Triumph 830 Wobbulator / Oscillograph
Here is a fascinating relic from the past, the manual was copyrighted in 1943. It is a 884 thyratron based repetitive sweep wobbulator / oscilloscope (or as termed then, oscillograph) having identical single-ended 6SJ7 pentode X and Y amplifiers. The sweep generator was horribly non-linear as received. I surmised that leaky timing caps would be the cause of that so I rebuilt the timing switch assembly and was rewarded with a respectably (for the period) linear sweep. Most of the capacitors required replacement and I padded critical carbon resistors back into tolerance discreetly using parallel resistors. I was not able to remove the shells of the paper-in-wax condensers (period terminology) and so I simply replaced them with Russian paper-in-oil capacitors. I was able to gut and re-stuff the main power supply electrolytic. It now works as well as it can, as an oscilloscope:
The term wobbulator refers to a signal generator that is frequency modulated or wobbled over a limited range either side of the set frequency. It is used to align tuned circuits such as superhet radio intermediate frequency stages. Some wobbulators used motor driven rotary capacitors and others, a voice coil / magnet linear pump arrangement to obtain the wobble. This one has neither, the wobble being generated from the 60Hz heater supply which is passed via a 6J5 control tube to modulate the 6K8 1MHz oscillator resulting in a frequency wobble around the 1MHz centre frequency, the wobble can be varied from zero to +/- 15kHz. The way the control tube works is that the 60Hz signal is arranged to modulate the Miller capacitance of the 6J5, the grid of the which is connected to the oscillator tuned circuit such that modulation of the Miller capacitance causes the oscillator frequency to wobble. (I would like to be able to claim that I figured this out for myself however, it was actually my friend Kenneth Kuhn who pointed out that the 60Hz is modulating the Miller capacitance for the 6J5. It was perplexing to see the take off to the RF oscillator from the grid of the 6J5.) Having understood this, it was now obvious to me that the reason it was not wobbulating was due to the cathode bypass capacitor on the 6J5 having developed high ESR and so the Miller capacitance variation would be minimal. Replacing it restored normal wobbulation service! In wobbulator mode, the oscilloscope part of this beast is used to display the wobbling frequency response against a 120Hz baseline that is synchronised to the 60Hz wobble resulting in a double peak display. If the centre frequency of the wobble is set at the centre frequency of the circuit undergoing adjustment, a single peak would be displayed:
If the setting of the tuned circuit is slightly off centre, a double peak would be displayed:
This instrument then, permits the technician to see the results of his adjustments (or fiddling if like me, you don’t know what you are doing) in real time. The 6K8 RF oscillator tube is a frequency changer or mixer type (as used in superhet radios to convert the RF signal to a fixed intermediate frequency); an input on the front allows a RF generator to be connected that can be set to mix with the wobbling 1MHz signal internally generated to produce a difference frequency at the centre frequency required, often around 465kHz for intermediate frequency stages of AM radios. This avoided the complexity and expense of having an accurate on-board variable RF generator. This is quite a cunning and well thought out piece of equipment.
Here is the schematic: (click on it for full size.)
Here is the left side showing left to right, 6X5 rectifier, 6J5 wobbulation tube and 6SJ7 X and Y amps:
Here is the right side showing left to right, timing switch (in can), RF oscillator coil, 6K8 RF oscillator / frequency changer and 884 sweep thyratron:
Top view dominated by 3AP1 CRT, you can clearly see the Russian replacement caps:
View showing shields re-installed, ugly!