APPLICATION OF LINEAR SWEEP IN PULSE PHASEMETER CIRCUITS

I. D. Zolotarev, A. N. Kolesov, G. B. Grigoriev

(Presented by the scientific seminar of the Faculty of Radio Engineering)

For scientific research it is often necessary to carry out continuous photographic recording of oscillograms of the process under study on moving film.

In this case, the use of a phasemeter with an oscillographic (linear) sweep can ensure minimal consumption of photographic materials and greater convenience in processing the results of observations.

Methods for using a sawtooth voltage to create a linear sweep when measuring the phase shift between harmonic signals have been discussed in sufficient detail in the well-known literature [1, 2].

However, if it is necessary to measure the phase shift between high-frequency fillings of radar signals, a number of the following difficulties arise in developing a phasemeter circuit:

a) it is necessary to create a circuit for a fast-sweep generator that provides (because of the short duration of the signal pulses) a high repetition frequency of the sawtooth-voltage pulses (in our case \(2 \cdot 10^6\) pulses/sec);

b) it is necessary that the phasemeter ensure measurement of the phase shift between weak signals comparable with noise.

If the signals whose phase shift must be measured exist for only a short time (for example, for meteor reflections the duration of the signal is on the average of the order of fractions of a second), additional complications are encountered in creating the phasemeter. In these cases, sufficient image brightness on the indicator screens can be achieved by using special high-voltage cathode-ray tubes (13LO2S, 13LO7A, etc.). The low sensitivity of these cathode-ray tubes leads to the need to generate deflecting voltages of large amplitude.

As a result of the experimental work carried out, a phasemeter circuit with a linear sweep was developed that satisfies the stated requirements.

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

The signals are applied to the input of the two-channel amplifying device 1. From the output of both amplifier channels the signals are fed to the

special circuits for forming short pulses for triggering the sweep 2 and the phase mark 4.

A description of the operation and a diagram of the shaping channels are given in [3].

The time of appearance of the pulses triggering the fast sweep and of the phase-mark pulses is rigidly tied to the phases of the corresponding high-frequency fillings of the pulse signals at the input of the amplifying device.

The formed train of positive sweep-triggering pulses, with a duration of the order of \(0.5 \cdot 10^{-7}\) sec and a repetition period of \(0.5 \cdot 10^{-6}\) sec, is fed to the input of sweep generator 3.

The train consists of 10 pulses when the duration of the received-signal pulse is \(\tau = 5\) μsec. The sawtooth voltage from the sweep generator is applied to the horizontal-deflection plates of the cathode-ray tube.

Phase-mark pulses of negative polarity, with a duration of the order of \(10^{-8}\) sec, are applied to the cathode of the cathode-ray tube of the phasemeter indicators, thereby causing the phase mark to appear on the sweep line.

Fig. 1. Block diagram of a phasemeter with linear sweep: 1—two-channel amplifying device; 2—4—circuits for forming the sweep-triggering and phase-mark pulses; 3—sweep generator.

The schematic circuit of the fast-sweep generator is shown in Fig. 2.

The principal element of the circuit is a sawtooth-pulse generator built around tube \(L_1\) (GMI-83). In the standby state, tube \(L_1\) is cut off at the control grid, and a voltage \(U_{a0}\), equal to 900–1000 V and determined by the position of the slider of potentiometer \(R_{14}\), is established at its anode. Diode \(L_2\) (6Ts5) is open at this time, and through it the stabilizer, built around tubes \(L_4\) (6N8) and \(L_5\) (6N5S), is supplied.

When triggering pulses arrive at input \(\Phi_1\) of the sweep generator, tube \(L_1\) is unblocked. A rapid discharge of the total capacitance in the anode circuit of the tube takes place. The anode potential decreases to \(U_{a0} = 200\) V. After the end of the triggering pulse, the capacitance in the anode circuit of tube \(L_1\) charges through resistances \(R_4\) and \(R_5\). The voltage rises to \(U_{a0}\), after which diode \(L_2\) opens, preventing any further rise of the voltage. The use, as the sweep voltage, of a fixed initial portion of the exponential makes it possible to obtain a sufficiently small change in scale along the sweep line on the screen of the phasemeter indicator.

During the formation of the sawtooth voltage, diode \(L_2\) disconnects the stabilizer from the \(+3000\) V power supply. Capacitance \(C_6\) prevents a change in the cathode potential of \(L_2\) during the action of the train of triggering pulses.

In order to ensure good operating quality of the phasemeter, it is necessary that the sawtooth sweep voltage retain its

Fig. 2. Schematic circuit diagram of the fast-sweep generator.

Visible annotations in the diagram include: “Trigger pulse,” “Single sweep to the cathode-ray tube,” and supply voltages \(-300\ \mathrm{V}\), \(+250\ \mathrm{V}\), \(+150\ \mathrm{V}\), \(+10\ \mathrm{kV}\).

a constant amplitude value and a constant shape, and also so that the position of the sweep line on the phasemeter indicator screen does not shift with time. The stated requirements are met by introducing additional elements into the sweep-generator circuit and by selecting the optimum operating mode of the circuit, namely:

1) The bias on the control grid of tube \(L_1\), which determines the discharge current of the total capacitance in the anode circuit of the tube, is selected by setting potentiometer \(R_2\) so that during the action of the triggering pulse a saturated mode of tube \(L_1\) is attained. In this case the minimum voltage at the anode of tube \(L_1\) \((U_{\alpha min})\) will depend to a lesser degree on fluctuations in the amplitude of the triggering pulses.

2) There exists an optimum value of the charging capacitance, which is determined from the following considerations (Fig. 3). With \(C_{\text{ch}} > C_{\text{ch.opt}}\), the amplitude value of the sweep fluctuation is less than \(U_{\alpha 0}\), and therefore the influence of destabilizing factors on the sawtooth amplitude is not eliminated; with \(C_{\text{ch}} < C_{\text{ch.opt}}\), the flyback time increases.

Fig. 3. Voltage waveforms in the sweep-generator circuit.

By adjusting the semi-variable capacitor \(C_3\), the value of the charging capacitance is chosen so that the instant at which the linearly increasing voltage reaches the value \(U_{\alpha 0}\) somewhat precedes (by the amount \(t_3\)) the instant at which the next triggering pulse arrives at the input of the sweep generator.

3) Constancy of the position of the sweep line on the phasemeter indicator screen under real conditions (when there is a noisy signal at the phasemeter inputs) is ensured by using the restoring diode \(L_3\) (2Ts2S).

4) Stabilization of the shape of the sweep voltage is achieved by using a stabilized \(+3000\ \text{V}\) source.

5) An electronic stabilizer for fixing the amplitude of the sawtooth voltage is assembled on tubes \(L_4\) and \(L_5\). In the standby state of the sweep-generator circuit, the stabilizer maintains a constant voltage \(U_{\alpha 0}\) at the anode of tube \(L_1\). The stabilizer is supplied from voltage sources of \(+3000\ \text{V}\) and \(+750\ \text{V}\). A stabilized voltage of \(-250\ \text{V}\) is used as the reference. The operation of the stabilizer circuit is conventional and requires no special explanation.

To check the operation of the sweep-generator circuit, radio pulses of duration \(\tau = 5\ \mu\text{s}\), with high-frequency filling frequency \(f_0 = 2\ \text{MHz}\), are applied to the input of the sweep-trigger pulse-forming circuit. Simultaneously, to the modulating electrode of the electron-beam—

...a synchronous sinusoidal voltage of frequency 48 MHz is applied to the beam tube of the phasemeter.

When the sweep generator operates properly, the calibration marks for each sweep period are superimposed on one another, producing a clear calibration pattern.

A photograph of the phasemeter indicator screen in the calibration mode is shown in Fig. 4b.

Fig. 4. Photograph from the screen of the phasemeter indicator:
a) appearance of the phase mark on the phasemeter indicator screen; signal pulse duration \(\tau = 5\ \mu\text{s}\); high-frequency filling frequency \(f_i = 2\ \text{MHz}\);
b) calibration pattern; calibration-voltage frequency \(f = 48\ \text{MHz}\);
c) appearance of the phase mark for a real signal at

\[ \frac{U_c}{U_{\text{noise}}} \approx 5. \]

The phase mark for a pulsed signal (without noise) is presented in the photograph in Fig. 4a.

The character of the phase mark of a real signal at a signal-to-noise ratio of 5 is shown in Fig. 4c.

As follows from the figures presented and from experiment, the probable error in measuring the phase shift with linear-sweep phasemeters is:

a) in the pulsed mode, for a signal pulse duration \(\tau = 5\ \mu\text{s}\), \(\Delta \varphi \approx 3^\circ\);

b) in the pulsed mode, for the same signal pulse duration and a signal-to-noise ratio of 5, \(\Delta \varphi \approx 4 \div 5^\circ\).

REFERENCES

1. Aseev B. P., Phase Relations in Radio Engineering, Svyazizdat, 1953.

2. The Modern Cathode-Ray Oscilloscope, Vol. III (collection of articles), Foreign Literature Publishing House.

3. Zolotarev I. D., Denisov V. P., “Measurement of Phase Difference in the Pulsed Mode by the Circular-Sweep Method” (published in the present collection).