This example of a superb synthesised RF signal generator was found to have numerous gaps in its RF output range. However, the generator appeared to be working normally with the front panel displaying output frequency and RF level. The instrument was in storage in a loft and was in danger of being broken for spares.
Initially having no idea what might have been the problem, the first task was to accurately plot the gaps in coverage. An hour was spent monitoring the generator’s output on a spectrum analyser. It soon became apparent that the gaps were harmonically related-the start and finish point of each gap was twice the frequency of the gap immediately below. Time to read the service manual to discover how the output frequencies are derived…
The gaps in frequency coverage were- 1.501MHz to 2.032MHz, 2.876MHz to 3.415MHz, 5.751MHz to 6.829MHz, 11.501MHz to 13.657MHz, 23.001MHz to 27.313MHz, 46.001 to 54.626MHz, 92.001MHz to 109.251MHz, 184.001MHz to 218.501MHz, 368.001MHz to 437.001MHz and 736.001MHz to 874.001MHz.
Pages 17 and 18 of Chapter 4 in the Service Manual (see below) reveal the clues: Four separate oscillators cover the range 260MHz to 520MHz. Only one oscillator operates at a time and the output frequency is derived by successive divisions of the oscillator output. For frequencies above 520MHz the oscillator output frequency is doubled. It was soon clear that Oscillator No3 was not functioning.
Access to the Oscillators
1. Remove the top cover by removing the two feet at the rear and sliding the top cover rearwards.
2. Unscrew and detach the rigid coaxial connection from the rear of the exposed ‘RF Box 1’.
3. Loosen and remove two bolts at the front and two at the rear of RF Box 1. The RF box can now be hinged upwards.
4. Remove the 8 screws from the exposed underside and remove the cover. The oscillators live in the screened section lower right.
5. Remove the 6 screws and the cover to expose the oscillator board.
Voltage checks confirmed the presence of the 11.5V supply and that the relevant line, LD2, on tag 4 went low when appropriate.
As the circuits of the oscillator sections are pretty well identical, comparative continuity and resistance checks were made but nothing obvious was found. A very close visual examination was made with magnification and good lighting. This revealed several soldered joints which looked suspect i.e. dry or cold joints (arrowed below). These were carefully reflowed.
After reassembly- the above steps reversed- I was rewarded with full continuous frequency coverage.