[AMRadio] Re: GB> AM vs SSB???


D. Chester k4kyv at charter.net
Sun Apr 8 14:21:48 EDT 2007


> SSB AM is not as good as full DSB AM but it is also not as poor as some
> believe it is. A properly set up SSB AM rig can sound pretty good.
> Modulation level does have to be kept down some as Don says or the 
> receiver
> detector distortion goes up rapidly as you approach full modulation. As 
> you
> near 100% modulation the receiver detector can't tell the difference 
> between
> the carrier and the modulation signal. Second harmonic distortion results.

The only reason it even sounds fairly decent is that with the human voice, 
the average amplitude level is equivalent to about 30% modulation.  The 
peaks that approach 100% are relatively infrequent, so the distortion is 
tolerable.


> When I say set up properly, there are few that know how to do that right.
> Most that run SSB rigs on AM, in the DSB with carrier mode have a hard 
> time
> making them sound good on AM and it is mostly the fault of not setting it 
> up
> properly. Some sound very nice.

Most modern transceivers run true DSB in the AM mode.  Theoretically, there 
is no reason these wouldn't sound identical to a broadcast transmitter.  The 
carrier level setting is critical, and the audio level is critical.  These 
level settings must be so that the carrier is modulated 100% on voice peaks, 
but enough headroom is left to allow the crest of the positive modulation 
peak to reach 100% without flat-topping.


> Want to see what good SSB AM sounds like? Just tune off to the side of a
> good sounding AM signal with your receiver so that the filter cuts one 
> side
> band off but not enough to reduce the carrier. The audio level will drop a
> little and distortion may rise slightly but it will still sound pretty 
> good.
> You are hearing the same thing as if it was transmitted with only one side
> band. This same trick is often done when there is heavy qrm on one side. 
> You
> listen to SSB AM and don't even realize it!

You still hear the distortion on heavy voice peaks.  This would be better 
described as equivalent to vestigial sideband, just like the video TV 
signal. The lower frequencies, where the heaviest peaks usually occur are 
not cut off with the filter, since the receiver is rarely tuned to one 
sideband to the point that the carrier is reduced, so at the lower 
frequencies with the most peak audio power, you are still demodulating DSB. 
The part of the signal that is effectively SSB is in the higher frequency 
range.  With the human voice, these frequencies are naturally low amplitude, 
so quadrature distortion from the SSB detection mode  is usually not 
apparent, unless the high frequency components are riding on top of  heavy 
peaks at lower frequencies.  In the latter case, the distortion is apparent, 
and comes through as intermodulation distortion.


> As to Collins S line receivers sounding better than some of the rice box
> receivers, I haven't heard one. I have several 75S-1, 75S-3, 75S3A etc.
> receivers. All have more distortion than my Kenwood TS430. I measured each
> with a distortion analyzer, and you can hear the difference, and the TS430
> runs rings around any of the Collins receivers in audio quality. So just
> because it's old doesn't readily mean that it is better.

>From my experience, the audio section of all Collins amateur equipment from 
the era of the 75A4 on, is an afterthought at best.  I made my 75A4 sound 
reasonably good on AM with several reversible modifications.  This inclues 
increasing coupling capacitance between low level stages, and using an 
outboard audio power amplifier, with the AF signal picked off the AF gain 
control.  A plug-in module replaces the 1st audio stage (12AT7 IIRC), and 
delivers  the audio to the outboard amp via a cathode follower stage. 
Mechanical filters are the limiting factor in the attainable audio quality 
on any receiver that uses them.  That is why the R-390 sounds so much better 
on AM than does the R-390A.


The distortion that occurs with SSB + carrier can be easily and clearly 
demonstrated using the standard rotating vector diagrams like the ones in 
the 1960's era ARRL Handbooks and their series Single Sideband for the Radio 
Amateur.  I first became aware of it years ago when I built an experimental 
SSB generator at 64 kHz using a telephone company multiplexing mechanical 
filter, and looked at  the envelope of the signal on a scope with zero, 
reduced and full carrier.  With full carrier, the familiar modulation 
envelope of AM modulated with a sinewave became increasingly distorted as 
the percentage approached 100%.  At 100% negative modulation, the waveform 
becomes a sharp point at the part of the envelope near the baseline.

Interestingly, quadrature distortion is almost a mirror image of the 
distortion heard with a square-law detector.  SSB + full carrier should come 
thorugh distortion free if the signal level fed into the detector is such 
that the true square-law curve extends from zero signal to the crest of the 
100% modulated voice peak.  Square law detectors were commonly used on late 
1920's-early 1930's broadcast radios using triode grid-leak detectors. 
That's what gives those radios that peculiar sound that is almost like 
selective fading distortion.  Later BC sets switched over to diode detectors 
because the latter has a linear characteristic and not square-law.

It would be interesting to experiment with SSB-AM received with a grid leak 
detector to see if it really does demodulate that mode without the usual 
distortion.

Usually, if I have to tune to one side of the AM signal to dodge QRM, I turn 
on the BFO and copy it as SSB.  The inserted carrier into the product 
detector produces a true product of the sideband and carrier, without 
quadrature distortion.

When SSB is tuned in with a diode dectector, that's the reason the rf gain 
has to be turned down to get a clear signal.  Think of the SSB output from 
the i.f. amplifier as the modulation, and the BFO as the carrier.  If  the 
signal being fed into the diode decector is strong, the BFO is being 
modulated at a high percentage, hence the distortion.  When the rf gain is 
reduced, the percentage of modulation is effectively reduced to a low value, 
and the distortion is not apparent.

Don k4kyv




More information about the AMRadio mailing list

This page last updated 23 Oct 2017.