Extended SSB (eSSB)

[Bazz4CQJ]

I was litening to a US radio station overnight and the subject of radio amateurs using e-SSB came up. I could not recall ever hearing of this before.

eSSB " any J3E SSB transmission that exceeds the audio bandwidth of standard or traditional 2.9kHz SSB J3E modes (ITU 2K90J3E) starting at 3kHz (ITU 3K00J3E), in order to support the fidelity required and desired for relative high fidelity, full range clean and articulate vocal audio.

Translation; it sounds like AM.

Discussed in more detail at http://www.nu9n.com/essb.html

Have I missed something

B

[Sideband]

Reading through all that waffle in the link, I got the impression that it was the equivalent of audiophoolary for amateur radio......! Maybe I've missed something as well!!

[Dave757]

Hi,

I think it's been a niche interest in amateur radio for some time.
I had a friend who was well into it and he had racks of very expensive
audio equipment to achieve the desired effect.

Personally, I can't see the point of 'hifi' voice communication on the
amateur bands, so to me it does smack of 'audiophoolery', but I'm
sure to those who like it it adds a new dimension to speech.

Kind regards
Dave

[AdrianH]

Wide bandwidth spatter to me, there can be quite a bit on the satellite QO100 at times, seems to be a fashion statement, many using studio style microphones. Only seems to be with later digital rigs where the audio bandwidth can be changed from the nominal 300Hz to 2.7kHz to wider to get the hi-fi sound. of course to get the full benefit the receiver has to have wider bandwidths as well. Personal feeling is amateur radio has gone away from it's original roots and well into cheque book radio.

But as I say personnel feelings only and I am sure many will disagree.

Now must build that AM topband TX!

Adrian

[Bazz4CQJ]

Having looked at it more deeply, I see now this is just about creating "Wide-band SSB".

So if you could just flick a switch, you could use "conventional" SSB for working DX and Wideband while working the local net. But you cannot just flick a switch, so it's somewhat less interesting.

B

[Radio Wrangler]

SSB is a subset of AM. It only has audio sidebands on one side hence the 'single'. There is no definite limit to how much audio bandwidth may be carried, it will still be SSB.

SSB also comes with a choice of full, no or reduced carrier. All have different alphabet soup designators.

I well remember a lecturer (An ex-RAF guy) talking about bandwidth "The wider you open the window, the more **** flies in".

DSB with full carrier is what we normally think of as AM, but AM is a whole family of variants.

David

[Restoration73]

J3E is only the first 3 characters of the Class of Emission code.

The 4th character indicates detail of signal(s) - G = Sound of broadcast quality mono

The 5th character indicates type of multiplexing and if used.

so High quality SSB = J3EG

[G6Tanuki]

'classic' AM with two sidebands and full carrier, in amateur-radio form, takes up 6KHz or so.

SSB 3KHz or so. Likewise single-sideband-with-full-carrier [sometimes called 'AM-compatible SSB'].

So if you use SSB but extend the frequency-response up to 6KHz you're back to the sprctral-occupancy of classic DSB-full-carrier AM, but with increased audio quality.

I use SSB-with-full-carrier for some 80M QSOs; it's receivable on an old receiver without a BFO.

There's also "Voodoo audio" which some people play around with; it seems to involve massive bass-boost.

Personally, I prefer maximising intelligibility over quality, so am happy to run a Datong RF speech-processor set to give about 10dB of processing when needed. In practice, the 'amplitude' component of the human voice contributes surprisingly little to intelligibility!

[GMB]

I never understood SSB with supressed carrier. It doesn't contain enough information to reconstruct the original signal so how did that ever get approval for anything??

I have always assumed that although the audio is still up to 3kHz it exploits the "starts at 300Hz" to make a guard band so you get to cram the channels that little bit closer.

My blood boils whenever people say of AM that "the carrier contains no information". It flippin well does. It gives you the exact centre frequency and it also gives you the signal strength.

If the carrier was just reduced it would now be possible to use clever software to make sure the tuning is spot on.

[G6Tanuki]

Originally Posted by GMB

I never understood SSB with supressed carrier. It doesn't contain enough information to reconstruct the original signal so how did that ever get approval for anything??

The carrier is removed at the TX and regenerated at the rx; it doesn't actually need to be passed over the air. SSB _does_ contain enough data to reconstruct the original audio sufficiently closely that it makes no perceptible difference. [In practice you don't need the carrier-reinsertion to be in-exact-phase with the carrier that was taken away at the transmitter; inded can be quite a few tens of Hertz away from the suppressed carrier and nobody will notice]

Originally Posted by GMB

If the carrier was just reduced it would now be possible to use clever software to make sure the tuning is spot on.

Not just 'now', the idea of 'pilot-carrier SSB' which allowed the receiver to regenerate the carrier at the right frequency/phase was proposed and trialled in the late-70s [by Pye, among other people] as a potential way to allow more channels to be fitted into the then-crowded VHF mobile-radio bands. Phase-lock the receiver's reinsertion oscillator to the vestigial carrier. In practice it showed very little practical benefit [apart from avoiding the need for occasional tweaking of a 'clarifier' control] to make much noticeable difference.

[G0HZU_JMR]

Some of the Uniden AM/SSB CB radios of the late 70s and early 1980s used a main crystal filter that was a compromise between AM and SSB. These filters were typically 4.5kHz wide. This was just wide enough for AM and much wider than it needed to be for SSB. This was done to save on cost for some AM/SSB CB radios built for the US market.

The result of this was that the IF transmit bandwidth for SSB extended to about 4.5kHz. These radios had a reputation for sounding very nice on SSB, especially if two CBers were using a similar SSB radio with 4.5kHz bandwidth.

The Uniden export CB models used this system as well and I think this was also used in the (even cheaper) cloned versions of these Uniden radios for many decades.

[m0cemdave]

I agree with the comments in posts 2 and 3. Maybe substitute "audio..." for "radio..." to create a new class of, er, enthusiasts.

There are a few of these stations on 80m in the evenings. Some of them have bandwidths of over 8kHz, which is rather more than the AM stations around 3615 - 3625.

People play around for hours tweaking their racks of "home studio" stuff. Quite amusing (unintentionally) when listening on an SDR and watching the waterfall. Last week one chap was playing around with 30Hz boost on his parametric...

One thing I've noticed is that their attempts at LF boost often seem to result in a growly noise and a kind of vestigial sideband effect. I think the low frequencies are upsetting the balanced modulators.

(and yes I have an RE20 in my shack, but only because it was left over from a studio I built in the 1980s... I didn't keep the U87.)

[GMB]

Originally Posted by m0cemdave

Some of them have bandwidths of over 8kHz, which is rather more than the AM stations around 3615 - 3625

Blimey! The AM folks get a lot of stick for using lots of bandwidth - but this takes it to a new level. So I wonder when we (won't) see the RSGB band plan's allocation for HiFi SSB?

(I still see SSB and HiFi as fundemantally incompatible thanks to the carrier uncertainty problem.)

[Synchrodyne]

The use of a pilot carrier (at -16 or -26 dB) was for a long time the norm for point-to-point communications purposes. Receivers used a very narrow filter band that was used to extract the carrier, which was then amplified and limited for demodulation purposes. Point-to-point transmission bandwidths were ±6 kHz, accommodating either two voice channels each side of the carrier, or one broadcast channel for relay purposes. Thus “extended bandwidth SSB” is hardly a new idea.

Professional point-to-point receivers were often available with a choice of sideband filters, typically either 3.5 or 6.0 kHz wide. With improved frequency stability at both the transmitting and receiving ends, the use of pilot carriers fell away, and the facilities therefor became optional rather than standard for receivers. For some applications, including I think aviation HF, carrier-less SSB was the norm from the start.

My understanding is that the 3.4 kHz “telephone” bandwidth was derived in the 1930s or thereabouts on the basis that that was the point at which further bandwidth reduction did not improve intelligibility in the presence of noise. That lower bandwidths were used in radio communications was a reflection of the need for spectrum economy, not to improve intelligibility.

The intelligibility vs bandwidth issue surfaced again around 20 or so years back with the increased use of telephone conferencing. There were several papers written on the subject (a couple of which are attached), and some carriers offered wideband teleconferencing circuits. The consensus seemed to be that 7 kHz bandwidth was about right. I can recall that when I was working in Sydney, we used teleconferencing circuits provided by both ATT Singapore and ATT USA, of which only the latter were wideband. The difference was quite stark, the more so when the conference call was around 2300 h local time, and most of the participants were together in a room in the USA or Europe, with a few lone remote callers at the end of the telephone line. Apparently, the intelligibility problem was worse for folks whose first language was not English – at the time our Japanese colleagues circulated a paper to this effect (which I did not keep).

On balance I think one could say that there is some science behind the extended bandwidth SSB concept. Clearly, in an HF radio context, any extension of bandwidth beyond 3.4 kHz is going to run into the tradeoff between improved intelligibility obtained therefrom and the reduced intelligibility caused by noise and interference, but in relatively “clean” conditions, some users might find it worth having.


Cheers,

[Radio Wrangler]

For anyone who wants a laugh, look up "Stenode Reception" It could only have been thought up by someone unsullied by thoughts of signal-to-noise ratios and Fourier transforms.

David.

[Bazz4CQJ]

Given that 2m is pretty much empty, enthusiasts of wideband SSB could perhaps revive that band . I wonder if that band, at least in part, could be lost to amateur use?

B

[Radio Wrangler]

The 2m band is empty enough and wide enough that it could handle several channels of full broadcast quality stereo FM. It would be fun to propose having a few in some April the first magazine, then listen to the moanings of all the people who don't even use 2m

David

[G6Tanuki]

Yes, remember they gave us another MHz (146-147) on 2M a few years back, the intention being that it was for wider bandwidth modes. Ideal for someone who is interested in, let's say, transmitting stereo at 44.1Kbits/Second using CD technology


I am sure I remember Angus Mckenzie G3somethingorother transmitting analog stereo on 70cm some decades ago!!

[AdrianH]

If you want wideband transmissions why not just use AM TV again? That used to use a large chunk of 70cms at one point.

Adrian

[Radio Wrangler]

Angus: G3OSS

David