Racal SA.97 Wobbulator page 2

The phrase ‘poke around’ is synonymous with fault finding, especially when you have limited information, However when there is potentially 4KV present, poking around takes on a whole new dimension. I found an open-circuit 1M resistor in the -2KV line. This however was not the cause of the dysfunctional Brilliance control. On the other hand, the cause of the restricted Y-shift was nothing more than a broken wire to the base of the CRT. Then I spotted the 100nF/2KV capacitor under the chassis that was covered in light oil! This was connected between the grid of the CRT and the Time-base circuitry. I replaced the capacitor with a modern high voltage one and this cured the fault with the Brilliance control. So, that was easy!

A quick test showed that the Sweep Oscillator worked and so did the Lin/Log Amplifier module. With the display now working all I had to do now was figure out how the SA.97 was actually used.

There are 4 connectors on the front panel; two BNC and two mini-Burndept. A short BNC lead can be used to link the output of the 40MHz Lin/Log Amplifier to the Y-Input on the display. The mini-burndept connectors are not the same as those found in the RA17 in that the inner part is a receptacle as opposed to a pin. I temporarily connected the output of the Sweep Generator (as Racal call it) to the input of the Lin/Log Amp via a tuned circuit that I knew had a response at 40MHz. This produced a response trace just as I had hoped. So I was definitely on-track for a working SA.97. However there is a distinct lack of means of calibration regarding the displayed output. There is however a small sub-chassis inside with four valves and a 40MHz crystal which I assumed formed a marker generator. Examination of the trace did indeed reveal a dark spot. While I was able to prove that this was coming from the circuit on the sub chassis, one marker was obviously insufficient. I later discovered that shortly after the unit was switched on there were actually two markers but one of them would slowly diminish after a few minutes.

As already said, the Marker Generator, employs four valves, one ECC81, two EF91s and an EFC80. The latter appeared to be faulty; the triode section causing the trip on my valve tester to assert. I didn’t have a spare, so I had to order one. In the interim I set about working out the circuit diagram ... not as hard as it looks. Once I had an idea of the different parts of the circuit, it was just a case of following the signals through each ‘block’ and trying to figure out why one of the pulses would diminish. I then decided to replace all the half-watt resistors under the little chassis. Some were about 50% high in value. I also replaced three 10nF decoupling capacitors and the 100nF output capacitor. The large 33K wirewound resistor is a bit of an enigma because it is in parallel with what was originally a half-watt 47K resistor. I think I have it in the right place! On putting it all back together again and with a new ECF80 fitted, the two visible markers

remained constant, so I must have done something right. But only two markers still didn’t ‘feel’ right, so I played around with the two variable inductors. I figured it would be logical to have a centre marker and at least one on either side. So I also played around with the sweep width controls and finally got the trace on the left. Definitely one centre marker with two either side. Not bad for no cct. Diag. and no idea how it worked!

Marker Generator underside

Marker Generator