The AM Forum
March 28, 2024, 12:35:47 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Deep Thinking about Modulation  (Read 8933 times)
0 Members and 1 Guest are viewing this topic.
w8khk
Member

Offline Offline

Posts: 1200


This ham got his ticket the old fashioned way.


WWW
« on: November 04, 2018, 12:10:32 AM »

My cardiologist recommends that I consume 6 to 8 oz. of red wine each night for heart health. Some nights I take a double dose and feel twice as good. With an extra hour to sleep tonight (but being retired for so many years, who is counting hours anyway) - I had a double dose of that good medicine. That usually makes deep thinking work so much better! Without it I try to think, but nothing happens!

My thoughts sometimes are borderline extra-terrestrial or super-natural, or maybe more rightly should be limited to the first of April, but please bear with me for a few sentences...

So I got to thinking about various transmission modes, modulation, and power levels. Some folks have commented that CW is just a very narrow form of AM. But the CW power limits are much different than those imposed on the analog-modulated AM mode of communication. Isn't it true that the bandwidth of a CW signal increases as the sending rate is increased? Actually, CW would really be a continuous carrier, unbroken by a telegraph key. If this were true, it would consume approximately zero bandwidth. Isn't CW an abbreviation for Continuous wave, anyway? But we refer to CW as a keyed carrier, (discontinuous wave) encoding the message with the Morse dot and dash combinations. This results in varying carrier on times for dits and dahs, and various off times for spaces between characters, etc. Kinda sounds like a human form of pulse duration modulation, or pulse width modulation, doesn't it??? The faster we send these short carrier pulses, the wider the bandwidth of the "CW" signal.

So, continuing on this line of thought, if CW is just a make and break of the carrier at arbitrary times depending upon the information being sent, is there any limit to how fast we can send in this manner? Aside from the issue of prohibited codes and ciphers, not applicable here and which will be readily apparent (later in this discussion) and not intended to camouflage the meaning of the transmission, shouldn't it be permissible to send as fast as the receiving station can reliably and comfortably copy our sending?

Unfortunately, high-level plate modulation, grid, screen, and cathode modulation, all are analog methods of modulation, clearly not CW. Thus there is an arbitrary, inequitable penalty imposed in power level allowed with this mode of modulation. But what about PDM, or PWM? Here we are simply turning the supply voltage to the RF amplifier on and off at a high rate of speed, but we vary the on and off times, in a manner similar to the variation of on and off times when sending Morse characters generating CW with a telegraph key. (I hear objections to this theory already, but please bear with me for another moment.) With Morse CW, if the sending is fast enough, the receiving mind actually "hears" the words as each character is recognized in succession. Perhaps the sound of each character, at such a high signalling rate, is interpreted as syllables instead of letters. If we modulate with voice, converted to a PDM string of on and off pulses, the mind "hears" the words spoken, but does not hear the individual on-off pulses generated by the modulator. Is this not very similar to how the OM copies high-speed CW? Is this analog, or is this really just a very fast method of switching the carrier on and off? CW??? Almost!

But wait, there is more! When sending CW, it is not good practice to cut the carrier on and off abruptly, as this creates key clicks. So what do we do to prevent this? We round off the waveform of the keying (modulation), gently turning the carrier on and off to avoid the key clicks, which then avoids the splatter that would otherwise result from true CW keying. So we do have a bit of an analog waveshaping of the modulation when we transmit clean CW due to the key-click filter. This is referred to in the handbooks as proper engineering practice.

How do we deal with this problem with a PDM rig? With the PDM filter, of course. We filter out the abrupt on-off sharp transitions, eliminating the splatter and wide signal that would be generated by this undesirable keying waveform, thus preventing a wide, spattering signal. So, from this point of view, it appears that a PDM modulated carrier is "identical" to keyed CW emanating from a modulated CW RF amplifier, albeit the transmitter is keyed at such a fast rate that the listener's mind does not even hear the letters, but the words are clearly formed whether or not the receiving op is proficient in copying fast morse code. So, not only would this method of transmission enhance the effectiveness and value of CW, it would get more ops on the air sending and receiving CW efficiently without a long arduous study to get their code speed up to snuff!  Is it acceptable to assume that a PDM filter performs a function analogous to the CW key click filter?  Perhaps!

There is still considerable controversy about the AM carrier and sidebands, whether the carrier really exists when modulated, or does it disappear when the modulation waveform reaches a trough and the supply to the modulated amplifier reaches zero? When you look at the RF waveform in the time domain, you get a very different impression than when you look at it in the frequency domain. So, perhaps a PDM signal is just very-very-fast CW, with a properly designed key click filter, that allows spatterless CW that can be copied by the lay person with almost no training at all. Actually, it is much easier to tune than either a sideband or CW signal. Perhaps looking at modulation in this light will strengthen it's position in the vast group of modulation methods available to the amateur operator. Perhaps it is much more advanced than anyone ever realized?

I assume some official from some unfriendly enforcement agency might mistakenly interpret the transmission envelope as some form of amplitude modulation, but clearly, it is simply a form of discontinuous CW, with a level of filtering that effectively removes all key-clicks that would otherwise produce splatter and interference to adjacent channels. I believe this represents true progress by amateurs in producing a CW modulation method (more efficient than sideband or other forms of modulation) that requires no special training or equipment to receive and comprehend!   The fact that sending also requires no special skills or training makes it a double bonus!  (Come to think of it, this mode of CW produces two identical sidebands, either of which may be clobbered by incidental or intentional QRM. The experienced op may choose to receive both sidebands when they are in the clear, or select only one sideband when the other one gets hammered by a lid.

If you believe this dissertation should have been tabled and revealed only on April first, please let me offer you a bottle of Merlot and a corkscrew, and please accept this technological discovery a few months early. The vino really does clarify the thinking process.

73 to all, and to all a good night.  (Don't forget to set your clocks back for that extra hour of sleep that they stole from you in the spring!)
Logged

Rick / W8KHK  ex WB2HKX, WB4GNR
"Both politicians and diapers need to be changed often and for the same reason.”   Ronald Reagan

My smart?phone voicetext screws up homophones, but they are crystal clear from my 75 meter plate-modulated AM transmitter
PA0NVD
Member

Offline Offline

Posts: 615


Nico and Chappie (Chappie is the dog...)


« Reply #1 on: November 04, 2018, 11:03:53 AM »

Hi Rick
Interesting. Actually I tried this many years ago. I made a normal audio PWM signal by comparing audio and a triangular wave. This signal I did feed to a diode switch using 4 diodes. . The switch did turn the RF input on and off with the PWM signal. It was quite wideband that resulted, so I had to filter with an FM crystal filter at 9 MHz (the RF signal was 9 MHz)
What I got was AM, quite clear. I did not have a spectrum analyser at that moment, so I could no analyse better than with a narrow band receiver.
Due to he fixed frequency required by a extensive filtering, the generation of AM this way was as complicated as SSB, so I never tested more than that interesting experiment
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #2 on: November 06, 2018, 12:39:30 PM »

One view of CW Bandwidth by Mark Amos

https://www.w8ji.com/cw_bandwidth_described.htm

Explanation of Pulse Width Modulation for AM:

http://www.classeradio.com/theory.htm


Phil = AC0OB

Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #3 on: November 07, 2018, 09:40:50 PM »

Posted by: w8khk
There is still considerable controversy about the AM carrier and sidebands, whether the carrier really exists when modulated, or does it disappear when the modulation waveform reaches a trough and the supply to the modulated amplifier reaches zero? When you look at the RF waveform in the time domain, you get a very different impression than when you look at it in the frequency domain. So, perhaps a PDM signal is just very-very-fast CW, with a properly designed key click filter, that allows spatterless CW that can be copied by the lay person with almost no training at all. Actually, it is much easier to tune than either a sideband or CW signal. Perhaps looking at modulation in this light will strengthen it's position in the vast group of modulation methods available to the amateur operator. Perhaps it is much more advanced than anyone ever realized?


My opinion without portfolio on the question of whether the carrier disappears at the trough of the modulation cycle and sidebands come into being when a signal is amplitude modulation is very simple. (Assuming a perfect signal generating system and resistive load, the output power will follow the output voltage) At a given instant, the output voltage will become the answer to the question. I will cite the high level modulation system because it is what I understand decently.

At trough of modulation envelope, the voltage output of the transmitter is zero. The carrier disappears, all the entire signal disappears. There is no power supply into the RF amplifier, and any power driven into the carrier excitation input element of the amplifier device (the grid/gate/base/etc.) is dissipated in the element. In a perfect system, nothing gets out the plate (or other).

At any other moment in time, a specific voltage exists at that time depending on the degree of modulation from zero to 100%. The waxing and waning of the RF waveform can be seen on a scope and a cursor can be placed at any point along the wave on a high-feature scope.

What shows up on a spectrum analyzer centered on the carrier frequency is the result of the detection of the AM signal in the frequency domain that is in the fashion of a homodyne receiver. Citing wikipedia (which is totally unreliable as far as scholars are concerned) "Homodyne detection is a method of extracting information encoded as modulation of the phase and/or frequency of an oscillating signal, by comparing that signal with a standard oscillation that would be identical to the signal if it carried null information. "Homodyne" signifies a single frequency, in contrast to the dual frequencies employed in heterodyne detection".

The analyzer sweeps a bandwidth encompassing at least twice the modulating frequency and so it shows a modulated signal as a lower side band, carrier, and upper sideband. The pip size for each would depend only on the instantaneous RF voltage, but due to the slow (compared to audio) sweep of 10-20Hz and the persistence of the (beautiful CRT) phosphor, the image written is the familiar type.

If we were to have an AM transmitter capable of audio response down to 0 Hz, and applied 1 Hz modulation, and we could have an analyzer with sufficiently fine frequency resolution, we would see on the analyzer a graphical image that would change over the audio cycle, that is, the carrier pip would increase and decrease and so would the sideband pips.

Of course, trying to view two 1Hz sidebands and the carrier so that they are separate, and do this with a 10Hz sweep speed is not going to work well, at least not on the gear I have. maybe some digital filter type units might do it.

To get around this issue, it is only necessary to provide a 100% modulated AM signal for example 10MHz modulated by a 1KHz audio sine wave, and then additionally amplitude-modulate the audio signal, before it goes into the RF amp modulator, at the low speed of 1Hz for viewing purposes. Modern audio signal generators should have a DC coupled modulation input and allow this.

The result, that I believe, is that the neither the carrier nor the sidebands ever disappear above the negative 100% modulation level. They are mixed and the products of the mixing is what is seen on the spectrum analyzer frequency-vs-amplitude plot and on the scope voltage-vs-time plot.

Some people may have said that that at zero modulation (carrier level) only the carrier pip is present, and that above or below that the audio pips are present, and that the carrier is constant and the audio signal vary with time, and that the 'tall' carrier pip we see is the result of the summing of the modulated and unmodulated components. I don't have an opinion on this.

It would be interesting to see a plot of frequency vs. time but that is another instrument and I don't have one, therefore the elaborate experiment above, to see through a glass darkly this mystery of the physical universe that we call AM.

Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
R. Fry SWL
Member

Offline Offline

Posts: 114

Broadcast Systems Engineer (retired)


WWW
« Reply #4 on: November 08, 2018, 07:36:50 AM »

Comments to consider/add to the discussion mix —

The graphic below shows the output of an SDR looking at the r-f spectrum of a local, 5kW, non-directional AM broadcast station located about three miles away from the receive setup.

Splatter is produced as a function of the very short gaps in the carrier output during excessive negative-going modulation, analogous to the key clicks that can be produced when using very short on-off transition times when keying a "CW" (Morse) transmitter.  


Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #5 on: November 08, 2018, 10:12:14 AM »

Actually that Radio Whirled article?
4.5kHz sounds "really good"?
REALLY?
In what friggin universe it that?

OOOPSY!! Silly Bear! It says +/-9.5kHz... geez, you'd think I could read.
Still it is a bit narrow for music...
...and doesn't quite explain why AM talk that is simulcast with FM still sounds like
MUD by comparison...

Anyhow...
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #6 on: November 08, 2018, 11:24:21 AM »

Fm talk radio sounding like mud when simulcast on am....

I'll suppose an idea.  Processing set up for fm?

I've never A/B with an am and fm processor, so I have no idea.

I know that with preemphasis the processor spectral output would be different than for am, no?

--Shane
KD6VXI
Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #7 on: November 11, 2018, 01:16:05 PM »

...and doesn't quite explain why AM talk that is simulcast with FM still sounds like
MUD by comparison...


Just about any modern AM transmitter out there will transmit audio out to at least 10kHz. I just ran a Proof-of-Performance on a local 10kW PDM modulated transmitter back in the spring and the transmitter, without processing, could re-produce audio (with at least 100% Modulation) out to 15 kHz before it started rolling off.

There are two limiting factors in play for decent AM audio:

1) The example Mr. Fry gave shows how the AM receiver, and not the transmitter, is the factor limiting AM bandwidth and audio quality because of its narrow bandwidth IF strip,

2) Processing. While there is a Pre-emphasis system in play, in an attempt to overcome the narrow bandwidth and noise, many commercial AM processors are set to limit the audio bandwidth for News/Talk/Sports formats.

Below is a file showing a simplified AM/FM system for simulcasting as for a station set up for AM-FM Translator operation. If the audio processor is set up properly (as I think I do), the only real difference, the real limitation in received/perceived audio quality, will be due to your receiver's IF bandwidth.  

Listen to stations with a rock format (and set up properly) with a GE III or other radios with a wide bandwidth setting and the difference will be astounding.

I.e, If the transmitted AM bandwidth were allowed to be 30 kHz, and the receiver Bandwidth was forced to be 30kHz, FM quality AM audio could be delivered.

Summary Opinion: The delivery of Quality AM audio is NOT a technical problem, but the problem purely rests with the FCC and its archaic technical standards and policies.


Phil - AC0OB

 



* Audio System Studio to Transmitter.pdf (20.88 KB - downloaded 324 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
R. Fry SWL
Member

Offline Offline

Posts: 114

Broadcast Systems Engineer (retired)


WWW
« Reply #8 on: November 11, 2018, 03:25:35 PM »

... The example Mr. Fry gave shows how the AM receiver, and not the transmitter, is the factor limiting AM bandwidth and audio quality because of its narrow bandwidth IF strip...

Sorry, but that is incorrect.

The SDR device used to produce that graphic was/is essentially flat across that entire r-f bandwidth.  The steep rolloffs shown for the upper and lower sidebands at about Fc ±9.5 kHz are the result of the audio processing needed for the transmitter to implement the NRSC-2 curve in its transmissions (see below).

The NRSC-2 analog AM audio spec includes pre-emphasis for program audio below about 9.5 kHz, and calls for appropriate de-emphasis in receivers.  However most consumer-level AM broadcast receivers do not pass much of the transmitted spectrum beyond Fc ±4.5 kHz, or so.

Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #9 on: November 11, 2018, 06:09:51 PM »

... The example Mr. Fry gave shows how the AM receiver, and not the transmitter, is the factor limiting AM bandwidth and audio quality because of its narrow bandwidth IF strip...

Sorry, but that is incorrect.



I am not sure to what you are objecting.

The "example" I was referring to is the whole article 'paste' including the spectrum sweep plus the text:

"...Most commercial-level AM receivers are not designed to receive/reproduce program audio having a ~9.5kHz bandwidth. But that limitation is not a function of the capabilities of amplitude modulation itself...."  Emphasis mine.

This implies, and rightly so, that it is NOT the transmitter proper that limits good audio, but the fault lies with the receiver's IF bandwidth that can make AM audio sound "muddy." Add to that the crummy processing "re-set" by non-engineering Program Managers and you have muddy audio.

Without a wide bandwidth for reception, any audio beyond ~ 4.5kHz is essentially lost.

The NRSC Pre-emphasis helped somewhat but did not solve the problem of receiver bandwidth; i.e., it was a bandaid at best.

The latest version of NRSC permits fixed audio pre-emphasis with an upper frequency limit of about 9.5 kHz. The pre-emphasis was intended to compensate for the sideband rolloff of the r-f filters (mostly the IF filters) in the receiver.

So a receiver manf. could design and market an AM broadcast receiver with an audio output that was fairly flat to about 9.5 kHz. That is nowhere equal to that of what FM broadcast receivers can do, but it would sound a lot better than typical present-day AM receivers with their 3 to 4 kHz audio bandwidth's.

Many in the industry felt that at the time of the standardization of C-QUAM stereo, the FCC was derelict in their duties to impose proper bandwidth standards for AM receivers. See file below.

Another good article to Google on this topic is: Opinion: 'Let's Keep AM Sounding Good' The Movement to Reduce AM Radio Bandwidth Ignores the Realities of Hearing and Hardware
January 19, 2005, by Dana Puopolo



Phil - AC0OB


* AM Stereo and FCC Deregulatory Breach of Duty.pdf (1331.23 KB - downloaded 1410 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
R. Fry SWL
Member

Offline Offline

Posts: 114

Broadcast Systems Engineer (retired)


WWW
« Reply #10 on: November 13, 2018, 05:41:21 AM »

...  So a receiver manf. could design and market an AM broadcast receiver with an audio output that was fairly flat to about 9.5 kHz. ...

That wouldn't be too useful at night due to the skywave signals of analog AM broadcast stations only 10 kHz offset from a desired station (see the graphic below).

The interference issue when using receiver rf/if bandwidths of ~ 19 kHz would be especially bad for North American "Class A" AM stations assigned on a set of 10-kHz-adjacent carriers, such as now done for 740/750/760/770/780 kHz.

Logged
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #11 on: November 13, 2018, 01:49:18 PM »

...  So a receiver manf. could design and market an AM broadcast receiver with an audio output that was fairly flat to about 9.5 kHz. ...

That wouldn't be too useful at night due to the skywave signals of analog AM broadcast stations only 10 kHz offset from a desired station (see the graphic below).

The interference issue when using receiver rf/if bandwidths of ~ 19 kHz would be especially bad for North American "Class A" AM stations assigned on a set of 10-kHz-adjacent carriers, such as now done for 740/750/760/770/780 kHz.



I am quite familiar with the "bandwidth" situation, and I see no offering of solutions to improved audio delivery for AM, which is one of my primary concerns if we are to make AM viable again.

And that is one of the problems. The FCC has allowed too many stations on the AM band with the resulting close a spacing for AM BC stations. In the FCC's mind, forget about good audio and let's squeeze as many stations into the AM band as possible. Let's roll-off audio with a steep curve to prevent adjacent channel interference caused by us, the FCC, when we went crazy and stuffed too many AM channels into the band.

With the demise of so many AM stations and them going dark, it's time to re-vitalize the AM band and "cull" the herd.

So are you a proponent of the 9kHZ RF bandwidth situation as used by many other countries? If so, that would certainly kill AM BC for good.

Again, NRSC pre-emphasis and sharp roll-off filters are simply a bandaid caused by the FCC (and greed) which reduced station separation vs. power.  

 


Phil
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #12 on: November 13, 2018, 02:02:25 PM »

Proof of Performance tests during the 1962-4 time frame for 500watts to 10kW transmitters in the midwest were done where we swept the audio and measured the modulation of the AM transmitter with various test tones from 50 hz to 15 kHz and they had to be within 3 DB from bottom to top of that audio range.

We were not limited at that time to a top end audio of 7.5kHz, and the audio was alive and sparkled.

Phil
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.081 seconds with 19 queries.