In the thread “Television Receiver Intermediate Frequencies”, at https://www.radios-tv.co.uk/community/black-white-tvs/television-receiver-intermediate-frequencies/ there has been some discussion of FM radio receiver intermediate frequencies, and in particular, the history of the 10.7 MHz number, which is at least a de facto worldwide standard.
This topic seems worthy of a separate thread, which I have entitled “Radio Receiver Intermediate Frequencies”, so as not to limit it to the FM case only.
Anyway, in respect of FM receivers, a look through the likely sources on hand does suggest that 10.7 MHz was adopted quite early on in respect of Band II reception, although there were some variations.
Rider & Ulsan (1) appears to be a good reference in terms of American practice. Therein is the commentary that a 2.1 MHz IF was used for the first FM receivers tuning the original 42 to 50 MHz band, but that this was moved up to 4.3 MHz to ensure that images were out-of-band. With the change to the 88 to 108 MHz band there was some use of 8.3 MHz, but as this allowed in-band images, 10.7 MHz was adopted and was expected to become an industry standard.
The change to the 88 to 108 MHz band was announced by the FCC in mid-1945, but evidently took a couple of years or so to implement; see: “FM Broadcasting Chronology” at: http://jeff560.tripod.com/chronofm.html .
One might reasonably infer that the first use of the 10.7 MHz IF was likely to have been somewhere in the 1946-1947 period.
Tibbs and Johnstone (2) simply stated that for the 88 to 108 MHz band, a 10.7 MHz IF would appear to be standard, although frequencies in the 8 to 8.5 MHz range had been used.
Langford-Smith (3) was also succinct: “Frequency modulation receivers covering the 40-50 Mc/s band generally use a 4.3 Mc/s i-f, and for the 88-108 Mc/s band they use 10.7 Mc/s. This latter value has been adopted as standard in the U.S.A., and in some other countries, for v-h-f receivers.” Perhaps more significantly though, in chapter 38 “Tables, Charts and Sundry Data”, it notes: ”The American RTMA has standardized the following intermediate frequencies (REC-109-B, March 1950): Standard broadcast receivers – either 260 or 455 Kc/s. V-H-F broadcast receivers – 10.7 Mc/s.” Also in that section is noted that the European Copenhagen Plan allocated 415 to 490 kHz and 510 to 525 kHz as IF bands.
In the UK it would seem that 10.7 MHz was well-entrenched by the time that manufacture of FM receivers started. That attached excerpt from Molloy & Hawker (4) would i think, capture most early British FM and FM-AM receivers, and one or two FM tuners as well, and shows a general adherence to 10.7 MHz.
The notable exception in that list is Bush, with 19.5 MHz for its VHF41 model. As 19.5 MHz was used as a TV sound IF for Band I-only receivers (corresponding vision IF being 16.0 MHz), it seems possible that Bush simply transposed this number, perhaps in order to make use of existing components or designs.
Not in that list, but also dating from the start of UK FM transmissions is the Leak Troughline tuner, with 12.5 MHz IF.
According to the Leak literature, the intermediate frequency was chosen as 12.5 MHz in preference to the usual value of 10.7 MHz, with the oscillator operating on the high side of the signal frequency. This ensured that the oscillator never operated in Band II, and so could not cause interference with other receivers; that no IF harmonics fell into Band II; and that there were no images in Band I.
Actually, the first two of those three were true only for a Band II range of 87.5 to 100 MHz. The tuning range of the original Troughline was quoted as 88 to 100 MHz. But I should not be too surprised if Leak, who was very export oriented, had offered an export model that covered 88 to 108 MHz. Anyway, one may see where the 12.5 MHz came from; it was simply the difference between the UK Band II limits, namely 87.5 and 100 MHz, which both happened to be multiples of their difference.
The Troughline II, Troughline 3 and Troughline Stereo all tuned 88 to 108 MHz, but retained the 12.5 MHz IF. Leak did amend its benefit claims to simply that of eliminating mutual interference with neighbouring FM and TV receivers. With the solid state Stereofetic 1969, Leak reverted to the standard 10.7 MHz, probably in part because this model used standard ceramic IF filters.
As far as I know, Armstrong, Chapman, Jason, Pye and Quad all used a 10.7 MHz IF for their early FM tuners, dating from 1955 or even earlier in some cases.
Another possible exception is Eddystone, which used 5.2 MHz for its 770R VHF receiver. Given that it was a VHF communications receiver, covering 19 to 165 MHz, then perhaps there is no good reason why 10.7 MHz might be expected. On the other hand, it did cover wideband broadcast FM, as well as narrow band (for the time) communications FM of up to ±15 kHz deviation. And there was some evidence that in the USA at least, 10.7 MHz had quite early on been adopted for non-broadcast VHF receivers. Certainly it did become an important number in VHF communications; by the end of the 1950s 10.7 MHz crystal filters were available for 50 and 25 kHz channelling, and I think in anticipation of 12.5 kHz channelling. That was a few years before ceramic filters became the norm in domestic FM receivers. Eddystone changed to the standard 10.7 MHz IF for its solid-state 990R VHF communications receiver, successor to the 770R. (The 990R appears to have been a tour-de-force in germanium pnp technology that appeared on the cusp of the silicon planar era, itself soon overtaken by the combination of mosfet front ends and IC-based IF strips.)
Anyway, a summation might be that 10.7 MHz became the standard (both actual in some territories and de facto elsewhere) for Band II FM receivers quite early on in their development, and before many countries started their respective FM broadcasting services. Nevertheless, there were certainly some interesting exceptions.
For the non-band II cases, as far as I know the 10.7 MHz IF was used in Japan, where the FM band was and is 76 to 90 MHz. I should guess that the OIRT countries used 10.7 MHz for their Band I FM broadcasts, except perhaps in the early days in Russia.
Cheers,
Steve P.
-
- (1) “FM Transmission and Reception”, John F. Rider and Seymour D. Uslan, Second Edition, November 1950.
(2) “Frequency Modulation Engineering”, Christopher E. Tibbs and G.G. Johnstone, Chapman and Hall, Second Edition 1956.
(3)
“Radio Designer’s Handbook”, F. Langford-Smith, Newnes, 1997 reprint of Fourth Edition 1953, ISBN 0 7506 3635 1.
(4) “Radio Servicing Pocket Book”, E. Molloy & J.P. Hawker, editors, Newnes, 1955.[/list:u:1tmzgbvo]
Absolutely agree on an excellent and informative post.....it's been made a sticky so it will stay where it rightly belongs.
Rich
Trevor and Rich, thanks very much for the kind words.
Sometimes it can be quite interesting to do a little light research on topics where there is a long-established conventional wisdom, particularly where the oft-repeated “what” is taken for granted to the extent that the “why” becomes somewhat hidden.
Anyway, in this case it is evident that the 10.7 MHz IF became a standard for Band II FM receivers quite early on in their history.
Why 10.7 MHz exactly, and not another number in the near vicinity may remain a mystery. Something just above 10 MHz was needed to avoid in-band images. Perhaps it was simply the case that 10.7 MHz was then a very quiet part of the HF spectrum, for example well removed from WWWV at 10.0 MHz.
I also thought to look in the Hallicrafters book (*) for some early examples, as listed below. In fact these add supporting evidence to what has been said before, nothing really contrary.
[list=]1. S-31 AM-FM tuner, 1940, which had FM coverage of 40 to 52 MHz with a 4.3 MHz FM IF.
2. S-36 VHF receiver, 1942, 27.8 to 143 MHz in 3 bands, AM, FM and CW with an IF of 5.25 MHz.
3. S-37 VHF receiver, 1944, 130 to 210 MHz single band, AM and FM, 16 MHz IF.
4. SX-42 receiver, 1946, MF, HF and VHF 27 to 110 MHz in 2 bands, AM, FM and CW, VHF IF 10.7 MHz.
5. SX-43 receiver, 1947, MF, HF and FM 86 to 109 MHz, FM IF 10.7 MHz.
6. S-47 receiver, 1947, MF, HF and FM 88 to 108 MHz, FM IF 10.7 MHz.
7. S-55 receiver, 1948, MF and FM 88 to 108 MHz, FM IF 21.75 MHz.
8. S-81 “Civic Patrol” receiver, FM, 152 to 173 MHz.
9. S-82 “Civic Patrol: receiver, FM, 30 to 50 MHz, FM IF 10.7 MHz.
Item (1) is an example of the 4.3 MHz IF associated with Band I FM in the USA, so that number foes back to the very early FM days.
Items (2) and (3) evidently predate any standardization for FM or VHF IFs, and could be ad hoc selections by Hallicrafters.
Items (4), (5) and (6) show early used of the 10.7 MHz number, going back to 1946, and in the case of (4), for wider VHF use and not just for the 88 to 108 MHz band. Possibly through the choice here was informed by what was best for the 88 to 108 MHz FM band.
Item (7) is a departure, with a 21.75 MHz FM. This was in the region of TV sound IFs of the time, nominally 21.25 MHz but with some variations around that number. So perhaps it was informed by concurrent TV practice. There were a few US TV-FM receivers in the 1940s that were based upon a common IF system for FM and TV sound; I suspect that they were pretty much eliminated by the widespread take-up of intercarrier sound from 1948 or thereabouts. It is interesting though as whilst I doubt that there is a direct link, it did incidentally presage the much later use of 21.4 MHz for non-broadcast VHF and UHF receivers.
Items (8) and (9) show the early use of 10.7 MHz for non-FM broadcast VHF receivers. The “Civic Patrol” units were said to be the forerunners of the later scanning receivers that seemed to be popular back in the 1980s.
Re the 21.4 MHz case, I am not sure when this came into general use for communications equipment. Presumably it arose because at UHF, better image rejection was required as compared with what could be achieved with 10.7 MHz, and simply doubling the existing standard achieved that.
In my earlier posting I mentioned the cases of the Eddystone 770R and 990R VHF receivers, and the fact that the latter adopted the standard 10.7 MHz. The corresponding UHF receivers were the 770S and 990S. The 770S was double conversion, with IFs of 50 and 5.2 MHz, the 5.2 MHz being the same as had been used for the (single conversion) 990R. So the 770S also looks very much like an ad hoc case in respect of IF selection. The 990S used single conversion with a 36.5 MHz IF. I suspect that in any case, the 21.4 MHz option was not widespread by the mid-1960s, if it had emerged at all. The 36.5 MHz looks to have been, and quite possibly was ad hoc. But it also happens to be the mean of the UK standard 625-line UHF TV IFs, 33.5 MHz for sound and 39.5 MHz for vision. That points to at least the possibility that the 36.5 MHz choice was informed by TV practice. Furthermore, the 990S had separate front ends for each of its two tuning ranges, 230 to 510 and 470 to 870 MHz. The circuits for each look very much like UK TV UHF tuner practice of the period, 4-gang with a pair of AF186 transistors respectively as RF amplifier and self-oscillating mixer. Perhaps that is exactly what they were, one standard and the other modified to cover a lower frequency range. (As an aside, I must admit that I was a bit surprised to see self-oscillating mixers used in a professional communications receiver.)
For the next generation, Eddystone used 21.4 MHz for the 1990R VHF receiver, which optionally tuned up to 500 MHz, as compared with the 240 MHz upper limit of the 990R. So that at least gives some indication of the point where moving up to 21.4 MHz from 10.7 MHz became advantageous. The 1990S UHF receiver retained 36.5 MHz number of the 990S, but had a second conversion to 21.4 MHz for FM only. The latter seems to have been to enable use of the CA3089 FM IF and demodulator IC, whose upper frequency limit was 20 MHz or so.
Cheers,
Steve
(*) “Radios by Hallicrafters”, Chuck Dachis, Schiffer, 1996, ISBN 0-88740-929-6
See Hans Knoll's article (retired Grundig Engineer, many leading designs in 1950s)
http://www.radiomuseum.org/forum/zwischenfrequenz_ihre_auswahl_und_besonderheiten.html?language_id=6&hl=de&sl=auto&tl=en&ie=utf-8&oe=utf- 8"> http://translate.google.com/translate?u ... 8&oe=utf-8
Apparently the EBU issued a report on radio receiver IFs, namely: “The Choice of Intermediate Frequency for Domestic Supersonic Heterodyne Receivers”, Document C.T. 45-E, 1952. But I cannot find it at the EBU website ( http://tech.ebu.ch/Jahia/site/tech/cach ... blications).
However, in the LF/MF case, I doubt that the EBU arrived at a single IF for pan-European use, but more likely discussed the trades-off and arrived a small selection of candidates. Whether it cover
ed the FM case one may only speculate, but in 1954 it could well have done so.
An interesting commentary is provided in the attached letter to the editor (from a BBC staffer) in Wireless World October, 1949.
Does anyone know whether BREMA ever made any pronouncements about radio receiver IFs in the same way as it did for UK TV receiver IFs?
Cheers,
Steve
With multiple conversion HF receivers, intermediate frequencies, particularly those for the first conversion, seem to have been determined more on an ad hoc basis according to individual manufacturer preferences. Even so, the “standard” numbers still appeared, such as 455 kHz for the final IF. No doubt the ready availability of ceramic, crystal and mechanical filters for 455 kHz pointed to this choice. An example from the 1980s was the JRC NRD525, which had a 70.455 MHz 1st IF and a 455 kHz 2nd IF. Curiously that model was current at a time when some receivers with MF-only AM sections, such as the Carver TX-11a tuner, had moved to 450 kHz, possibly because that was easier with synthesized tuning. I do not know whether the 450 kHz number has been recognized by RETMA as a standard in the same way that was done for 455 kHz.
The “FM” IF of 10.7 MHz was also found in HF receivers, either as a 1st or 2nd IF. An example from the 1970s was the McKay Dymek range, including the DR33C. These were triple conversion, with IFs of 30 MHz, 10.7 MHz and 455 kHz. I should imagine that the reasons for using 10.7 MHz included the fact that it was a “designated” IF, and so could be expected to be free of interference, and the availability of a good selection of crystal filters, as used for VHF RT applications, including standard bandwidths of ±3.75 kHz, ±7.5 kHz and ±15 kHz.
In some cases smaller manufacturers might have chosen the same IFs as had been used by their larger counterparts because of filter availability. For example, the Liniplex F2 high-quality HF broadcast receiver had a 35.4 MHz 1st IF using a Cathodeon crystal filter whose availability might well have resulted from Racal’s earlier choice of the same 1st IF for its landmark RA1772 professional receiver.
Another apparently “standard” IF number is 9 MHz for HF SSB receivers, single-conversion as far as I know, for which standard crystal filters were/are available. 9 MHz is mentioned in some of the Plessey literature for its SL600 series ICs, but I am not sure whether it was much used in commercial consumer or amateur level equipment. Also unknown is when it first appeared. In that sense it is like the 21.4 MHz number previously mentioned; both seemed to have slipped in quietly, and probably in the solid-state era.
Perhaps not surprisingly, some manufacturers, having chosen an ad hoc IF for a particular purpose, then continue to use it over a long period, and example being Eddystone and 720 kHz. This was first used for the 850 series of valved VLF/LF receivers in the 1960s (these tuned from 10 to 600 kHz), then in the EC10A2, which was a marine version of the EC10, and then again in the 1004, which was a marine version of the 1000 series. In the case of the 850, up-conversion was probably the simplest approach, and as the chosen 720 kHz number was in the gap between the marine LF and MF bands for the later receivers, it was probably seen as logical to use it, particularly as it was not too far removed from the 455 or 465 kHz of the parent designs in each case. The 1004 had ceramic IF filters; possibly they were made especially for Eddystone, although maybe there were others who also used this IF.
Prior to the advent of upconversion as a widespread technique, with a 1st IF say above 30 MHz, marine main receivers, which had to cover from 10 or 15 kHz to around 28 MHz, presented some difficulties as compared with general coverage receivers tuning upwards from around 500 kHz. Two examples come to mind, the Marconi Atalanta and the IMR 54. The former used a mix of double- and single-conversion, varying by band, with a 1st IF of 700 kHz and a 2nd IF/single-conversion IF of 85 kHz. The IMR 54, designed by Eddystone, had a dual-frequency IF strip that operated at 110 or 465 kHz. It was single-conversion on all bands, with 465 kHz IF on the bands above this frequency and 110 kHz on the lower bands except for that which included 110 kHz.
Thus duality, with one of two IFs selected according to tuning range, appeared again in a quite different context, namely the ICOM R7000 VHF/UHF receiver. This had upconversion to a 1st IF of 778.7 MHz for frequencies below 512 MHz, and to 266.7 MHz for frequencies above 512 MHz. Both 778.7 and 266.7 MHz were then down converted to a 2nd IF of 10.7 MHz, which was also the final IF for wideband FM, and then to a 3rd IF of 455 kHz for narrowband FM, AM and SSB. As an aside, a TV adaptor for the R7000 was also available that had a 10.7 MHz sound IF with a vision IF of either 15.2 MHz (Systems M,N) or 16.2 MHz (Systems B,G,H). If I have this right, the 778.7 MHz 1st IF had a SAW filter, perhaps suggesting that the same number was used by others, whilst a lumped L-C filter was used for 266.7 MHz, suggesting that this number was very much ad hoc.
Cheers,
Steve P.
I think I read somewhere that some radio makes had different AM IF frequencies for north & south UK, I presume to reduce interference, or did I dream it ?..
No I think you are right but it was much later in the evolution of radio and was to do with the Japanese using 455Khz IF in their pocket portables. Second harmonic used to beat with (what was then) the Home Service on 909Khz in the London and South East area. The radio's destined for London were retuned to 450 or 460Khz to minimise the whistle. Anyone in the South East who had a radio with a 455Khz IF also had a nice 1Khz whistle on 909Khz!
Rich
I think that it was both and old and a new problem, which was probably why there was never a single standard AM receiver IF in Europe. See post #6 above.
It was not unique to Europe and the UK, either. Before the change to 9 kHz channelling, Wellington, NZ had AM stations on 570, 660, 800, 980 and 1130 kHz, which made it a bit awkward for receivers with IFs at or close to 470 kHz.
Cheers,
I have found some background on the 9 MHz IF used in some HF SSB receivers, at: http://www.wb4hfn.com/DRAKE/DrakeArticl ... %20TR4.pdf.
It appears that Drake was an early user, back in the valve era in the 1960s. The rationale for 9 MHz was that it allowed a single VFO with 5.0 to 5.5 MHz range to cover both the 20 and 80 metre amateur bands, oscillator low for the 20 metre case and oscillator high for the 80 metre case. This looks to have been a case where an IF was chosen for receiver layout convenience, rather than from exhaustive analysis of the various potential interference modes. Evidently though there were no major issues with 9 MHz. Then I suppose that once the crystal filter makers started offering 9 MHz SSB filters, there was incentive for other receiver and transceiver makers to adopt this IF.
ICOM used 9.0115 MHz in its R-70 and R-71 HF receivers of the 1980s. Presumably the slight offset suited its complex quadruple conversion scheme and synthesis approach. It is maybe like the variations around 455 kHz that one finds for regular AM receivers.
Changing direction, the American RTMA Standard REC-109-B mentioned in an earlier post now exists as ANSI/CEA 109-D, “Intermediate Frequencies for Entertainment Receivers”. I do not have a copy, as it is not available as a free download; one has to pay. But I have attached an excerpt of the available preview.
Cheers,
Steve
Some great research there!
Yaesu used 8.9MHz single conversion on many models, all the long lived F101 series and FT901 etc. They used 5 to 5.5 MHz VFO. Actually very many models 1960s to 1980s used between 4 and 7 for VHF. An IF and VFO both outside the normally received range is desirable. Hence 1.6MHz on Military sets (below 1.7MHz to 30MHz) which still allowed reception on MW & LW on many too (usually Transmission only above 1.5MHz).
Later on Dual conversion IF is chosen above all SW bands, usually between 68 & 70MHz (originally only used in Eastern Europe for FM broadcast) which is why so few Amateur models are suitable for 4m / 70MHz Amateur Band even if they do all others 1.8MHz up to 440MHz.
Later 1.2GHz becomes a 1st IF on many Cable TV modems, Cable setboxes, scanners, communications receivers/tranceivers etc and of course 950 to 1700 on Analogue Sat 1st IF later 950 to 2200 Analogue and Digital, the requirement being to be above TV (862Mhz), SDR (864 -868MHz), USA 915MHz & most 900MHz mobile.
Initially IF was as low as possible to have gain (127KHz), then as high as practical for Image rejection, but avoiding either a band to be received or likely strong local signals. Hence the approximately 39MHz vs 44MHz for Europe vs USA. Of course Cable Set boxes and Cable Modems being largely US originated designs use 44MHz IF if they use a conventional IF at all, as of 2005 Thomson was using so called "Zero IF" DSP but Arriss was using 44MHz 2nd IF to the DSP and 1.2GHz 1st IF.
480KHz and 70MHz are now both Amateur channels/bands, creating difficulty adapting older equipment and in case of 70MHz almost any suitable new or old equipment other than ex Fleet PMR.
Some SAW era NICAM PAL-I the NICAM IF SAW filter may be good for "home brew" 1st IF at 33.5MHz approx or at the PAL B/G NICAM IF. Many Crystal / SAW vendors still stock these and of course they can be salvaged from scrap VHS and TVs.
The 1983 November issue of the magazine “Radio & Electronics World included mention of HF receiver IFs at the end of an article entitled “Communications Building Blocks: Front Ends for All - Part I”.
A pertinent excerpt is:
“For an HF receiver, the choice of 9MHz as an IF arose, as far as I can see, by the accidental fact that a local oscillator tuning 5.0-5.5MHz works for both 80m and 20m in one go. But I cannot for the life of me imagine how 10.7MHz arose, apart from the fact that there doesn't appear to be a lot happening on that frequency. If any reader knows the secret of 10.7MHz, perhaps they would write and let me know.
“The use of 21.4MHz seems equally arbitrary, and then above 30MHz, things have really gone haywire with 34.5MHz, 35.4MHz, 40MHz, 45MHz, 48.005MHz, 70MHz... 48.055MHz can be derived from the R1000 oscillator and mix-down process, and seems to make some sense, while 45MHz is a nice round number though it is potentially a problem in the lands of Band 1 VHF TV.”
That confirms the rationale for the 9 MHz IF. I am puzzled by the author’s puzzlement about the use of 10.7 MHz in HF receivers though. Its origin as an FM broadcast receiver IF in the late 1940s, migration to VHF radiotelephone applications and then to HF receivers is fairly transparent, I think. That said, I am not sure quite when it arrived on the HF scene. I have a vague notion that it was used in some Japanese portables of the late 1970s and/or early 1980s, as reviewed in WRTH. Another apparently early application was in the McKay Dymek series of HF receivers, starting with the DR22 circa 1977. These were triple-conversion, with IFs of 30 MHz, 10.7 MHz and 455 kHz.
Cheers,
Steve
In connection with the use of 10.7 MHz as an IF in HF receivers, I overlooked the Plessey PR155, which dated from 1967. This was triple-conversion, with IFs of 37.3 MHz, 10.7 MHz and 100 kHz. One may suppose that the 100 kHz final IF was chosen in part because of the ready availability of bandpass filters for that frequency, bearing in mind that it (100 kHz) was also used for commercial point-to-point receivers. The need for an intermediate conversion might have stemmed from the chosen frequency selection method. Essentially the PR155 was of the pre-mixer type (like the Drake R4 series), albeit with the actual 1st oscillator controlled from the pre-mixer system by a PLL, but instead of multiple crystals providing the coarse frequency selection, it was done by a comb of 1 MHz harmonics, one of which (48 MHz, from about the middle of the comb) was used for the 2nd conversion, thus forming a drift cancelling system. I imagine it would have been difficult to use this process to go from a high (>30 MHz) 1st IF direct down to 100 kHz (or even to 455 kHz), and that once an intermediate IF was determined as necessary, 10.7 MHz was found to be suitable. (Possibly the 2nd IF needed to be a non-integer number of MHz to avoid any IF harmonics conflicting with the comb, and if so, this would have outruled 9 MHz.) In turn, the actial 1st IF numner was derived from this, effectively by subtraction from a suitable 1 MHz harmonic. The 1st IF (37.3 MHz) filter in the PR155 had a 12 kHz bandwidth, and switched variable selectivity was provided by a bank of 100 kHz 3rd IF filters. I do not know what bandwidth of 10.7 MHz IF filter was fitted, but I should guess 15 kHz. At the time “standard” 10.7 MHz crystal filters for VHF radiotelephone equipment came in 30, 15 and 7.5 kHz bandwidths, for nominally 50, 25 and 12.5 kHz channel separations.
At the time that the PR155 was designed, the 1.4 MHz final IF for commercial HF receivers, with a concomitant good range of crystal filters, might not have been in widespread use. The Racal RA1772 of 1974 seems to have been an early example with the 1.4 MHz final IF, and it was then used by Plessey in its later receivers such as the PR2250. More evidence that 1.4 MHz was not a contender for the late 1960s crop of solid-state commercial HF receivers is provided by the fact that for example the Eddystone EC958, Marconi Apollo and Marconi Hydrus all had 100 kHz final IFs.
Cheers,
Steve
The Plessey PR155 example was a case where the first and second IF numbers were interdependent and had to be related in a way that allowed the second local oscillator frequency to be an integer multiple of 1 MHz. And it looks as if the 1st IF of 37.3 MHz was derived after the 2nd IF of 10.7 MHz had been chosen. This kind of interdependency seems to have been a characteristic of other early solid-state upconversion HF receivers, as well. The Racal RA1772 had a 1st IF of 35.4 MHz and a 2nd IF of 1.4 MHz. Here it appears that the 1.4 MHz 2nd IF was chosen first, with the 35.4 MHz 1st IF being that which allowed a 34.0 MHz 2nd local oscillator frequency, which was derived as a harmonic of the synthesizer 1 MHz reference.
Then one may ask why 1.4 MHz for the final IF. Hitherto, 1.6 MHz, or just above it, had been common as the 1st IF in “conventional” dual-conversion receivers. That number was at the top end or just above the MF broadcast band (at least until its upward extension in the Americas) but still underneath the HF bands. On the other hand, 1.4 MHz was within the MF broadcast band. By the 1970s, receiver design may have advanced to the point where internal screening was good enough that an “in-band” final IF was not a problem. Racal’s reason for using 1.4 MHz, as stated in this article ( http://www.keith-snook.info/wireless-wo ... %20DCD.pdf), included the fact that at this frequency, crystal bandpass filters were reasonably compact. Presumably this was as compared with filters for say 100 kHz or 455 kHz.
The later Racal RA1792 had IFs of 40.455 MHz and 455 kHz. The 2nd local oscillator frequency was 40 MHz, again an integer multiple of 1 MHz. In this case the choice of 2nd IF might have been prescribed by a customer (government or quasi-government) specification.
1st and 2nd IF interdependency goes back to the valve era, at least to the Wadley Loop-based Racal RA17.
Cheers,
Steve
I have since ascertained that the 1.4 MHz final IF for professional/commercial HF receivers goes back at least to the Redifon R550/551 of circa 1969. Thereafter it was used by Racal, Plessey, Eddystone (1650) and Marconi (Oceanic), and probably by others. So it may be seen as a de facto standard, at least for professional receivers. However, it seemed not to influence consumer/amateur HF receiver practice, in which the established 455 Hz final IF remained widespread. I suspect that the relative cost of filters may have had something to do with this. The available 1.4 MHz filters were probably professionally oriented, with commensurate prices. Each of the above makers using the 1.4 MHz final IF had its own choice of 1st IF, though.
As something of a “curiosity corner”, I looked again at consumer-level VHF/UHF level communications receivers. The ICOM R7000 has already been mentioned. Contemporary with it (i.e. from circa 1986) but perhaps pitched at somewhat lower overall performance and price levels, were the AOR AR2002 and Yaesu FRG-9600. These looked as if they were hybrids between scanners and communications receivers. The AR2002 used upconversion to a 1st IF of 750 MHz for the RF range 25 to 550 MHz, followed by downconversion to a 2nd IF of 45.03 MHz. The RF range 800 to 1300 MHz was converted directly to 45.03 MHz. (There was no coverage from 550 to 800 MHz.) The final IF was 455 kHz for narrow FM and AM modes, and 5.5 MHz for the wide FM mode. The latter was certainly unusual, although it was a standard “IF” in that it was the intercarrier sound frequency for 625-line TV systems B, G and H. One wonders why AOR did not simply use 10.7 MHz. The AR2002 used a pair of Motorola MC3357 ICs, one each for the wide FM and narrow FM channels. The MC3357 was typically used to down-convert from 10.7 MHz to 455 kHz. Evidently the input side of this IC had no trouble with 45 MHz. But perhaps the IF amplifier and demodulator, normally used at 455 kHz, would not stretch to 10.7 MHz. If so, then AOR might have opted to use the highest usable frequency for which suitable standard ceramic IF filters were readily available, which would have pointed to one or other of the TV sound intercarriers (any of 4.5, 5.0, 5.5, 6.0 or 6.5 MHz). The MC3357 datasheet is mute on the subject of maximum operating frequencies.
The Yaesu FRG-9600 tuned continuously from 60 to 905 MHz. It had a 1st IF of 45.754 MHz. This was the American standard television (System M) vision IF of 45.75 MHz, with a small (4 kHz) offset. Thereafter the narrow FM/AM/SSB channel saw conversion to a 2nd IF of 10.7 MHz, with conversion to a 3rd IF of 455 kHz taking place within an MC3357 IC, which also provided narrow FM IF amplification and demodulation. The wide FM channel had conversion to a 2nd IF of 10.7 MHz within an MC3356 IC, which also handled IF amplification and demodulation. The MC3356 was said to be able to handle input frequencies up to 200 MHz, and a 2nd IF up to 50 MHz, although with an expected value of 10.7 MHz. That Yaesu chose an MC3356 for wide FM (at 10.7 MHz) when it used an MC3357 for narrow FM (at 455 kHz) lends some credence to the notion that the MC3357 would not handle 10.7 MHz.
The FRG-9600 had an optional (System M) internal video board that received the 1st IF and demodulated it to baseband video. It differed from the ICOM R7000 in that the video board did not also deliver accompanying audio; presumably the wide FM channel was used for that. Thus in IF terms, the sound carrier of a tuned-in TV channel would have been at 45.754 MHz, with the vision carrier thus at 50.254 MHz, making it non-standard. Hence it would appear that the 45.754 MHz 1st IF was not chosen because of the TV reception option, but for other reasons.
So the AR2002 used two of the standard IFs, namely 5.5 MHz and 455 kHz, but 5.5. MHz was used in an unusual manner. The FRG-9600 used three of the standard IFs, namely 45.75, 10.7 and 455 kHz, but 45.75 MHz was used in a very unusual way.
The AR2002 converted direct from around 45 MHz down to 455 kHz for all but the wide FM modes, in which aspect it followed prevailing HF receiver practice. On the other hand the FRG-9600 stepped through 10.7 MHz on the way from around 45 MHz to 455 kHz, which made it more like minority HF receiver practice.
A point about these receivers, including the ICOM R7000, is that for narrow FM, AM and where applicable, SSB, they used a final IF of 455 kHz, whereas professional VHF-UHF receivers of the same era were more likely to have used 10.7 MHz as the final IF, with a selection of crystal filters for various bandwidths and channel spacings. I should imagine that such 10.7 MHz filters were relatively costly as compared with their 455 kHz counterparts, hence the preference for another conversion to this frequency for consumer market equipment.
On 455 kHz itself, whilst it was an American IF number for AM broadcast receivers, with European practice having been to use somewhat different numbers, it did eventually become common as the final IF for consumer level European HF receivers. Examples were the Lowe HF-series, the AOR AR7030 and the Eddystone 1000 series (except 1004), EB35 Mk III and 1570/1590. Previously Eddystone had used 450 kHz for its valve receivers that had a single IF or a final IF in this range. I imagine that the ready availability of a wide range of standard filters, ceramic, crystal and mechanical, for 455 kHz was a motivator, as designs moved away from LC filters. There were, for example, 470 kHz ceramic filters, but I suspect that the range was more limited, and addressed predominantly the needs of LF/MF receivers.
Cheers,
Steve
The attached item from Wireless World 1951 September comments upon the effects on receiver IF of the implementation of the Copenhagen frequency allocation plan for European LF and MF broadcasting.
The previous commonly used IF of 465 kHz was no longer suitable for use in the UK, with both 422 and 470 kHz being recommended by BREMA. Perhaps that is why 470 kHz became reasonably popular. On the other hand, 470 kHz was not always a good choice for overseas territories that then used 10 kHz channelling in the MF band.
Cheers,
Steve
The previous commonly used IF of 465 kHz was no longer suitable for use in the UK, with both 422 and 470 kHz being recommended by BREMA.
That accounts for why the KB FB10 'toaster' set from 1950 (at least the version I have) uses what I thought was an odd IF of 422Khz! Mystery solved!
A (non-VHF) AM-related application for the 10.7 MHz IF was found in the Motorola MC13030 IC. This provided the basis for a dual-conversion AM receiver for car radio and other applications, covering the stages from 1st mixer to demodulator. It was designed for use primarily with an agc’d fet/bipolar cascode RF amplifier and an agc-controlled pin diode RF attenuator, two RF agc outputs being provided accordingly.
In general the 1st IF used with the MC13030 was not fixed, but was usually, although not always, set above the highest desired receive frequency. In practice, judging by the Motorola data, a 1st IF of 10.7 MHz was normally used for MF reception. This allowed the use of stock 10.7 MHz IF crystal and ceramic filters, and being a standard IF, was less likely to be troubled by interference. The 2nd IF was usually the standard 455 kHz. Motorola also illustrated a 27 MHz CB receiver with initial down conversion to 10.7 MHz, thence to 455 kHz. Something that has just occurred to me in writing this is that 27 MHz CB might well have been an early use for the 10.7 MHz IF in (non-VHF) AM equipment.
I imagine that single-conversion AM receivers with the 10.7 MHz IF are quite scarce, but one example was the BBC RC3/10 LF-MF monitoring receiver of the mid-1970s. This was of the crystal-controlled, pre-tuned type with PLL fully synchronous demodulation. It had a 12 kHz bandwidth 10.7 MHz IF filter, but as can be done with this kind of demodulation, additional selectivity was provided by a steep-cut low pass filter in the AF domain, to give an audio bandwidth of 5 kHz. From circa 1973, the BBC had been using its FL4/55 filter to restrict MF transmitted bandwidth to 5 kHz, so the RC3/10 was a match in that respect.
The RC3/10 was an interesting example of compact receiver design, facilitated by the use of the 10.7 MHz IF, at which frequency crystal filters were quite small. It had a bandpass tuned input, a 40673 dual-gate mosfet RF amplifier with a single-tuned interstage to a BF115 mixer and thence to the 10.7 MHz crystal filter. Thereafter it used standard TV IF ICs of the period in what was no doubt quite an economical combination. IF gain and agc generation, including delayed RF agc, was taken care of by a Motorola MC1352 vision IF IC. Demodulation was done by an MC1351, nominally a TV intercarrier FM sound IC, but equally capable of being used as an AM demodulator. The reference input came from the PLL section, which used a discrete oscillator in conjunction with another MC1351, which handled phase detection for the PLL. The audio processing circuitry was discrete; the balanced line output was provided by a pair of LM380N. (There was also auxiliary circuitry for a carrier fail alarm and a carrier output.)
Initial upconversion to tune across the LF and MF bands in one sweep was not new, though. Armstrong had used it in the AM tuner section of its 600-series hi-fi tuners and tuner-amplifiers of the early 1970s. In this case the 1st IF was 3.1 MHz, and the 2nd IF 455 kHz.
Still connected somewhat with 10.7 MHz is the attached (highlighted) item. It refers to early UK dual-standard (405/625) TV-FM combination receivers in which the FM IF was 6 MHz, that is the same as the TV intercarrier. The notable comment was “As no one seems to remember why 10.7Mc/s was chosen for the FM intermediate frequency, B.R.C. were a little wary of using the 6Mc/s intercarrier i.f. of television, but no snags have been found.” I suspect that that may have been somewhat tongue-in-cheek, but on the other hand, if the professionals who are supposed to know these things in fact did not, then no wonder it is difficult for a layman to build a post-facto history on the topic.
But there was evidently a precedent for using the intercarrier as an FM IF. According to the following item, 5.5 MHz was used in Germany for TV/FM receivers.
And this provides conformation that 10.7 MHz was standardized as the FM receiver IF in Germany at an early stage:
Cheers,
Steve
The previous commonly used IF of 465 kHz was no longer suitable for use in the UK, with both 422 and 470 kHz being recommended by BREMA.
So that's why KB used it on some of their sets... it seemed such an odd number - I never realised it was a recommended frequency.
It's a pity that the BREMA missives on intermediate frequencies (and other topics) appear to be unobtanium. I think that they'd be a useful resource.
Cheers,
Steve
