625-Line Television System Origins

[Synchrodyne]

This is more-or-less a sequel to the long ago and now closed thread: https://www.vintage-radio.net/forum/...ad.php?t=36027.

Subsequent occasional internet searches have found some additional information, although not yet a complete picture.

The origination of the 625-line television system is sometimes attributed to Walter Bruch in 1945, and then as being simply a 50 Hz derivative of the 1941 NTSC 525 line 60 Hz system. But it does seem that the Russians had by then pretty much defined their version of the 625/50 system. There is a paper, in Russian, at: http://625.625-net.ru/files/587/511/...895912fb8bb262.

Babel Fish translations of likely-looking excerpts indicates that the basic parameters, including 625 lines with 6 MHz video bandwidth in an 8 MHz channel had been established by 1944, apparently with the objective of matching 16 mm film definition, and that these were adopted as a standard in 1946, with regular broadcasts starting in 1948.

So that leaves open the possibility that the Russians and Germans (under Walter Bruch) worked completely independently on optimized 50 Hz TV systems, although in both cases informed by the NTSC system, and arrived at essentially the same conclusion.

But information at: http://www.radiomuseum.org/forum/45_...elevision.html puts a somewhat different complexion on it. The short biography of Walter Bruch therein includes the following:

“After WWII he worked until the beginning of 1946 under Soviet command in a former research laboratory in Berlin, where he contributed to the development of the 625 line television standard as well as various UHF projects. Later he was once again able to assume a leadership position in television development at Telefunken.”

Thus it appears that there was overlap and probably coordination between the Russian and German efforts on 625-line television, although it does not outrule the possibility that each group had independently arrived at more-or-less the same solution before Bruch was involved with the Russians.

Some confirmation that this might have been the case is found at: http://www.radiomuseum.org/forum/all...yclopedia.html.

The last-mentioned site in turn leads to: http://www.scheida.at/scheida/Televi...CCCP_UdSSR.htm.

A Babel Fish translation from German, without any editing, of an included note about Walter Bruch comes out as:

“Note: Which are to have received walter break with a Soviet officer the USSR as relief supplies during the war received a European standard adjustment to accomplish after the war the order, to American TV devices: 525 lines with 60 heart on the 50Hz reticulated system in Europe, which led finally (over Dutch Philips attempts with 567 lines had proceeded only from the channel width with 6Mhz) to 625 lines with 8 MHz channel width. And this again over Swiss Mr. Gerber to 7 MHz special solution for defeated Europe will should.”

From this one certainly gets the impression that Walter Bruch and his German group were thinking in terms of 6 MHz video bandwidth and 8 MHz channels, and that it was the work of Gerber of the CCIR who brought the reduction in video bandwidth to 5 MHz and channel width to 7 MHz. And perhaps that was a compromise to save bandwidth, so allowing some economies in receiver design. What is not apparent though is just when the work on and lobbying for the 7 MHz channel actually happened; maybe it was in the 1949-50 period.

Mention of the Philips 567-line system and its use of a 6 MHz channel is interesting. This seems to have been alternative approach to developing a 50 Hz NTSC derivative, and perhaps one in which the channel parameters were set first (i.e. the same as in the NTSC case) and then the “best-fit” number of lines derived with 50 Hz field rate as a given. The same process was used by the NTSC to determine the number of lines for its system, with 60 Hz field rate as a given. On the other hand in the 625-line case, just speculation, but perhaps the German work was done on the basis of more-or-less matching the NTSC line frequency, and working out the other parameters from there.

Anyway, the evidence at least points to the fact that what is/was probably the most common 625-line transmission system, that is CCIR System B, was somewhat compromised in terms of video bandwidth. So in that light it is not so surprising that British and French developers sought to do better when working on their respective 625 line transmission variants in the late 1950s.

Finally, by way of digression, there is an item in this previously mentioned article: http://www.radiomuseum.org/forum/45_...elevision.html that also speaks to another earlier and closed thread, https://www.vintage-radio.net/forum/...d=1#post317595. To quote:

“Since SECAM was originally oriented toward the 819 line b/w system used in France at that time (using a 8.37 MHz color subcarrier), and the fact that a suitable delay line to cache a full scan line was not available at a reasonable price, SECAM remained a laboratory curiosity for several years. Even in France, the state-owned TV station showed little interest, and as a result, no one paid much attention to this system outside of France.”

Cheers,

[emeritus]

My translation of the note is as follows:

Note:

After the war, Walter Bruch, with a Soviet officer, had the task of adapting American television apparatus which had been received by the USSR during the war as aid supplies, to a European standard: 525 lines at 60 Hz to the European 50 Hz power supply system, that eventually led (via the Dutch Philips research with 567 lines that was achieved with a channel width of only 6 MHz) to 625 lines with an 8 MHz channel width. And this further led to the 7MHz channel spacing for occupied Europe that was proposed by the Swiss Herr Gerber.

Our office used to have a booklet, published shortly after the war by the Custodian of Enemy Property, that gave details of otherwise unpublished wartime German research into televisions systems for the benefit of anyone who might be interested. While there were some outline proposals for higher definition systems than pre-war, I don't recall seeing 625 lines specifically mentioned. I don't remember the precise title and I fear the booklet must have got thrown out during an office relocation in the early 1990's.

[dominicbeesley]

Wow, so by the sounds of that, more digestible translation, 625 really was chosen to make modification of NTSC sets easier. i.e. the biggest hurdle is the LOPT so just choose the closest odd multiple number to the 15kHz line frequency....well I never!

[FERNSEH]

This Book "Fernsehen fur Alle" was published in 1951 and outlines the 625 line system that will be the norm in post-war Germany.
If you think about it, 625 lines is the most logical sytem of all. Can this system be attributed to one or two developers? Although the name "Gerber" turns up often.

DFWB.

[FERNSEH]

Originally Posted by dominicbeesley

so just choose the closest odd multiple number to the 15kHz line frequency....well I never!

Dead easy: 15,625Hz divided down by 5 X 5 X 5 X 5. = 50Hz for the field frequency.

DFWB.

[dominicbeesley]

Yes, but there are lots of other multiples of small odd integers that would work. What has alwas intrigued me is why they chose ones that came out with a line frequency so close to NTSC. If you were starting this as a paper exercise there is no reason to. However now we know the people working on the system had a load of old NTSC equipment to play with it becomes obvious whey they'd like to keep a roughly similar line frequency...

[emeritus]

Glad the translation was useful. For me, it begs the question as to why the US authorities would have shipped a consignment of their TVs to Russia as part of the aid programme: hardly the most useful thing to assist fighting the Nazis!

[dominicbeesley]

From earlytelevision.org:

Russia: According to an article in the British Vintage Wireless Society Bulletin, television was used to help in the air defense of Leningrad in 1941. A radio detection system called Redut could locate aircraft within 100 kilometers, and displayed their position on an electronic screen. A television camera was located above the screen, the the picture was transmitted to receivers located around Leningrad at anti aircraft sites. The article also claims that the system used the same frequency as an experimental London station used to transmit TV programs to hospitals, and that the receivers in Leningrad received pictures from that transmitter. The Russians were known for exaggerating their technical achievements (see the story of the TK-1 and the 17TH-1), so it is possible that this story is all or partially fiction.

Also intriguing but going slightly off topic but still pertinent to the American influence in Russian development of TV: http://www.earlytelevision.org/rca_rr-359_russia.html

[emeritus]

Just spent a wet lunchtime exploring the threads and using "Reverso" to translate from the original Russian. The following may be of interest:

From http://www.radiomuseum.org/forum/all...yclopedia.html

a Post points to http://www.chipinfo.ru/literature/ra...08/p24_25.html

from which the following comes:

In the beginning of 1941 on the МТЦ and ОЛТЦ commenced reconstruction of the equipment to standard 441 line. However his realisation was prevented by Great Domestic war. Television broadcasting in the country was terminated. After war specialists counted up that to continue reconstruction of the МТЦ begun earlier does not make sense: the fact is that Americans as early as 1941 developed standard to 525 lines and in 1943 broadcasting was begun on it. Moreover in 1944 at us the project of the standard to 625 lines was developed. Technically ОЛТЦ equipment was easier to modernise to standard to 441 line. Therefore television center restored by Vniityj in Leningrad went on the air November 7 1947 [5], and regular broadcasting began August 18 1948 [17].

In October 1945 USSR government accepted resolution of reconstruction of the МТЦ to the standard of 625 lines pursuant to 1944 standard project Experienced equipment was developed by Special design office in situated near Moscow settlement of Fryazino (chief А. А. Fedorov) together with German specialists (team leader E. Gunther). In this work actively as well the chief of the АСБ МТЦ М. И. Кривошеев [M E Krevoshcheev] participated.

World's first television center to the standard of 625 lines had five studio chambers and four telemovie cameras. In studio chambers with optical view-finders they were used domestic ikonoscope Whether- 1 with opportunity of transition hereinafter to more sensitive tubes - superikonoscope. Sound equipment was developed by the Institute specialists of radio reception and acoustics (ИРПА, Leningrad, chief Н. С. Kupriyanov). The МТЦ equipment provided opportunity of transition to two-program broadcasting [18].



On the fig. 2 the schedule of dependence of the number of различимых vertical lines from videochannel pass band width for different European modifications of the standard to 625 lines is presented (at q=0,82 number N=105 fmax, where fmax is in megahertz). Videochannel band reduction on 1 MHZ in standards B, G lowers resolution horizontally N by 20% on comparison with standards D, K, L, but at the same time she remains higher, than on vertical.

No mention of Herr Bruch!

[tubesrule]

At the convention at the ETF last weekend, Dr. George Lemaster gave a presentation on the early development of color television. This came about due to the discovery of the only known CPA prototype television found to-date. The set dates from 1951.

Part of his presentation centered on Hazeltine Labs who were crucial in the development of compatible color television. A few interesting excerpts from his presentation in relation to PAL follow:

--------------------------
Hazeltine originally called CPA “Oscillating Color Sequence” (OCS)

Described at Hazeltine in Sept 1950

Tested operation at Hazeltine January 1951

Loughlin patent 2943142 filed 22 Jan 1951

Hazeltine CPA patent 2943142 “Color Television System” filed 22 Jan 1951.

Oscillating (‘swinging’) burst described in Hazeltine patent 2754356 (24 Apr 1952)

Line Rate CPA was evaluated at Hazeltine and at RCA
- delay line averaging in a receiver was not considered practical
- for 525/60, experimentally determined that for reasonable phase errors the spurious flicker of CPAF was more tolerable than line crawl pattern of CPAL

“Hazeltine engineering memo submitted to panel 13 of NTSC in 1951 proposing a
signal for field test was written as if either field rate or line rate might be proposed.”

“An early version of this memo exists with line rate written in, but just before submission to NTSC we decided in favor of recommending field rate."

Hazeltine then recommended CPAF (field rate CPA) to the NTSC

"PAL operates using a principle invented in the U.S.A. in 1950, called Color Phase Alternation. With color phase alternation the color phase sequence of the chrominance signal is periodically reversed so that phase errors represent opposite hue errors during the successive periods.“
Louglin, “The PAL Color Televsion System”, IEEE, 1966

Due to cost of a 1-H delay, Line Rate CPA was not considered further by NTSC -1953. (HCD, p21-1)

“Patents filed overseas require an annual maintenance fee. So, four or five years after the NTSC decision, while Europe was playing with investigating color TV, Hazeltine decided to drop European CPA patents”. (HCD, p21-1)

“Hazeltine’s European patents would have clearly covered PAL” (HCD, p21-1)*

1965 Walter Bruch at Telefunken announces “Phase Alternating Line” system.

1966 Bruch’s IEEE paper “The PAL Color Television Transmission System”

No references to US CPA work in Bruch’s 1966 IEEE paper

Only one minor reference to CPA mentioned by Bruch in Telefunken papers on PAL
--------------------------

It can be seen that two key PAL innovations, Color Phase Alternation at line rate and swinging burst were first suggested at Hazeltine Labs in the early 1950's. Reading through the article http://www.radiomuseum.org/forum/45_...elevision.html it states that PAL came to Bruch while dreaming at the opera, yet the Hazeltine patents were well known at the time.

At the convention in 2009, Dr. Reinhard Bruch, Walter Bruch's son gave a presentation on the development of PAL. There was an interesting discussion on the role Hazetine Labs played. I believe this presentations is available on DVD as will be Dr. Lemaster's sometime in the future.


Not to go OT, but RCA supplied broadcast equipment and 75 receivers based on the RR-359A to Moscow in 1937. This was 343/60i equipment. The receivers were copied and became the Russian TK-1, although wildly varying estimates of how many were made seem to be floating around.

Darryl

[tubesrule]

The link in my prior response was incomplete. It should have been:
http://www.radiomuseum.org/forum/45_...elevision.html

Darryl

[emeritus]

Loughlin's patent US2943142 [ Filed Jan 22, 1951, but not published until June 28, 1960] is acknowledged as prior art in the German master patent DE 1252731, but no mention is made of the earlier Espenlaub patent US2754356.

The article refers to the idea of the implementation of the delay line having been thought of at the opera, and the claims of the master UK patent GB 1066798 do indeed include a restriction to a supersonic delay line. However, neither the claims of the German PAL master patent DE 1252731 or any of Bruch's other related German patents include this limitation.

[Synchrodyne]

Thanks for all of the additional research!

Probably the dots can be connected in several ways, but the picture emerging is one where the 625-line system and its initial transmission parameters originated in Russia. Clearly, much of the detail, such as negative modulation, FM sound, vestigial sideband (lower), inclusion of equalizing pulses, etc, was informed by the work of the NTSC, which in turn had been informed by the earlier RMA 441/60 proposal. It might be an open question as to whether the line number was chosen primarily to match the NTSC line frequency, but it could be that it was the logical starting point and was then found to provide adequate definition – and in a reasonably future-proof sense - when used with a 6 MHz video bandwidth. What is interesting is that the 525 line number of the NTSC system was a late decision, and a compromise number proposed by Fink that sat between the two camps, one wanting to stay with 441 lines, and the other wanting something a lot higher to take better advantage of the 4 MHz (later 4.2 MHz) video bandwidth available, bearing in mind that 2.5 MHz was originally seen as acceptable for 441/60. Fink had realized that the line number/video bandwidth trade-off head a reasonably wide range of acceptability, rather than a single point that was clearly better than all others.

So it would look as if Herr Bruch, in his immediate post-WWII work with the Russians, encountered a fully-developed 625-line system inclusive of transmission parameters, and perhaps saw the logic of adopting it for European use, particularly given that circuit techniques and components developed for NTSC receivers would be equally applicable.

Gerber’s role, heading up a CCIR subcommittee, then seems to have been one of obtaining a Western European consensus on adoption of the 625-line system, including the 7 MHz channel compromise. Whether or not he proposed the 7 MHz channel is unknown, but one assumes that he championed it.

Backward extrapolation from there suggests that the 6 MHz NTSC channel had created a strong precedent for Europe, and that there was reluctance to adopt the 8 MHz Russian channel, probably because it was seen as being expensive in bandwidth terms when the available spectrum looked constrained when weighed against the needs of national networks. Back-of-the-envelope calculation (using gross not net line numbers) shows that 5.0 MHz video bandwidth (requiring a 7 MHz channel) at 625/50 gives the same proportional increase in horizontal definition as vertical definition when using 4.2 MHz with 525/60 as a baseline, so that may have been an argument used by Gerber starting from an assumption that the NTSC system per se had an acceptable balance. In comparison the same calculation for the Philips 567-line system shows that it would require a video bandwidth essentially the same as NTSC for the same proportionality, so it does very much look as if that proposal was developed around the 6 MHz channel.

Nevertheless, if one uses the Blumlein/EMI 405-line system (with 3 MHz video bandwidth) as a starting point, then 625/50 would require something like 7 MHz video bandwidth for proportionality. (But did 405/50 start out with 3 MHz video bandwidth, or something a bit less, then “growing” to the 3 MHz that was available once 0.5 MHz became the acceptable guard band between highest vision sideband and sound carrier?)

Notwithstanding the CCIR apparent compromise on video bandwidth and channel width, in 1951, fairly soon after the CCIR had completed its deliberations, Argentina adopted another 625-line variant, in this case further compressed into an NTSC 6 MHz channel. This concatenation of events, which saw video bandwidth move downwards from 6.0 to 5.0 then to 4.0 (4.2) MHz seemed to confirm Fink’s viewpoint on the somewhat rubbery nature of the video bandwidth/line number tradeoff.

I have since found another couple of sites that tend to confirm the general story, although not decisively so, namely:

http://www.ebha.org/ebha2007/pdf/Gaillard.pdf
http://de.wikipedia.org/wiki/Fernsehnorm

Cheers,

[Synchrodyne]

It seems likely that as well as setting the basic 625-line vision parameters, the early Russian work might also have determined the FM sound channel details, with +/-50 kHz deviation and 50µs pre-emphasis.

NTSC TV sound started with +/- 75 kHz deviation and 100 µs pre-emphasis, the same as for US FM broadcasting at the time. Later the deviation was reduced to +/-25 kHz as being sufficient to equalize sound and vision transmitter coverage areas. And the pre-emphasis was reduced to 75 µs. When these changes occurred, and whether they happened at the same time I do not know. But perhaps it was circa 1945, when the channel frequencies were reworked. US FM broadcasting also changed to 75 µs pre-emphasis. This might have happened when the FM band was moved from 42 to 50 MHz up to 88 to 108 MHz. Tibbs and Johnstone mention early BBC FM tests with both 100 µs and 50 µs, with the latter giving the better results overall. The Russians could well have reached the same conclusion, but earlier on. And then CCIR accepted the Russian parameters without change for its Gerber system.

Cheers,

[Synchrodyne]

I have since found another article that looks at this question, at:

http://www.scheida.at/scheida/Televi...aus%20Russland

This is in German so I am able only to get a rough gist of it. But it does seem to confirm:

The Russian origins of 625/50

That line frequency was the link-point between 525/30 and 625/50

That the Philips 567/50 proposal was the logical 50 Hz counterpart to NTSC 525/60, using the same 6 MHz channel and respecting a general concern over available bandwidth for TV broadcasting.

That the role of Gerber in promulgating the CCIR 625/50 standard is not clear.

Nor is the exact rationale for choosing a 7 MHz channel for the CCIR system.

I suppose that as logical as the Philips 567/50 proposal might have been, it was unlikely that Western Europe would have chosen 567 lines when the USSR had already chosen 625 lines. That would be too visible a point of difference in the public eye whereas the difference in video bandwidth would escape general notice as a below-the-line technicality. And the Russian antecedent could be – and was – ignored given that 625/50, whichever way you look at it, was essentially informed by NTSC 525/60.

Regarding Gerber, perhaps it was simply a case of “being there”.

Cheers,

[emeritus]

An interesting article. The press release dated 11.08.1950 {page 6 of the PDF version} on the Dutch switchover from 567 to 625 lines in 1950 notes that it was going to be very much a DIY exercise:

"Television amateurs in Holland will shortly be informed that the experimental transmissions of the Dutch television transmitter will be changing from 567 to 625 lines. Such a change of the number of lines will of course require the modification of the approximately 400 television receivers that are presently in use. This modification to the new number of lines involves no great difficulties and, in Holland, will be carried out by the users of the equipment themselves because it has been established that 90% of all equipment that is presently in use was constructed and reconstructed [repaired/rebuilt] by amateurs. "

[ENGLISH VICTOR]

Fernseh Suggests that the 625 line rate of 15625Hz /5/5/5/5 = the field rate of 50Hz. Not so. Fernseh's formula results in the Frame rate of 25Hz. to obtain the field rate we need to take twice the 625 line rate which is of course 31,250Hz. and divide it by five four times to obtain 50Hz. I am sure this is in fact what David meant. Of course 5*5*5*5=625 what a coincidence.
Victor.

[dominicbeesley]

There _are_ 625 25 times a second...don't forget its interlaced so every 50th of a second you only get half the picture!

[ENGLISH VICTOR]

When discussing T.V. systems in detail it is a good idea to nominate fields and frames. To be specific two successive vertical scans or fields produce one frame. Interlacing can only be achieved when the total number of lines in a frame is an odd number. In simplistic terms this is because the vertical scans are derived from exactly twice the line frequency, this is how the half line offset is produced that is required for two to one interlacing. Interlaced scanning only produces a fully detailed picture on a stationary image, we tend not to see the reduced vertical definition on movement because the picture when moving denies us a stable reference point for comparison.
Victor.

[FERNSEH]

Originally Posted by ENGLISH VICTOR

Fernseh Suggests that the 625 line rate of 15625Hz /5/5/5/5 = the field rate of 50Hz. Not so. Fernseh's formula results in the Frame rate of 25Hz. to obtain the field rate we need to take twice the 625 line rate which is of course 31,250Hz. and divide it by five four times to obtain 50Hz. I am sure this is in fact what David meant. Of course 5*5*5*5=625 what a coincidence.
Victor.

Hi Victor,
yes, the twice line rate 2H of 31,250Hz divided by 5X5X5X5 yields 50hz, The 2H frequency is for the for the five broad field pulses and the equalising pulses. Further division by two creates the line frequency.
I hope that I have got it right.

DFWB.