited.

The block diagram of the equipment is shown in Fig. 1.

When recording amplitudes, carried out by channel \(A\), the recorded pulses are applied to the input of a “change amplifier,” where, depending on the amplitude, polarity, and magnitude of oscillations in time, they are limited and amplified. The pulses are then increased in duration, with preservation of amplitude, by “expanders” and are fed further to a loop oscillograph. The durations of the expanded pulses are brought to a value \(t_{\text{exp}} \div 1000\ \mu\text{sec}\) greater than the time constant \(\tau\) of the vibrators used (type N-135-1 ÷ IV).

As an example, Fig. 2 presents a schematic diagram of channel \(A\) when recording oscillations of the amplitude of the injection-voltage pulse. Since the injection voltage is taken from a divider and in this case has a value on the order of hundreds of volts, the pulse is not preamplified, but is directly limited by the diode circuit of the limiter,

performed with diode \(D_1\). By pulses limited in amplitude by the charging circuit, assembled according to a common-collector circuit on transistor \(T_1\), through diode \(D_2\) the memory capacitor \(C\) is charged to the amplitude value; from this capacitor, through the readout circuit, also assembled on transistor

Fig. 1. Block diagram of the device

\(T_2\) according to a common-collector circuit to provide a high input resistance, the voltage level is taken off with small leakage. The pulsed voltage is expanded into a “quasi-constant” one [4]. At the same time a monostable multivibrator (\(T_4\) and \(T_5\)) is triggered; through the discharge circuit (\(T_3\)), by the trailing edge of the pulse formed after \(1000\ \mu\text{s}\), it discharges capacitor \(C\).

Fig. 2. Schematic diagram of a channel for recording changes in the amplitude of the injection voltage pulse

The voltage taken from the emitter resistor of transistor \(T_2\) cannot be fed directly to the vibrator loop, since the resistance of the loop varies within the range \(1\text{–}15\ \Omega\), and it would disturb the normal operating mode of transistor \(T_2\). To match the low resistance of the loop with the comparatively high resistance of the expander, a balanced circuit implemented on transistors \(T_6\) and \(T_7\) is included.

Circuits for recording the amplitudes of the inflection-voltage pulses, the current from the layer, the current in the bias winding, the radiation intensity, etc., differ from the circuit for recording oscillations of the injection-voltage pulse only in the presence or absence of limiters, preliminary and main amplifiers, and in the limiting polarity.

When recording duration oscillations (channel B), the recorded pulses are normalized in amplitude by a Schmitt trigger, and changes in the duration of the resulting pulses are converted into proportional changes in the amplitude of the “sawtooth” by a linearly varying voltage generator; that is, the pulse duration is converted into amplitude, \(t_{\mathrm{imp}} \to A\) (Fig. 1). The resulting sawtooth pulses are limited in order to isolate the “changes,” are expanded by the method described above, and are fed to the loop oscillograph.

Fig. 3. Part of an oscillogram: I — mains voltage \(U_{\text{mains}}\) at the level of 98%; II — magnet voltage \(U_M\) at the level of 98%; III — injection voltage \(U_i\) at the level of 96%; IV — inflection voltage \(U_f\) at the level of 95%; V — current from the layer \(I_{\text{layer}}\) at the level of 96%; VI — amplitudes of the intensity of \(\gamma\)-radiation at the level of 60%.

Recording of short time intervals between pulses (channel C) is carried out in the same manner as for durations, with the only difference that a two-cycle trigger is connected instead of a Schmitt trigger.

The apparatus also makes it possible, simultaneously with pulsed quantities (injection voltage, inflection voltage, etc.), to record non-pulsed quantities (filament current, etc.), thereby establishing the relation between them.

If the duration of the recorded pulse is greater than \(t_{\text{exp}} = 1000\ \mu\text{s}\), the expander is excluded from the circuit. The main technical characteristics and capabilities of the apparatus developed are as follows:

1. The amplitude of the input recorded pulses with a duration of \(0.1\)—\(100\ \mu\text{s}\) may be \(0\)—\(100\ \text{V}\).

2. The recording limits for durations and time intervals are \(0.1\)—\(100\ \mu\text{s}\).

3. The maximum error is determined mainly by the diameter of the focused light beam on the photographic film and is estimated at 2.5%.

1. The maximum repetition frequency of the recorded pulses is 1000 Hz.

2. Using an eight-channel loop oscillograph, with normal recording resolution, a photographic-film transport speed of 2.5 cm/sec, and a photographic-film reserve of 5 m, it is possible to record, on each channel, up to \(10^4\) pulses following at a frequency of 50 Hz.

Figure 3 presents part of an oscillogram of the pulse parameters of the high-current betatron “Luch,” obtained with the oscillograph at a photographic-film transport speed of 2.5 cm/sec.

1. The apparatus described makes it possible to document simultaneously up to 8 or more processes that are important from the standpoint of investigating the causes of radiation instability not only in a betatron but also in other accelerators.

The results of the investigations being carried out will be published.

REFERENCES

1. V. M. Razin. Dissertation. Tomsk, 1954.
2. V. A. Moskalev, V. G. Shestakov, et al. Proceedings of the Fourth Interuniversity Conference on Electron Accelerators, 1961, pp. 204–209.
3. N. I. Balamatov, B. I. Goryachev. Proceedings of the Fifth Interuniversity Conference on Electron Accelerators, 1964, pp. 182–186.
4. Z. V. Magrachev. Voltmeters for Single Pulses. Moscow, “Energiya,” 1967.