FM stereo

[Ted Kendall]

Originally Posted by Radio Wrangler

...though the Optimod wars have rendered all but R3 unlistenable.

It's a bit late, now, but people are beginning to learn just how good what we used to have was.

David

Not just Optimod, though that should be dumped somewhere inaccessible in perpetuity, but the balance rules for some networks introduced in the eighties. I did, by my own admission, a lovely job on a brass band session - I think it was Grimethorpe, playing a Wagner arrangement of huge dynamic range, which I fitted into the prescribed 22dB window by oh-so-careful gain riding. I recorded the transmission off-air. Somebody or something downstream pulled all my carefully preserved pps and ffs to a constant mf. This was bad enough, but then the screaming climax twenty seconds from the end of the piece hit a limiter and the level dropped 6dB. Shouldn't have bothered...

[Jon_G4MDC]

Excellent thread. I need some more days to work my way through Post#119.
I don't know Optimod - but if that is what is to blame for the current situation
(and quietly I know for the most part it is) >> dustbin

[Graham G3ZVT]

Optimod, the consensus is that the youngsters are attracted to the stations that are perceived to be the "loudest", but I'm sceptical if it's really true.

My wife likes to listen to one of the worst Optimod offenders, "Smooth". I can hear the volume pumping along with the sound content, reminiscent of a second generation copy compact-cassette or VHS audio. I am more aware of the unwanted artefact than I am of the music and I find it stressful to listen to.
My wife hasn't a clue what I am talking about.

Odd that the same company also brings us Classic FM which does a good job with limiting the dynamic range. Yes, it's a lot more aggressive than Radio 3, but in a noisy car R3 needs a bit more.

[Electronpusher0]

There is a good explanation about Optimod, the drive for ever more compressed sound and the "loudness war" in Greg Milners excellent book "Perfecting Sound Forever" - a really interesting read.
The Optimod was an audio processor for radio introduced by Orban in 1975.

"It wasn't a multiband processor, but it was designed specifically to address some problems inherent in processing an FM signal and put many functions in one box. The Optimod was a huge success and gave stations another loudness weapon"

Peter

[Ted Kendall]

However you dress it up, Optimod is a weapon for gaining loudness at the expense of signal quality without it being too obvious to the punter. For listening in cars, some contraction of dynamic range is useful, but the logical place to put this is in the receiver. Note I say the logical place, not the easy place...

[Synchrodyne]

Definitely the receiver is the logical place in which to do any compression that is perceived as being desirable by listeners. I don’t think that it would be too difficult. Suitable ICs that allowed economical implementation probably could have been made available at about the same time that gain-riding devices were developed to assist QS and SQ quadraphonic “decoding” (early 1970s) and Dolby B encoding/decoding, as the underlying techniques were probably not a lot different. And audio domain automatic gain control was used in some valved American car radios in the 1950s in order to level out differences between louder and softer stations. Perhaps those circuits were a little crude, but if you’re going to take a wrecking ball to fidelity by seriously compressing dynamic range in the first place, maybe it does not matter too much what collateral damage is done along the way.


Cheers,

[Pfraser]

Years ago, I had a Roberts Gemini 55 (RD55) portable MW, FM & DAB set. In amongst the menus was DRC - Dynamic Range Control.

It seemed pretty crude and brutal in the way it operated...much like the way broadcast audio processors are often misused! I generally left it switched off.

But if the set were to be used in a noisy environment, DRC would definitely help with audibility.

[Ted Kendall]

Originally Posted by Pfraser

But if the set were to be used in a noisy environment, DRC would definitely help with audibility.

And this is an important point - if the listening environment is noisy, then quite crude processing will give a subjective improvement, with the extraneous noise masking the side-effects. Ergo, put it in the receiver rather tham making the more serious (or fortunate) listener suffer the nasties.

[brightsparkey]

Regarding valve decoders, my Grundig 4070M (1964) has an all valve (1 ecc81) decoder - work of art! The transformers are impressive in themselves. There are germanium diodes, and there is a germanium transistor to drive the stereo mono/stereo relay.

My SABA radiogram had provision for a decoder (1965) but didn't have one fitted - the appropriate decoder is all germanium transistors and diodes. Transmission in Germany preceded the BBC by some years.

The operating principle of both is to filter and amplify the 19kHz pilot, double it and then use the resulting 38kHz carrier to mix the left and right channels out of the multiplex. In the SABA there are two independent mixers for best separation.

The FM signal, viewed in this way (and not as described to me in my electronics degree) is very simple - left and right channels are multiplied by positive and negative half cycles of the 38kHz carrier..

On another sub-topic - DAB is not as simple as some seem to believe. There are layers of decoding - the signal is a digital transmission, in so much as a bit stream is encoded into OFDMA - a thousand sub-carriers, each with QPSK modulation. The received signal is decoded and error corrected to produce an mpeg-2 audio stream (actually a multiplex of many streams).

MPEG2 is a sub banded audio compression, using the principle of 'masking' to throw away the information you can't hear. This system is capable of impressive results at 320kbps as originally transmitted for R3, but has been wrecked by commercial greed as stations are stuffed into too little bit-rate.

If your DAB signal bubbles, then you are very marginal for received SNR - the performance is restored to near perfect for a few more dB's - this is the sharp transition of digital reception caused by error correction. FM quality degrades gracefully due to its analog simplicity - the C/N leads to the output S/N more directly. With digital it either works or it doesn't.

Kevin.

[G8HQP Dave]

Originally Posted by brightsparkey

The FM signal, viewed in this way (and not as described to me in my electronics degree) is very simple - left and right channels are multiplied by positive and negative half cycles of the 38kHz carrier..

This is the difference between the Zenith and GE stereo systems. One used DSB with a 38kHz carrier; the other used 38kHz TDM. Someone noticed that they were more or less identical apart from a factor of pi/4, so they were merged.