|
Vintage Television and Video Vintage television and video equipment, programmes, VCRs etc. |
|
Thread Tools |
28th May 2016, 10:10 am | #1 |
Octode
Join Date: Jul 2006
Location: Solihull, West Midlands and Beaford, Devon
Posts: 1,626
|
Mathematical basis for setting channel numbers
Morning all.
I recently purchased a set of 4 modulators designed for hotels etc. (Philips HeadLine Agile model SIG8172 - If that means anything to anybody.) These modulators are synthesised and any channel can be chosen - but I was wondering what the rule was for choosing a selection of 4 channels. I've found that the best spacing from channel 21 upwards is 4-3-3 (21, 25, 28 and 31). Any other combination of these causes patterning etc. I'm sure I read somewhere that these were vestigial sideband modulators but I'm not absolutely sure of that now - and I can't find where it was that I read it either so I'm going to assume that they are double sideband. So, does anyone know what the rule is for setting channels? Just spreading them out doesn't seem to be ideal either - and I'd like, if at all possible, to be able to use the 4 channels and retain the Sutton Coldfield Freeview multiplexes in reasonable order. I remember learning all about it at tech college - but that was 25 years ago and most of it has drained away from my brain. Thanks all. Kind regards. From Mike. *PS. The reason I'm asking is that even with this 4-3-3 spacing, I do still see a little bit of patterning on the top channel (31) which I'd like to try and eliminate.
__________________
G7TRF |
28th May 2016, 10:53 am | #2 |
Nonode
Join Date: Jun 2009
Location: Herefordshire, UK.
Posts: 2,495
|
Re: Mathematical basis for setting channel numbers
Mike,
Like you I can't remember the fine detail all these years later - but it was to do with the choice of TV IF frequencies acting in conjunction with the image response of the superhet TV tuner. But 4-3-3 was what was decided for the UK Transmissions (with one or two exceptions). |
28th May 2016, 11:01 am | #3 |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: Mathematical basis for setting channel numbers
Sutton Coldfield was 3-3-4 (40 = BBC1, 43 = ITV1, 46 = BBC2, 50 = CH4).
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |
28th May 2016, 11:08 am | #4 |
Nonode
Join Date: Jun 2009
Location: Herefordshire, UK.
Posts: 2,495
|
Re: Mathematical basis for setting channel numbers
Julie, you are right about SC and more generally it was 3-3-4, eg Ridge Hill CH 22,25,28,32. These are the analogue era allocations.
|
28th May 2016, 11:10 am | #5 |
Dekatron
Join Date: Aug 2013
Location: Wigan, Greater Manchester, UK.
Posts: 9,427
|
Re: Mathematical basis for setting channel numbers
What's the tuner in the TV, the channel spacing was done with a specification requiring a tuned aerial input stage. Some tuners , Thorn and some Continental used aperiodic and they will not give enough rejection to second channel, ie channel 21 causing problems with 31, 40 troubling 50 etc.
In tbe early UHF days it did not matter too much until channel 4 took up the highest channel in the group. If the modulators are giving out a strong signal it will show up much more. Frank |
28th May 2016, 10:08 pm | #6 |
Hexode
Join Date: Nov 2010
Location: Market Drayton, Shropshire, UK.
Posts: 483
|
Re: Mathematical basis for setting channel numbers
The spacing on The Wrekin was 3-3-3-4
Ch 26 Bbc1 Ch 23 Itv Ch 29 Ch4 and Ch 33 Bbc2 W/h had 4-3-3-3 Ch 55 Bbc1 Ch59 Itv Ch62 Bbc2 Ch65 Ch4 regards Derrick |
28th May 2016, 11:23 pm | #7 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Mathematical basis for setting channel numbers
The mathematical basis for assigning the European analogue UHF TV channels for a given transmitter site was apparently:
Either: n, n + 3, n+ 6, n + 10; Or: n, n +4, n + 7, n + 10. This was documented in several publications; a couple of examples are attached. From the Wireless World 1962 December article, one has the impression that the first of the above sequences was primary, with the second as an alternate. I have not seen any comment as to how the choice between the two sequences was made. These came out of the ITU Stockholm 1961 planning meeting. In the UK a subsequent decision was made to move the IF upwards from the CCIR standard 38.9 MHz to 39.5 MHz, and that in turn required much higher image rejection performance from UHF tuners, as documented in WW 1962 October. In practice that meant that four-gang tuners (typically single-tuned input and bandpass interstage, but occasionally untuned input and triple-bandpass interstage) were required, whereas the three-gang type (typically untuned input and bandpass interstage) sufficed for most of Europe. That seemed to catch out some makers; for example I think that Cyldon’s original UT model was three-gang. I have never found an explanation for the IF “move” from 38.9 to 39.5 MHz. A possible rationalization is that it was done to simplify the design of dual-standard receivers, allowing the use of a basic IF bandpass curve of the double-Nyquist type – for those makers who wanted to go this route - without overly restricting 625-line vision bandwidth. With the 38.9 MHz IF, the -6 dB points of a double-Nyquist curve were 34.65 and 38.9 MHz, allowing a 625-line vision bandwidth of just 4.25 MHz. In fact Pye appeared to have done just this on its initial dual-standard receiver, which had a 38.9 MHz IF, as shown in the attached WW item. Moving the IF upwards provided incremental vision bandwidth, and in that regard the further upwards the better. But the upwards movement would have been limited by the increasing image rejection requirement and the need to avoid any serious overlap of the IF channel with channel B1. One assumes that 39.5 MHz was the practicable limit, allowing a 4.85 MHz vision bandwidth for the double-Nyquist case, not good but evidently acceptable to some. This also put the adjacent channel sound on 41.5 MHz, allowing the same trap to be used for this and channel B1 sound. I don’t think that there was any fundamental reason, such as IF harmonic interferences, not to use 38.9 MHz, and in fact South Africa used 38.9 MHz for system I receivers. Cheers, |
29th May 2016, 2:18 pm | #8 |
Pentode
Join Date: Jul 2009
Location: Hucknall, Nottinghamshire, UK.
Posts: 223
|
Re: Mathematical basis for setting channel numbers
Wonder why the UK didn't change over to 38.9 mhz IF after 1985 when 405 was gone for ever? I'm presuming a South African TV would have worked perfectly well here. Did Irish VHF/UHF system I TV's also use 39.5 mhz IF?
|
29th May 2016, 3:33 pm | #9 |
Heptode
Join Date: Oct 2015
Location: Gloucestershire, UK.
Posts: 719
|
Re: Mathematical basis for setting channel numbers
We always worked on the principle that you should not use +-1 or +-5 to avoid patterning. However, we were usually dealing with grotty cheap double sideband modulators in Sky boxes and VCRs.
I gather that for digital, you can use +-1 without issue. |
29th May 2016, 4:54 pm | #10 |
Nonode
Join Date: Jun 2013
Location: Surbiton, SW London, UK.
Posts: 2,801
|
Re: Mathematical basis for setting channel numbers
Having converted some B/G receivers to system I, those with a sharp shaped e.g. SAW
filter for 38.9 could mean the 6MHz sound was at a lower level, worsened when the sound transmitter power changed from 0.2 to 0.1 of vision power. Fortunately NICAM came in making such conversions uneconomic. |
29th May 2016, 5:18 pm | #11 |
Hexode
Join Date: Mar 2014
Location: Dublin, Ireland
Posts: 396
|
Re: Mathematical basis for setting channel numbers
Irish system 'I' sets did indeed use 39.5MHz vis IF freq
|
30th May 2016, 12:17 am | #12 | |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Mathematical basis for setting channel numbers
Quote:
The South African 38.9 MHz IF system I case was for example covered in the initial Plessey SAWF range with the SW450. The SW150 was the UK version, and the SW170 was the system B/G 38.9 MHz version. Getting back to the UHF channel assignments, the attached page from the ITU Stockholm 1961 report provides a little extra background as to potential problems, although it is not particularly explicit. One may though work from the earlier American work on UHF channel assignments, for example as shown in Benson & Whitaker, and recalculate for the European case. Local oscillator interference at (n+5) for systems G & H and I suggests that the ITU was here using the CCIR standard IF of 38.9 MHz. The same would apply for 39.5 MHz, but I doubt that that was yet in play in 1961. In the Italian system H case, oscillator interference at both (n+5) and (n+6) indicates that the ITU did its calculations using both the CCIR standard IF and the then Italian standard IF of 45.75 MHz. For the French case, oscillator interference at (n+4) is consistent with the system L IF of 32.7 MHz. This had been arrived at by setting the sound IF to be the same as for system E, namely 39.2 MHz, which then eased dual-standard IF strip design. System K with oscillator at (n+4) indicates the then Russian standard IF of 34.25 MHz; the later 38 MHz number was still in the future. The double-conversion case is a mystery, though – meaning that I have not attempted to fit some plausible numbers to it. Cheers, |
|
30th May 2016, 1:04 am | #13 |
Nonode
Join Date: Jan 2004
Location: 1966-1976 Coverack in Cornwall and Helston Cornwall. 1976-present Bristol/Bath area.
Posts: 2,965
|
Re: Mathematical basis for setting channel numbers
For analogue TV the RF output of STB's and videos would be set to avoid using channels +/- 5 or +/- 9 to avoid 1st and 2nd IF interference effects with local transmitters and/or other RF outputs. I think this was very much taken into consideration when they planned the UHF TV's network to avoid any interference effects between the allocated 4 channels to each transmitter. Some TV's had very poor 1st and 2nd IF image rejection which did cause some problems in areas where the UHF band was congested with strong signals available from two or more high powered transmitters as well as from surrounding relay stations.
__________________
Simon BVWS member |