[AMRadio] AM Bandwidth and Audio Quality (was Source Broadcast Xmtrs)

[D. Chester]

To avoid further hijacking the thread, I am starting this new one. Due to an 
e-mail problem, I was only able to download these message this morning, so 
here is my response to some points in the discussion I found of particular 
interest .

>   Thirdly, USA HAMS are limited to 375 Watts unmodulated carrier, (1.5 kW 
> P-P) a one kW AM
> transmitter can produce one thousand Watts of unmodulated carrier.  (so 
> you need to cut it back)

>   Ron KA4INM  73

Show me anywhere in Part 97 where the phrase "375 watts" even appears in the 
text.


> From: "Bernie Doran" <qedconsultants at embarqmail.com>
>of course now
> someone will chip up and say we can only use 3 kc bandwidth, or that we 
> need
> to conserve bandwidth, does it wear out after a period of time?

Nor is there anything in Part 97 that prescribes ANY specific bandwidth 
limitation. The FCC left it vague for a purpose: to give amateurs the 
maximum flexibility for experimentation and self-instruction in the radio 
art.  The rules do require one to operate according to "good amateur and 
engineering practice". That means that we should adjust our transmit 
bandwidth according to band occupancy and band conditions, just as we adjust 
our receiver selectivity. If a band like 75m or 160m is lightly occupied as 
is often the case during daylight hours or in the wee hours of the morning, 
there is nothing wrong with opening up the audio to full high fidelity and 
transmitting a signal that is 30 kc/s wide. But during peak prime time 
operating hours, as during the evening on 75m, I switch in a sharp cut-off 
low-pass audio filter with a brick wall cutoff at approximately 3400 cps. 
Most of the time I use a more gradual cut-off that begins to roll off at 
about 5000 cps. Most of the undesirable splatter heard on the bands, whether 
from AM or SSB, results from spurious distortion products, not the frequency 
response of the audio used for modulation.


> Bandwidth for communications quality AM is +/- 3KC. Commercial AM 
> broadcast
> transmitters have more bandwidth than this. Close to +/- 5KC.
>
> Bob Macklin
> K5MYJ

True enough, but there is nothing in Part 97 that specifies that amateur AM 
must be "communications quality".


> From: "Jim Tonne" <Tonne at Comcast.Net>

> And speaking of hi-fi, that is nonsense about
> two ham signals closer than 10 kc to each
> other causing a heterodyne to both.   I routinely
> listen to AM on my Collins 51J4 using the
> 3.1 kc mechanical filter.   I have yet to hear
> a heterodyne over about 4 kc...

My 75A-4, which is modified to accommodate additional mechanical filters, is 
equipped  with a Sherwood SE-3 synchronous detector, which allows me the 
full advantage of product detection to be used with the double-sideband AM 
reception. I have at my disposition filters with 3.1, 4.0, 6.0 and 8 kc/s 
selectivity for voice reception. I have found that even under congested band 
conditions, the signal is usually more readable with the 8 kc/s filter.  The 
additional sibilance response makes the voice easier to understand.  There 
is a lot of information contained in the voice frequencies above 3000 cps. 
"Telephone quality audio" is designed for efficiency rather than audio 
quality, a compromise between bandwidth and intelligibility. Thanks to the 
wonders of the human brain, the ear is capable of filling in the gaps in the 
audio spectrum that occur with "communications quality", but the result can 
be grating and fatiquing to listen to for extended periods. Only when there 
is strong or devastating adjacent channel interference do I find the 
narrower filters helpful in rendering the received signal more readable.


> ...And it is treacherous to use a ricebox to make
> an AM signal and follow it by a linear.   The
> linear has to be adjusted   *just right*   or it will
> cause splatter.
>
> - Jim W4ENE

It can be done, but AM linear operation, or other forms of low-level 
modulation like control or screen grid modulation, are much less forgiving 
than is plate modulation.  With linear amplification/grid modulation, the 
grid drive level to the final, and the degree of coupling from the final to 
the antenna are critical, to achieve a linear modulation characteristic all 
the way to 100% modulation, plus the audio level has to be adjusted just 
right  to fully modulate the carrier without overmodulating.

With plate modulation, grid drive to the final and antenna loading are much 
less critical. As long as the grid of the tube has enough drive to make the 
tube reach saturation and operate as a true class-C amplifier over the 
entire audio cycle, and the antenna coupling is not excessive to the point 
where the output capability of the tube(s) is not exceeded, good modulation 
linearity and 100% modulation capability can be easily achieved.  Of course, 
it is still important to maintain the proper audio level to avoid 
overmodulation while peaking close to 100%, and the final must be clean and 
free of parasitics and other spurious products. Not to mention that the 
audio that is being fed into the modulator must be clean and distortion-free 
to begin with.

> From: Ron Youvan <ka4inm at tampabay.rr.com>

>   No, would you like to have your audio bandwidth out to 5 kc?  Have a 
> signal 10 kc wide? *
> even with the rule:
>
> Part 97 : Sec. 97.307 Emission standards
> (a) No amateur station transmission shall occupy more bandwidth than 
> necessary for the information
> rate and emission type being transmitted, in accordance with good amateur 
> practice.
>
> If you want REAL fidelity, try FM, see you on 29+ MHz.  (or 2 meters)


But again, Part 97 contains no specific bandwidth limits for amateur phone 
operation. If audio bandwidth out to 5 kc/s is "more bandwidth than 
necassory for  the information rate being transmitted" on 75m or 160m, why 
is it not on 29+ mc/s or 2 metres as well?  What does the transmit frequency 
have to do with information rate being transmitted?

I routinely operate 75 and 160m with audio bandwidth out to 5 kc/s and 
rarely draw any complaints.  I narrow up the audio bandwidth to 3400~ under 
congested conditions, and usually keep the 3400~ filter in line when 
operating 40m. That's what the FCC had in mind with the phrase "good 
engineering and amateur practice".  Regarding the "information rate being 
transmitted", full fidelity voice runs at a greater information rate than 
does restricted "telephone quality" voice, so therefore the additional 
occupied bandwidth is legal under that rule cited above. OTOH, if one is 
transmitting telephone quality audio at 300-3000 cps, but distortion 
products cause the signal to occupy a total of 10 kc/s, the signal would be 
violation of the same rule because the signal would be occupying excessive 
bandwidth for the audio being transmitted.

>   Is your MIC is flat to 15 kc?  And your voice goes out to 15 kc?

No, so for precisely that reason, if the speech amplifier is reasonably 
distortion-free and the final amplifier is operating with good modulation 
linearity, there is really no need for any kind of frequency response 
limiting. The amount of energy in the human voice that falls above 3000 cps 
is so low that it seldom causes harmful interference. It may be audible, but 
we in the amateur radio service operate on an interference-expected basis, 
and no-one is guaranteed a clear channel.

Look in the older Handbooks (both RADIO and ARRRL) that date back to the 
days when AM was the predominant voice mode. Do you see low-pass audio 
filters in the circuit diagrams of all the speech equipment featured?  In 
Canada, they have had a specific bandwidth limitation of 6 kc/s written into 
their rules for decades, but few, ifany Canadian AMers have any special low 
pass audio filtering, and I have NEVER heard of a Canadian AMer running a 
normal transmitter properly adjusted being cited for "excessive bandwidth". 
This whole thing is a non-issue.


>...BTW, do you tell SSB folks that they should limit >their bandwidth to 
>1.8 Khz? Because 1.8Khz is all that is necessary for the "inforamation rate 
> >and emission type" of SSB. The part that says "in accordance with good 
>amateur pratice" is somewhat subjective but it is known that good >amateur 
>practice among the A.M. operators is "pleasing armchair ragchew fidelity". 
>Good amateur practice also means if the band is crowded to >tighten up the 
>bandwidth but if it is not crowded to allow for a more pleasing fidelity 
>that does not fatigue the listener.
>
> 73, John KX5JT AMI# 1684

If the sole purpose of amateur radio is to transmit as narrow a bandwidth as 
possible, then AM, SSB, FM and all other forms of phone should be outlawed, 
and everyone limited to CW or PSK. But many hams (I'm not sure about the 
HAMS) operate SSB instead of CW because they prefer to talk into a 
microphone to manipulating a key, even though SSB takes up more bandwidth. 
By the same token, others operate AM instead of SSB because we prefer the 
quality of AM voice to the scratchy, pinched, distorted sound of SSB, even 
though AM takes up more bandwidth. So where is the arbitrary line to be 
drawn between bandwidth and personal preference?  At the maximum necessary 
bandwidth of CW? of SSB? of telephone-quality AM? of near-broadcast quality 
AM?  And who draws that line?  Would we want the FCC to further micro-manage 
amateur radio (even if they were interested in doing so)?

The present rules have been working just as they are for decades. Back in 
the 50s, 60s and 70s,  there was a lot of obsession with "bandwidth", but 
the HF phone bands were a lot more congested back then, plus there were 
markets to be created for new products. Levels of activity have been 
dropping for a couple of decades or more, first with the advent of 2m FM and 
repeaters, which took a lot of pressure off the HF bands for local ragchews, 
then along came computers, the internet and cell phones, things that now 
compete with amateur radio for one's time. And in the past few years, 
further pressure has been taken off with the expansion of the 75m phone 
band, and the removal of most 40m broadcast activity between 7.1 and 7.2 
mc/s.  With the relief in congestion on the phone bands and drop-off of 
amateur HF activity in recent years, signal bandwidth has pretty much become 
a non-issue.

Don k4kyv


_______________________________________________________________

This message was typed using the DVORAK keyboard layout.

http://www.mwbrooks.com/dvorak/
http://gigliwood.com/abcd/