System A Front Porch Timing

[Synchrodyne]

Originally Posted by Doctor Hue

Found this in EBU Tech Review 1996

Thanks for that.

There was also an earlier EBU item on Krivocheev that I found and downloaded a while back, attached. He looks to have been someone in or approaching the Blumlein class, a rare talent.

Origins of the 625-line system were discussed in this thread some time back: https://www.vintage-radio.net/forum/...ad.php?t=83653.

Cheers,

[Doctor Hue]

Originally Posted by Synchrodyne

There was also an earlier EBU item on Krivocheev that I found and downloaded a while back, attached.

Thanks for That.

Going back to the original Front Porch question.

I've found a C.C.I.R. Rep. 308-2 that shows some timings from 1970 also
the Rep. 624 from 1974 has the same timings but omits the percentages.

Also attached from a 1938 document showing Front Porch of 1/200 th the line scan as Peter Scott's drawing was a little indistinct ( mine's not much better )

[SwanseaSteve]

Originally Posted by Synchrodyne

But it seems that back in the late 1940s and early 1950s, the Russian origins of 625/50 were not mentioned in polite company. One needs to use deduction to work out what the Gerber (compromise) was all about.

As far as I can deduce, around 1950 Herr Gerber was the chairman of the CCIR committee tasked with standardising the soon-to-be-deployed 625 television system for western Europe. It would seem that the scanning parameters were just lifted from the Soviet work (well, there was no point in re-inventing all that, it was already being transmitted in the USSR by that time).

The so-called "Gerber System" was really the CCIR's standard system (later labelled CCIR System B). The CCIR's main contribution was to specify that the channel width would be 7 MHz rather than the Soviet system's 8 MHz.

This, I suspect, was down to western Europe needing to cram more channels into the available VHF space, and to be able to get four channels into Band I. After all, western Europe was a patchwork of small countries all wanting to transmit their own national networks, the USSR was a mega-state with only a need to transmit their one or two state-owned networks. Quite different spectrum demands!

By 1950 it must have been well known that Band III was going to start up "real soon now", but the 7 MHz channel spacing was good there too.

Only by the 1960's with UHF being the next big thing did the CCIR develop an 8 MHz channel-spacing concept. At least this time they got it more right, dictating the vision carrier frequencies for all the UHF channels. They *should* have done that (and enforced it) on VHF, but didn't. Maybe they realised that with Britain and France running non-625 systems on VHF, enforcing standardised vision carrier frequencies was impossible. Many countries did choose to use the "E" channels, but not all.

[Synchrodyne]

Indeed that appears to have been the case, although I have never come across any literature or documents that verify it.

There must have been a constituency in favour of a 6 MHz channel, following the American lead, and that accounted for Philips’ development of its 567-line system, which it considered to be the optimum (amongst nnn/50 systems) for a 6 MHz channel. But that was effectively dead-on-arrival. 567 lines in Western Europe against 625 lines in Eastern Europe just would not have flown. One may also infer that 625/50 in a 6 MHz channel evoked technical discomfort in the minds of many European participants.

But the channel count gain with 7 MHz as compared with 8 MHz was quite small. Band I (41 to 68 MHz) allowed just three 7 MHz channels (47 to 68 MHz) and three 8 MHz channels (44 to 68 MHz). Nevertheless, channel E1 was also mapped out as covering 40 to 47 MHz, initially not usable because it dropped below the Band I lower limit of 41 MHz, but on the basis that at some time in the future, Band I might be extended downwards, then allowing its use. It did not occur to me before, but I wonder if this was a bit of “window-dressing” to support the idea that the 7 MHz choice did indeed allow an additional channel.

Almost certainly the use of Band III in Western Europe must have been in view in the late 1940s, if nothing else by virtue of the American precedent. (The UK was I think behind the curve in terms of Band III use, although it benefitted from the fact that by the time it got there, all of the required development work on low noise, high gain front ends had been done.) The French used Band III from the start for 819 lines. Channel F8A (174 to 188 MHz) was part of the original channelling plan (as Channel I, I think) that was abandoned early on for the tête-bêche system for which Band III was extended downwards to 162 MHz.

The original Band III (174 to 216 MHz) accommodated seven 7 MHz channels and six 8 MHz channels. The subsequent upward extensions to 223 and 230 MHz did not change the relativity, and the 7 MHz advantage remained at one channel.

One wonders whether much consideration was given to the vestigial sideband size in the early days. It may not have been that simple, but it does look as if the Russians simply copied over the NTSC (I) choice of 0.75 MHz to its 625-line system, and that this was again used for the CCIR 615-line system. The French chose 2.0 MHz for the 819-line system, so that was presumably the result of technical deliberation. Then in the UHF era the matter was evidently revisited, as we saw 1.25 MHz adopted for 625 lines by the British for System I, the French for Systems K’ and L, and by I don’t-know-who (but maybe Belgium) for System G. Which of these preceded the others is not clear.

Now returning to 405-line parameters, is there a record as to when the decision was made to adopt the vestigial sideband system for future transmitters, and the basis for the choice of 0.75 MHz, which also followed the NTSC (I) precedent. Clearly it was done before the Sutton Coldfield transmitter was opened in 1949.

Cheers,

[Synchrodyne]

Originally Posted by Synchrodyne

The original Band III (174 to 216 MHz) accommodated seven 7 MHz channels and six 8 MHz channels.

Correction!

It was six 7 MHz channels and five 8 Mhz channels.

Cheers,

[SwanseaSteve]

Originally Posted by Synchrodyne

567 lines in Western Europe against 625 lines in Eastern Europe just would not have flown. One may also infer that 625/50 in a 6 MHz channel evoked technical discomfort in the minds of many European participants.

Too right! Back in those immediate pre-Cold War days, the idea of having "the Russkies" running a TV system with more of anything compared with "ours" would have been a non-starter.

Congrats to the CCIR for not just kneee jerking and trying to outdo "the Russkies", but to join in with their standard (albeit with slightly modified transmission parameters). After all, their standard had been transmitted for about two years before Phillips Labs's first demo of what was to became System B in about 1950.

Originally Posted by Synchrodyne

But the channel count gain with 7 MHz as compared with 8 MHz was quite small. Band I (41 to 68 MHz) allowed just three 7 MHz channels (47 to 68 MHz) and three 8 MHz channels (44 to 68 MHz). Nevertheless, channel E1 was also mapped out as covering 40 to 47 MHz, initially not usable because it dropped below the Band I lower limit of 41 MHz, but on the basis that at some time in the future, Band I might be extended downwards, then allowing its use. It did not occur to me before, but I wonder if this was a bit of “window-dressing” to support the idea that the 7 MHz choice did indeed allow an additional channel.

The CCIR could easily have persuaded the WARC (or whoever) to tweak band I. They just never got round to it, hence the odd missing channel 1, and erratic use of channel 1A and 2A.

Originally Posted by Synchrodyne

One wonders whether much consideration was given to the vestigial sideband size in the early days. It may not have been that simple, but it does look as if the Russians simply copied over the NTSC (I) choice of 0.75 MHz to its 625-line system, and that this was again used for the CCIR 615-line system. The French chose 2.0 MHz for the 819-line system, so that was presumably the result of technical deliberation. Then in the UHF era the matter was evidently revisited, as we saw 1.25 MHz adopted for 625 lines by the British for System I, the French for Systems K’ and L, and by I don’t-know-who (but maybe Belgium) for System G. Which of these preceded the others is not clear.

The Russians obviously wanted as much video bandwidth as they could get into the channel. So they went for 8 MHz spacing, with a sound carrier at 6.5MHz and with a mighty 6.0MHz of video bandwidth. That pretty much forced them to adopt 0.75MHz for the vestigial sideband or they'd have had bad adjacent vision on sound, and (in the other direction) adjacent sound on vision - which might have been worse.

I've read somewhere that the BBC's research in the 1950's favoured vestigial sideband width over full-on video bandwidth (to a point) and this influenced the design of System I to have 1.25MHz vestigial at the cost (compared with system D) of 0.5MHz at the top end of the vision signal. The BBC's argument (as I remember it) was that wider vestigial bandwidth gave better rendition of low-frequency picture elements, including cleaner sync pulses. The Irish were first to adopt System I, a full two years before the BBC, but there must have been a bit of cross-border argreement on transmission standards I suspect.

Notice that, compared with system A, system I had a proportionally narrower vestigial sideband!

Originally Posted by Synchrodyne

Now returning to 405-line parameters, is there a record as to when the decision was made to adopt the vestigial sideband system for future transmitters, and the basis for the choice of 0.75 MHz, which also followed the NTSC (I) precedent. Clearly it was done before the Sutton Coldfield transmitter was opened in 1949.

No knowledge here, sorry. It was probably just a straight pinch of the American design, and maybe even extended to obtaining actual sideband filters off the Americans with which to equip those 1st gen vestigial sideband transmitters. It would have made good sense.

[Karen O]

Very true - both 405 and 625 go straight from active video into the frame synch. sequence, presumably to get the frame retrace started as quickly as possible.

[Ray Cooper]

Originally Posted by Synchrodyne

Now returning to 405-line parameters, is there a record as to when the decision was made to adopt the vestigial sideband system for future transmitters, and the basis for the choice of 0.75 MHz, which also followed the NTSC (I) precedent. Clearly it was done before the Sutton Coldfield transmitter was opened in 1949

I've never seen a specific date that this decision was made, but I suspect immediately after the War. Before the war, strange beliefs were entertained about the possibilities of VHF propagation - at one point, it was believed that you could run a Birmingham station on the same channel as Alexandra Palace so long as you used horizontal polarisation instead of vertical, but experience of propagation earned the hard way during WWII put the tin hat on that idea. With the realisation that geographical separation wasn't going to work as well as they'd believed, and the simple fact that the only had five Band I channels to play with to cover the entire country (early service planning in 1946-47 was done on the basis of just four channels being available, but the scheme simply couldn't be made to work) the need to use Vestigial Sideband became over-riding.

With regard to the original Sutton Coldfield system, the sound/vision combiner was a direct pinch of an RCA design but built by Marconi's. Regrettably, it didn't perform all that well and had severe maintenance constraints (the fact that it had to be mounted at the top of the mast rather than at ground level in a nice warm building didn't help), so the original was replaced in the early 'fifties by a Marconi design and build Maxwell combiner.

With regard to the VSB filter, this seems to have been built to a design by E.C. Cork of the EMI Research Laboratories. The operation of this filter was quite satisfactory and it survived until the end of 405 in 1985, when it was taken away by the scrap merchants (mostly copper and brass, and weighed about 900 pounds)...

[Synchrodyne]

Originally Posted by SwanseaSteve

The Russians obviously wanted as much video bandwidth as they could get into the channel. So they went for 8 MHz spacing, with a sound carrier at 6.5MHz and with a mighty 6.0MHz of video bandwidth. That pretty much forced them to adopt 0.75MHz for the vestigial sideband or they'd have had bad adjacent vision on sound, and (in the other direction) adjacent sound on vision - which might have been worse.

As I understand it, the Russians chose 6 MHz video bandwidth in order to achieve generally the same definition as was available from 16 mm film.

Nevertheless, the French chose 6 MHz video bandwidth and a 1.25 MHz vestigial sideband for Systems L and K’, still in an 8 MHz channel. But then very narrow guard bands and even overlaps were established practice in France. System E had a nominal 14 MHz channel width, and the original Band III channelling plan was based upon this. But the definitive tête-bêche channelling had 13.15 MHz separation between successive vision carriers.

The ITU documents covering the outcomes of the 1961 Stockholm meeting note that at the time, the proponents of System K were considering extending the vestigial sideband to 1.25 MHz.

Originally Posted by SwanseaSteve

I've read somewhere that the BBC's research in the 1950's favoured vestigial sideband width over full-on video bandwidth (to a point) and this influenced the design of System I to have 1.25MHz vestigial at the cost (compared with system D) of 0.5MHz at the top end of the vision signal. The BBC's argument (as I remember it) was that wider vestigial bandwidth gave better rendition of low-frequency picture elements, including cleaner sync pulses. The Irish were first to adopt System I, a full two years before the BBC, but there must have been a bit of cross-border argreement on transmission standards I suspect.

Notice that, compared with system A, system I had a proportionally narrower vestigial sideband!

Yes, I have also read about the BBC’s findings in one or other books, although right now I cannot recall where. Curiously though, despite the wealth of BBC Research reports and like items now available on line, I have not found anything that deals with the development of the System I parameters. The work must have been done in the late 1950s, as the 1960 TAC report recommended what became System I for UK 625-line broadcasting, and for a while it was known – in the pages of Wireless World at least - as the TAC system.

The European UHF system letter designations, G through L, appear to have been assigned at or proximate to the 1961 Stockholm meeting, on the basis of increasing vision bandwidth as the primary determinant, with increasing vestigial sideband size as the tie-breaker.

At the time, the European UHF channel assignment planning assumed the use of System I for both Ireland and the UK albeit with the following provisos:

Ireland: No decision has been taken on the standards, including the number of lines, to be adopted for Bands IV and V in Ireland. But for planning purposes without prejudice to such decision, the parameters preferred are those shown in the Table under "Standard I”.

United Kingdom: The parameters preferred for use in the planning of a 625-line system; the standards, including the number of lines, to be adopted in the United Kingdom for Bands IV and V, have not yet been decided.

Also included as a footnote was:

“Ireland is providing for the possibility of using a Standard similar to Standard I in Bands I and III.”

At the time, systems G through L were being treated as being exclusive to UHF, which would explain the “similar to” wording used.

One imagines that the RTE and any other involved authority in Ireland had followed the UK developments and had concluded that adopting the same 625-line variant as was likely to be used in the UK was the prudent approach.

Originally Posted by Ray Cooper

With the realisation that geographical separation wasn't going to work as well as they'd believed, and the simple fact that the only had five Band I channels to play with to cover the entire country (early service planning in 1946-47 was done on the basis of just four channels being available, but the scheme simply couldn't be made to work) the need to use Vestigial Sideband became over-riding.

And even then the BBC planning had assumed the use of some Band III channels to achieve national coverage. I am not sure exactly when that idea was stopped; perhaps after the 1953 TAC Report, but it was then necessary to squeeze even more transmitters into Band I, really getting a quart into a pint pot. The situation reversed in the early 1960s, when the BBC was allowed some Band III channels, perhaps after it became apparent that ITA would not fully utilize the band.

Another 405-line question: did the aspect ratio change, from 5:4 to 4:3, occur before Sutton Coldfield came on line?

Cheers,

[SwanseaSteve]

Originally Posted by Synchrodyne

Also included as a footnote was:

“Ireland is providing for the possibility of using a Standard similar to Standard I in Bands I and III.”

That's interesting. I'd read that "System I" got its letter designation because it was the system initially proposed by Ireland for their new TV service. But the quote above kills that idea off, as it seems that letter designation was in use before Ireland or the UK had decided on anything much.

It's odd too. The BBC had decided not to proceed with 405-line colour by the late 1950's, planning for colour TV to use the "new standard" which I'd understood was well-known to be some sort of 625-line system.

For the UK to state at the CCIR 1961 meeting that they "hadn't decided on a line-scan structure" seems like a misdirection. Of course they knew! The BBC launched BBC2 on System I with 625 lines only three years later! The newsletters to manufacturers informing them of the build-parameters for the new generation TVs must have been going out in 1961 at the latest, along with procurement specs for transmitters, aerials, combiners, filters etc etc

[Synchrodyne]

I suppose that notwithstanding the development work done by the BBC and the 1960 TAC report, nothing was official until the 1962 Pilkington Report was presented, digested and acted upon by the Government of the day. Nevertheless, in 1961 the probability was surely very high that 625 lines UHF transmissions according to the System I parameters were in the near-term future. In this case there seems to have been a reasonably smooth pathway from the BBC work on 625-lines and UHF transmissions to implementation. This contrasts with the case for VHF sound broadcasting. The BBC had decided around 1945 that FM was the way to go, and wanted to use VHF-FM for the Third Programme from the start. The BREMA-promoted digression into VHF-AM evaluation was something of a delaying tactic, although not the only one.

That the System I designation was derived from I = Ireland might be an urban legend. I have never seen any hard evidence to support it. The CCIR designations G through L followed a logical pathway, and applying Occam’s razor it would appear that it was pure happenstance that the “TAC” system came out as “I”. Possibly the CCIR manipulated the classification so that it came out that way, but if so, then one would want to know why it did so for one country’s benefit and what that manipulation actually was. Or the burden of proof is on the proponents of I = Ireland, given that the circumstantial evidence suggests otherwise.

The VHF System classifications seemed to have had increasing line number as a primary determinant. Thus A was 405 lines; B, C and D were 625 lines; and E and F 819 lines.

B, C and D followed the pattern of increasing vision bandwidth, with negative/FM systems preceding otherwise similar positive/AM systems. E and F had a reversed pattern, in that F had much lower bandwidth than E. Here it would seem that E was chosen for the French 819-line system by its chronological precedence. Had country associations been a factor, then F might have been more appropriate for it. But as it happened, perhaps F for the Belgian 819-line system was F-for-fail.

M and N were evidently afterthoughts, not originally included because they were not used in Europe.

Cheers,

[Synchrodyne]

Re the 405-line waveform and its history, the attached item from Wireless World 1956 January records some of the changes made.

Cheers,

[Restoration73]

During the last years of System I in the UK, the vestigial sideband was reduced to 0.75MHz to reduce interference to the interleaved DVB transmissions - although I do not
know if this applied to all or just the main analogue transmitters.
Some viewers said this change was noticeable.

[Synchrodyne]

That was an interesting move when measured against the original rationale for System I. One supposes that protecting the future (digital broadcasts) was perceived as being more important than maintaining the integrity of the past (analogue broadcasts).

The System I parameters were chosen as being the best utilization of the 8 MHz channel, and better than Systems D/K. The loss from reducing the main vision sideband from 6.0 to 5.5 MHz was judged to be less than the gain from increasing the vestigial sideband from 0.75 to 1.25 MHz. The latter seemed to be predicated on two main advantages. One was that the resultant shallower Nyquist slope reduced vision carrier relative amplitude errors due to local oscillator drift, which had the effect of moving the vision carrier away from the -6 dB point on the IF bandpass curve. The other was that the shallower slope reduced phase errors – that had visible effect - in the vicinity of the vision carrier. Another outcome would have been smaller errors in the intercarrier, although probably only decremental. (The fundamental flaws in the basic intercarrier system were not fixed until QSS came along.)

The System I parameters were set in the days of distributed LC IF selectivity and diode demodulators, the latter of course producing distortion whenever they were presented with a signal with asymmetric sidebands. The VSB vision signal was asymmetric to start with and made more so by the receiver Nyquist slope.

One could perhaps argue that the march of technology diluted the original arguments for the 1.25 MHz vestigial sideband. The widespread adoption of AFC largely eliminated the need to allow for oscillator drift. Quasi-synchronous demodulation (from circa 1970 in consumer receivers) reduced errors in the vision channel (although not materially in the intercarrier channel, hence the development of QSS when multi-channel sound arrived). Then SAWFs (from circa 1977) allowed the vision IF channel amplitude and phase responses to be set more-or-less independently. Fully synchronous vision demodulation was a further improvement. It seemed to have been used more widely late in the analogue era, although consumer ICs suitable for its implantation had been around from the late 1970s.

But the fact that adequately good results could have been obtained from the available technology in the 1980s with a 0.75 MHz vestigial sideband would not have been in and of itself reason to change a long-established standard. And probably until near the end of production of such, System I-only analogue receivers were most likely fitted with SAWFs that followed the original System I standard in having a Nyquist slope that assumed a 1.25 MHz vestigial sideband. Multistandard receivers may have been different. I think that there was some tendency to use the same SAWF, with 0.75 MHz vestigial sideband, for multiple systems.

So when the change was made from 1.25 MHz to 0.75 MHz for the vestigial sideband, there would likely still have been a large population of analogue receivers that were designed for the former. On these the change could well have been visible; if not, then the original premise for the 1.25 MHz vestigial sideband would look a little shaky. The effect would have been to create a depression in the video response, around 6 dB maximum, in the 0.75 to 1.25 MHz range.

One wonders if, for example, the BBC adjusted all of its rebroadcast and monitoring receivers to compensate for the reduced vestigial sideband. The available on-line information for these indicates that it took great pains to get the Nyquist slope exactly right.

Getting back to the System I parameters and the trade-off between the main and vestigial sidebands as compared with Systems D/K, it is interesting to note that the French elected to have it both ways with Systems L (domestic) and K’ (Outré Mer). These combined a 1.25 MHz vestigial sideband with a 6.0 MHz main sideband. But then the French had their own precedent for using narrower or even negative guard bands. System E was originally designed for a 14 MHz channel, and three such were envisaged in Band III, of which what became F8A was the lowest. But the definitive tête-bêche channelling plan, which included Band I as well as Band III, was based upon 13.15 MHz separation between successive vision carriers. The System L channel was also subjected to destructive surgery to accommodate stereo/two-channel sound, with the main vision sideband trimmed to 5.1 MHz to accommodate a NICAM carrier at 5.85 MHz. I am not sure when this was, but I think that France might have been on the tail end of European countries in the implementation of a modern stereo/two-channel sound system. (Although actually it had been the first with two-channel (bilingual) sound with System E in Algeria in the 1950s.)

In this general vein, System H (used by Belgium, I think, amongst others) also had a 1.25 MHz vestigial sideband. But I wonder how many domestic receivers actually took advantage of this, given that they were also required to receive Systems B/G. Not many, I’d imagine. It would have been easier to use the same vision IF channel, based upon a 0.75 MHz vestigial sideband, for all of Systems B, C, F, G and H.

Cheers,

[Synchrodyne]

Regarding the System A vestigial sideband, the following comment was made in BBC Monograph #39, “Twenty-five Years of BBC Television”:

“The total band (1.5 Mc/s) occupied by the `Nyquist flank' in the receiver characteristic is a larger fraction of the video bandwidth than in any of the other television systems now in use, and the distortion due to vestigial-sideband working is therefore lower.”

That might have been a case of making a virtue out of happenstance or perhaps necessity. As discussed earlier in this thread, it does appear that the BBC simply copied the NTSC (I) precedent of a 0.75 MHz vestigial sideband, as did the Russians for their 625-line system. Quite why the NTSC chose 0.75 MHz is not completely clear, but it could have been connected with what was practicable at the transmitting end. Thus it may have represented a reasonable minimum. The preceding RMA standard (441/30) had been 0.5 MHz, though.

Cheers,