80-Bus News

1) A few words telling me what to expect (e.g. Naspen file ‰ 1200 baud, or 48tpi Poly-Dos disk). and

2) a stamped addressed label (or envelope) for the return of your media if so required.

So put your fingers to the keyboard and send the results to 80-BUS News.

Video Output Stages

I was recently doing some work with a video output stage, and, having a reasonable ‘scope to hand as a result, I decided to look at the ouput stage of Gemini IVC. There were two points I wanted to deal with, a) the low level of output ( < standard 1V p-p), and b) ‘The marching sands of time’. (Background interference patterns that are noticeable when using large areas of inverse video).

The IVC output stage is shown in Fig 1. It consists of an emitter follower driven by the video signal from IC3a (via R5), and the mixed syncs from IC5f (via R1). The ideal output waveform is shown in Fig 2. This consists of a three level signal whose peak-to-peak amplitude is 1V (when driving into a 75 ohm load). The three levels of the signal are white, black, and sync, and
Fig 1 their relative amplitudes are also given in Fig 2. Assuming a perfect emitter follower with an output impedance of 75 ohms, we want to aim for a 2v p-p signal at the base of the transistor. I started by selecting arbitary voltage levels of White=3V, Black=1.6V, Sync=1V. Let us now consider how the three signals are generated: For the ‘White’ level the output of IC3a is high, (dot on), and the output of IC5f is high (no sync signal). For the ‘Black’ level we have the output of IC3a low (dot off), and the output of IC5f is high (no sync signal). Finally for the ‘Sync’ level the outputs of both IC3a and IC5f are low. Armed with this information it is only a case of applying Ohm’s Law and Kirchoff’s Law to the circuit…​or is it? One problem is that the high level output voltage from a TTL gate is not very well
Fig 2 defined, and it also has an appreciable output impedance. With the ICs fitted to my card the high level seemed to be around 3V, so, as I just happened to have picked 3V for my white level, I could conveniently ignore R1 and R5 and calculate a suitable value for R3. This then left me with two unknowns, and two equations (for the Black level and the Sync level). The result of the calculations (followed by a small adjustment after an actual trial) is shown in Fig 2. (Compare the values against those of the IVC circuit diagram). These values resulted in a reasonable video signal driving out from the card.

One other small change I would recommend, especially if you use inverse video or occasionally have large white areas on your screen, is to short out C7. This should not have any harmful effect, as all monitors I have encountered have their own DC blocking capacitor. (They normally have a 75 ohm terminating resistor, the top-end of which is coupled via a small capacitor, – 10uF, into a high impedance input stage. I note also that the BBC Micro has no