[Sticky] Television Receiver Intermediate Frequencies

I now have the Wireless World December 1955 article on the BREMA TV IF recommendations. This does not refer to a specific BREMA document or publication, though. It includes a chart to show the possibilities and prohibitions, with the conclusion that a vision IF in the vicinity of 35.0 MHz was the best choice. The actual number chosen was 34.65 MHz, which put the fifth harmonic of the IF between channels 8 and 9.

Also interesting was an article by G.H. Russell in the April 1954 issue of WW making the case that with Band III TV imminent in the UK, it was appropriate to develop a standardized IF. The author came up with 35.25 MHz (vision), which was in general agreement with the later BREMA figure. The difference may be accounted for by the fact that the author considered that IF harmonics up to the 4th only needed to be taken account of. He did concede though that there was a body of opinion that the 5th harmonic could also be problematical, and indeed as it turned out BREMA was in this group.

So if nothing else, that is another little piece of System A history.

Russell also made the case for pan-European agreement, but then went on to note that some unilateral decisions had already been made, with the American and Italian examples quoted.

In the American case, the RMA number of 45.75 MHz is footnoted as having been recorded in the Electronics journal for November, 1950. So that gives us a definite date. I should infer that the REC 109-C revision of the pertinent RMA document followed this announcement. The previous REC 109-B issue is known to have covered FM and AM radio receiver IFs, but probably not TV IFs.

In Italy, there was a government decree that protected the 40 to 47 MHz as the TV IF band. This was footnoted as being recorded in Gazetta Ufficiale della Republica Italiana (Part 1), 1952 April 03. So the Italians were relatively early with a standard “high” IF. That an IF channel mostly within Band I could be used in Italy reflected the fact that the lowest Italian channel, IA, occupied 61 to 68 MHz, corresponding with E4. One wonders whether the choice of the 40 to 47 MHz IF channel was influenced by the American choice of 41 to 47 MHz.

Whether, within the 40 to 47 MHz channel, the actual IFs were defined is not recorded in the Russell article. One might expect 45.75 MHz vision, 40.25 MHz sound as the logical numbers following the usual System B channel pattern, albeit inverted. But the numbers used by Philips for its early Italian-channel turret tuner, namely 45.9 and 40.4 MHz, were still in-band.

That the Italian IF channel was government-decreed stands in contrast to the US and UK cases, where it was left to the appropriate trade associations to develop and make the recommendations. However, in both cases, the numbers so developed were then used by governmental organizations for channel assignment planning purposes. It may have been the same in France. At least the System L IFs were embodied in a SCART standard, and possibly that approach reached back to the System E IFs.

In both the American and Italian cases, the “high” IFs were to some extent pushing the technology. In 1950, a typical American TV tuner, with “low” IF output, would have used a 6AG5 pentode RF amplifier and a 6J6 double triode mixer-oscillator. The pentode RF amplifier was rather noisy at Band III, but was easy to use than quieter triodes, particularly in what was a mass-produced item of relative precision. RCA had seen the situation as being unsatisfactory back in 1948, and its development work culminated in the 6BQ7 cascode double triode RF amplifier announced early in 1951. This would have happened more-or-less independently of the move from “low” to “high” IF, which seems to have been an idea that came after 1948. I do not know the exact mixer history, but once the “high” IF was in sight, it must have become clear that the established 6J6 was not going to be an easy fit (looked at from a mass-production robustness viewpoint) with an IF that sat just below channel A2. Hence the pentode mixer, and RCA had its 6X8 triode-pentode available at about the same time as the 6BQ7 was released. But the pentode mixer could be used at Band III only if it were preceded by a high-gain, low-noise RF amplifier, so its deployment was essentially dependent upon the availability of the 6BQ7 and like valves.

In Europe, the corresponding valves, ECC84/PCC84 and ECF80/PCF80 did not arrive until mid-1953, which was around a year after the Italian TV IF decree. Perhaps Italian TV makers had access to American valves? Or possibly, they started with tuners using the typical EF80 + ECC81 combination; the greater separation between their standard IF and lowest channel as compared with the American case would have made that a somewhat easier.

Now to circle back to the previously mentioned Cyldon Teletuner TV.12; as advertised in WW 1953 August, it was stated to have an IF output in the range 40 to 47 MHz. Given that it was, at the time, aimed at the export market, it could have been that Cyldon picked an IF range that covered what may have been the only two national standards that were known to be established at the time, namely those in the USA (41 to 47 MHz) and Italy (40 to 47 MHz). I do not know exactly when the TV.12 was released, but the TV.5 Band I-only model was announced and advertised in WW 1953 June, so it was likely after that. The TV.5 came late in the Band I-only era, and one suspects that it did not find too many takers.

The Russell article made no mention of any other national TV IF standards or recommendations. Whilst one cannot be sure that this was because nonesuch existed when it was written (probably some months before it was published), it does seem likely that had any been established, they would have been mentioned. Still, pertinent information may not have been readily available in those days. Even today, in the “information age” obtaining the information required for this thread has not been easy. In the French case, the need for multichannel receivers probably arrived with the third transmitter, Strasbourg, channel F5, which started in October 1953 (according to WW 1953 December). Previously, Paris and Lille had both used what became channel F8A. The same WW item noted that the Lyons and Marseilles transmitters were scheduled to come on line in the second half of 1954. Conceivably the System E standard IFs (28.05 + 39.2 MHz) had been established by late 1953, in time for the Strasbourg transmitter start-up.

Talking of national standards, I have since found my copy of NZS 6551:1973, “Specification for Television Reception”. This stated that the CCIR IFs, 38.9 + 33.4 MHz, should be used in New Zealand as being the best compromise for the channels allocated in NZ. NZS 6651 was a revision (in anticipation of the arrival of colour) of the earlier NZS 1712:1963 (which I don’t have), which I should guess contained the same statement in respect of IFs.

Cheers,

Steve

In Italy, there was a government decree that protected the 40 to 47 MHz as the TV IF band. This was footnoted as being recorded in Gazetta Ufficiale della Republica Italiana (Part 1), 1952 April 03. So the Italians were relatively early with a standard “high” IF. That an IF channel mostly within Band I could be used in Italy reflected the fact that the lowest Italian channel, IA, occupied 61 to 68 MHz, corresponding with E4. One wonders whether the choice of the 40 to 47 MHz IF channel was influenced by the American choice of 41 to 47 MHz.

Well I need to correct that error. In fact it was Italian channel IB that covered 61 to 68 MHz. Channel IA ran 52.5 to 59.5 MHz, with the vision carrier at 53.75 MHz. So the Italian IF channel was certainly nudging its lowest Band I channel from the underside.

I was planning to attach the WW 195412 article on the BREMA IF choice, but I cannot make the file small enough. The main chart is a double-page spread, and that might be the reason. But I could reduce the individual page scans for the WW 195404 Russell article, so that is attached.

Cheers,

Well, by devious means I managed to reduce the WW 195412 BREMA article enough to post it:

(Irfanview would not get it down to the desired size, so I used Microsoft Office Picture Manager to get it down to around 250 kB, small enough for the old Microsoft Photo Manager to open, then used the latter to reduce it to a tad under 100 kB.)

Cheers,

Steve

Maybe it is time to put the information discovered to date into some kind of tabular form, and I’ll ponder that possibility.

Attached is the first draft of the tabulation, with the IFs presented in ascending vision frequency order. When I have as much information as I think that I am likely to find, which point I suspect that I am approaching, I’ll also include a second list by CCIR System.

Probably it is not very readable as attached; the primary document is a spreadsheet, from which I “printed” a .pdf, converted that into a .jpg, then reduced it.

Some very recently found information on late-era SAWFs has thrown up some more numbers, particularly in respect of multistandard receivers, but I’ll discuss those separately.

Cheers,

Steve

Courtesy of terrykc, I can now provide a very readable version of the tabulation of analogue TV receiver IFs.

Thanks much for your help, Terry.

Cheers,

Steve

Early on in this thread, I mentioned this document:

“Choice of Intermediate Frequencies for Domestic Television Receivers” (Union Europeenne de Radiodiffusion Tech 3062-E, April 1954).”

Although I have not since found a copy, I have recently found an article in Wireless World 1954 July that discusses this document, namely “Television I.F. Inquiry: Aspects of Receiver Design in Various Countries”, by G.H. Russell.

The EBU report was the outcome of a questionnaire distributed in December, 1952 to television receiver manufacturer’s organizations in various countries in 1952 December, and to which replies were received during 1953 and 1954. I have attached a copy of the article.

The additional information certainly requires changes in some of the assumptions that I had previously made.

Firstly, “low” IFs, around 25 MHz vision give or take, were used in the early days several European 625-line countries, including Italy.

Secondly, “high” IFs were also used in some, but not all of those countries in the early days, although all projected future use of “high” IFs.

Thirdly, although the use of 38.9 MHz was projected for future use by two countries, Germany and Netherlands, it was not reported as already being in use. As the dates of the submissions from the various participants are unknown beyond being in the 1953-54 period, one cannot be sure as to when the use of 38.9 MHz started. But certainly it was not the first “high” IF used for the 625-line system, even if one discounts the Italian exception.

Excluding the Italian case, the initial “high” IFs for the 625-line system were:

41.75 MHz Belgium, Denmark
39.75 MHz Denmark
39.5 MHz Sweden, Switzerland

Of these, 39.75 MHz was projected for future use in Denmark, and 39.5 MHz for future use in Switzerland.

Notwithstanding these plans, it would appear that the 38.9 MHz number adopted by Germany and Netherlands quickly became the standard number, possibly or probably with CCIR backing. Back-of-the envelope calculations show that amongst 38.9, 39.5, 39.75 and 41.75 MHz, 38.9 MHz was the most favourable in terms of positioning of the fifth harmonic of the IF, putting it in the guard band between the upper edge of the vision sidebands and the sound carrier of channel E7. So this may have been a factor in its selection. 41.75 MHz was the worst case, with the fifth IF harmonic falling on the channel E9 sound carrier. It was also just inside Band I, which was undesirable. Also of note is that the 39.5 MHz number associated with UK 625-line practice had a prior existence in Sweden and Switzerland.

Likely then this previously-mentioned article:

“W. Holm and W. Werner, “Choice of an intermediate frequency for television receivers to suit the C.C.I.R standard,” Funk und Ton 8, 1954.”

was pivotal in the 38.9 MHz story, and I should guess would explain why in the first instance, it was adopted in Germany and the Netherlands. Holm and Werner were both Philips engineers and technical writers as far as I know.

Turning to the Italian case, the “low” and “high” IFs of 25.75 and 45.75 MHz appear to have been drawn directly from American practice. As previously observed, 45.75 MHz fitted the Italian declared IF channel of 40 to 47 MHz, and it becomes evident that the 45.9 MHz number used by Philips was in fact a slight deviation from “standard”.

The Italians seemed to have realized reasonably early on that the unilateral declaration of a protected TV IF channel was not a complete answer. In Wireless World 1954 November it was reported that Italy had asked the CCIR to study the question, and that the above-referenced EBU document was submitted as a contribution to the enquiry. So somewhere there might be a pertinent CCIR document on the subject. This request by Italy may have marked the start of the process of its change from 45.75 to 38.9 MHz.

The French contribution to the EBU report evidently predated the adoption of standard IFs in that country. The typical numbers quoted include, as well as some in the “high” class, others that might be described as being “very high”, up to 80 MHz. One may rationalize the latter by recalling that the original French 819-line channelling plan involved Band III only, not Band I, so that there was no apparent need to keep the IF channel below Band I. The Band I channels, F2, F3 (not used) and F4 evidently arrived with the tête-bêche channelling scheme, whenever that was announced. The IF numbers quoted for France appear to be inconsistent in respect of the vision and sound carrier spacings, which should have been 11.15 MHz. Just possibly separate conversions were used for the vision and sound carriers, but that seems unlikely. Perhaps there was a data transposition glitch somewhere along the way.

Wireless World 1954 August included an article about a French TV receiver using printed circuits. It was described as a single-channel (evidently F8A, as used at Paris and Lille) superhet with oscillator low, although the actual IFs were not quoted. This receiver had a 12AT7 cascode RF amplifier and a 6X8 frequency changer, which looks to have been an unusual combination. RCA introduced the 6X8 triode pentode frequency changer in 1951 as the companion valve to its 6BQ7 cascode double triode, for use in TV VHF tuners where the IF was around 40 MHz. It developed the series cascode circuit because the original Wallman (shunt) cascode was not easy to use with variable tuning; rather it was better suited to fixed frequency band IF use. Applications for the 12AT7/ECC81 included shunt cascode amplifiers. In the case of this French TV receiver, its fixed, single channel nature might have been why 12AT7 was used.

Returning to the Wireless World 1954 July article, this also delineates the US UHF channel allocation guidelines predicated on the general use of the standard 45.75 MHz IF.

Stepping away from the EBU document to the Australian case, I have found a comment in a 1970 Australian publication to the effect that the (vision) IF would eventually be changed to 36.875 MHz (from 36.0 MHz) following an Australian Broadcasting Control Board recommendation. Unfortunately no further explanation was provided.

Back-of-the-envelope calculations suggest that the original 36.0 MHz might have been chosen in part to avoid having Band I channel local oscillator frequencies fall within some of the Band II channels. With a 38.9 MHz IF, the local oscillator frequencies for channels 1 and 2 would have fallen within the vision sidebands of channels 4 and 5 respectively. This conflict was avoided with 36.0 MHz. But it did happen that the local oscillator frequency for channel 5, 138.25 MHz, was right on the vision carrier for channel 5A, which was a later addition. Whether the change to 36.875 MHz was at least in part to alleviate this is unclear; it still resulted in the channel 5 local oscillator frequency falling not too far above the channel 5A vision carrier.

Cheers,

Steve

I have updated the tabulation to include all of the information available to date. In large part this update is based upon the Wireless World 1954 July article that in turn was based upon EBU Technical Document 3062.

Additional to my earlier comments, the French numbers given in that WW article may reasonably interpreted as referring to the 14 MHz IF “channel” limits, rather than to the IFs themselves. From these I have inferred probable actual IFs based upon System E channelling practice.

In my February 25 posting, I said: “Some very recently found information on late-era SAWFs has thrown up some more numbers, particularly in respect of multistandard receivers, but I’ll discuss those separately.”

In particular, there were dual-Nyquist slope SAWFs that had vision IFs at 38.9 and 33.9 MHz. The 38.9 MHz number accommodated Systems B, D, G, H, I, K, K’ and L. The 33.9 MHz number was for System L’, with oscillator high. Thus I have included 33.9 MHz in the list, as it seems likely that this would have had reasonably significant use for System L’. The dual-Nyquist SAWFs appeared to have had a rather steep slope at the 33.9 MHz end, perhaps to minimize incursion into the primary vision response. Insofar as the SAWF group delay response could be tailored separately, this was probably not a major issue, although it would have increased the errors with conventional quasi-synchronous demodulation. One has the impression that System L’ was treated as something as a second class citizen in multi-standard receivers.

Switched SAWFs for multistandard receivers throw up some other IF oddities that I’ll mention in a subsequent posting.

Some of the questions that remain unanswered include:

Exactly when did the CCIR promulgate the 38.9 MHz standard for System B? I am fairly sure that it was sometime in 1954.

When was the French 28.05 MHz standard for System E established? Most likely later in 1954 or 1955. It might have been a SCART standard, if SCART existed back then.

Why did Australia choose its 36.0 MHz number for System B, and why did it later on change to 36.875 MHz?

Why did Japan stay with its “low”, 26.75 MHz number for so long?

When did BREMA publish its 39.5 MHz number for UK System I? The system itself was recommended in the 1960 TAC report, so BREMA may have done its “homework” soon after that.

Origins and timings of the 38.0 and 37.0 MHz numbers used for Systems D/K in Russia/Eastern Europe and China respectively.

Cheers,

Steve

Now that we're digital (well most of us) is there a difference in the IF's for the different systems? I'm thinking of DVB, DVB T2, ATSC, ISDB ect. I know of two countries that use VHF as well as UHF for digital TV (Australia & USA) (there's probably others I don't know about) would that affect the choice of IF?

Strangely, I was looking at SAW filters for the IF in digital receivers recently and found a difference, albeit a small one, in the centre frequency for SAWs intended for Cable system* use and receivers for off-air reception ...

This didn't make any sense to me ...

* Interestingly, I came across SAW filters with switchable 7/8MHz bandwidths for continental UHF/VHF DTT use ...

Oh yeah Terry I'd forgot about the different (7/8 megs) channel widths at VHF & UHF, also I think Australia uses 7 megs channels at UHF so I presume that would complicate things. Then there's the 625 lines digital system N (6 megs I believe) to throw a spanner in the works.

As far as the Australian and American systems are concerned, it is merely a question of selecting the appropriate SAW to match the system. It is the switchable SAW for sets in countries that formerly used system B/G that I found interesting.

These are probably for use in CATV set top boxes as the European cable systems have inherited a large number of 7MHz channels in a continuous sequence up to 300MHz (32 channels). http://en.wikipedia.org/wiki/European_c ... requencies.

I believe that VHF has been dropped for DTT in Europe but that is not a choice open to the cable operators due the vast amount of bandwidth they would lose. This will be further complicated by the retention of a large number of analogue channels alongside the new digital ones. (Unlike UK cable networks which are now completely digital.)

Perhaps, in due course, they will re-engineer the entire frequency range as 8MHz channels ...

When SAWs were used for IF shaping in analogue sets, of course, every system required a different SAW.

The exception was again System B/G receivers where one SAW suited both systems. This, like its discrete component predecessors, merely needed extra traps to suit both channel spacings: no switching was necessary.

The attached short item from Wireless World 1955 October confirms the original Australian standard analogue TV IF numbers of 30.5 MHz sound, 36.0 MHz vision. Evidently these were established well before the start of TV transmissions in Australia. Whilst the rationale for the exact numbers chosen was not given, a reasonable inference is that they were a “best fit” to the Australian VHF channel plan, and for example better than the then-recently emerged European 625-line system standard numbers of 33.4 and 38.9 MHz.

Cheers,

Steve

The attached WW items are interesting in that they show that the consideration of and debate about a British standard TV IF go back to 1949, and that a "high" IF was also suggested back then.

Cheers,

Steve

By way of some more background on the French TV IF situation, the attached Wireless World summary of the ITU Stockholm 1952 May VHF meeting mentions both the tête-bêche channelling system and the French use of the 162-to-174 MHz band just below Band III.

So one may assume that the French had worked out the details of their tête-bêche channelling system before the Stockholm meeting. The earliest mention of planning for this meeting that I have is in WW 1952 January. On the other hand, what was apparently not done was the simultaneous development of accompanying standard IFs, at least judging by the French input to the EBU TV IF enquiry. Given that some of the IFs reported to be in use circa 1954 were relatively high and implied Band III-only receivers, I don’t think that we could outrule the possibility that the use of Band I channels might not have been part of the original tête-bêche plan. That question might be answered if we could find a copy of the ITU Stockholm 1952 report. Of course another possibility is that with Band III channels in the majority, some of the setmakers simply chose high IFs for what were single-channel Band III receivers.

The Belgian TV situation was also noted as a complication. The earliest record I can find of the decision to use its own 625- and 819-line variants is in WW 1952 April. It did not, though, have a major bearing on IF choices insofar as whatever eventuated for the CCIR 625-line standard would also be appropriate for use in Belgium, subject to accommodating the French 819-line requirements in multistandard receivers.

Cheers,

Steve

Some clues to early French practice are provided in the attached item from Wireless World 1951 November. Surprising is that there were some dual-standard 441/819 line receivers.

Unfortunately I don’t have the continuation page, but I think what was being said was that 441-line receivers were often of the TRF type (for channel F1), whereas 819-line receivers (at the time for channel F8A) were superhets. I think though that dual-standard receivers would likely have been superhets on both systems with the IF channel not overlapping channel F1. The only IF information I have for 441-line superhet receivers is in respect of the Philips TF390, 13 MHz vision and 9 MHz sound with oscillator low. I doubt though that 441-line IFs were ever standardized.

Cheers,

I recently came across the attached item in Wireless World 1963 April. It shows some preferred TV IF numbers as recorded at the CCIR 1963 February Geneva meeting.

In the main, it confirms the numbers already recorded in this posting.

Of note:

Italy was still using 45.75/40.25 MHz, so its transition to the European standard 38.9/32.4 MHz was yet to come.

The USSR was using 34.25/27.75 MHz. That implies that the 38.0/31.5 MHz combination came later, possibly with UHF, possibly with the advent of colour.

The 34.25/27.75 MHz combination was mentioned in an early posting in this thread as having been found at: http://www.scheida.at/scheida/Televisio ... raster.htm. So the CCIR listing provides confirmation that it was used.

Japan was still using 26.75/22.25 MHz, its transition to a “high” IF probably not contemplated at that time.

Not mentioned was the UK System I IF of 39.5/33/5 MHz, but I have still not found information as to exactly when it was established; somewhere in 1962 or perhaps early 1963 would be my estimate. On the other hand the French System L IF of 32.7/39.2 MHz was listed.

That the USSR 38.0/31.5 MHz combination was, as it were, a second iteration that most likely arrived in the 1960s means that it might well have postdated the Chinese 37.0/30.5 MHz numbers. The working assumption is that 37.0/30.5 MHz was adopted at the start of TV broadcasting in China.

Cheers,

Steve

An item that I have so far failed to pin down is exactly when BREMA announced the standard IFs, 39.5 MHz vision and 33.5 MHz sound, for System I in the UK.

The available evidence places it somewhere between the 1961 and 1962 UK annual radio shows.

The WW 1962 October report on the 1962 show noted that many of the setmakers were displaying dual-standard and convertible receivers, and the inference is that these were using the BREMA standard IFs. Certainly the highlighted commentary in the following page shows that the BREMA numbers were in place by then.

The WW 1961 October report on the 1961 show also referred to dual-standard and convertible receivers from the setmakers. It noted the difference between the “TAC” and “CCIR/Gerber” 625-line variants of the 625-line system and that CCIR/Gerber signals had been piped into the show. But also recorded was the fact that both Ekco and Pye had displayed dual-standard receivers intended to receive the TAC variant. Pye had used the CCIR standard vision IF of 38.9 MHz, with sound therefore at 32.9 MHz. Thus seems unlikely that 39.5 MHz was “in play” at the time.

Back to the 1962 show report, the commentary about the consequences of the 39.5 MHz choice are interesting. As it reads, it is suggested that with AFC to minimize oscillator drift, the European standard number of 38.9 MHz could have been used, in which case receiver image rejection requirements would have been much more relaxed. That the 39.5 MHz choice actually created some difficulties does I think lend some credence to the notion I mentioned in my original posting that it owed something to easing of IF strip design in dual-standard receivers, and was not based upon System I requirements alone.

The Pye receiver with 38.9 MHz 625-line vision IF had a video bandwidth of 4.25 MHz. On the face of it, this was rather low, given that 4.5 MHz was needed to match 405-line horizontal definition (at 3 MHz) bandwidth) and that the TAC parameters implied that the 5 MHz of the CCIR/Gerber system was not enough. But one may see where the 4.25 MHz came from in the IF passband diagram in the WW article. The basic IF passband was of the double Nyquist type, with 6 dB points at both 34.65 MHz (405-line vision IF) and 38.9 MHz (625-line vision IF). This no doubt simplified the required switching between the two systems, but also predetermined the 625 vision bandwidth. Possibly some of the setmakers favoured the double-Nyquist basic IF curve because of its simplicity. And thus perhaps the compromise to accommodate this option, but with more acceptable 625 video bandwidth was to move the 625 IF out to 39.5 MHz. This allowed a 4.85 MHz 625 vision bandwidth, hardly stellar, but somewhat better than 4.25 MHz. 39.5 MHz may have been seen as the practicable upper limit; anything higher would probably have placed the image too close to the (n+10) channel. Also, if the 625 IF “channel” were not to overlap channel B1, then 39.5 MHz was close to the upper limit in this regard as well.

The 53 dB image rejection requirement associated with the 39.5 MHz IF choice in turn demanded the use of four-gang UHF tuners, whereas European practice had been to use 3 gangs. So assuming that the receivers at the 1962 show had 4-gang tuners, the 39.5 MHz IF and 53 dB image rejection numbers must have been known at least a few months earlier to allow time for tuner design/redesign. As I understand it, Cyldon’s initial UHF tuner, the UT, was 3-gang, and I imagine that Philips also had 3-gang designs for European use.

Irish TV started System I transmissions on November 01, 1962. So if Irish 625-line and dual-standard receivers used the 39.5 MHz from the start, then it must have been known quite a bit in advance of this date to allow adequate initial production of receivers.

The System I parameters were essentially defined by the TAC in its 1960 report. The 1961 ITU Stockholm UHF meeting allocations were based upon the use of System I by both the UK and Ireland, although in both cases the system choices were noted as being provisional. But the go-ahead in the UK would not have come until appropriate Government actions following the Pilkington Report of 1962 June. So it would seem very likely that BREMA did its homework on the IF choice (and its consequences) in advance of the go-ahead, in anticipation that System I at UHF would come into use sooner or later.

Cheers,

Steve

As mentioned in the Radio Receiver Intermediate Frequencies thread, I have recently come across a couple of articles in “Radio News” magazine for 1947 which provide useful background information on American domestic receiver IF choices. This magazine is available at the excellent American Radio History site, at: http://www.americanradiohistory.com/Rad ... _Guide.htm.

The two articles are in fact part of a long series entitled “Practical Radio Course” by Alfred A. Ghirardi, and the two at interest here are: Part 53, in the 1947 May issue, and Part 56, in the 1947 November issue. Part 53 discusses IFs for AM, FM and TV receivers. Part 56 elaborates on FM IFs.

In respect of TV receiver IFs, it was stated that the RMA had recommended that the sound IF be in the range 21.25 to 21.9 MHz, and that the vision IF be 26.4 MHz, with oscillator high. This does not quite add up, as a sound IF range of 21.25 to 21.9 MHz implies a vision IF in the range 25.75 to 26.4 MHz, rather than 26.4 MHz alone. It was also noted that pre-1941 TV receivers generally used IFs of 8.25 MHz sound and 12.75 MHz vision. The standard IF range was said to (just) place the images outside of the TV VHF channel range, spanning 44 to 50 MHz (channel A1), 54 to72 MHz (channels A2 through A4) and 76 to 88 MHz (channels A5 and A6) for a total range of 44 MHz. The high-band channels A7 through A13 were not mentioned, but with a span of 42 MHz (174 to 216 MHz), the RMA IF was also just high enough to keep the images out-of-band.

Other contemporary sources allow some crosschecking. “Radio Craft” magazine ( http://www.americanradiohistory.com/Rad ... _Guide.htm), in its 1946 October issue, ran Part V of an article series “Television for Today” by Milton S. Kiver. This included comment on intermediate frequencies. It was noted that until quite recently, TV IFs were 8.25 MHz sound and 12.75 MHz vision, but that the RMA had recommended a change to 21.25 MHz sound and 25.75 MHz vision. As with the Radio News article, it was observed that these RMA numbers placed the image band outside of the 44 MHz band occupied by TV channels A1 through A6.

A different perspective though was provided in “Radio-Electronics” magazine ( http://www.americanradiohistory.com/Rad ... r_Page.htm) for 1950 April in an article “Television Equipment Standards” by Mathew Mandl. The author asserted that there was no such thing as a standard intermediate frequency in receivers, although most were in the same general range. This is confirmed in a tabulation of then-current receiver IFs. Excluding the outliers, sound IFs lay in the range 21.25 to 22.25 MHz, and vision IFs in the range 25.75 to 26.75 MHz. The outliers were in the 30 MHz range.

A later Radio-Electronics article, 1952 January issue, “44-MC I.F. Amplifiers for TV”, by David T. Armstrong, said differently, and it is easiest to quote the opening paragraphs:

“FM sound i.f. has climbed the stairway to the stars from 2.1 to 4.3 to 10.7 to 21.75, and now to 41.25 mc. Video i.f.'s have moved from 8.25 to 12.25 to 25.75, and now to 45.75.

“Practical field experience with the 24-mc i.f. has been exceptionally good except for the peculiar problems caused by spurious oscillator radiation. There have also been minor disadvantages, in-cluding direct i.f. interference from hams and from medical and industrial equipment, from powerful FM stations inducing image interference, and from the international short-wave distortions. It was mainly the problem of oscillator radiation that caused the RTMA to advocate the new 44-mc standard.”

My take on this is that most likely the RMA did recommend the “24 MHz” IF in the 1940s, perhaps quite soon after WWII. And that it recommended the narrow ranges, 21.25 to 21.9 MHz sound and 25.75 to 26.4 MHz vision, rather than specific numbers. But for convenience and brevity, some commentators referred only to the lower edge 21.25/25.75 MHz numbers. Receiver manufacturers mostly stayed within these ranges, or did not move far from them. As observed previously, the 22.25 and 26.75 MHz combination used by a small number of the American setmakers may well have informed Japanese practice.

The opening sentence of the 1950 article, in referring to FM IFs, appears to have conflated both the radio and TV sound cases.

The previously mentioned outliers in respect of the “24 MHz” era included the combinations 31.625/36.125 MHz and 32.8/37.3 MHz. The makers concerned evidently preferred a “high” IF, but they may have been limited as to how high they could go by the need to accommodate the then channel A1, 44 to 50 MHz, something that was not a factor when the RTMA (née RMA) set its “44 MHz” standard. Still, one may see these American 30-something MHz IFs as presaging later European practice, where “high” IFs generally had to keep below the 41 MHz lower edge of Band I.

Cheers,

Steve

When I was in the Army our artillery radios tuned from 23.5 to 38 Mhz & the infantry/tanks/cavalry ect from 36 to 60ish Mhz. We used to use around 37 Mhz to talk to the infantry if we didn't have an infantry set or it had conked out. What effect (if any) we had on the local TV's I never found out...

P.S. I got threatened with a charge for listening to 'top of the pops' on 41.5 Mhz on an infantry set when on exercise near Aldershot in the late 70's.

By way of background information on the original Australian standard TV receiver IFs, I found this comment in Amalgamated Wireless Valve (AWV) “Radiotronics” of 1956 December. This is available at: http://frank.pocnet.net/other/AWV_Radiotronics/ .

“The Australian Broadcasting Control Board has allotted a frequency spectrum for use by television receiver manufacturers for the i-f channel which will be kept as free from interfering signals as is possible. The recommendation is to use 36 Mc/s as the i-f vision carrier frequency and 30.5 Mc/s as the i-f sound carrier frequency.”

No further detail was provided, but then the article was primarily about TV IF amplifiers, and not about the IF choice itself. It does suggest though that in part, the selection was made on the basis of the IF channel, 30.5 MHz sound and 36.0 MHz vision, was and could be maintained relatively free of other activities. Perhaps the European standard of 33.4 and 38.9 MHz was not so favourably placed.

Cheers,

Steve