Grid Control of a Pulsed Three-Phase Full-Wave Rectifier with Discontinuous Valve Operation

V. V. IVASHIN
Submitted 1969 | SovietRxiv: ru-196901.09547 | Translated from Russian

Abstract Generated abstract

This paper addresses grid-pulse control for a pulsed three-phase full-wave thyratron rectifier operating with discontinuous valve current in the power supply of the Sirius synchrotron. It describes a pulse-forming circuit using an induction phase regulator, a six-secondary transformer, saturating reactors, resistors, grid transformers, and control thyratrons to produce steep firing pulses of at least 60 electrical degrees with equal spacing during pulsed operation. The article derives approximate relations for selecting saturating-reactor parameters and for adjusting pulse timing through winding turns, with attention to phase unbalance and valve load distribution. The proposed system is reported to have operated for an extended period in pulsed service and to simplify regulation and maintenance of the rectifier.

Full Text

GRID CONTROL OF A PULSED THREE-PHASE FULL-WAVE RECTIFIER WITH DISCONTINUOUS VALVE OPERATION

V. V. IVASHIN

The rectifier charging the storage capacitance of the power-supply system of the “Sirius” synchrotron at \(1.5\ \text{GeV}\) is assembled on TR1-85/15 thyratrons according to the Larionov circuit. The rectifier is switched on for \(0.7\) seconds every second and, during accelerator tuning, must operate at large control angles. The rectifier valves operate in a discontinuous-current mode. In this case the firing pulses must satisfy the following conditions.

  1. The pulse duration must be not less than \(60^\circ\).
  2. All pulses must be shifted by equal time intervals relative to one another and must have steep fronts.

The second condition proved especially difficult to fulfill because the rectifier operates in a pulsed mode. The transient processes that arose in the pulse-forming circuit disrupted the rectifier operating mode. The first of the pulses being formed arrived at the rectifier valve several degrees early, which led to overloading of the valve by current and to operation of the overcurrent protection.

Figure 1a shows a circuit for forming firing pulses that satisfies the above requirements. The circuit consists of an induction phase regulator 1, a three-phase transformer 2 with six secondary windings, saturating reactors 3, resistors 4, grid transformers 5, and control thyratrons 6. The six-phase system of voltages on the secondary windings of transformer 2 is converted, by means of reactors 3 and resistors 4, into pulses shifted relative to one another by \(60^\circ\). These pulses are formed across resistors 4 and can be transmitted through grid transformers 5 into the grid circuits of the rectifier valves when the control thyratrons 6 are open.

The connection, in parallel with resistors 4, of chains consisting of series-connected primary windings of transformers 5 and valves 6 makes it possible to fulfill the second requirement.

Pulses can be transmitted to the rectifier only when thyratrons 6 are open, which makes it easy to implement the control circuit for a rectifier with pulsed operation. The pulses transmitted to the rectifier cannot lag behind one another by more than \(60^\circ\). This is explained by the remagnetization of transformers 5 only over partial

cycles, and also because pulses are formed continuously at resistors 4, irrespective of the operation of the rectifier.

Grounding the cathode circuits of the control thyratrons makes it possible to make the control system simple and easy to regulate.

As a rule, the parameters of individual manufactured saturation reactors always differ somewhat. The voltages induced in the six secondary windings of transformer 2 may also be unequal, for example because of phase unbalance of the supply voltage, which is often encountered in practice. All this affects the time displacement between the pulses being formed and the uniformity of loading of the valves of the controlled rectifier.

Fig. 1

Fig. 1. a—circuit for forming grid pulses; b—formation of pulses with a steep front from a sinusoidal voltage curve

For calculating the parameters of reactor 3, the voltage drop across active resistance 4 may, to a first approximation, be neglected. In this case,

\[ u = U_m \sin \omega t = \frac{d\Psi}{dt}, \tag{1} \]

where \(u\) is the voltage on the secondary winding of transformer 2,
\(\Psi\) is the flux linkage of the reactor.

We shall assume that the reactor is completely remagnetized during the time \(t_1\), and that in doing so its induction changes from \(+Bs\) to \(-Bs\) (Fig. 1, b).

After integrating expression (1), the parameters of the reactor can be determined:

\[ wS = \frac{U_m}{2Bs \cdot \omega}\,(1-\cos \omega t_1). \tag{2} \]

From (2) it is not difficult to choose the number of turns of the reactor winding \(w\) and the cross section of its magnetic core \(S\) according to the profile of the laminations from which the reactor magnetic core is to be made. To ensure a width of the igniting pulses at the level of a grid displacement of more than \(60^\circ\) (the \(60^\circ\) condition), the remagnetization time is chosen from the condition

\[ t_1 = \frac{1}{\omega}\left[\frac{\pi}{3}(3-\kappa_1)-\arcsin\frac{U_0}{U_m}\right], \tag{3} \]

where \(U_0\) is the constant bias voltage on the grids of the rectifier thyratrons, referred to the primary winding of transformer 5;

\(k_1\) is a coefficient indicating how many times the voltage pulse across the resistor at the level \(U_0\) is wider than \(\pi/3\).

Figure 1b shows the voltages \(u_g\) across the choke and the voltage across the resistor (the shaded region).

For uniform distribution of the load among the valves of the rectifier, it is necessary to be able to vary the remagnetization time \(t_1\) in any of the six chokes. This can be done by changing the cross sections of the choke magnetic core and by changing the number of its turns. Changing the number of winding turns is considerably simpler, since for this purpose several taps can be provided on the winding and the number of turns can be changed without disassembling the chokes.

To determine the sensitivity of the change in the control angle to the change in the number of turns, we write expression (1) in the form

\[ U_m \cdot \sin \omega t = 2B_s \cdot S \cdot \frac{dw}{dt} \tag{4} \]

and, after transformations taking (2) into account, we obtain

\[ \frac{dt}{dw} = \frac{1 - \cos \omega t}{\omega w \sin \omega t} \ \frac{\mathrm{s}}{\mathrm{turn}}, \tag{5} \]

or

\[ \frac{dN}{dW} = \frac{360^\circ(1 - \cos \omega t)}{2\pi \cdot W \cdot \sin \omega t} \ \frac{\mathrm{deg}}{\mathrm{turn}}, \]

where

\[ dN = \frac{360^\circ}{T} \cdot dt . \tag{6} \]

Usually, a shunt in the rectified-current circuit feeds a signal to the oscillograph which, during discontinuous operation of the rectifier, has the form of current pulses with a repetition period of six pulses. By switching off one of the control thyratrons 6, it is possible to determine which pulse on the oscillograph corresponds to the operation of any pair of valves, which facilitates maintenance of the rectifier.

A rectifier with the grid-pulse formation system considered here has been operated for a long time in pulsed mode.

Submission history

Grid Control of a Pulsed Three-Phase Full-Wave Rectifier with Discontinuous Valve Operation