US2752490A - Method of producing short pulses - Google Patents

Description

June 26, 1956 RQTHSTEIN 2,752,490

METHOD OF PRODUCING SHORT PULSES Original Filed Oct. 2, 1950 FIG. I

FIG. 2

1 w 22] n ,L 4 2'5 2'2 T 232 FIG. 3

FIG. 4

INVENTOR.

JEROME ROTHSTEI N United States Patent 2,752,490 METHOD OF PRODUCING SHORT PULSES assignor to the United Jerome Rothstein, Belmar, N. J.,

the Secretary of States of America as represented by the Army Serial No. 188,042.

Original application October 2, 1950,

29, 1953, Serial Divided and this application January No.334,768

The invention described herein may be manufactured and used by or for the Government, for governmental purposes, without the payment of any royalty thereon.

The invention as disclosed herein is a division of the copending application of Jerome Rothstein, Serial No. 188,042, filed October 2, 1950, for Method of Producing Short Pulses.

This invention relates to a pulse generating circuit and particularly to a circuit for generating extremely short pulses.

For some applications of radar, extremely short pulses are desirable for improving precision ranging as well as aiding in reduction of ground clutter. The pulse generating circuit or network of the type to be considered here employs, generally, an initiating wave front and a delayed terminating wave front combined in a utilization circuit, such as the output load of a vacuum tubeeither a plate or a cathode load, depending on the gain or impedance match required. The terminating wave front may be from a separate source or may be taken from the leading edge of the initiating wave front, delayed in a delay line before being applied to the utilization circuit to terminate the pulse.

Pulses may also be formed by a multivibrator, tripped by the initiating pulse and terminated by the return of the multivibrator to its initial position after the interval of time prescribed for this function.

The foregoing systems all require, in effect, two wave fronts, one specifically delayed with respect to the other, to determine a pulse length. The conventional delay line used to achieve this specific delay seems to reach its limit near a few hundredths of a microsecond, residual capacitances and inductances apparently setting the limit. The present invention alfords a simple method of getting pulses shorter than this by a factor of perhaps or more, with a simple means of controlling pulse width.

It is, therefore, an object of this invention to provide a pulse generating circuit for extremely short pulses.

It is a further object of this invention to provide a pulse generating circuit whose initiating and terminating pulses are fed simultaneously to the control tube.

It is a further object of this invention to provide a pulse generating circuit that does not require a delay line or delay circuit for controlling the length of the pulse.

It is a further object of this invention to provide a pulse generating circuit whose pulse length can be controlled by the grid bias of the control tube.

It is a further object of this invention to provide a pulse generating circuit wherein the pulse initiating and terminating functions are accomplished by the leading edge of one input pulse.

This invention can be more clearly understood by reference to the drawing wherein Figs. 1, 2, 3, and 4 are species of this invention. Similar elements in these figures are similarly numbered.

In Fig. 1, the leading edge of a steep wave form 16 from any conventional pulsing circuit or square wave generator is fed into a phase inverter 10, at 12. The

2,752,490 Patented June 26, 1956 ice outputs 13 and 14 from the phase inverter are fed to separate grids of a multigrid tube 11 whose output, at 15, will be the desired pulse.

In operation, the steep leading edge 16, which may be either positive or negative going at 12, is formed by the phase inverter into simultaneously positive-going and negative-going steep leading edges 17 and 18 which are applied to the grids of the tube 11. The positive-going leading edge 17 is applied to the control grid 20, in this embodiment, and the negative-going leading edge 18 is applied to the suppressor, or other grid 21. The tube 11 is normally biased beyond cut-ofl at the grid 20, which is accomplished here by the bias battery 19.

When the positive-going leading edge 17 has raised the potential of the grid 20 above cut-off the tube will start to conduct, and electrons will flow towards the plate 22.

However, the other grid 21, between the grid 20 and plate 22, is simultaneously being driven negative by the negative-going leading edge 18. The initial voltage of grid 21 is established by the voltage divider 23 that may be adjustable. The more positive this grid is held, the longer it takes the negative going leading edge 18 to drive the tube 11 to cut-off, and the duration of the electron current that will reach the plate can be controlled by the initial voltage of the grid 21. The voltage divider 23 may be connected from the B+ to the negative bias 19 to allow a wide range of adjustment of the voltage on grid 21. The grid 21 should be decoupled from its voltage divider-which would normally be of low impedanceby a conventional grid resistor 40.

This particular method of controlling a pulse width is not dependent on external timing or delay circuits and can produce pulses as short as can be utilized. The pulse width can be controlled electrically-instead of manually -so that even successive pulses can have different widths if desired. Also continuous variations of pulse width are obtainable. The pulse width can be modulated by superimposing a signal voltage on the width determining voltage. The negative-going grid 21 can be adjusted to cut off the tube before the other grid is positive enough to pass electrons, or can be adjusted, theoretically, to allow only a few electrons to reach the plate.

The circuit of Fig. 2 shows a dual control tube 111 and a regenerative circuit 128. The regeneration is through the pulse transformer 126127. The first control grid is 120 and 121 is the second control grid. A bias voltage 129 can be placed in the first control grid lead for pulse width control. A negative pulse is applied to the cathode in this circuit at 112 rather than a positive one to the grid 120. A relatively long positive pulse can be taken from 132 in addition to the ultra-short pulse at point 115 from the plate of tube 111. This circuit, in the regenerative amplifier part, is similar to a blocking oscillator. In operation, the grid of the control tube is normally biased to cut-01f. The second control grid 121 is normally positive with the regenerative amplifier 128 normally conducting. A negative pulse is applied at 112 to the cathode of tube 111 causing it to conduct. Onset of conduction triggers the regenerative amplifier, driving it to cut-off. Its grid 133, goes strongly negative, cutting off the second grid 121 of the control tube 111, which is also connected to the amplifier grid 120 through condenser 134, thereby terminating the pulse.

The pulse duration depends on how rapidly the second grid can be driven to cut-01f. If 10 volts are needed to cut it off and the rate of change of grid potential of the regenerative amplifier is 2,000 volts per microsecond (obtainable in blocking oscillators), the duration would be .005 microseconds. Simple control of pulse width is had by varying the bias on the first control grid, the rate of change of the second control grid potential, or both. The regenerative amplifier could be replaced by a gas tube, but then a certain amount of jitter, or fluctuations in firing time of the tube, would probably result, giving rise to fluctuations in pulse width.

Fig. 3 shows a variation of this circuit using a multivibrator type circuit instead of a single tube regenerative amplifier. Thernultivibrator 210 can be two halves of a twin triode, the second half 235 normally conducting and connected to the control tube 211 to cut it ofi when 235 is driven to cut-off by the initiating pulse at 212.

In this circuit, the pulse width can be controlled by the bias voltage at 229 and, as in the circuit of 2, a rectangular pulse at 232. is also available.

Fig. 4 shows another variation of the basic circuit actuated by a sine wave input 336 instead of a multivibrator or phase inverter. Theeffect is the same and the amount and the length of time that the tube will be conducting can be controlled by the bias voltage 329 of the two grids. In this circuit, the sine wave input at 336 is applied to the grids through transformer 337 to allow step-up or step-down of the sine wave input voltage, as well as decoupling the mean input voltage from the mean D. C. grid potential.

The foregoing circuits are typical for utilizing the basic concept described herein. Other circuits will be apparent to those skilled in the art from this disclosure.

What is claimed is:

A circuit for generating pulses comprising a first vacuum tube having an anode, a cathode, a first control electrode adjacent to said cathode and a second control electrode further removed from said cathode, means for biasing said first control electrode to render said first vacuum tube normally non-conducting, a second vacuum tube having an anode, a cathode and a control electrode, a third vacuum tube having an anode, a cathode and a control electrode, the cathodes of said second and third vacuum tubes connected together and to a common cathode impedance, supply means for said tubes including a load impedance connected in the anode circuit of each tube, means coupling the cathode of said first vacuum tube to the control electrode of said second vacuum tube, means coupling the anode of said second vacuum tube to the second control electrode of said first vacuum tube and the control electrode of said third vacuum tube, means for applying positive pulses to the said first control electrode of said first vacuum tube and an output circuit connected to the anode of said first vacuum tube.

References Cited in the file of this patent UNITED STATES PATENTS 2,190,504 Schlesinger Feb. 13, 1940 2,556,934 Mulligan et al June 12, 1951 2,572,038 Killne Oct. 23, 1951

Images