FM Multiplexing for stereo

[Sideband]

I have a book describing the workings of FM stereo from the transmission end and the receiving end. It's an American book but the technology is the same as ours so virtually all the technical side applies to our own multiplex system.

Very briefly, the book describes how the multiplex signal is encoded using sum and difference techniques, how it's applied to the 38Khz sub-carrier and the function of the pilot tone. Then it describes how the decoder proceses the signal back to stereo, how the sub-carrier regenerator works and how the pilot tone syncronises it all together finally ending with the special ring demodulator that provides the left and right signal.

It also describes how some of the 'latest' decoders work. In the theme of all things vintage, valve technology is used throughout...not a single transistor is mentioned. Some of the decoders use some interesting American valves like a 6AR8 which is a beam tetrode with two anodes used for demodulation.

It doesn't use any really heavy math, just a few vectors.

Now the point of all this is the fact that this forum is for the distribution of knowledge and my idea was to scan this book in stages so that the information can be used by everyone...that is assuming that anybody wants to know about this technology. The book is likely to be irreplaceable which is the reason I want to hang on to it. However the information within is likely to be of great interest to users of these forums hence my reason for wanting to share it.

There are some sections that won't be worth copying as they involve conversion of mono tuners for MPX output and also the same section covers what were currently available multiplex adaptors.

Anyway I will need to play with the scanner and decide the best file format and try to keep the filesize down whilst retaining good resolution.

Firstly though, is there enough interest in stereo valve decoders to start this project? The book was printed in 1964 (first edition, third printing) so I don't think copyright will be a problem. The moderators may like to comment on this if they think it will be a problem.


SB

[murphyv310]

Go for it SB, I for one would save it and read through the "book" when I have some time.

[dominicbeesley]

Me too - I keep meaning to build an FM stereo tuner from scratch...

Dom

[Roy]

I would certainly be interested to see this. I'm currently building a stereo tuner, to go with the stereo 5-10 amplifier which I built a couple of years ago. The design of the multiplex decoder has been an interesting exercise. It's actually not too difficult to build a valved decoder that works, if you don't mind winding a few coils using very fine wire. Enhancements such as reliable automatic mono/stereo switching, and a stereo beacon, are more of a challenge. Of all the circuits I have tried, one using the American 6JH8 (which I believe is an "improved" 6AR8) has emerged as the clear winner.

Depending on the level of interest, if you do decide to scan part of the book, I would suggest 300 or 400 dpi in black-and-white (not greyscale or colour) for best clarity combined with reasonable file size. This is what I use for scanning service manuals, and an average A4 page generally comes out at around 60-70 KB when saved as a PDF.

Thanks for bringing this to our attention.

Roy

[AlanBeckett]

I'd be interested as well. I've got a couple of (valve) VHF Tuners and a VHF3D which I could use one with.
Alan

[Sideband]

Ok it seems that this may be worthwhile. I'll post a couple of trial pages. Roy. I've been playing with resolutions and find that using straight BW/300 is ok for plain text and simple 'line' diagrams but no good for more detailed diagrams. I may use two resolutions depending on content and the best compromise seems to be greyscale 150 for these. They will be in .gif file format. I can't do PDF's on this printer/scanner and Jpeg is a good bit larger.

I won't say what is what at this stage. Just pick the scans that give the best resolution.


SB

[Sideband]

Interesting. The forum software has also changed the resolution...I may have to try others.



SB

[Sideband]

Here's one I did earlier.


If this one is OK, I'll start scanning the book tomorrow and then zip the files. It'll take a while (there's 160 pages) although I wont be scanning all of them...just the one's that are important for understanding the principles and operation.


SB

[Roy]

This seems an interesting publication, and well worth sharing, if you have time to do the scanning.

There have been one or two discussions on the subject of document scanning in the past, including this one:

https://www.vintage-radio.net/forum/...ead.php?t=7577

I found Paul's advice in post #4 very helpful when I had a lot of scanning to do.

Hope this helps,

Roy

[dominicbeesley]

Look forward to having a read!

Have you tried saving as TIFF files?

Dom

[Sideband]

Hi Dom. Yes I did try .tif but once uploaded to the forum, the quality seemed to suffer presumably because the forum software also compresses the files.


SB

[MrElectronicman]

Don't know if this would help, but I have found a website that, maybe? has the book in question.
http://audioclassics.com/detail.php3...=20199&nav=cat

There are no other details, so I am unable to determine if it is actually the same book.

[Sideband]

OK as a taster, here's the first two chapters. I've zipped it to further reduce file size. It looks OK directly on the PC but not sure if the forum software will further change the resolution. I used B/W and 150 dpi I may have to change that for some of the diagrams though.

If this is OK, I'll copy the other relevant parts as time allows over the next few days/weeks and hopefully post them together probably over several posts.

SB

MrElectronicman. Very interesting. It could be the same book....same publisher. It's quite likely the book is still available in the States on the second-hand market. Does this mean I shouldn't be copying it?

[Roy]

Many thanks for sharing this with us.

The scans are perfectly readable, and look about right for 150dpi in B/W. As you mentioned, a higher resolution might be needed if there are more detailed diagrams in later chapters.

It should be easy to assemble the pages into a PDF in due course.

Roy

[Synchrodyne]

The book “FM Multiplexing for Stereo” by Leonard Feldman seems to have had several editions, including a 3rd edition in 1972, and a Foulshams UK edition with preface for the UK reader, as Foulshams was wont to do at the time. I would think that the timing of the 3rd edition should have just caught the of beginning PLL IC decoder era, with the MC1310 and CA3090. Thus the various editions might nicely cover the period from early valve decoder designs through discrete bipolar designs to the point where began something of a plateau in basic technique, albeit with continual improvement.

Cheers,

[MrElectronicman]

Hi SB,

It's quite likely the book is still available in the States on the second-hand market. Does this mean I shouldn't be copying it?

I am not a lawyer so I can't comment, but there is another thread on this topic that may be relevant.

Out of Print Publications - Legitimate Copying?

[Sideband]

OK having some spare time this evening, I've scanned some more of the book. There is a problem with some of the pages because it's not possible to hold the pages completely flat whilst scanning. If some of the wording is too blurred, I can probably re-scan some of the pages. I don't want to break the spine of the book obviously as it's getting a little delicate...and the covers are already being held on with Sellotape.

This is quite a long chapter and deals with multiplex decoder circuits.


Hope this is of interest.

The final chapter deals with servicing.


SB

[Roy]

Many thanks once again for taking the time to scan this publication.

I find this chapter most interesting, as it compares the various approaches to decoder circuit design, and confirms much of what I had already discovered "the hard way" as a result of my experiments.

Roy

[Synchrodyne]

Thanks again for the items posted. What is interesting is the early emphasis on frequency division (matrixing) decoders, even though the conventional wisdom is that time division was an easier approach. Perhaps it reflects the fact that early descriptions of the GE-Zenith system usually described it in frequency division multiplexing terms.

I have also looked at a couple of other editions of this book. One is the 1968 Foulsham UK edition of the 1962 US version. This shows a mix of valve and solid state decoder circuits. In it the early popularity of the frequency division decoder is acknowledged, with the comment that the time division system by then had become more popular. It also includes a description of the same Zenith decoder as used in the MJ1035W stereo table radio mentioned in the thread: "Were there any all valve FM stereo radios?" at: https://www.vintage-radio.net/forum/...ad.php?t=69972.

The 1972 3rd edition just touches the IC era. Surprisingly, it is the non-PLL Motorola MC1304 that is described, and not one of the early PLL types, although it is mentioned that both RCA and Sprague had decoder ICs available as well. There is also a description of the Sound Technology 1000A signal generator. I recall that this was highly regarded, and was reviewed by one of the UK Hi Fi magazines in the 1970s. I also saw one in use on the Quad stand at one of the 1970s UK audio shows, being part of a direct vs via-the-FM3 tuner comparative setup.

Cheers,