GB586176A - Arrangements for selecting electric pulses from a pulse train in accordance with their amplitude - Google Patents

Abstract

586,176. Pulse modulation; radio receiving circuits. STANDARD TELEPHONES & CABLES, Ltd. Sept. 27, 1944, No. 18521. Convention date, Dec. 15, 1942. [Class 40 (v)] In a filtering arrangement for removing from a pulse train pulses of greater amplitude than the wanted pulses, the pulses of greater amplitude are distorted by producing sharp amplitude maxima defining the leading and trailing edges thereof, the pulses are then filtered to eliminate the higher harmonics, thereby reducing the amplitudes of the distorted pulses below the amplitude of the wanted pulses and the unwanted pulses are then removed by clipping at a level above their maximum amplitude. As shown, the pulses are applied to an amplifier 10, Fig. 1, which may comprise any arrangement giving a characteristic as shown at 15, Fig. 2, i.e. where an increase of input potential causes the output potential to increase to a maximum 41 and then to decrease toward zero at 42. The amplifier 10 comprises an input triode 17 coupled by condensers 25, 29 to push-pull connected triodes 21, 22. The triodes are separately biassed over resistors 27, 30, triode 21 being in the absence of a signal, biassed beyond cut-off and triode 22 being biassed to permit a normal plate current. As the grid 14 of tube 17 is made increasingly negative by an increasing input pulse potential, the grid 24 of tube 22 is made more negative, thereby reducing its anode current and increasing the output potential of point 33 as indicated between points 40, 41 on curve 15. At the same time the grid 23 of tube 21 is made less negative and as tube 22 reaches the cut-off point tube 21 begins to conduct. The anode current of tube 21 then increases with further increase of the input potential causing the potential of point 33 to fall as indicated between points 41, 42 of curve 15, Fig. 2, until at 42 the tube 17 reaches the cut-off point. The characteristic curve 15 need not be symmetrical as shown, however. The gain of the amplifier 10 is adjusted so that when a series of pulses 51 ... 54, derived e.g. by demodulation of a carrier wave, is applied thereto, the amplitude of the wanted pulse 51 is brought near and preferably slightly under the maximum point 41 of curve 15. The output pulse 51a is thereby produced in substantially the same shape as the input 51. The unwanted pulses 52, 53 of greater amplitude than 51, extend beyond the point 41 and cause sharp amplitude maxima 52a, 52b, 63, 64, corresponding to the leading and the trailing edges of the input pulses 52, 53. The unwanted pulse 54, of lesser amplitude than 51, produces a reduced output pulse 54a. By suitably filtering out the higher harmonics at 12, Fig. 1, the sharp pulse portions 52a, 52b ... are reduced in amplitude as indicated by the wave 70. The latter is then clipped at 13, Fig. 1, along line 71 of wave 70 to give the final output wave 80 containing only pulse energy 81, corresponding to the wanted pulses 51.