Early UHF D/S TV performance
I found this article on UHF field strength tests on UHF from PW March 1963.
The tests were conducted be Aerialite and used a number of types of equipment to conduct the tests, the TV being a KB VV20, I have no experience of this set but presume it used the PC86/PC88 valves introduced in 1962.
The article claims 'good quality' pictures from as little as 250 microvolts of signal, the channel used was 44. I appreciate that my signal strength meter from RBM may well have not been as accurate as the ones used in the tests but 250 microvolts would not have produced a good quality signal on the sets I serviced, I.e. Pye/Ekco and RBM using valve tuners. I know those tuners were not the best, the Philips ones were much better but I don't think they would have produced good quality on that low a signal. The Winter Hill transmitter that I received used group C channels and the sets would not have been as sensitive at those frequencies as group B.
Has anyone any experience of these KB sets on UHF to give an indication of their low noise performance and also anyone know how much of a difference there was in sensitivity in valve UHF tuners in the different channel groups.
When the first transistor UHF tuners arrived the difference in sensitivity was enormous.
I'd wondered just how good valve UHF tuners were in practice (well, I wonder about a whole load of odd things....) Yonks ago, I had a PC88/86 tuner ex a GEC D/S set as an object of interest as much as anything (long gone, unfortunately!- I like good examples of dead-end technology). The valves were obviously stretching the art of "conventionally" based valves as far as possible with electrode structure sat right down against the button base with several paralleled grid pins- trying as hard as possible, so to speak. It struck me that the disc-seal triode (e.g. DET22, etc.), devised some two decades earlier, would have been a more elegant and possibly more effective solution- this type of valve lends itself particularly well to mounting on the screen between tuning lines, keeping grid connection inductance to the absolute minimum with cathode and anode poking into their respective sections. Of course, this type of valve is more expensive to make and requires appropriate mounting in the tuner assembly, also it would be more difficult and time-consuming to replace. In other words, a no-no for a device that had to have minimal production and servicing costs.
This era is long before my time, but I hear again and again that valve UHF tuners were somewhat deaf (although I'd also heard that the Philips ones were supposed to be quite good- some folk swear by Philips, others swear at them, but they did seem to be adept at getting good results from electronics) and that even early transistor tuners were something of a revelation. Possibly some blame could be put on tuners that had been in situ for years without being selected (HT applied) and had developed cathode interface, but I get the impression that UHF sensitivity was a step (or at least half a step) too far for valves, really.
Thanks for the reply.
With enough signal, say 1.5-2mV even the deaf valve UHF tuners would produce a good noise free picture, if you could put more signal in it would give some lea way for the valves ageing, very difficult to overload them.
It would be good to know how well the UHF system worked in the USA. The UHF tuners didn't in general use an RF amplifier but a (crystal) diode mixer and a valve oscillator, gain being made up in the IF amplifier. Oscillator injection level critical for best signal to noise ratio.
I suppose another issue with picture quality is that it is subjective.
As you say, one wonders just how well any of these tuners were coping at the high end of the UHF TV band relative to the low end. There was some discussion on UKVRRR on the subject a while back, the gist seemed to be that US UHF TV coverage tended to be more localised in nature, so maybe there was less expectation of tuner sensitivity than with our culture of national TV coverage. A "passive" UHF front end (apart from LO) must have needed a particularly low-noise IF amp- presumably making use of a high-gain VHF tuner switched to act as the first stage? Perhaps this was a result of UHF TV in the US having started somewhat earlier than over here, in that consumer UHF amplifier valves were yet to be developed. I've heard it said that an active device (be it valve or transistor) will typically be useful as an oscillator around an octave or so beyond where it is useful as an amplifier, so maybe the first US UHF tuners made use of existing valves that were OK as Band III amplifiers as the LO and it wasn't felt worthwhile to develop consumer valves that would be effective as amplifiers at 800MHz+.
It's difficult to think of many other examples of continuously-tunable UHF front-ends- military aviation used 225-400MHz comms, I think from late in WW2 on, but there wouldn't have been such a tight limit on expenditure for the necessary performance, also good sensitivity and noise performance would have been far easier to achieve in such a narrow-band system. Eddystone produced a continuous-tuning UHF receiver, the 770S, with 500-1000MHz coverage using "tunable cavities", but I'm not sure what devices were used in the front-end, or just how good it was. A few were apparently supplied with IF tailored for TV purposes (rather than military monitoring). I think the later solid-state replacement (990S?) simply used an off-the-shelf Mullard/Philips varicap "tin",
I have just downloaded some BBC technical papers from the library, G-074 and T-074.
Interesting read, in one were tests on reception using a USA UHF tuner which had a measured NF of 19db, the other from 1958 was a test of valve UHF amplifiers for TV. These amplifiers would have be too expensive for domestic use but realised NF's of around 10db.
A transistor from 1965 (AF186) amp for aerial preamplifier had a NF of 7db. These transistor amps greatly reduced noise in sets using valve UHF tuners so I would assume that the PC88/PC86 tuners would have been at least 10-11 dB probably a lot more.
I used to have a fully dual standard McMichael MT763 DS-T 19" model. The UHF performance was surprisingly good when used with an aerial fed by an aerial amplifier.
This was not an isolated finding. Some sets had superior gain to others. The two Thorn 3000 sets were poor in this respect unless used via an aerial amplifier. On the other hand the Thorn 1500 series has excellent gain and only benefits little when used with an aerial amplifier.
Something to do with noise levels I believe but doesn't suggest transistorised tuners were always better. Depends on the model they were fitted in at the end of the day.
Is it possible the 1500 had a slightly narrower IF bandwidth the the 3000 CTV?
One problem with Thorn tuners was a broadband aerial input on their CTV's, probably on their BW TV's as well, a practice common on Continental Europe but was not recommend in the U.K., must find a reference.
A aerial amplifier will only help if it's noise factor is less than the tuner, putting it on the masthead well help by overcoming cable losses, a preamp with a larger NF than the tuner will make matters worse.