DEVICE FOR MEASURING THE CHARACTERISTICS OF ALTERNATING AND PULSED MAGNETIC FIELDS

V. A. KASYANOV, A. G. SEDOY, M. M. SHTEIN

(Presented by the Scientific Research Institute
of Nuclear Physics, Electronics, and Automation)

The start-up of any accelerator is associated with carrying out a considerable volume of magnetic measurements. The betatron is no exception in this respect.

Of the entire variety of methods of magnetic measurements, for investigating betatron fields the induction method has become the most widespread; it makes it possible to study fields over a wide range of inductions with high accuracy, while the measuring apparatus itself is comparatively simple. Since the magnetic materials in

Fig. 1. Block diagram of the apparatus for magnetic measurements

the electromagnets of modern betatrons are loaded to the maximum and the field distribution in the accelerator magnet gap may change with increasing induction, in many cases it is necessary to know the field dynamics during the acceleration cycle.

It should be noted that almost always it is necessary to know only the relative characteristics of the betatron control field. In this case, a null method of measurement can be successfully applied, in which the measured field is compensated by part of the reference field.

The block diagram of the apparatus for investigating alternating magnetic fields, developed by the authors, is shown in Fig. 1.

Let us consider the operation of the apparatus when measuring the relative magnitude of the field. The reference sensor \(SW_2\) is fixed immovably in the field under study, while the measuring sensor can be moved in the field. The difference signal from the sensors is fed to a compensation circuit consisting of \(R_0\), \(R_n\), \(R\), and then to the input of the integrator. The output signal of the integrator is supplied to a comparison circuit, which determines the ratio between the instantaneous values of the integrator output voltage at times corresponding to the zero and the level of the reference field at which the measurements are made.

A built-in pointer indicator is used as the recording instrument. The measurement process consists in bringing the comparison circuit to zero by changing \(R_n\). Knowing \(R_n\) and \(R\), it is easy to determine the ratio

\[ \frac{B_{\mathrm{ref}}\cdot SW_2}{B_{\mathrm{meas}}\cdot SW_1} \]

at a specified instant of time. Here \(B_{\mathrm{ref}}\) and \(B_{\mathrm{meas}}\) are, respectively, the instantaneous values of the induction of the reference and measured fields, and \(SW_2\) and \(SW_1\) are the flux linkages of the corresponding coils.

It should be noted that in similar previously known instruments the function of the comparison device was performed by an oscilloscope, on whose screen the curve of the integrator output voltage was reproduced [1].

The comparison device proposed by the authors makes it possible to minimize the effect of integrator drift on the measurement error and makes operation with the instruments more convenient. In addition, this circuit solution also provides measurement of absolute field values. In this case the sensor is a calibrated coil, and, together with the integrator output voltage, an adjustable reference voltage \(U_{\mathrm{ref}}\) is fed to the comparison circuit. The measurement process consists in bringing the comparison circuit to zero by varying \(U_{\mathrm{ref}}\). Naturally, in this case the zero drift of the integrator has a substantial effect on the accuracy of the measurements.

To set the measurement instant, the instrument has a delay unit, which includes a four-decade decimal summing counter \(CCh\), a counting-pulse generator \(GSI\) at a frequency of \(1\) MHz, a control trigger \(Tg_u\), and pulse shapers \(\Phi1\), \(\Phi2\), \(\Phi3\), \(U\Phi\).

The delay unit can operate in two modes:

1. The delay is counted from the instant of time corresponding to the passage of the field through zero.

2. The delay is counted from the instant of arrival of an external synchronizing pulse.

The indicator of the field crossing zero (“zero sensor”) is a peak-transformer sensor (peaker). In the initial position \(Tg_u\) is in the zero state, and \(GSI\) is not operating. When the delay unit operates in the first mode, the peaker pulse switches \(Tg_u\) to the “1” state. Pulses from \(GSI\) are fed to the counter. When the required delay is reached, the amplifier-shaper \(U\Phi\) produces a marker pulse. This pulse blanks the counter. \(Tg_u\) is reset to “0.”

In the second mode the unit operates similarly, but \(Tg_u\) is switched to “1” at the instant of arrival of an external synchronizing pulse. The instrument provides for the possibility of using an oscilloscope as the comparison and recording device; for this purpose there is a mixer, to one of whose inputs the integrator output is connected, while marker pulses from \(\Phi3\) and \(\Phi4\) are fed to the other two inputs.

Fig. 2. Schematic circuit of the integrator and comparison device

The instrument is made entirely with semiconductor devices. Switches Kl1 and Kl2 are assembled using KP301B field-effect transistors and have a resistance in the open state of not less than \(10^9\ \Omega\), which made it possible to obtain a very short time for holding the instantaneous value of the integrator output voltage (\(1.5\ \mu\mathrm{s}\)).

In designing the instrument, special attention was paid to the development of the integrator dc amplifier (DCA). By using integrated circuits, it was possible to create a comparatively simple and reliable DCA, the circuit of which is shown in Fig. 2. The DCA consists of a main amplifier, for which a type 1UT401B integrated amplifier with a gain of 2000 is used, and an automatic zero-adjustment circuit, which in turn consists of a circuit for summing the input and part of the output signals (\(R_{61}, R_{64}, R_{66}\)) and an M–DM channel. The M–DM channel includes an ac-voltage amplifier on a 1UT401A integrated circuit, with a modulator and demodulator based on RES-49 relays.

The relay switching frequency was chosen to be 20 Hz. Owing to synchronous switching of the modulator and demodulator and a reduction in the on-state time of the relay, it was possible to extend the passband of the M–DM channel to 5 Hz; moreover, with the feedback loop closed, the ripple at the integrator output does not exceed 3 mV.

An emitter follower is connected to the output of the main amplifier in order to reduce the influence of interference arising during commutation of the switches in the comparison circuit.

Technical characteristics of the instrument

1. Range of measured fields: \(0.02 \div 2\ \mathrm{T}\).
2. Error in measuring the radial field falloff index \(n\), introduced by the integrator and the delay device: \(\le 0.1\%\).
3. Accuracy of measuring the absolute value of the field induction: \(\le 1\%\).
4. Delay range: \(0 \div 10000\ \mu\mathrm{s}\).
5. Delay-time setting discreteness: \(1\ \mu\mathrm{s}\).
6. Weight: \(\approx 8\ \mathrm{kg}\).
7. Overall dimensions of the instrument: \(480 \times 400 \times 230\ \mathrm{mm}\).

Fig. 3. 1 — phase of the supply voltage \(\Phi = 18^\circ\); 2 — \(\Phi = 72^\circ\); 3 — \(\Phi = 90^\circ\).

The instrument was manufactured at the NIIYaF EA at TPI and is being successfully used for magnetic measurements in betatrons. With its aid, in particular, the field dynamics in a betatron with pulsed power supply at \(6\ \mathrm{MeV}\) was measured, and the cause of electron “self-dumping” in the production PMB-6 betatron was also determined. As an example, Fig. 3 shows the pattern of variation of \(n\) with the phase of the supply voltage for the PMB-6 betatron.

References

1. A. G. Alekseev, L. V. Vaulin et al. Methods and apparatus for measuring the magnetic field of the Yerevan synchrotron. Proceedings of the All-Union Conference on Charged-Particle Accelerators. Moscow, Atomizdat, 1970.