US2444438A

US2444438A - Pulse generation method

Info

Publication number: US2444438A
Application number: US548369A
Authority: US
Inventors: Donald D Grieg
Original assignee: Standard Telephone and Cables PLC
Current assignee: STC PLC; Federal Telephone and Radio Corp
Priority date: 1944-08-07
Filing date: 1944-08-07
Publication date: 1948-07-06
Anticipated expiration: 1965-07-06
Also published as: FR941510A; FR57500E; BE474667A; BE472524A

Description

July 5, 19443- D.. D. GRIEG 2,444,438

PULSE GENERATION METHOD Fild Aug. '7, 1944 .l 5 4 6 l PULSE 5477/005 INVENTOR. 7o/wm 0. @ff/f6 Patented July 6, 1948 PULSE GENERATION METHOD Donald D. Grieg, Forest Hills, N. Y., assignor to Federal Telephone and Radio Corporation, New York, N. Y., a corporation of Delaware Application August 7, 1944, SerialNo. 548,369

lIhis invention relates to the generation of electrical pulses and more particularly to the generation of symmetrically shaped electrical pulses.

`In multi-channel pulse systems and in the production of synchronizing pulses for television, as well as the production of pulses in other pulse systems,itis often desirable to produce symmetrical pulses of a given duration and/or of a given steepness for the leading and trailing edges thereof. l It is not always possible to obtain a consistency of duration or symmetry for pulses produced, especially where their production is dependent on a source of pulses, the duration and/ or repetition rate of which vary.

It isan object of my invention to provide a method and means for producing a substantially symmetrically shaped pulse from another pulse equal to or greater than the duration of the desired pulse.

Another object of the invention is to provide a method and means for producing substantially symmetrically shaped pulses hawng a given duration and/or a given steepness of slope for the leading and trailing edges thereof, from a source of pulses regardless of any variation that may occur in the duration and/or repetition rate of the pulses of such source.

According to a feature of my invention use is made of the properties of a short circuited passive network or other pulse reflecting means to produce pulses whose duration is determined by the retardation characteristics of the reflecting means independent of the repetition rate and duration of the input pulses. The duration of the input pulses, however, should be at least equal to the desired duration of the output pulses. The input pulse is impressed at the input end of the network and travels through the network to the short circuited portion thereof from which itis reflected. The net result of this action is the appearance of a reflected pulse at the sending end impedance possessing substantially the shape characteristics of the input pulse, but reversed in polarity and delayed by an amount equal to twice the forward delay of the network. The resulting pulse energy existing across the sending end impedance of the network is the sum of the input and reflected pulses. The sending end impedance is preferably made equal to the charr acteristic impedance of the line so that the reected energy is absorbed, thereby clearing the network of reflections prior to the occurrence of the next input pulse. By suitably clipping a selected portion of the resulting pulse energy an output pulse of thedesired duration and polarity is obtained.

pulse, as the case may be. As will'be described in further detail hereinafter the leading and trailing edges of the output pulses are caused to correspond-directly tothe steepness of the slope of either the leading or trailing edge of the input pulses, depending upon from which of the two pulse portions resulting across the input impedance, the output pulse energy is obtained.

Ihe above and other objects and features of the invention will become more clear upon consideration of the following detailed description to beread in connectionv with the accompanying drawings, in which,

Fig. l'is a schematic wiring diagram with parts in block form of anembodiment of my invention, and,

Figs. 2 and 3 are graphical illustrations used in explaining the operation of the embodiment shown in Fig. l.

The pulse producer l of the system may comu priseany source of pulses available so long as the pulses thereof are of a duration equal to or greater than the duration of the desired output pulses. In graph 2ar of Fig. 2 a rectangularly shapedipulse is shown by way of example as the character of the pulses obtained from the producer l. These pulses are applied to a cathode follower, 3 the output circuit of which contains an impedance load ll. Connected across the-impedance load 4 is a passive network cr artificial delay line 5 comprising a series arangement of inductance coils 1l5 and parallel arrangement of condensers l. `The load impedance 4 together with the equivalent cathode follower impedance, as hereinbefore stated, is preferably chosen so as to match the characteristic impedance of the network 5. For controlling the retardation characteristics of the network a shorting device 8 is providedfor adjustment or selective connection at points along the length of the network.l For any input pulse the energy thereof travels along the network until it comes to a short circuited portion thereof, such as formed by the short circuiting device 8. The short circuited portion refleets the pulse energy resulting in its delayed re-appearance across the load impedance 4 but with the polarity thereof reversed. This delayed reversed reflection pulse effect is indicated by pulse 9 in graph 2b. As shown in graph 2c, the

IIJ- or negative pulse il whichever is desired, a

clipper I2 is provided across the load impedance 4 and is provided with a suitable bias for clip-` ping the desired pulse energy from the resulting f wave indicated by graph 2c. Inl graph 2c, the clipper I2' is indicated as having a threshold clipping level I3 whereby energy of the positive pulse l l I0 is obtained as shown'in graph 2d. If desired, the clipper may be provided with a cathode follower M, whereby output pulses may be obtained at a lowered output impedance.

Assume, for example, that the input pulses vary in duration between the solid line and the dot-dash lines indicated for pulse 2 of graph 2a. The reected pulses 9 of graph 2b will likewise be varied as indicated at I6. This variation does not alter the pulse Iwidth or position of the positive pulse Iii. The negative pulse, however, is shifted in time in accordance with the variation in width of the input pulse, but its Width remains the same, as indi-cated at l1.

Where a given slope is desired for the output pulse as indicated by graphs 3d and 3e of Fig. 3, the leading or trailing edge of vthe input pulse is caused to have the desired slope. As shown in graph 3a the leading edge I8 of the input pulse I9 is given the desired slope for output pulse 20a. The leading edge I8a of the reflected pulse |90, of graph 3b corresponds substantially directly to the slope of the leading edge I8 of input pulse I9.l The summa-tion of the two pulses result in positive pulse 2G, whose leading edge is the same in slope as the leading edge I8 and whose trailing edge is the saine in slope as leading edge I8a. By clipping the pulse 2li at a level 2| an output pulse a is obtained as desired.

It will be observed that by shaping the leading or trailing edge of the input pulse andselecting a certain retardation characteristic for the network 5 symmetrically triangular -pulses can be obtained. This is illustrated by the trailing edge 22 olfinput pulse I9, graph 3a. As shown in graph 3b, the total retardation of travel through the network to and from the short circuited device 8 is taken equal to the duration of the trailing edge 22 as indicated by the time intervait. The summation of the pulses I9 and I9a therefore results in a negative pulse 23 of symmetrical. triangular shape, the leading edge thereof corre- 4 spending to the trailing edge 22, while the trailing edge thereof corresponds to the trailing edge of pulse I9a. The triangular pulse is obtained by clipping the resulting pulse energy along a level rEll thereby producing output pulse 23a shown in graph 3e. The slopes of the pulse 23a, of course, may be varied in steepness by varying the steepness of the trailing edge'of-{the input pulse. It will also be observed that symmetrical triangular pulses may be obtained from the leading edge `oi the input pulse by proper selection of the retardation .characteristics of network 5 with respect to the duration of the leading edge of the input pulse.

AWhere lthe Ainput pulse varies in duration as indicated in graph 3a. by broken lines 25, no eifect on the resulting negative pulse is obtained other than a corresponding displacement in time.

In pulse yproducing devices such as, :for example, a multivibrator,y there is usually a vari ation in the pulse duration of the output pulses for variation in repetition rate of these pulses, due to the fact that such variation changes the mode of functioning of the pulse rproducing circuit. Such variation in repetition rate, however, does not eiect the output of the circuit shown in Fig. 1 since each output pulse is the direct summation of its input and reflected energy.

While I have shown and described particular forms and variations of the above-described invention it should be understood that the forms herein mentioned are given by lway of illustration of the invention only an-d not as restricting the invention as set forth in the objects and the appending claim.

I claim:

A method .of producing a symmetrical triangular pulse, comprising propagating a trapezoidal pulse, the leading or `trailing edge of which Ahas a slope corresponding to the slope desired for REFERENCES CITED The following referencesy are of record in the file of this patent:

UNITED STATES PATENTS Name Date Vl-lumlein Dec. 16, 1941 Number