An early version of the Racal RA-17.
There is so much written about this famous receiver on the 'Net that it seems pointless adding any more here, however a brief summary seems in order.
Coverage: 500 KHz to 30 MHz continuous. Triple conversion superhet, using the Wadley Loop anti-drift technique. The only 'snags' with the RA-17 are its size and its weight, but if you do want a receiver that is a joy to operate and will just about hear anything that is radiated from 500 kHz to 30 MHz - then you get a Racal RA-17! The RA-17 is derived from the Collins 51J-4 . . . . .

History.
As I understand it, the story goes like this. Collins wanted to market a receiver based on the 51J-4 in the U.K. but added that it had to be comprised of USA-sourced components. This condition was not acceptable to the U.K customer (the M.O.D.) so two guys set up their own company to build the RA-17 to meet that need. Although the U.K. navy was the original prime customer, the RA-17 soon became established as THE general purpose receiver for government short-wave monitoring services; GCHQ being one of the prime users - who used them in huge quantities during the Cold War.

► Maintenance notes.
There is a great deal of info. covering maintenance on the 'Net. There are a number of things that you should do before attempting any maintenance of the RA-17 though. The first is get a copy of the manual and read it. Then read it again. Make sure that you thoroughly understand the principles on which it works.

► Test gear: apart from the usual multimeter, a quality 'scope and an accurately calibrated signal generator are essential. If you are going to attempt to re-align the 40 MHz band-pass filter, you must have the correct specialist kit (a Rhode and Schwartz polyskop is what I used to use, many years ago)- or a quality spectrum analyser with its own synchronised tracking generator.

► Known 'special' faults: A common fault is a defective 1 MHz crystal; commonly corroded pins: poor contact in the crystal's chassis-mounted crystal socket. With almost any symptoms of poor sensitivity; no reception abobe a certain freq., etc., always check the 1 MHz crystal osc. first! Do not adjust the coil in the oscillator's L/C cct. for maximum output; the L/C circuit must present an inductive load, not a resistive load that will result by tuning the L/C cct. to resonance.
The crystal calibrator is known to be unreliable - and it must be set up properly: not a simple matter.

► Modifications.
There are several that have been published on the 'Net. A favourite is a mod. to convert the BFO to a product detector. Reports have been received that this is fairly simple to implement and improves the resolution of ssb tremendously.

Skywave. January 5th., 2012.

A104 advert I recently spotted. Having got one I still cannot work out how (or why) you'd want to hang it on a wall. Andrew

Eddystone 770U Mk.1. Vintage: 1955 - 1963.
Use: as per 770R (q.v.)
UHF comms. receiver: 150 - 500 Mc/s. Twin s/het., 6 bands; turret tuner.
15 valves, 1 RF, 2 IF @ 50 Mc/s, 2 IF @ 5.2 Mc/s, S-meter, AM/FM.
Many variants on a common theme of this receiver were built, primarily to the requirements of the M.O.D.

► Collector's notes.
As per comment for 770R Mk.1: avoid mistaking a 770U for a 770R.
A 770S is known to have been built: 500 Mc/s - 1,000 Mc/s. Totally different in appearance and very rare. There are two versions of the 770S: 1961 and 1962 build. They are quite different in appearance.

--------------------------------------------------------------------------------
Photo. & comment by Al. (Skywave); Dec. 20, 2011.

HP Vector Network Analyzer - model 8752A.

Kit like this is rarely seen outside of professional R&D labs. and test houses.
The only reason I've added it is simply so that those who were wondering what a VNA (vector network analyzer)looks like can here see one being used in anger.

What does it do, how do you use it?
It can do lots of clever measurement things. Here's a common use. A VNA consists of two ports: one is an output port that you connect to one input (or output) terminal of the Device Under Test (DUT), and the other to the output (or input) terminal. The VNA output port sweeps over a predetermined freq. range at a predetermined level and the resultant response from the DUT is displayed on the screen. Hence, a graphical display of freq. versus amplitude is displayed. In the photo above, the response of a VHF multi-pole filter is being examined from 30 MHZ to > 1GHz. But a VNA can do a lot more than things like that. It can measure insertion loss, input impedance, plot a Smith Chart, measure & display S-parameters, etc. - and all these measurements over an extraordinarily wide freq. range.
Prior to using a VNA, it must be 'calibrated'. This involves taking it through a series of set-up routines so that is will accurately measure in, say, a 50-ohm system and also compensate for the co-axial cables that connect it to the DUT. (That way, the performance of the cables can effectively be taken into account by the VNA when making measurements ~ clever, eh?)

Note, a VNA is not the same 'animal' as a spectrum analyzer, although a VNA can be used for this, but it is not recommended on the grounds of possible damage (very expensive damage) to it if used in this way.

Finally, kit like this is prohibitively expensive. It's not unusual for a new one to cost in excess of £100,000! :eek: Ouch!

--------------------------------------------------------------------------------
Photo. and comment by Al. (Skywave): Dec. 19, 2011.

Manufacturer: Grundig. Model 3028; “Marlborough”

--------------------------------------------------------------------------------
Photo. submitted by maitu: Dec. 2011.
Comments by Al.(Skywave): Dec. 2011