Sometimes I really get curious and want to know about something.

I haven't seen the Ham Radio article, but I'm thinking if the whole
idea had any merit it would be a popular mode by now.

Bruce-

It has been about 35 years since I had a class in school where SSB-FM was
discussed. I recall that if you derive the equations for both AM and FM SSB,
they are identical for practical purposes if the FM signal has low deviation
(low modulation index?).

Looking at Two Meter FM, the deviation typically peaks at about 5 KHz. If you
listen to your local repeater with an SSB rig such as the IC-706, it will be
obvious that it isn't a clean signal! However, a 3 KHz deviation FM signal on
HF (below 29 MHz) will sound much cleaner when tuned as SSB, and you may not
notice it isn't AM-SSB.

With this in mind, consider that AM-SSB and FM-SSB might just be two ways to
generate an SSB signal, assuming you use a filter to eliminate the carrier and
other sideband.

By the way, an IC-706, especially one with the TCXO, often has a more accurate
frequency read-out than a typical Two Meter rig. Therefore you can use it to
check a repeater's frequency by tuning it as if it were an SSB station while
someone is speaking. It is easy enough to check the IC-706 against WWV on HF.

73, Fred, K4DII

Fred McKenzie wrote:
It has been about 35 years since I had a class in school where SSB-FM was
discussed. I recall that if you derive the equations for both AM and FM
SSB, they are identical for practical purposes if the FM signal has low
deviation (low modulation index?).

You're probably thinking of AM vs. narrow band FM. Although the equations
look very similar on paper and the MAGNITUDE spectrum is identical, the
phase spectrum is different in that -- in the phasor domain -- AM always
sits at 0 degrees and just grows and shrinks with modulation (overmodulation
pushes it over to 180 degrees, BTW). NBFM, on the other hand, still has the
carrier at 0 degrees but grows and shrinks along the imaginary axis. Hence
the angle of the phasor is small but time-varying (which implies that the
instantaneous frequency is varying as well -- but of course you already knew
that since we called this whole mess 'frequency modulation'). The angle is
about 15 degrees for a modulation index of 0.3 (what my notes claim as a
good cutoff for NBFM) and about 5 degrees at 0.1.

See the message I posted earlier tonight for a discussion of whether or not
you can recover NBFM with an envelope detector as of course one often does
with AM (the difficulty is due to that phasor's wiggling...). I think not,
but there's plenty I don't have a clue about... yet!

What's the modulation index on two meters anyway?

---Joel Kolstad
....who does know that a wideband FM receiver has no problem whatsoever
receiving NBFM...

Looking at Two Meter FM, the deviation typically peaks at about 5 KHz.
If you listen to your local repeater with an SSB rig such as the IC-706,
it will be obvious that it isn't a clean signal! However, a 3 KHz
deviation FM signal on HF (below 29 MHz) will sound much cleaner when
tuned as SSB, and you may not notice it isn't AM-SSB.

You're probably thinking of AM vs. narrow band FM. Although the equations
look very similar on paper and the MAGNITUDE spectrum is identical, the
phase spectrum is different

Joel-

Perhaps that is what I'm remembering. Now, if you use a filter to eliminate
the other sideband, the higher frequency components and the carrier, don't you
have a nearly identical remainder?

See the message I posted earlier tonight for a discussion of whether or not
you can recover NBFM with an envelope detector

Somehow I missed that one. It seems that AOL does not post messages in the
order in which they were originated!

I think we are in agreement that you can't recover FM modulation with just an
envelope detector, but there is another approach. Again, you need a filter,
but maybe one that is not as sharp as above. If you tune the radio so the
carrier is just outside the passband, an amplitude variation will occur as the
signal slides up and down the shoulder of the filter. The result is a pseudo
AM signal that is detected by the envelope detector. I recall that this
approach is called "slope detection".

73, Fred, K4DII

Fred McKenzie wrote:
Perhaps that is what I'm remembering. Now, if you use a filter to
eliminate the other sideband, the higher frequency components and the
carrier, don't you have a nearly identical remainder?

At that point I don't think you could tell the difference since there's no
longer any local phase reference (i.e., the carrier) to compare with. I
suppose this is why your SSB-AM rig is able to (somewhat) receive low
frequency (and thereby presumably narrowband) FM broadcasts; this is what
you were saying in your last post, correct?

I think we are in agreement that you can't recover FM modulation with
just an envelope detector

Yes, at least you can't recover a signal that directly corresponds to what
you transmitted. It does appear that you can recover the signal's square,
however, so this approach might be useful for, e.g., remote command
transmissions. (But probably just for the novelty of having said you did
it... since it's probably not much harder to build the slope detector you
describe!)

---Joel

Along the same line consider that the envelope of an SSB signal has no
direct relationship to the original modulation the way that an AM
signal does.

This is why you can not use RF derived ALC to control the audio stage
of an SSB transmitter the way you can with an AM transmitter.
Or audio clipping that works on AM but does not work the same on SSB.

Transmit a square wave on an AM transmitter and you see a square wave
in the AM envelope. Do the same with an SSB transmitter and you only
see sharp spikes in the envelope.

73
Gary K4FMX


On Thu, 23 Oct 2003 12:08:31 -0700, "Joel Kolstad"
wrote:

Fred McKenzie wrote:
Perhaps that is what I'm remembering. Now, if you use a filter to
eliminate the other sideband, the higher frequency components and the
carrier, don't you have a nearly identical remainder?

At that point I don't think you could tell the difference since there's no
longer any local phase reference (i.e., the carrier) to compare with. I
suppose this is why your SSB-AM rig is able to (somewhat) receive low
frequency (and thereby presumably narrowband) FM broadcasts; this is what
you were saying in your last post, correct?

I think we are in agreement that you can't recover FM modulation with
just an envelope detector

Yes, at least you can't recover a signal that directly corresponds to what
you transmitted. It does appear that you can recover the signal's square,
however, so this approach might be useful for, e.g., remote command
transmissions. (But probably just for the novelty of having said you did
it... since it's probably not much harder to build the slope detector you
describe!)

---Joel

In article , Gary Schafer
writes:

Along the same line consider that the envelope of an SSB signal has no
direct relationship to the original modulation the way that an AM
signal does.

This is why you can not use RF derived ALC to control the audio stage
of an SSB transmitter the way you can with an AM transmitter.

You can't use ENVELOPE detection on SSB the same way it is
done on conventional AM.

But, you CAN use RF-derived feedback - if mixed with a steady
carrier to recover the modulation content - to do that very well.

Or audio clipping that works on AM but does not work the same on SSB.

? Wrongly-done audio clipping on AM is just as bad as on SSB.

RF clipping circuits are quite another thing from audio.

Transmit a square wave on an AM transmitter and you see a square wave
in the AM envelope. Do the same with an SSB transmitter and you only
see sharp spikes in the envelope.

That depends on the frequency of this square wave. That also depends
on what is being used to view the RF envelope. A 50 MHz scope will
show the RF envelope of any HF rig.

Put an electronic keyer on the SSB transmitter and transmit only dots
at a high speed setting. The SSB envelope will show the dots as
dots.

Conversely, if you put a high-purity sinewave audio into a SSB xmtr,
a spectrum analyzer display will show only a single frequency signal.

No one can interchange frequency and time domains directly and
get an explanation. Envelope viewing is time domain. Spectral
analysis is frequency domain.

Len Anderson
retired (from regular hours) electronic engineer person

On 24 Oct 2003 01:00:44 GMT, (Avery Fineman)
wrote:

In article , Gary Schafer
writes:

Along the same line consider that the envelope of an SSB signal has no
direct relationship to the original modulation the way that an AM
signal does.

This is why you can not use RF derived ALC to control the audio stage
of an SSB transmitter the way you can with an AM transmitter.

You can't use ENVELOPE detection on SSB the same way it is
done on conventional AM.

But, you CAN use RF-derived feedback - if mixed with a steady
carrier to recover the modulation content - to do that very well.

Agreed.


Or audio clipping that works on AM but does not work the same on SSB.

? Wrongly-done audio clipping on AM is just as bad as on SSB.

RF clipping circuits are quite another thing from audio.

In SSB RF clipping, the signal being peak limited is the same as that
being transmitted.
If you do audio clipping on an SSB transmitter you can not limit the
peak output as you can in AM or in RF clipping of SSB, because the
output of the SSB transmitter has no direct relationship to the audio
in.


Transmit a square wave on an AM transmitter and you see a square wave
in the AM envelope. Do the same with an SSB transmitter and you only
see sharp spikes in the envelope.

That depends on the frequency of this square wave. That also depends
on what is being used to view the RF envelope. A 50 MHz scope will
show the RF envelope of any HF rig.

With a 1000hz audio square wave into an SSB transmitter you will not
see a square wave envelope as you would from an AM transmitter. You
will see sharp spikes.
My point is that what you get out of an SSB transmitter is not
directly representative of what you put in, as it is in AM.
Even though the modulation process in the SSB transmitter starts out
the same as it does in the AM transmitter. Much phase modulation takes
place in the SSB process.


Put an electronic keyer on the SSB transmitter and transmit only dots
at a high speed setting. The SSB envelope will show the dots as
dots.

Same as a CW transmitter.


Conversely, if you put a high-purity sinewave audio into a SSB xmtr,
a spectrum analyzer display will show only a single frequency signal.

Put two pure equal amplitude audio tones into the SSB transmitter and
the envelope out at first glance looks like a 100% modulated AM
transmitter envelope that is modulated by a single tone. But a closer
look will reveal that the envelope is not a pure sine wave. The
envelope is folded over so as to produce a sharper crossover.

If this signal is looked at or listened to on an envelope detector the
detector will produce high 2nd harmonic distortion. (that's another
story)

Look at the signal on a spectrum analyzer and it will show 2 carriers
seperated by the seperation of the two tone frequencies.


No one can interchange frequency and time domains directly and
get an explanation. Envelope viewing is time domain. Spectral
analysis is frequency domain.

Agreed.

Len Anderson
retired (from regular hours) electronic engineer person


73
Gary K4FMX

On 24 Oct 2003 01:00:44 GMT, (Avery Fineman)
wrote:

In article , Gary Schafer
writes:

Along the same line consider that the envelope of an SSB signal has no
direct relationship to the original modulation the way that an AM
signal does.

This is why you can not use RF derived ALC to control the audio stage
of an SSB transmitter the way you can with an AM transmitter.

You can't use ENVELOPE detection on SSB the same way it is
done on conventional AM.

But, you CAN use RF-derived feedback - if mixed with a steady
carrier to recover the modulation content - to do that very well.

Agreed.


Or audio clipping that works on AM but does not work the same on SSB.

? Wrongly-done audio clipping on AM is just as bad as on SSB.

RF clipping circuits are quite another thing from audio.

In SSB RF clipping, the signal being peak limited is the same as that
being transmitted.
If you do audio clipping on an SSB transmitter you can not limit the
peak output as you can in AM or in RF clipping of SSB, because the
output of the SSB transmitter has no direct relationship to the audio
in.


Transmit a square wave on an AM transmitter and you see a square wave
in the AM envelope. Do the same with an SSB transmitter and you only
see sharp spikes in the envelope.

That depends on the frequency of this square wave. That also depends
on what is being used to view the RF envelope. A 50 MHz scope will
show the RF envelope of any HF rig.

With a 1000hz audio square wave into an SSB transmitter you will not
see a square wave envelope as you would from an AM transmitter. You
will see sharp spikes.
My point is that what you get out of an SSB transmitter is not
directly representative of what you put in, as it is in AM.
Even though the modulation process in the SSB transmitter starts out
the same as it does in the AM transmitter. Much phase modulation takes
place in the SSB process.


Put an electronic keyer on the SSB transmitter and transmit only dots
at a high speed setting. The SSB envelope will show the dots as
dots.

Same as a CW transmitter.


Conversely, if you put a high-purity sinewave audio into a SSB xmtr,
a spectrum analyzer display will show only a single frequency signal.

Put two pure equal amplitude audio tones into the SSB transmitter and
the envelope out at first glance looks like a 100% modulated AM
transmitter envelope that is modulated by a single tone. But a closer
look will reveal that the envelope is not a pure sine wave. The
envelope is folded over so as to produce a sharper crossover.

If this signal is looked at or listened to on an envelope detector the
detector will produce high 2nd harmonic distortion. (that's another
story)

Look at the signal on a spectrum analyzer and it will show 2 carriers
seperated by the seperation of the two tone frequencies.


No one can interchange frequency and time domains directly and
get an explanation. Envelope viewing is time domain. Spectral
analysis is frequency domain.

Agreed.

Len Anderson
retired (from regular hours) electronic engineer person


73
Gary K4FMX

In article , Gary Schafer
writes:

Along the same line consider that the envelope of an SSB signal has no
direct relationship to the original modulation the way that an AM
signal does.

This is why you can not use RF derived ALC to control the audio stage
of an SSB transmitter the way you can with an AM transmitter.

You can't use ENVELOPE detection on SSB the same way it is
done on conventional AM.

But, you CAN use RF-derived feedback - if mixed with a steady
carrier to recover the modulation content - to do that very well.

Or audio clipping that works on AM but does not work the same on SSB.

? Wrongly-done audio clipping on AM is just as bad as on SSB.

RF clipping circuits are quite another thing from audio.

Transmit a square wave on an AM transmitter and you see a square wave
in the AM envelope. Do the same with an SSB transmitter and you only
see sharp spikes in the envelope.

That depends on the frequency of this square wave. That also depends
on what is being used to view the RF envelope. A 50 MHz scope will
show the RF envelope of any HF rig.

Put an electronic keyer on the SSB transmitter and transmit only dots
at a high speed setting. The SSB envelope will show the dots as
dots.

Conversely, if you put a high-purity sinewave audio into a SSB xmtr,
a spectrum analyzer display will show only a single frequency signal.

No one can interchange frequency and time domains directly and
get an explanation. Envelope viewing is time domain. Spectral
analysis is frequency domain.

Len Anderson
retired (from regular hours) electronic engineer person

Along the same line consider that the envelope of an SSB signal has no
direct relationship to the original modulation the way that an AM
signal does.

This is why you can not use RF derived ALC to control the audio stage
of an SSB transmitter the way you can with an AM transmitter.
Or audio clipping that works on AM but does not work the same on SSB.

Transmit a square wave on an AM transmitter and you see a square wave
in the AM envelope. Do the same with an SSB transmitter and you only
see sharp spikes in the envelope.

73
Gary K4FMX


On Thu, 23 Oct 2003 12:08:31 -0700, "Joel Kolstad"
wrote:

Fred McKenzie wrote:
Perhaps that is what I'm remembering. Now, if you use a filter to
eliminate the other sideband, the higher frequency components and the
carrier, don't you have a nearly identical remainder?

At that point I don't think you could tell the difference since there's no
longer any local phase reference (i.e., the carrier) to compare with. I
suppose this is why your SSB-AM rig is able to (somewhat) receive low
frequency (and thereby presumably narrowband) FM broadcasts; this is what
you were saying in your last post, correct?

I think we are in agreement that you can't recover FM modulation with
just an envelope detector

Yes, at least you can't recover a signal that directly corresponds to what
you transmitted. It does appear that you can recover the signal's square,
however, so this approach might be useful for, e.g., remote command
transmissions. (But probably just for the novelty of having said you did
it... since it's probably not much harder to build the slope detector you
describe!)

---Joel