SSB -- an novice SWL-ing intro, please

[Al (astral highway)]

Please can someone explain in a paragraph what SSB actually means and sounds like on the old AM shortwave bands.

Before you fall about laughing, I'll just say that I don't know the first thing about shortwave listening termology and my listening experience is limited to a few random trawls through the night-waves, but only picking up massive KW transmitters such as those in Moscow, Beijing and so on.

I do understand the theory of how AM comms receivers, transmitters, modulation and antennas work, so you can plunge straight in to the 'dark arts' of the propagation and listening experience.

You could say: 'Go away and read a book', but I'm forever doing that, and instead I want to the distilled clarity of all those 1,000s of hours of forum expertise instead. Much more interesting!

Mods, no offence if you move this to somewhere more appropriate. Couldn't think where!

[ppppenguin]

The Wikipedia entry makes the maths far too complex (heaven knows why, you can do the maths pretty simply) but if you ignore them you can find a reasonable introduction to SSB.

http://en.wikipedia.org/wiki/Single-sideband_modulation

As for what it sounds like in practice I'm the wrong man to ask, having never been an active SWL or ham.

[MichaelR]

An AM signal has carrier present with the upper and lower sidebands.Your broadcast receiver in AM mode has an envelope detector in essence to extract the audio signal.

SSB is when a single sideband is transmitted , not the carrier. Less frequency space is taken up ( no upper or lower sidebands plus carrier). To recover the audio the carrier is inserted in effect at the receiver (using the BFO) and the mixed output then detected.

SSB on an Am receiver sounds very "garbled" no matter how you tune you cannot tune it in.

[PJL]

From one novice to another - the SW stations you are listening to are AM not SSB. SSB sounds like low frequency squaking without a product detector (BFO). As it has no constant carrier the AGC messes it up to. I have a couple of old military receivers which sort of work...

[Hermit6345]

Hi,

SSB sounds, on an A.M radio, a bit like Donald Duck. The reason for this is that no carrier is transmitted with the sideband so an A.M radio cannot mix the carrier with the sideband(s) in the detector as it can when a carrier is transmitted. To resolve SSB, you have to re-insert the carrier in exactly the right place and at a suitable level into the received signal spectrum. Hence the need for a very stable oscillator. Older communications receivers had an oscillator, usually at IF frequency, which carried out this function. This was known as a Beat Frequency Oscillator (BFO). There is something called the "Ten Megs Convention" which means that below 10 MHz the lower sideband is transmitted (LSB) and above 10 MHz the upper sideband (USB) is used. Exceptions to this are generally the military and some others. This convention is now very old but amateurs still use LSB below and USB above 10 MHz.

Listen on about 7.1 +/- MHz and you will hear the Donald Duck transmissions from all over Europe.

[Ian - G4JQT]

Have a listen between between 7050 and 7100 kHz. This is an amateur radio band and will have plenty of SSB signals at any time of day. (There are plenty of other frequency bands with SSB, but this is near enough to an AM brodcast band to be easily received by an basic short-wave receiver at any time.)

That's what they sound like on an ordinary AM receiver.

Ian

[Al (astral highway)]

Originally Posted by Hermit6345

Hi,

To resolve SSB, you have to re-insert the carrier in exactly the right place and at a suitable level into the received signal spectrum. Hence the need for a very stable oscillator. Older communications receivers had an oscillator, usually at IF frequency, which carried out this function. This was known as a Beat Frequency Oscillator (BFO).

Shoot me down in flames if this is wrong, but could I 'inject' such a signal into the set by inductively coupling exactly the IF frequency into the set, from a separate source -- I was thinking I could easily use my old signal generator, which (when I've finished restoring it will be very stable.

[Station X]

Here's my reply which I typed up last night in anticipation of this thread. It looks like the rest of you have got the subject well covered.

------------

In an AM Transmitter a carrier wave (CW) is modulated with the audio waveform. This is a mixing process, so two sidebands are produced in addition to the carrier wave. These are known as the upper sideband (USB) and lower sideband (LSB). If the maximum audio frequency is say 3 KHz, then the sidebands will extend 3 KHz either side of the carrier. The only reason for transmitting a signal like this is that it is very easy to detect (recover the audio) using just a diode and capacitor. This sort of modulation ahs been around since the days of Crystal sets.

However the carrier wave contains no information and the sidebands each contain exactly the same information. So we can remove the carrier and one sideband. Only the remaining sideband need be transmitted hence Single Sideband (SSB). The means that the transmitter can be used much more efficiently putting all its power into the sideband.

On a normal AM receiver an SSB signal sounds like Donald Duck and is unintelligible. To recover the audio we need to put the CW back, albeit at very low level. This is done by means of a Beat Frequency Oscillator (BFO) in the receiver. The BFO operates close to the IF and by careful tuning it can be spaced the correct distance from the IF and the audio waveform resolved.

Most Communication Receivers have a BFO, but in sets built up until about the end of WW2 it was designed for the reception of morse signals. These sets will resolve SSB, but a more modern Comms Receiver with a Product Dectector wil do a better job.

---------------

Is anyone able to post a link to a wav file of the Donald Duck sound?

[GMB]

Originally Posted by astral highway

could I 'inject' such a signal into the set by inductively coupling exactly the IF frequency into the set, from a separate source -- I was thinking I could easily use my old signal generator, which (when I've finished restoring it will be very stable.

Yes, that's what a BFO is effectively. I think it will also help to disconnect the AGC from the detector and instead just feed it an adjustable voltage to control the gain.

[GMB]

Originally Posted by Station X

However the carrier wave contains no information

Everyone says this but I suggest it is not true. The carrier contained the information about depth of modulation. By loosing it you cannot tell how strong the signal is, hence the problem with AGC.

[Station X]

We'll have to agree to disagree on that one. AGC will work fine on an SSB signal, so long as the circuit is designed to react to the sideband and not the re injected carrier. If this is not the case it may be better to turn the AGC off. The AGC also has to have a longer time constant to give satisfactory operation on SSB.

[Al (astral highway)]

Originally Posted by Hermit6345

Hi,


Listen on about 7.1 +/- MHz and you will hear the Donald Duck transmissions from all over Europe.

Hmm. This is around the 40 metre band, right? And guess what I picked up when I tuned in....

...Just various sounds a bit like the sound of a propeller-driven aircraft just before take off.

I was about to post this finding on the forum when I noticed that every time I moved my mouse or clicked on a different part of the browser, I could 'modulate' this prop. driven aircraft noise.

By a process of elimination, I discovered that either the following devices or interaction with them, 'modulated' the 40 metre noise that I detected on my receiver:

1. The optical mouse and pad (not wireless). Capacitatively coupled to me and changes the f of the radiated interference.

1. The USB 2 hub (not wireless)

1. The laptop itself.

This is a bit of a shocking finding. I could forgive the off-brand USB 2 port and optical mouse, but the laptop is a leading brand starting with an 'H'. I'd expect perfect suppression of RFI into the mains, and certainly not interference like this.

Putting it into perspective, 40 metres is around 7.5 Mhz, right?

And the operating frequency of the laptop is 1.5 Ghz, around 200 times higher.


I guess that's a good example of broadband interference!!

(sorry about the numbered list - something went wrong there).

[MichaelR]

If you just want to resove the ssb and not worried about the technicalities of quality etc yes you could inductively couple at IF , you will have to adjust the frequency +/- 3Khz to get the sideband as per upper or lower sideband being used on the frequency you are set to.Again messing about with injection levels will be needed

[Station X]

...Just various sounds a bit like the sound of a propeller-driven aircraft just before take off.

PCs are notorious for creating RF interference. The worst offender in my shack is the document scanner, even when it's not scanning anything.

Amateur Bands are generally referred to by their wavelength, but the wavelength is not exact. So:-

160 metres 1.81 to 2.0 MHz.
80 metres 3.5 to 3.8 MHz.
40 metres 7.0 to 7.2 MHz.
20 metres 14 to 14.35 MHz.
15 Metres 21 to 21.45 MHz.
10 metres 28 to 29.7 MHz.

[jjl]

Originally Posted by astral highway

Shoot me down in flames if this is wrong, but could I 'inject' such a signal into the set by inductively coupling exactly the IF frequency into the set, from a separate source -- I was thinking I could easily use my old signal generator, which (when I've finished restoring it will be very stable.

I used a technique along these lines when I was about 13 or 14 but instead of a signal generator I used two radios both more or less covering the required SW range. My Russian Vega tranny was used to pick up the SSB signal and my late '50s valve RGD placed a couple of metres away was used as the BFO by virtue of its local oscillator radiating the required carrier. By very careful tuning of both receivers it was possible to resolve SSB transmissions on the 40 metre band.

John

[RobinBirch]

I also used this method with a simple oscillator built round an IFT and a small variable capacitor. Coupling was a length of wire from the collector of the oscillating transistor wrapped round the detector valve.

I think I was 13 when I built this, it kinda worked but not very well.

Robin

[Station X]

Everyone says this but I suggest it is not true. The carrier contained the information about depth of modulation. By losing it you cannot tell how strong the signal is, hence the problem with AGC.

The other "missing" information contained in the carrier is its spacing from the transmitted sideband. That's why the receiver must be carefully tuned to make the speech intelligible. Even when the speech is intelligible it's impossible to tell whether the pitch is the same as that used by the operator speaking into the microphone.

[Al (astral highway)]

Originally Posted by RobinBirch

I also used this method with a simple oscillator built round an IFT and a small variable capacitor. Coupling was a length of wire from the collector of the oscillating transistor wrapped round the detector valve.


Robin

Good thinking, esp as you were 13 at the time

[Al (astral highway)]

Originally Posted by Station X

Amateur Bands are generally referred to by their wavelength, but the wavelength is not exact. So:-

160 metres 1.81 to 2.0 MHz.
80 metres 3.5 to 3.8 MHz.
40 metres 7.0 to 7.2 MHz.
20 metres 14 to 14.35 MHz.
15 Metres 21 to 21.45 MHz.
10 metres 28 to 29.7 MHz.

Thanks. That's interesting. Is/was that strictly 'policed', or is there creep either side of the cut-off points?

I'm thinking that if the RF power stage of someone's homemade xmitter wasn't particularly well built or sharply tuned, it could be a bit too 'broadband' and spill over. (And no, not since my early teens have I am been remotely interested in building such a transmitter illegally)

[Hermit6345]

Originally Posted by GMB

Everyone says this but I suggest it is not true. The carrier contained the information about depth of modulation. By loosing it you cannot tell how strong the signal is, hence the problem with AGC.

I can't agree about the carrier containing any information. It's a carrier wave and serves no other purpose. Take a look at an AM DSB signal on a spectrum analyser. The carrier does not vary whatever the depth of modulation of the sidebands up to 100%. What confuses some people is what a double sideband, full carrier signal looks like on an oscilloscope. The amplitude appears to vary. This amplitude variation is cause by the vector addition and subtraction of the sidebands above and below the carrier. The carrier amplitude remains constant at all times.

The STRENGTH of the carrier (and sidebands) as received, may vary because of propagation conditions etc and this may affect AGC. However AGC is quite possible with SSB and usually involves an AGC circuit that can provide a fast attack and slow decay so as to deal with RF which varies at a syllabic rate. Most, if not all, amateur transceivers are so equipped.

This is how I see it anyway!