Homelab Signal Generator

(s/n 4303)

Homelab Signal generator s/n 4303 Black crackle finish cast aluminium box housing triode 400Hz AF oscillator and triode connected pentode HF oscillator (6 ranges). AF / CW / MCW UK Now working after quite a lot of work, see below 1946? (home consructed from ex WD parts?)

Black crackle finish cast aluminium box housing triode 400Hz AF oscillator and triode connected pentode HF oscillator (6 ranges). AF / CW / MCW. The cast outer box contains two smaller cast boxes each with a sheet aluminium cover. The larger of these houses the HF oscillator and yet another box which encloses the coils and band cvhange switch. The second houses the output attenuator switch and resistors and potentiometer.

This seems to be a home consructed kit and most of the parts appear to be UK government surplus. Probably made in 1946. Both valves are of American manufacture, the AF oscillator uses a hf triode Hytron type 2A4 with octal base and two top caps. The HF oscillator was a type 9003 with B7G base.

The unit produced an out put on two of the HF ranges when it was given to me some years ago. When I opened up the case and looked inside, the valve heaters seeemed to be glowing too brightly and the mains lead to the transformer was connected to the 210volt tapping. After correcting this and replacing the smoothing capacitors which I was unable to reform well enough, I found that the copper oxide rectifier was not doing a very efficient job. A silicon rectifier solved this problem.

Oscillations now on the three lower ranges, but nothing much on the higher ranges. Replacing the components in the HF section had little effect, but replacing the bakelite vave holder with a nylon one helped extend the range over which stable oscillation was possible. However in the end I gave up on the 9003 thinking that perhaps it may have been adversely affected by many hours of high heater voltage. The valve base was rewired and an EF91 was substituted (not quite as easy as it may seem since the glass envelope is slightly longer). Eureka! Oscillations of reasonable purity available over all six ranges. The schematic diagram of the unit as it was before the latter modification is shown below.

This one was sold on ebay in 2004 and a similar one without the meter is currently for sale (April 2005) at www.crowthornetubes.com/testers.htm

I have now found in a 1953 copy of "Wireless World" an advertisement for a 100Kc/s-100Mc/s signal generator type 10 manufactured by Homelab Instruments Ltd, 615-617 High Road, Leyton, London E10. I assume that mine is an earlier model from the same firm.

Here are two pictures of the type 10 owned by Ralph which show how the mechanical design has been simplified and how it still used similar ex government components

Circuit Diagram