Home | | TELEVISION AND VIDEO ENGINEERING | Video Amplifier Circuit

Chapter: Television and Video Engineering : Monochrome Television Transmitter and Receiver

Video Amplifier Circuit

A transistorized video amplifier circuit with emitter follower drive and partial dc coupling is shown in Fig.

VIDEO AMPLIFIER CIRCUIT

 

A transistorized video amplifier circuit with emitter follower drive and partial dc coupling is shown in Fig. The salient features of this circuit are :

 

Signal from the video detector is dc coupled to the base of Q 1 . This transistor combines the functions of an emitter follower and CE amplifier. The high input impedance of emitter follower minimizes loading of the video detector. The sync circuit is fed from the collector of this transistor, where as signal for the sound section and AGC circuit is taken from the output of the emitter follower.

 

The output from the emitter follower is dc coupled to the base of Q 2 . This is a 5 W power transistor, with a heat-sink mounted on the case. The collector supply is 140 V, to provide enough voltage swing for the 80 V P-P video signal output.

 

In the output circuit of Q 2 , contrast control forms part of the collector load. The video output signal is coupled by the 0.22 μF (C 2 ) capacitor to the cathode of picture tube. The partial dc coupling is provided by the 1 M (R 2 ) resistor connected at the collector of Q 2 .

 

The parallel combination of L 1 and C 1 is tuned to resonate at 5.5 MHz to provide maximum negative feedback to the sound signal. This prevents appearance of sound signal at the output of video amplifier.

 

The neon bulb in the grid circuit provides protection of a spark-gap since the neon bulb ionizes and shorts to ground with excessive voltage. The ‘spark gaps’ are employed to protect external receiver circuitry from ‘flash overs’ within the tube. The accumulation of charge at the various electrodes of the picture tube results in the appearance of high voltages at the electrodes, which if not discharged to ground, will do so through sections of the receiver circuitry and cause damage.

 

(vi) Note that dc voltages at the base and emitter of the two transistors have been suitably set to give desired forward bias.

 

Vertical retrace blanking pulses are fed at the grid of the picture tube through C 3, and the grid-return to ground is provided by R 3 .

 

Brightness control. The adjustment of average brightness of the reproduced scene is carried out by varying the bias potential between cathode and control grid of the picture tube. In the circuit being considered a 100 KΩ potentiometer is provided to adjust dc voltage at the cathode.


This bias sets correct operating point for the tube and in conjunction with the video blanking pulses cuts-off the electron beam at appropriate moments. The setting of grid bias depends upon the strength of signal being received. A signal of small amplitude, say from a distant station, requires more fixed negative bias on the grid than a strong signal. The dependency of picture tube grid bias on the strength of the arriving signal is illustrated in Fig. For a weak signal, the bias must be advanced to the point where combination of the relatively negative blanking voltage plus the tube bias drives the tube into cut-off. However, with a strong signal the negative grid bias must be reduced, otherwise some of the picture details are lost. Since the bias of the picture tube may require an adjustment for different stations, or under certain conditions from the same station, the brightness control is provided at the front panel of the receiver. The effects of brightness and contrast controls described earlier overlap to some extent. If setting of the contrast control is increased so that the video signal becomes stronger, then the brightness control must be adjusted to meet the new condition, so that no retrace lines are visible and the picture does not look milky or washed out. Too small a value of the negative grid bias allows average illumination of the scene to increase thus making part of the retrace visible. In addition, the picture assumes a washed out appearance. Too low a setting of the brightness control, which results in a high negative bias on the picture tube grid, will cause some of the darker portions of the image to be eliminated. Besides this overall illumination of the scenes will also decrease. To correct this latter condition, either the brightness control can be adjusted or the contrast control setting can be advanced until correct  illumination is obtained. If the brightness control is varied over a wide range the focus of the picture tube may be affected. However, in the normal range of brightness setting made by the viewer, changes in focus do not present any problem. It is now apparent that despite the fact that video signal, as received from any television station, contains all the information about the background shadings of the scene being televised, an optimum setting of both contrast control and brightness control by the viewer is a must to achieve desired results. Many viewers do not get the best out of their receivers because of incorrect settings of these controls. However, to ensure that retrace lines are not seen on the screen due to incorrect setting of either contrast or brightness control, all television receivers provide blanking pulses on the grid electrode of the picture tube




Study Material, Lecturing Notes, Assignment, Reference, Wiki description explanation, brief detail
Television and Video Engineering : Monochrome Television Transmitter and Receiver : Video Amplifier Circuit |


Privacy Policy, Terms and Conditions, DMCA Policy and Compliant

Copyright © 2018-2024 BrainKart.com; All Rights Reserved. Developed by Therithal info, Chennai.