Television on a Tek Scope
This started with an idle conversation with a colleague (thanks Neil!), who when training with our company in the early 70's, spent some time at our test equipment calibration centre in the Midlands. At certain times of the year (Wimbledon fortnight for example), a Tektronix 545 oscilloscope was pressed into service as a television set.
Intrigued by this idea, I wondered how it was done..........
At the time we also used a number of Eddystone 990 receivers, which cover the UHF television band, and have sufficient IF bandwidth to be used as TV receivers (two are needed, one for sound, and one for vision). Having got a video signal, I wondered how a scope could be used to display it. The video signal applied to the 'scopes CRT grid, at sufficient level, will modulate the intensity of the electron beam, giving a brilliance signal, a Tek 545 needs between 20 and 30 volts of video to go from nothing to full brilliance, this suggests that an external video amplifier may be needed.
Now, how to get horizontal and vertical scanning.
The Tek 545 (and the 535A that I used to prove this), have two seperate timebases, which can be used independantly, or togethether in a number of ways. The A timebase also has an external sawtooth ouput. Both timebases run at all times, no matter what the setting of the horizontal display selector. Both timebases have independant trigger circuits which can be switched to internal, external or line, triggering on the positive or negative edges.
Building on this, if the B timebase is switched to the 'scopes horizontal plates, and set to cover a scan length of about 4cm at a repetion rate of about 15KHz for a 625 line picture (10KHz for 405). The A timebase sawtooth ouput is connected to the Y amplifier input, and the amplifier controls set to give a picture height of about 3cm (this gives the standard picture aspect ratio of 4:3), and the timebase repetition rate set to cover 3cm at a repetition rate of 50Hz - the rate for a half frame in an interlaced picture (this equates to about 16ms / division). This will give a blank raster on the 'scopes tube.
Triggering for the timebases is provided by connecting the video signal to the external trigger input for each timebase, as well as to the CRT grid connection. I'm not sure if the 'scope can sort out the synch pulses on its own (the A timebase (vertical) does have a high frequency rejection on its trigger input, but I think the cut-off may be too high, more work is neede here). The B timebase will trigger from the line synch pulses with no trouble.
See Here for a modern version, and here for a 1937 version (look for the "Use Your Oscilloscope for Television reception" article, it is a little way down the page).
This shows the current state of play - a simple vertical bar pattern, this is generated by feeding a square wave into the system at a multiple of the line frequency. This also demonstrates that the B timebase will synchronise to the horizontal (line) synch pulses, it is not possible to see if the vertical signal is locked, as vertical rolling is not detectable with this type of pattern.
Copyright J.Beacon & M.Wroe-Parker 2004