Radio 4 Long Wave Switch Off

[radiolistener]

The lower frequencies mean that you need longer or fancier aerials which is not good for portable equipment. In an age of internet radio and on demand services long wave is several steps away from the imagination of modern users in thew UK. Yes there will always be a bunch of old farts and enthusiast that complain but things move on.

I think this is mixing up the requirements for transmitters and receivers.

On the receiver side, long-wave reception does not necessarily require a large external antenna. Ferrite rod antennas have been used successfully in portable LW/MW radios for decades and can provide surprisingly good reception despite their compact size. On the transmitter side, the situation is more complex than simply comparing antenna sizes.

Yes, an FM transmitting antenna is physically much smaller than a LW antenna. However, FM broadcasting still relies on tall towers or elevated sites to achieve useful coverage. The antenna itself may be smaller, but the tower and supporting infrastructure is often of a comparable scale.

More importantly, a single high-power LW transmitter can provide coverage over an entire country or even several countries, whereas FM typically requires a large network of hundreds transmitter sites distributed across the coverage area.

And each of those FM sites requires its own tower, transmitter equipment, power supply, building, maintenance, and operational support. From an infrastructure perspective, one centralized LW facility can therefore be surprisingly efficient because the costs are concentrated in one location rather than replicated across a large network.

The same principle can be seen in power generation. A small portable generator is much smaller and cheaper than a large power station like nuclear power plant, but supplying the same amount of power with a millions of small generators would be far less efficient than using one centralized facility.

The same principle applies to large-area radio coverage: judging a broadcasting system solely by the size of an individual antenna can be misleading if you ignore that the alternative requires many separate transmitter sites to provide the same overall service.


Another point that is often overlooked is that the large antennas and high transmitter powers commonly associated with LW broadcasting are primarily a consequence of the very large coverage areas these stations are designed to serve. If the goal were only to provide local coverage comparable to that of a typical FM station, there would be no need for a massive antenna system or hundreds of kilowatts of transmitter power. A much smaller transmitting antenna and a relatively modest transmitter could be used.

In that case, the LW installation could be significantly smaller than the large national broadcasting facilities that people usually associate with LW transmission. It could also operate at a fraction of the power of a typical high-power FM broadcast station while still covering a substantially larger area.

The enormous antenna structures seen at traditional LW stations are largely a result of their exceptional coverage requirements rather than an inherent property of the frequency band itself.

[radiolistener]

The bandwidth for AM broadcast radio is 4.5kHz, which comes from the 9kHz channel spacing agreed by Europe, Africa, the Middle East, Asia, and Australasia (not the Americas) in the Geneva Plan 1975.

Or were you thinking of something else?

I mentioned the Geneva Plan several posts ago. There still needs to be gaps between the channels, to avoid clashes, so it can't be 4.5kHz. I mentioned 3kHz, because of my subjective experience of listening to AM broadcasts, but I accept it could be a little more than that.

That is incorrect information.

The bandwidth of broadcast AM is defined by regulatory and engineering standards, not by subjective listening experience.

For example, NRSC-2-C specifies the spectral mask for broadcast AM transmissions, where the signal is allowed to occupy ±10.2 kHz RF bandwidth around the carrier (with defined amplitude limits relative to the carrier level).

Also, FCC 47 CFR § 73.44 specifies ±10.2 kHz RF limit for AM.

This ±10.2 kHz RF occupancy is explicitly based on a 10 kHz bandwidth-limited audio signal as stated in the standard:

4.1 Purpose
The purpose of this RF mask specification is to control the spectral energy of an analog AM transmission resulting from modulation and intermodulation products that fall outside the necessary bandwidth of an analog AM broadcast signal modulated with 10 kHz (or less) bandwidth-limited audio.

Narrower occupied bandwidth in AM broadcast is not a fundamental limitation, but rather a result of transmission and processing choices. In some cases it can be influenced by older or simpler transmitter chains and old audio processing equipment that are not optimised to fully utilize the spectral mask up to ±10.2 kHz while maintaining efficient modulation and required emission constraints.

However I have never seen a broadcast AM station in practice operating with an occupied bandwidth narrower than ±4.5 kHz. For example, the discussed BBC Radio 4 LW at 198 kHz, as shown in the spectrum screenshots provided earlier, operate with an occupied bandwidth of approximately ±5.5 kHz.

At the same time, modern AM broadcast stations are operating close to the defined spectral mask up to ±10.2 kHz limits as specified by relevant standards.

[radiolistener]

Actually FM results in a much wider bandwidth per channel: 150-200kHz.

FM occupied bandwidth is typically estimated using Carson’s rule applied to the composite baseband signal:

B = 2*(Δf + f_max)

FM peak frequency deviation Δf is 50 kHz or 75 kHz depending on region, see ITU-R BS.450-3.

For CCIR FM broadcasting, the composite baseband f_max depends on transmission mode:
1) Mono: M component audio (L+R), f_max=15 kHz
2) Stereo: S component audio (L-R) as DSB-SC around 38 kHz ±15 kHz, f_max=53 kHz
3) Stereo+RDS: RDS component as BPSK on 57 kHz ±2 kHz, f_max=59 kHz

Thus for different modes we have different RF bandwidth::
- mono FM station bandwidth = 2*(75+15) = ±90 kHz = 180 kHz
- stereo FM station bandwidth = 2*(75+53) = ±128 kHz = 256 kHz
- stereo FM with RDS bandwidth = 2*(75+59) = ±134 kHz = 268 kHz

[paul cotter]

The theoretical analysis of wideband FM gives rise to a Bessel function- don't ask me for equations as this is beyond my pay grade!

[Simon]

We were comparing it to broadcasting. Every TV is becoming a smart TV, so basically a monitor with a SBC running android and I presume that the transmissions are now streamed or do they have a way of doing it "broadcast style", I doubt it as that requires the network to cooperate and I'm sure non of that exists.

[radiolistener]

Some of this is because of the deliberately superior sound quality

The sound of FM receivers is not inherently "superior".

AM is a linear amplitude modulation scheme, so noise and interference in the RF channel directly appear as amplitude variations at the output.

FM encodes information in frequency deviations, so channel noise affects the instantaneous frequency estimation in the demodulator, resulting in frequency/phase variations that can result as distortion-like audio artifacts rather than simple additive noise. That's the key difference.

As a result, AM provides a more faithful representation of the received RF conditions — interference and fading are directly audible as noise and level fluctuations. FM in contrast tends to convert part of the channel noise into demodulation artifacts, which can subjectively sound less noisy, but represent a different mechanism of added audio artifacts. They are less noticeable than noise, but still distort the audio.

In some ways, FM reception is somewhat analogous to MP3 compression. Both can appear cleaner because the listener is exposed to less obvious noise, yet neither is free from impairments. The mechanisms are very different. MP3 introduces artifacts through lossy compression, while FM can transform channel noise and interference into demodulation artifacts. However, the end result is similar in one respect: instead of hearing only straightforward noise, the listener hears a different form of distortion.

The human brain often perceives these artifacts as less objectionable than continuous noise and may largely ignore them. As a result, the sound can seem cleaner, even though it is still being altered by distortions that were not present in the original audio.

Another practical consequence is that once noise has been transformed into distortion-like artifacts, recovering the original audio becomes more difficult than simply reducing additive noise, because the information has already been altered rather than merely obscured.

[radiolistener]

It is however an anachronism, using that much brute force power.

Reading that, I cannot help thinking of a similar argument: that using a 1 GW nuclear power plant is an anachronism because it uses so much brute-force power, while a 3 kW portable gasoline generator is smaller, cheaper, and easier to deploy.

Of course, that comparison ignores the service being provided. The relevant metric is not the size or power of a single installation, but how efficiently it delivers coverage or energy to the intended area and population. The same principle applies to broadcasting systems in general.

The very large antennas and high transmitter powers traditionally associated with broadcasting are primarily a consequence of coverage requirements. High transmitter power is used to extend service range, while large antenna systems are used to maximize radiation efficiency and reduce system losses to make it highly efficient.

However, the “brute force” characterization becomes less convincing when looking at real FM infrastructure. Many high-power FM broadcast transmitters operate at powers on the order of 100–250 kW. For example, BBC FM transmitters such as Rowridge (Isle of Wight, 88.5 MHz) and Blaenplwyf (Aberystwyth, 88.7 MHz) operate at 250 kW power levels, yet still provide coverage typically on the order of tens of kilometers to roughly 50–100 km in practice, depending on terrain and conditions.

In that context, FM is not a low-power alternative to long-range broadcasting, but rather a highly distributed network requiring substantial aggregate transmitter power to achieve national coverage.

So the comparison is not between "high power LW vs low power FM", but rather between a single high-power long-wave site covering a very large area and a large number of high-power FM sites collectively consuming multi-megawatt levels of power while still relying on line-of-sight constrained coverage.


In practical terms, we are comparing a single LW 500 kW transmitter providing near-national or even multi-country coverage at radius ~3000 km, with a single FM 250 kW transmitter providing a service radius typically on the order of ~50–100 km.

When you compare energy consumption against actual covered area, the efficiency argument becomes very one-sided in favour of LW broadcasting at the system level.

From that perspective, the phrase "using that much brute force power" in relation to LW broadcast station is somewhat ironic.

[tom66]

Of course, the superior mechanism is 320kbps MP3/AAC or FLAC over an internet connection, which is why Spotify sounds so much better than any broadcast radio.  DAB, FM, AM etc are all inferior in terms of audio reproduction to even 160kbps MP3.  The difference then between 320kbps and FLAC is slight for me, some people may notice it better, but it's so few extra bits that you might as well turn it on unless you're data limited.

[radiolistener]

For reference, below is a list of FM broadcast transmitters in the UK operating at power levels 100 kW and above.

MHz kW Station Name Location
88.1 250 BBC R 2 Sandale, Cumbria
88.1 160 BBC R 2 North Hessary Tor, Devon
88.3 110 BBC R 2 Sutton Coldfield, W.Mids
88.5 250 BBC R 2 Rowridge, Isle of Wight
88.5 130 BBC R 2 Pontop Pike, Co.Durham
88.7 250 BBC R 2 Blaenplwyf, Aberystwyth
88.7 150 BBC R 2 Meldrum, Aberdeenshire
89.1 250 BBC R 2 Wrotham, Kent
89.3 250 BBC R 2 Holme Moss, W.Yorks.
89.7 250 BBC R 2 Tacolneston, Norfolk
89.9 250 BBC R 2 Black Hill, Lanarkshire
89.9 250 BBC R 2 Wenvoe, Cardif
90.1 125 BBC R 2 Divis, Belfast
90.3 250 BBC R 3 Sandale, Cumbria
90.3 160 BBC R 3 North Hessary Tor, Devon
90.5 110 BBC R 3 Sutton Coldfield, W.Mids.
90.7 250 BBC R 3 Rowridge, Isle of Wight
90.7 130 BBC R 3 Pontop Pike, Co.Durham
90.9 250 BBC R 3 Blaenplwyf, Aberystwyth
90.9 150 BBC R 3 Meldrum, Aberdeenshire
91.3 250 BBC R 3 Wrotham, Kent
91.5 250 BBC R 3 Holme Moss, W.Yorks.
91.9 250 BBC R 3 Tacolneson, Norfolk
92.1 250 BBC R 3 Black Hill, Lanarkshire
92.1 250 BBC R 3 Wenvoe, Cardiff
92.3 125 BBC R 3 Divis, Belfast
92.5 250 BBC R 4 Sandale, Cumbria
92.5 160 BBC R 4 North Hessary Tor, Devon
92.7 110 BBC R 4 Sutton Coldfield, W.Midlands
92.9 250 BBC R 4 Rowridge, Isle of Wight
92.9 130 BBC R 4 Pontop Pike, Co.Durham
93.1 250 BBC R Cymru Blaenplwyf, Aberystwyth
93.1 150 BBC R Scotland-a Meldrum, Aberdeenshire
93.5 250 BBC R 4 Wrotham, Kent
93.7 250 BBC R 4 Holme Moss, W.Yorks.
94.1 250 BBC R 4 Tacolneston, Norfolk
94.3 250 BBC R Scotland Black Hill, Lanarkshire
94.3 250 BBC R 4 Wenvoe, Cardiff
94.5 125 BBC R Ulster Divis, Belfast
94.7 250 BBC R Scotland-sw Sandale, Cumbria
95.3 150 BBC R 4 Meldrum, Aberdeenshire
95.3 120 BBC R Wales Aberystwyth (Blaenplwyf)
95.8 200 BBC R 4 Black Hill, Lanarkshire
96.0 125 BBC R 4 Belfast (Divis)
96.8 250 BBC R Cymru Cardiff (Wenvoe)
97.7 250 BBC R 1 Sandale, Cumbria
97.7 160 BBC R 1 North Hessary Tor, Devon
97.9 110 BBC R 1 Sutton Coldfield, W.Midlands
98.1 130 BBC R 1 Pontop Pike, Co.Durham
98.2 250 BBC R 1 Rowridge, Isle of Wight
98.3 250 BBC R 1 Blaenplwyf, Aberystwyth
98.3 150 BBC R 1 Meldrum, Aberdeenshire
98.8 125 BBC R 1 Wrotham, Kent
98.9 250 BBC R 1 Holme Moss, W.Yorkshire
99.3 250 BBC R 1 Tacolneston, Norfolk
99.5 250 BBC R 1 Black Hill, Lanarkshire
99.5 250 BBC R 1 Wenvoe, Cardiff
99.7 125 BBC R 1 Divis, Belfast
99.9 250 Classic FM Cumbria (Sandale)
100.0 160 Classic FM SW England (N Hessary Tor, Devon)
100.1 220 Classic FM West Midlands (Sutton Coldfield)
100.3 250 Classic FM S England (Rowridge, Isle of Wight)
100.3 125 Classic FM NE England (Pontop Pike, Co.Durham)
100.5 150 Classic FM Aberdeen (Meldrum)
100.9 250 Classic FM London/SE England (Wrotham, Kent)
101.1 250 Classic FM N England (Holme Moss, W.Yorks)
101.5 250 Classic FM Norfolk (Tacolneston)
101.7 250 Classic FM Cardiff (Wenvoe)
101.7 250 Classic FM Glasgow/Edinburgh (Black Hill)
101.9 250 Classic FM Belfast (Divis)
104.0 250 BBC R 4 Blaenplwyf, Aberystwyth
104.2 150 BBC R Nan G/Scotland Meldrum, Aberdeenshire

The list is intentionally limited to FM transmitters 100 kW and above, because including lower-power sites (1–100 kW) would make it excessively long.

Taken together, these FM high-power transmitters represent an aggregate installed power on the order of many megawatts. Yet this is rarely characterized as "anachronistic" or described as "using brute force power", even though each individual of these FM transmitter typically serves only a very limited geographic area due to line-of-sight constraints.

In contrast, a single long-wave transmitter operating at a fraction of that total power can provide continuous-area coverage spanning an entire country and beyond, including cross-border propagation.

From that perspective, the criticism of LW broadcasting as "using that much brute force power" appears somewhat inconsistent when compared to the distributed power requirements of FM broadcast networks providing equivalent national coverage.

[paul cotter]

Quick question for you, radiolistener: are those figures which you kindly quoted, real power up the spout or ERP? A lot of tv transmitters will have 500k levels but the transmitters are often in the 50k region with aerial(antenna) gain making up the difference.

[radiolistener]

Quick question for you, radiolistener: are those figures which you kindly quoted, real power up the spout or ERP? A lot of tv transmitters will have 500k levels but the transmitters are often in the 50k region with aerial(antenna) gain making up the difference.

The source simply states “transmitter power in kilowatts”, so I assume these values refer to transmitter output power. However, this is not fully clear, since some sources list the same stations and power using ERP instead.


But even if one assumes that the listed values are ERP rather than true transmitter output power, and that the actual transmitter power is typically in the range of ~50–100 kW, the aggregate picture does not change significantly. In that case, only 5–10 FM transmitters correspond to a total installed transmitter power comparable to a single ~500 kW LW transmitter.

However, even this small subset of FM infrastructure still only provides fragmented, line-of-sight limited coverage over relatively small regions, typically serving individual cities or regional clusters. In contrast, a single LW transmitter provides continuous-area coverage over an entire country and beyond.

This highlights the key point: the comparison is not about individual transmitter power, but about system-level coverage efficiency. In that sense, a long-wave broadcast transmitter is the clear total winner as an efficient solution in terms of coverage per unit of installed transmitter power.

[paul cotter]

The figures seem rather high to me, not that I am disputing them. I know from experience that 3kw into a Phelps Dodge 8 bay circular polarisation array(can't remember the gain figure, long, long time ago!) from a elevated location ~1000ft gave excellent reception out to about 35miles with acceptable reception way beyond that.
Another way of looking at it is kW/listener and under this criterion I reckon the figure is rising for FM too as virtually everyone I look at is glued to their smart phone and broadcast radio is facing a dismal future.

[SteveThackery]

So the comparison is not between "high power LW vs low power FM", but rather between a single high-power long-wave site covering a very large area and a large number of high-power FM sites collectively consuming multi-megawatt levels of power while still relying on line-of-sight constrained coverage.

I still think you are missing a point. That LW transmitter provides one speech-grade channel. The FM network provides over 400 hi-fi music-grade channels.

You aren't comparing like with like.

[radiolistener]

The figures seem rather high to me, not that I am disputing them. I know from experience that 3kw into a Phelps Dodge 8 bay circular polarisation array(can't remember the gain figure, long, long time ago!) from a elevated location ~1000ft gave excellent reception out to about 35miles with acceptable reception way beyond that.

Which station are you referring to? If you know the site name or frequency, I can look up how it is listed in the transmitter databases.

Transmitter powers vary enormously. There are plenty of FM broadcast sites operating at 1–10 kW, and many operating even below 1 kW.

However, coverage figures need to be interpreted carefully. A few kilowatts from an elevated site can indeed provide excellent coverage over tens of miles under favourable conditions, especially across flat terrain. But once you introduce hills, dense urban development, indoor reception requirements, or difficult terrain, the coverage area can shrink dramatically, which is why large FM networks still require a considerable number of transmitter sites and many kW power to achieve reliable regional or national coverage.

As far as I know, many FM broadcast sites carry multiple stations from a single tower, often on the order of ten or more services. Typical FM transmitter output powers are around 30–33 kW per transmitter. This is based on information I observed on an engineering monitoring system displaying real-time transmitter status, including output power at the antenna feed and VSWR for individual transmitters.

Consequently, the aggregate transmitter power at a single major FM broadcast tower typically on the order of several hundred kilowatts, approaching the same 500 kW power in some cases. And importantly, these are actual transmitter output powers delivered to the antenna systems, not ERP figures inflated by antenna gain.

[paul cotter]

Radiolistner, That station I mentioned is long gone, it was actually a land based pirate station for which I did the engineering.

[radiolistener]

For example, here is a single FM broadcast tower equipped with two transmitting antenna systems. The engineering monitoring display shows that one antenna is being fed with 127 kW of transmitter power, while the other is being fed with 120 kW. In other words, the total transmitter power delivered to the antenna systems is approximately 247 kW.

[u666sa]

One thing I noticed is that this often seems to happen before war at specific areas. About a year or two before the war here, many long-wave and medium-wave transmitters had already been shut down. And it wasn't only here.

Right before the war, same week, perhaps some days before, you'd hear a good spinnage and panic over TV and radio, they'll tell you the other country is planning to attack. Then war happens. If you want to go back sometime, perhaps as much as 5+ years, 8+ years, a lot of demonization and propaganda happens.

I'm sure turning off broadcast at 198 kHz has nothing to do with a war. Frankly, I'm not convinced Britain has the capacity to attack anyone. It's a marginal country with a lot of immigration and social problems these days.

I'm sure reason for stopping broadcast on long waves at 192 kHz has more to do with the fact that nobody listens to that anymore. Nobody listens to BBC to begin with, and much less people listen to BBC on long wave AM radio. 

People nowadays have these:



Stream apps, or online radio apps. You have all your radio on your phone, and when you get into your car it plays right on your car stereo system. At home, you can stream a station onto your music center or TV.

FM is almost never used.

AM, still has some uses to get traffic and emergency weather, but that's about it.

Here, though, we are talking about long waves. That's much longer waves than regular AM. One would have to go find a radio receiver from 1990's to listen to that.

[radiolistener]

I still think you are missing a point. That LW transmitter provides one speech-grade channel. The FM network provides over 400 hi-fi music-grade channels.

I think "speech-grade" is somewhat misleading in this context. Traditionally, speech-grade audio refers to telephone-quality transmission with an audio bandwidth 3.4 kHz or 2.7 kHz for SSB radio.

Broadcast AM is a different matter. AM broadcasting standards were developed for audio bandwidths 10 kHz, which is far beyond traditional speech-grade communications and is suitable not only for speech but also for general audio and music programming.

While FM does offer up to 15 kHz audio bandwidth, describing a 10 kHz audio AM broadcast service as merely "speech-grade" significantly understates the audio quality that such systems are capable of providing.


I'm sure, that if you listened to music on an AM broadcast station with a 10 kHz audio bandwidth and a good SNR, it would be difficult to distinguish it from FM in many cases, since a significant number of FM broadcasts are actually of poorer overall quality.

The main practical limitation of AM would be the lack of stereo sound.

[tom66]

I'm sure, that if you listened to music on an AM broadcast station with a 10 kHz audio bandwidth and a good SNR, it would be difficult to distinguish it from FM in many cases, since a significant number of FM broadcasts are actually of poorer overall quality.

But that's the fundamental issue with AM: it's so much more difficult to get a good SNR because RF noise directly impacts the audio signal.  That means anything from your neighbour's pet motorbike project with worn points to powerline Ethernet and EMI emissions from badly designed SMPSes all impact it. This is especially problematic for long-wave as this type of interference is more commonly conducted down powerlines than radiated directly and so tends to impact reception indoors particularly significantly.  With a large external antenna, many of these problems go away, but that's often impractical.

FM is far less susceptible to this type of interference since it sits in the VHF band and the typical behaviour is for FM signal reception to 'fall off a cliff' once the signal is too weak to demodulate successfully.  You do have some degradation in SNR near this, but for about 90% of the transmission range the signal fidelity is virtually unchanged. It is true that demodulation errors might be occurring below this point but they're not perceptible to the average listener.

DAB exhibits this effect even more significantly, with the audio becoming nearly completely inaudible once the bit error rate exceeds zero.  I remember it sounding a bit like bubbling soup.

[u666sa]

AM is anywhere from 176 meters to 561 meters in length, your window is only about 0.8 meters tall. Waves are too big to come into your dwelling. FM at 3.5 to 2.5 meters is much more suited for city use. For AM you need a proper external antenna, which will of course pickup all kinds of noise from within the city.

Plus AM is not really efficient. A typical 1500 Watt rated station only send about 250 Watts of modulated signal on upper and 250 watts on lower, while carrier, unmodulated signal is 1000 Watts. It's simply a waste. 

[radiolistener]

But that's the fundamental issue with AM: it's so much more difficult to get a good SNR because RF noise directly impacts the audio signal. 

FM is far less susceptible to this type of interference

Yes, urban RF environments are indeed very noisy today, and this affects all broadcast systems, not only LW.

However, it is not quite accurate to say that SMPS noise and similar interference mainly impacts LW while leaving FM largely unaffected. These noise sources are present across a very wide frequency range and can affect reception in both bands. The main difference is that FM systems typically maintain higher link margin due to high transmitter powers and relatively short receiver distances, so the SNR remains sufficient over most of the service area.

For LW (and AM in general), a similar effect applies: when the received field strength is sufficiently high (for example, close to a high-power transmitter), many forms of interference become much less significant relative to the wanted signal. In other words, the issue is not strictly modulation-dependent, but SNR-dependent.

It is also worth noting that FM is not immune to interference, it can degrade quite rapidly once the signal falls below the threshold for proper demodulation. The difference is more about the nature of degradation rather than its absence.

As a practical example, during power outages I have also observed that FM reception from neighbouring cities can be received cleanly, but when I connected my phone to a power bank it immediately raised the noise floor and caused the FM signal to disappear into noise. Before the blackout, I did not even notice how much EMI my power bank was producing in the FM band, because the environmental noise floor was high enough to mask it completely.

This behaviour is not fundamentally different for AM/LW, it is just that the balance between wanted signal and interference shifts depending on field strength and local noise sources.

To some extent, you may be right: the LW band is closer to frequency ranges where a large amount of SMPS-related switching noise energy is present, which can make it a bit more susceptible to interference from such sources. However, in practice I have observed that SMPS-related interference seems to affect the MW band and the lower part of the SW band more noticeably than LW. It is an interesting observation, but in my local environment the noise level on LW is surprisingly lower than on MW, at least in my city.

Noise is certainly present on the LW band, but in practice it is not as strong as on higher frequency bands. I find this somewhat counterintuitive, since most SMPS systems operate in the 40–60 kHz range.

[radiolistener]

AM is anywhere from 176 meters to 561 meters in length, your window is only about 0.8 meters tall. Waves are too big to come into your dwelling. FM at 3.5 to 2.5 meters is much more suited for city use. For AM you need a proper external antenna, which will of course pickup all kinds of noise from within the city.

I think there are a couple of misconceptions here.

The ability of a signal to enter a building is not limited by the physical size of a window relative to the wavelength. In fact, longer wavelengths generally penetrate buildings and other obstacles more effectively than VHF signals. This is one of the reasons why LW and MW broadcasting have traditionally provided good indoor coverage. FM reception is often much more sensitive to building construction, reinforcement, low-emissivity windows, and similar attenuation mechanisms.

Plus AM is not really efficient. A typical 1500 Watt rated station only send about 250 Watts of modulated signal on upper and 250 watts on lower, while carrier, unmodulated signal is 1000 Watts. It's simply a waste. 

As for efficiency, conventional AM does indeed devote a significant fraction of its power to the carrier. However, FM also transmits a full-power carrier continuously. The carrier is not switched off when there is no modulation. From an energy perspective, FM is not inherently a low-power alternative to AM.

More importantly, transmitter efficiency should be evaluated at the system level. A high-power LW transmitter may consume hundreds of kilowatts, but it can provide coverage over an entire country. Achieving comparable coverage with FM requires a large number of transmitters operating simultaneously, so the aggregate power consumption can be much higher.

In fact, a broadcast FM signal occupies a much wider RF bandwidth than a conventional AM broadcast signal (roughly 20–25 times wider), which means the receiver also collects proportionally more channel noise.

A wider RF bandwidth generally requires a higher received signal level to achieve the same channel SNR, which is one of the reasons why many FM broadcast transmitters operate at power levels comparable to LW transmitters, despite covering only a tiny fraction of the area served by a typical LW station.

The reason is that transmitter power is distributed across the occupied bandwidth. A narrower signal concentrates more power into each Hertz of spectrum, while a wider signal spreads the same transmitter power over a much larger bandwidth. As a result, the power available per Hertz is lower, and achieving the same channel SNR generally requires either a stronger received signal or, in practice, higher transmitter power levels.

[radiolistener]

I don't know what the numbers are for the environmental cost re. power useage of the transmitter, versus distribution of the station programmes via FM installations, (which days are also numbered), and via streaming services, but with the latter the cost is no longer entirely that of the broadcaster.

Roughly speaking, a single high-power FM broadcast site typically operates at transmitter power levels of around 100–250 kW and is of the same order as a high-power LW transmitter, typically operating at around 250–500 kW.

However, an LW site can provide a single mono service covering an entire country and even beyond its borders under favourable propagation conditions. In contrast, an FM site often provides up to 10–15 stereo programme services, but only within a limited line-of-sight area, typically a city or part of it.

Since FM broadcasting requires hundreds of such sites in order to achieve national coverage comparable to a single high-power LW transmitter, the overall FM infrastructure footprint and total energy consumption are correspondingly much higher, making the system less efficient in terms of energy per unit of coverage area and delivered service.


As result, both broadcasting approaches have their own advantages and trade-offs.

LW broadcasting is generally more energy-efficient at the system level and can provide very wide-area coverage, often spanning an entire country and beyond under favourable propagation conditions. It also offers greater resilience, since reception can remain available over very large areas even when local infrastructure is disrupted by major power outages, communication failures, natural disasters, or other emergency situations. However, it is typically limited to mono transmission and usually delivers a single programme service per transmitter.

FM broadcasting, on the other hand, provides stereo audio and a much wider selection of programme services. However, this comes at the cost of a significantly higher number of transmitter sites required to achieve national coverage, resulting in higher aggregate energy consumption, a larger infrastructure footprint, and reduced coverage efficiency per unit of deployed transmitter power. In addition, the service depends on a local infrastructure elements, making it more vulnerable to widespread disruptions affecting power or communications networks.

Therefore, the most balanced solution is a hybrid approach: LW broadcasting for wide-area, energy-efficient and resilient coverage, and FM broadcasting for higher programme diversity and stereo audio quality, albeit with more localized coverage and lower system-level energy efficiency.

[steve30]

I don't think you can win.

If $dictator decides to bomb Boston Park (nr Rotherham), we lose Greatest Hits Radio (boo hoo ). There are so many localised variants of GHR which can still be picked up from nearby transmitters that you will probably cope.

If $dictator bombs Droitwich, then the whole country loses Radio 4 LW (provided they do it in the next fortnight), but here in Rotherham, we get a perfect BBC FM signal from Holme Moss so probably no biggy.

Last time we were at war, we only had one TV channel and only two or three radio channels, many of which were switched off anyway to avoid Hitler's lot from using them for direction finding. We won said war. So, whatever happens now, I'm sure we'll cope.

[Ranayna]

We were comparing it to broadcasting. Every TV is becoming a smart TV, so basically a monitor with a SBC running android and I presume that the transmissions are now streamed or do they have a way of doing it "broadcast style", I doubt it as that requires the network to cooperate and I'm sure non of that exists.

Beyond live streams from sport events, there is not much possible to reduce with streaming.
Livestreams, or linear TV over the internet, which is essentially a livestream, is, at least by some providers, already multicast. The tech exists.

But individual streams, like Netflix or any other video on demand stream, can't conceivably be multicast. Large providers will have dedicated cache servers though, so that the data for popular shows won't have to travel many times around the world.