On the tropical 60m band AM was the best I ever experienced. My DIY receiver at the time had a 12 KHz bandwidth and synchronous demodulation so I could tune to a -6dB carrier point and enjoy audio up to 12 KHz. At the time, early 1970s, that rivalled cassette recorder quality. Often, it could be perceived that the audio spectrum was wider than 12 KHz as percussion was detectable at more than 15 KHz from the carrier. This "HiFi SW" mostly came from Central America, like R. Colosal de Neiva, R.Sutatenza (Colombia), R. Lara, R. Continente (Venezuela), la Voz del Norte (Ecuador), R. Rebelde, Cuba. Even Costa Rica had such a service (forgot the name). These stations ware playing music, at the time unavailable in the Netherlands. Nowadays there still are some SW stations providing audio up to 10 KHz, on AM the requirement for it is low power, I think <= 5 KW ERP.
I first heard AM stereo being "announced" in the 1970s, and I think it finally came out (in the USA) in the 1990s, but by then almost no one cared - even the remaining AM classical stations were converted to talk radio by then. I don't know how many, if any, AM stations actually transmitted in stereo.
AM Stereo was a failed plan to "save" AM radio.
It also gave equipment manufacturers something new to sell (both transmitter exciters, and receivers).
AM radio is 'saving' itself.
How many of you remember the Elcaset tape recordings and playback machines?
How about SACD, where you could have a few old recordings in 3 channels?
4 track 1/4 inch 4 channel tape, anybody?
Shibata stylus and 4 channel Records, with a Mutiplex demodulator?
CBS SQ 4 channel matrix records?
Hafler 4 channel matrix records?
It also gave equipment manufacturers something new to sell (both transmitter exciters, and receivers).
AM radio is 'saving' itself.
How many of you remember the Elcaset tape recordings and playback machines?
How about SACD, where you could have a few old recordings in 3 channels?
4 track 1/4 inch 4 channel tape, anybody?
Shibata stylus and 4 channel Records, with a Mutiplex demodulator?
CBS SQ 4 channel matrix records?
Hafler 4 channel matrix records?
Apparently this voluntary preemphasis/deemphasis with a notch filter is from the early 1990's:
https://en.m.wikipedia.org/wiki/AMAX
https://en.m.wikipedia.org/wiki/AMAX
AM radio is supposed to be mandatory for vehicles in the USA again, or did I misunderstand it.
AM for Every Vehicle Act
AM for Every Vehicle Act
@stonegreen Senator Markey, sponsor of the legislation in the Senate, has been a "Communications Guru" for eons.
The threat of his legislation resulted in Ford backing down, and General Motors has no present plans to remove AM radio functionality.
With this being said, the quality of AM reception in the new Ford is atrocious -- and if you get near an EV the reception can be knocked out. I wonder if someone has thought to file an FCC "Part 15" Complaint against EV's which spew RFI. (Note, we already discussed LED's which are not covered by FCC Part 15).
The threat of his legislation resulted in Ford backing down, and General Motors has no present plans to remove AM radio functionality.
With this being said, the quality of AM reception in the new Ford is atrocious -- and if you get near an EV the reception can be knocked out. I wonder if someone has thought to file an FCC "Part 15" Complaint against EV's which spew RFI. (Note, we already discussed LED's which are not covered by FCC Part 15).
AM radio done correctly, and listening with a good radio, can sound very good. Correctly means that the transmitter is passing the full 10 KHz bandwidth and the receiver can pass that as well. The first big problem is that car radios usually don't pass anywhere near the full 10 KHz of bandwidth. Most are about 5 KHz or less. Why is that? This was the first nail in the coffin for AM. Next is the PPM watermark encoding for radio listener ratings. I was told that the most important listener is the Neilson PPM ratings meter that is carried in a woman's purse on her way to and from work in the car. The stations audio processing was then supposed to be made exceptionally dense, as in compressessed, in the 3 KHz area to produce more valid PPM watermark codes being received. More PPM codes received equals higher Neilson radio ratings scores. So the large market stations are really screwing with the audio to ekk out the last 0.1 share of the ratings score. The next problem for AM was the adoption of MDCL or Modulation Dependant Carrier Reduction. The idea is that you can reduce your transmitter's electric utility bill by 20% by using MDCL to turn down the carrier for short periods such as pauses between words during speech. For a 50,000 Watt transmitter that can be a lot of money. The problem is that it screws with the listening experience. HD radio, also known as Hybrid Digital and analog on AM, was a real thing for a while but it had problems of its own. If you listened on a good quality wide band analog radio you could hear the hiss of the nearby digital sidebands and it was bad. Also if you drive around listening to the HD on AM, in any place that the reception gets marginal, such as highway underpasses or below large power lines, the receiver would fail back to the analog AM. A lot of that failover switching on your radio sounded terrible. At night when the sky wave propagation increases and meets with the groundwave it would be unlistenable because of the switching back and forth that the radio did.
AM processing was a meat grinder throughout my 40 years in the industry. In the early 80's we were already running a 7 band (analogue!) processor. If anything decades of psychoacoustic and DSP advances reduced some of the more grievous damage.
Besides low quality receivers a big reason for AM's characteristic sound is asymmetry. AM modulation has a hard negative limit set by waveform pinch off. Conversely positive modulation is limited by transmitter dynamic power and your tolerance for distortion. AM processing allows for distorting the modulated waveform - 'stretching' the waveform in the positive direction - to gain ~1 dB more level. Typical asymmetry employed is 25%, or roughly 25% peak distortion. There's your 'crunch'.
When I retired the PPM argument was still active. Is it better to PPM encode pre-processor and let compression and limiting raise the embedded signal's level, or PPM encode as late as possible to minimise code alteration? The requirement to feed multiple destinations with different processing requirements from a single encoder - cable systems, transmitter, FM HD - usually mandates pre-processing injection. Typically the only processing accommodations I made was to select multi-band processing crossover frequencies to keep the PPM signal in a single band and minimise alteration.
Besides low quality receivers a big reason for AM's characteristic sound is asymmetry. AM modulation has a hard negative limit set by waveform pinch off. Conversely positive modulation is limited by transmitter dynamic power and your tolerance for distortion. AM processing allows for distorting the modulated waveform - 'stretching' the waveform in the positive direction - to gain ~1 dB more level. Typical asymmetry employed is 25%, or roughly 25% peak distortion. There's your 'crunch'.
When I retired the PPM argument was still active. Is it better to PPM encode pre-processor and let compression and limiting raise the embedded signal's level, or PPM encode as late as possible to minimise code alteration? The requirement to feed multiple destinations with different processing requirements from a single encoder - cable systems, transmitter, FM HD - usually mandates pre-processing injection. Typically the only processing accommodations I made was to select multi-band processing crossover frequencies to keep the PPM signal in a single band and minimise alteration.
The last 20 Years of US FCC regulations for classically modulated AM radio:
The occupied bandwidth of the station is to be 10kHz, with steep cutoff to -80dBc beyond that.
The means there are up to +5kHz sidebands, and down to -5kHz sidebands.
That restricts the AM bandwidth to 5kHz, Not 10kHz.
Math and Physics.
Digital Modulation of AM transmitters:
The highest order of digital modulation (QPSK, 16 QAM, etc.) and the maximum symbol rate (dictated by the 10kHz channel bandwidth) trades off the audio bandwidth, versus:
The signal to noise ratio
The bit error rate
The maximum distance range,
etc.
Math and Physics all over again.
The above is just my opinion.
There is No Free Lunch in a narrow US 10kHz wide AM channel.
When is the last time you saw a Digital AM receiver in WallMart, etc.?
Which US AM station is advertising their wonderful digital modulation?
I must be living in the wrong US location.
The occupied bandwidth of the station is to be 10kHz, with steep cutoff to -80dBc beyond that.
The means there are up to +5kHz sidebands, and down to -5kHz sidebands.
That restricts the AM bandwidth to 5kHz, Not 10kHz.
Math and Physics.
Digital Modulation of AM transmitters:
The highest order of digital modulation (QPSK, 16 QAM, etc.) and the maximum symbol rate (dictated by the 10kHz channel bandwidth) trades off the audio bandwidth, versus:
The signal to noise ratio
The bit error rate
The maximum distance range,
etc.
Math and Physics all over again.
The above is just my opinion.
There is No Free Lunch in a narrow US 10kHz wide AM channel.
When is the last time you saw a Digital AM receiver in WallMart, etc.?
Which US AM station is advertising their wonderful digital modulation?
I must be living in the wrong US location.
Last edited:
I don't miss modern TV or radio. I wouldn't of guessed it'd be less stressful day to day personally over all but that came as a realization.
Modern broadcasting and the drowning of us all in over commercialization, the death counts, the drama. All of it. It all surpassed out of control therefore I refuse to sign back up.
Modern broadcasting and the drowning of us all in over commercialization, the death counts, the drama. All of it. It all surpassed out of control therefore I refuse to sign back up.
The old Clear Channel stations were permitted to have wide channel bandwidth (wide modulation sidebands that made for wide audio bandwidth).
As I mentioned, some receivers had deep 10kHz notch filters to notch out the adjacent channel carriers (absolutely necessary), and a selectable Wide / Narrow IF bandwidth to be able to receive the wide bandwidth modulation.
I believe a lot of regulations applied, for both the clear channel station, and the adjacent channel stations. Day/night power; day/night directivity; etc.
I believe most of that went away in later years.
Perhaps what helped to kill it was the attempt to have Stereo AM in a 10kHz wide channel.
Yes you can spread digital modulation and make it look like noise to a standard receiver, but that just moves listeners away from listening to AM.
Do not ask my about my local FM Jazz and Classical Channels. They sound terrible, so I have to stream them.
Do you remember the El Cassette; Beta Max; Shibata phono cartridge and multiplexed vinyl; etc.
Some things do not get widely accepted to keep in business.
As I mentioned, some receivers had deep 10kHz notch filters to notch out the adjacent channel carriers (absolutely necessary), and a selectable Wide / Narrow IF bandwidth to be able to receive the wide bandwidth modulation.
I believe a lot of regulations applied, for both the clear channel station, and the adjacent channel stations. Day/night power; day/night directivity; etc.
I believe most of that went away in later years.
Perhaps what helped to kill it was the attempt to have Stereo AM in a 10kHz wide channel.
Yes you can spread digital modulation and make it look like noise to a standard receiver, but that just moves listeners away from listening to AM.
Do not ask my about my local FM Jazz and Classical Channels. They sound terrible, so I have to stream them.
Do you remember the El Cassette; Beta Max; Shibata phono cartridge and multiplexed vinyl; etc.
Some things do not get widely accepted to keep in business.
Then every AM site I managed over decades has been out of compliance with full approval of the regulatory authorities.
In the mid 90's a station I worked with had AM stereo. We could only find one person who had an AM stereo radio. That guy had it in a Chrysler LeBaron.
The management wanted to save money by not having to pay the phone company for two equalized audio lines to the transmitter so the decision was made to go mono.
Turning off the stereo exciter allowed us to increase the modulation somewhat as well. I suppose that's how it went at other stations as well.
The last AM HD station I knew of turned the HD off last December. The HD exciters were problematic and finally failing and no longer supported or being made.
The management wanted to save money by not having to pay the phone company for two equalized audio lines to the transmitter so the decision was made to go mono.
Turning off the stereo exciter allowed us to increase the modulation somewhat as well. I suppose that's how it went at other stations as well.
The last AM HD station I knew of turned the HD off last December. The HD exciters were problematic and finally failing and no longer supported or being made.
Actually our tv news broadcasts are terrible imo (except weather info ) and my local AM radio provides the best critical information
in day to day listening with far more accuracy on any particular subject with in depth reporting with no serious time constraints most of the time
This never happens with tv news with their 3 to 5 minute segments all stuffed in their 30 min allotment and all the channels broadcast the same basic thing… what a mess
Never cared about the quality of sound on AM just the news please
rdf,
I had a typo.
I said 5 kHz AM bandwidth.
I meant 5kHz Audio bandwidth.
Perhaps I am missing some fact that makes things better than they are.
Amplitude modulation requires an upper sideband and a lower sideband that have the same width as the Audio bandwidth (5kHz lower sideband and 5kHz upper sideband, which occupies a single 10kHz wide RF channel bandwidth (but the Audio bandwidth is only 5kHz).
In the same 10kHz wide RF Channel bandwidth, on Shortwave, there are special transmitters that transmit AME (AM "Equivilant").
A carrier and only one Sideband can occupy a 10kHz RF channel width, and alsohave 10kHz Audio bandwith.
That AME "miracle" can not happen in the US AM 540 to 1700kHz channels, they are all spaced on 10kHz RF channels, with the carrier in the middle of the RF channel.
I had a typo.
I said 5 kHz AM bandwidth.
I meant 5kHz Audio bandwidth.
Perhaps I am missing some fact that makes things better than they are.
Amplitude modulation requires an upper sideband and a lower sideband that have the same width as the Audio bandwidth (5kHz lower sideband and 5kHz upper sideband, which occupies a single 10kHz wide RF channel bandwidth (but the Audio bandwidth is only 5kHz).
In the same 10kHz wide RF Channel bandwidth, on Shortwave, there are special transmitters that transmit AME (AM "Equivilant").
A carrier and only one Sideband can occupy a 10kHz RF channel width, and alsohave 10kHz Audio bandwith.
That AME "miracle" can not happen in the US AM 540 to 1700kHz channels, they are all spaced on 10kHz RF channels, with the carrier in the middle of the RF channel.
Your analysis was rigorous and correct but the permitted audio bandwidth on AM is still 10 kHz. The Wikipedia entry is surprisingly useful: https://en.wikipedia.org/wiki/AM_broadcasting
"To allow room for more stations on the mediumwave broadcast band in the United States, in June 1989 the FCC adopted a National Radio Systems Committee (NRSC) standard that limited maximum transmitted audio bandwidth to 10.2 kHz, limiting occupied bandwidth to 20.4 kHz. The former audio limitation was 15 kHz resulting in bandwidth of 30 kHz. Another common limitation on AM fidelity is the result of receiver design, although some efforts have been made to improve this, notably through the AMAX standards adopted in the United States."
Given the 10 kHz spacing of primary station carriers, a 5 kHz modulation limit seems obvious but regulatory agencies instead control interference through RF spectrum management. Transmitter locations, power levels, carrier frequencies and day/night patterns are carefully juggled by highly paid engineering consultants to keep things civil. At least in Canada competing broadcasters can opt to negotiate deals limiting audio bandwidth in scenarios where these solutions are impractical.
When skywave cooperates both 810 New York and 1210 Philadelphia activate my Sangean HD portable.
rfd,
I am talking about Classical Amplitude Modulation of a US AM station.
Not Stereo AM, Not Digital AM.
Please refer to the attached file, It shows the extremely steep drop off of the sidebands that are larger than 5kHz offset from the carrier.
The NRSC-2-A calls out certain RBW. Resolution BandWidth are spelled out as 300Hz and 1kHz (The plots I am attaching to this thread)
A Gaussian filter has the relationship of rise time versus bandwidth as:
0.35 / bandwidth = rise time
With an RBW of 300Hz, the rise time is 1.17milli sec; With an RBW of 1kHz, the rise time is 0.35milli sec.
Those slow rise times very severely reduce the amplitudes of the sidebands of musical programs on an AM station.
So, that makes these two things happen:
1. Good News for the Station Engineer: the station easily passes the NRSC-2-A measurements
2. Bad News!
If you actually reduced the rise time of the AM modulator to 1.17msec or even to 0.35msec, the musical program would be severely compromised.
If some doubt that, then just imagine:
A Hi Fi or Stereo amplifier that has 0.35msec rise times, it would get thrown out of every Audio Magazine Test Report.
That points out the very large compromise of the NRSC-2-A measurements. The spectrum amplitude of those sidebands is much larger than the measurements reflect. As a result, a person that listens to an adjacent station hears the sidebands of the other station.
As I said, that is Not Hi Fi.
I am talking about Classical Amplitude Modulation of a US AM station.
Not Stereo AM, Not Digital AM.
Please refer to the attached file, It shows the extremely steep drop off of the sidebands that are larger than 5kHz offset from the carrier.
The NRSC-2-A calls out certain RBW. Resolution BandWidth are spelled out as 300Hz and 1kHz (The plots I am attaching to this thread)
A Gaussian filter has the relationship of rise time versus bandwidth as:
0.35 / bandwidth = rise time
With an RBW of 300Hz, the rise time is 1.17milli sec; With an RBW of 1kHz, the rise time is 0.35milli sec.
Those slow rise times very severely reduce the amplitudes of the sidebands of musical programs on an AM station.
So, that makes these two things happen:
1. Good News for the Station Engineer: the station easily passes the NRSC-2-A measurements
2. Bad News!
If you actually reduced the rise time of the AM modulator to 1.17msec or even to 0.35msec, the musical program would be severely compromised.
If some doubt that, then just imagine:
A Hi Fi or Stereo amplifier that has 0.35msec rise times, it would get thrown out of every Audio Magazine Test Report.
That points out the very large compromise of the NRSC-2-A measurements. The spectrum amplitude of those sidebands is much larger than the measurements reflect. As a result, a person that listens to an adjacent station hears the sidebands of the other station.
As I said, that is Not Hi Fi.
Last edited: