#2 - Analogue Television Modulation - NICAM Stereo Audio

[@altebander2767]

I've been to a technical tour of a transmitter site of Bayerischer Rundfunk (one of the few broadcasters in Germany to operate their own transmitters). At least back when I was there, tehy did feed their FM transmitters via NICAM over an E1 line, as you described. This must have been around 2008 or so. However back then they were in the process of moving from leased lines to their own microwave radio links. Bayerischer Rundfunk is in the unique position in Germany that they are able to see all their transmitter sites from one central location.

[@Richardincancale]

I was an enthusiastic early adopter of Nicam as soon as it reached Scotland. What was impressive was the amazing robustness of the signal - I could receive Nicam signals from remote transmitters even when the video was basically unviewable

[@digitalmediafan]

Amazing. I use to fo DXTV and here is Anglia with a weak signal but nicam was fine. Very clever when you think back to have digital audio with analogue video kzbin.info/www/bejne/eWTImHudlNFjg6ssi=AvlrAKVsjyX7iAav

[@cromulence]

I've been fascinated with NICAM since I was a kid - my brother rescued a Delta 150 decoder from his friends dad after he stopped using it - to me it was eye (well ear) opening, being able to hear high quality audio pumped through a hi-fi instead of a flat mono soundtrack from a TV. It's lovely to see another one on KZbin finally. It was a great device and pretty robust in what it could receive, even when the signal was poor. Really cool to see all of the kit in action - you might be interested in going down the D-MAC rabbit hole (look up British Sky Broadcasting, or BSB) - it uses a modified NICAM audio system that could do a lot of very clever stuff, including a mix of sound channels, up to 6 mono tracks, or a mix of stereo/mono/data.

[@marria01]

There were effectively three types of SIS used in the UK. Mono 'linear' SIS which used simple binary encoding on the video signal, and then RE-DSIS and Varian DSIS, which I believe both used quaternary encoding, but the data payload was different. Originally both the BBC and the IBA used the same mono SIS systems, which meant that they could interoperate easily. But with the advent of stereo (DSIS) the BBC used Varian and the IBA used RE. So they could no longer exchange video feeds natively, one side would always need to accommodate the other. I recall whenever Granada did any live broadcasts for the BBC out of their Quay Street studios (happened quite a lot back in the day), one TX path would go to NTA/DTA at TVC on BT's TVNET via the Tower and the other path would go on an analogue BT local end between Quay Street and the BBC's site at Oxford Road in Manchester via the local TV NSC. Once on site at Oxford Road, a RE DSIS decoder would convert the audio back to analogue and then fed into a 34 mbit ETSI coder as discreet baseband audio and video, or re-encoded into Varian DSIS and sent on a 140mbit REC-656 circuit and sent down to TVC via the BBC's (mostly) Energis supplied MBN. From what I can remember of the few Tx sites I've seen. The BBC TV network feeds would be handed over on site as Varian DSIS encoded composite video, which would then be decoded back to analogue audio, before being NICAM encoded and modulated on site. They'd still need analogue audio to feed the mono subcarrier, plus they'd have kit like programme fail detectors and the like that used to listen out for pilot tones on the audio. In fact ISTR the DSIS kit used for transmission (as opposed to contribution) had some sort of mod to allow the passing of 19kHz tones, not sure how, as that would most likely get knobbled by the nyquist filtering on the encoding stage. It did seem like the NICAM stuff was often a bolted on addition at these sites, as opposed to a significant reworking of the existing PIE kit. The NICAM rollout was 'way' before my time. ITV, on the other hand might've done the NICAM encoding at their playout centres, as I remember seeing the BT PIE bays on one site having separate video plus a pair of 'data' and 'clock' signals for the incoming network feed. There might even have been a discreet associated audio circuit for the mono subcarrier, but I'm not 100% sure. So it seemed like only the modulation was done on site. Regarding the NICAM 'data mode', I believe Channel 4 might've used it for network talkback for outside broadcasts. But I only read that somewhere, I've no direct knowledge of that.

[@mattstvbarn]

Thanks for the info. I actually have that equipment here, it will be demo'd in the next video. I'll even plug an RE-DSIS into a Varian and see what happens :) From my look at it the data payload are the same between the two, but the RE-DSIS packs it into a shorter burst than the Varian.

[@marria01]

@@mattstvbarn I never had the chance to investigate much since we only had a decoder on site and I was never around when those shows went out, which would be the only time we'd see any RE-DSIS on a circuit. It'll be interesting to see what you discover, but I'm going to assume it'll just refuse to decode anything!

[@debranchelowtone]

France had AM mono signal, never had stereo analog on television. Starting around 1995 some channels had NICAM stereo signal, and it was implemented slowly, transmitter by transmitter. Some channels like Canal+ never had NICAM audio. Arte which is a channel funded by France and also Germany, used this stereo system to transmit some programs with the two languages, just had to select language I or II. Once i watched a whole movie with the two languages in stereo on purpose, it was quite weird. Stereo televisors and VCRs were widely available and had NICAM decoder.

[@yahnsolo]

Yes, Arte but also M6 if I remember right was the first french tv chanel to advertise about it, being the only chanel that would broadcast regularly music clips and concerts, as well as simultaneous live duplex with FM stations. TF1 and France2 were also Nicam broadcasters. Good times when we were just a few audio freaks to start linking tv and Hi-Fi VHS to sound systems 😍

[@debranchelowtone]

@@yahnsolo Back then i had a music teacher who ordered a stereo TV and also VCR to record cable channel Muzzik, and Mezzo in stereo.

[@testcardsandmore1231]

When I started my TV DX'ing in 1989 I'm quite sure that TV2 Denmark were using NICAM. I didn't have a NICAM capable TV set but when tuned slightly off channel a characteristic pattern would appear on the screen, the NICAM carrier.

[@Bus2000]

NICAM was used for DMAC AND D2MAC as the main audio mainly on Satellite broadcast around the 90s

[@kentsmith7507]

TVNZ enabled the NICAM 1988 I believe. It was very gradual and it was mostly just mono on the stereo NICAM channels. I think it was originally just Shortland Street, some commercials and a late night music show that we're actual stereo for quite a long time. TV3 launch in 1989 was first to go full time stereo and was amazingly exciting time for a tech geek. Probably easier for TV3 S starting with new infrastructure rather than having to retrofit. I purchased a Mitsibishi E70 VCR and recall paying about $500 just to get the optional NICAM decoder module installed before any NZ broadcasting. Was pleased when it started getting some use. :)

[@mattstvbarn]

It was my understanding that it didn't officially launch until 1989? It might interest you know that the equipment we were using to code and modulate NICAM in NZ was the brown faced Philips PM5687 I demo'd in the video!

[@Pepek94]

In late 1980s NICAM has been introduced in Scandinavian countries. In early 1990s extended into Belgium. In Poland NICAM broadcasting has been commenced in 1994 with trials and went regular in 1997, and switched-off in 2013 on ATT. PS: I had Samsung CRT TV receiver with Stereo NICAM from 2001 but in 2019 has been forced to move to my deceased grandparents in order to replace old worn out Philips Monaural CRT TV from ca. 1998. Samsung CRT TV still stands to this day and has not been used for 3 years.

[@MePeterNicholls]

BBC we’re using digital backhall for radio as far back as 1970

[@enojelly9452]

Marvelous videos, absolutely love them. One of my last projects was implementing both my own PAL generator and decoder from scratch (starting with the math, from first principles), both in hardware and software. So yeah, this is right up my alley. I trust you have a Sony PVM? It has a "delay" button that allows direct viewing of the sync gaps (both horizontal and vertical), which can be super useful and interesting. Would be great to see the NICAM signal! Also, have you seen Jim Easterbrook's compute-intensive but absolutely amazing BBC PAL decoder? Blew my mind when I read about how it works (just search for "The BBC transform PAL decoder"). As for the triggering of the scope in the IBA video: It's hard to see what's plugged into the scope, but I wonder if one of those cables isn't just an external sync signal from the modulator (or whatever component superimposes the NICAM signal), going into the scope's external trigger input.

[@mattstvbarn]

With regards to the triggering you are probably correct. When I use a sync pulse / video trigger to view the SIS data it does not look as clean, i.e. it's all jumpy and didn't make for a good video. So I ended up trigging inside of the SIS data its self which was a bit more fiddly.

[@Tooby001]

Nicam transmissions from the Swedish public service TV (SVT) became pemanent in the three largest citys here in Sweden by february 1988, and according to SVT; first in the world... ;-)

[@virgogreg]

NICAM was in NZ in 1989 it was set up by JDA

[@digitalmediafan]

When did TVNZ start using nicam ?

[@joshjones3227]

Greetings from NTSC land. Love you videos, and I have a question. Normally with dual amplification transmitters in the analog days we had a (using your terminology) vision and sound transmitter which were combined using a diplexer. How was the NICAM transmitted? Was it simply added as a subcarrier to the vision carrier?

[@mattstvbarn]

NICAM was generally done using the intercarrier method i.e. all three carriers amplified together. Some countries may have done it with split sound, but did they have a third amplifier for the NICAM? I don't know. Hopefully someone will tell us that some day!

[@mtc2300]

I'm not sure, when DR started NICAM in Denmark, but if I'm not mistaken, the testcard had 1Khz tone in mono and DR Radio P3 in stereo some time in 1988. I think Sweden was first, as they changed the music loop on Swedish TV2 in early 1988. And why do I remember that? Because the music loop on SVT2 contained the "Star Wars theme": kzbin.info/www/bejne/Y2etholtoMSsadE

[@mattstvbarn]

@@mtc2300 the date I gave was from a Philips Denmark staff magazine. It was the earliest I knew of at the time. I deliberately didn't say it was "the earliest" :)

[@liam3284]

Australia used A2 stereo, 2 subcarriers of FM. Technically it was better than FM radio broadcast.

[@pete3897]

Are you familiar with "The Wellington VHF group" branch of NZART? They have people who did some of the first amateur DTV tests in NZ and I'd bet they have ex-BCL engineers who could shed more light on our NICAM approach. The VHF group used to have an annual swap-meet sale that was open to the public; not sure if it still happens...

[@mattstvbarn]

I wasn't no. I haven't lived in NZ for a very long time unfortunately. I think I have a reasonably good representation of what was there. Most of it came from a technician who dismantled the analogue transmitters who recalled how it was put together. I am certain someone will add something more in future!

[@pete3897]

@@mattstvbarn keep an eye out for magazine called "break in" - you'll find load and loads of NZ radio and tv history in the back issues of that too. Nice kit, great vid and hello from Nelson :)

[@liam3284]

This is close to how a "wideband" telephone codec worked. Its sad that today, many digital channels are using 160kbps MPEG Layer II audio. It sounds worse than NICAM

[@MePeterNicholls]

Uk had teletext - data over the tv signal.

[@pete3897]

Yes so did NZ

[@user-yo6sy3ih7j9]

I LOOKING for CHEAP nicam tv MODULATOR pls

[@andydelle4509]

NTSCer here! Could baseband NICAM be recorded on 1in type C tape? Meaning the H interval is fully captured so why not? I realize the TBC would remove or trash it but was it usable from the VTR demod output? And what happened in the vertical interval? Was NICAM also on the equalizing pulses or was if just held in a digital buffer until the next active video line? How about FEC or other error correction? Are there ITU (SMPTE like) documents that go into deeper NICAM engineering details?

[@mattstvbarn]

Ah. You may have missed an important point there. NICAM in the HBI was only done inside of broadcaster networks. It was stripped out at the transmitters and re-combined as a DQPSK subcarrier. Few VCRs were ever able to record NICAM. If they did, it was by some other means.

[@andydelle4509]

@@mattstvbarn No I don't mean on the consumer VTR side. I mean within the broadcast / production plant for improved audio quality. Same idea as the FM audio channels 3 & 4 on Betacam. NICAM is a baseband signal so it could be be carried through distribution routers and anything that didn't reprocess sync. So why not record it on a highband VTR? Now I can think of some issues such as rise time exceeding the 4.2mhz (NTSC) VTR bandwidth assuming type C. And of course there are tape dropouts to deal with. And as I said before, it's not going top make it through the TBC unless there is a processor to deal with it. But the VTR demod out might be stable enough for good recovery? Another glaring problem is audio or video only edits - same problem Betacam had with audio channels 3/4. But for playout to air, that isn't a problem. And for some material like concerts, the audio was often laid back from recording studio grade audio tape. So in this application, there would be benefits over even a typc C VTR's linear audio tracks which were barely HiFi.

[@mattstvbarn]

@@andydelle4509 That is such an interesting question. I would guess the answer is no. A 1-Inch VTR (commonly used) is already capable of recording stereo sound in analogue form, which would be a great deal easier to deal with than SIS encoded into the video signal. There is quite a bit of hoo-ha around splicing/synchronizing NICAM signals as it was, adding VTRs into the mix would have made it even more difficult requiring equipment which wasn't commonly found in studios. Better deal with it as analogue then convert it to NICAM as it leaves the studio. That said, perhaps it was occasionally done in limited circumstances but I have not heard of it.