SYSTEM A MODULATOR

With the impending close down of the 405-line transmitter network those of us with collections of early TV sets are having to make new arrangements to supply them with suitable signals. Whatever the scheme used, some method of converting baseband audio and video into a form suitable for feeding into the set's aerial socket is required.

The modulator described in this article was designed to perform this task with little degradation of the signal. Separate carrier generators and modulators are used for the sound and vision for best possible sound-on-vision and vision-on-sound performance. Crystal control of the oscillators was chosen to ensure stability and reduce alignment problems. The choice of Channel 1 (41.5MHz sound, 45MHz vision) was made because the vast majority of 405-line TV sets, including all pre-war models, can operate on this channel. Those with one of the few Ch. 4 "Birmingham" models of the late 40s should be able to modify the design for 58.25MHz sound and 61.75MHz vision without much difficulty.

Circuit Operation

Fig. 1 shows the vision modulator circuit. The 1V peak-to-peak video input is first amplified by Trl/2 and then fed via the emitter-follower Tr3 to the d.c. restorer Tr4. The potentiometer in Tr4's base circuit varies the modulator bias and is adjusted so that the carrier is just extinguished on the sync pulse tips.
Tr5 is connected in a Butler oscillator circuit operating at 45MHz. The output from the modulator chip is transformer coupled to the output socket via a lOdB pad and a 6dB combining network. The lOdB pads in this and the sound modulator circuit act as isolators to protect each modulator from the other's r.f. output.

The sound modulator circuit (see Fig. 2) is similar, with IC2 driving the modulator chip in push-pull to reduce distortion to a minimum. The audio sensitivity of 1V r.m.s. = 100 per cent modulation is set by the value of Rl, which can be varied if desired. For example, with Rl at lOOKohm 100 per cent modulation is achieved with a 200mV r.m.s. input; with Rl at 1Mohm 2V r.m.s. equals 100 per cent modulation. The quiescent carrier level is set by the two biasing networks connected to the audio input pins 1 and 4 of theMC1496.

Both modulators generate 8OOmV of r.f. across the secondary windings of the output transformers T2 and T4 under maximum modulation conditions (peak white for video and the positive peak of a 1V sinewave for audio). Each modulator thus produces about 130mV of r.f. at the output socket, so about 20dB of extra attenuation will be needed when this is connected directly to the aerial socket of a set. This allows several sets to be driven via a passive splitter network if required.

It will be noticed from the above that the peak vision to average sound power ratio is 6dB (4:1) rather than the correct 7dB (5:1). This has never caused problems with any set I've used, but purists may if they wish increase the loss in the sound modulator's output pad from lOdB to lldB.

Construction

The prototype was constructed on a single Veroboard Eurocard with a "colander" ground plane. An alternative would be to use "solid construction" on a piece of plain copperclad board about 4 x 6in., preferably glassfibre. Start by fixing the coil formers to the copper side of the board, then glue the MC1496s to the board upside down. All the wire-ended components can then be soldered directly to the appro- priate i.c. pins. A small soldering iron and a steady hand are essential. The video amplifier/d.c. restorer and the audio amplifier can be built on a small piece of Veroboard which is then mounted on the main board.

The transistor types are not critical. I used those specified because they were to hand. Any r.f. types should be suitable for the oscillators and any small-signal silicon general-purpose types for the video section. The non-electrolytic capacitors are all miniature ceramic types and the polarised capacitors bead tantalum or low-leakage electrolytics. The resistors are 0.25W, 5 per cent carbon or metal film types and the crystals were made to order by IQD Electronics..

The transformers are constructed as follows. Tl and T3 have a primary consisting of 15 turns of 26 s.w.g. wire close wound on a 6mm former with a dust core. The secondaries consist of two turns of 26 s.w.g. wound over the "cold" end of the primary. T2 and T4 have primaries consisting of 12 turns of 26 s.w.g. close wound on a 6mm former with a dust core and a centre tap. The secondaries consist of three turns wound over the centre of the coil.

Alignment

Three alignment methods are possible depending on the equipment available.

The first method requires a video source (preferably a line repetitive waveform covering the full grey scale, such as a wedge or step-wedge pattern), an audio signal generator with an output adjustable over the range 0-1V, and an oscilloscope with a bandwidth of at least 50MHz. Connect the scope across the secondary of T2 and apply 12V d.c. to the vision modulator only. Adjust Tl and T2 for maximum r.f. output. These adjustments are very broad and not difficult to make. If the maximum signal is reached with the cores fully in or out it may be worth trying a different value for the tuning capacitor Cl or varying the number of turns.

Fig.3: Alignment traces :- upper, video input, 0.5V/cm ; lower, r.f. envelope, 200mV/cm.

Connect the video pattern generator to the video input and synchronise the scope's timebase to this input. Adjust the video bias control so that the r.f. output drops to zero, or as close to zero as possible, during the sync pulses. The scope should now display the modulated r.f. envelope shown in Fig. 3.

Transfer the scope to the secondary of T4 and the 12V supply to the sound modulator. Adjust T3 and T4 for maximum r.f. output as above. Connect the audio signal generator to the audio input: with about IV r.m.s. of input drive it should be possible to produce an envelope display as shown in Fig. 4.

Fig. 4: Alignment traces :- upper, sound sinewave input, 0.5V/cm; lower, r.f. envelope, 200mV/cm.

The second method requires a video pattern generator as above, a scope with a limited bandwidth (say 5MHz), and a good working TV set that can be tuned to Ch. 1. This set should preferably have manual r.f. gain control, i.e. no a.g.c., and a chassis isolated from the mains. Connect the modulator's r.f. output to the TV set via an attenuator of about 20dB and the scope across the set's vision detector load resistor. Apply 12V to the modulator. Tune the TV set for maximum d.c. across the load resistor, then peak Tl and T2. Apply the pattern to the modulator and synchronise the scope with the generator. The pattern should be visible on the scope: adjust the video bias so that the sync pulses crush, then back off the adjustment until the pulses regain their maximum amplitude. Check that the whites are not compressed - if necessary reduce the TV set's r.f. gain to ensure that this does not happen.

Connect an audio tone to the modulator's audio input socket. This tone should be heard from the speaker. Tune T3 and T4 for maximum output.

The third method requires a source of 405-line video, e.g. a pattern generator or tape, and a working 405-line TV set. Connect the modulator to the video source and to the TV set (via a 20dB pad). Apply 12V to the modulator and tune it in on the TV set. Peak Tl and T2 for maximum contrast and adjust the video bias control for best grey-scale reproduction consistent with reliable synchronisation. It may be necessary to adjust the set's contrast control to make this possible. Connect an audio source, preferably a tone, to the modulator and adjust T3 and T4 for maximum volume.