And by the virtue of shortwave propagation, it could be heard across the world. For the past month and a half (from when the news of its impending shutdown was revealed) I was regularly picking it up in Australia right up until the bitter end.
In the MF (AM broadcast) band, you can observe this at night - in Australia I can pick up the 50kW Melbourne ABC station (public broadcaster) at 774kHz with a good radio, just about across the entire country.
In the LF (longwave) band, the earth’s surface and the ionosphere start to behave more like a waveguide than skywave. This is actually more reliable/consistent than even HF, but you need massive transmitting antennas due to the large wavelengths involved.
HF also generally wins for distance covered per watt - despite the massive power of Radio 4 longwave, I’d have no chance of hearing it reach Australia.
Bounce off ionosphere
Radio stations are usually measured by the last of those: Effective radiated power.
You can have a radio station with a 50,000 watt ERP, but running only a 2,500 watt transmitter.
For FM radio stations, it's all about the height of the transmitter above average terrain. For AM, it's about the ground conductivity and frequency.
I once worked at a 1,000-watt AM station that had a signal much larger and clearer signal than the 5,000-watt AM station a few miles away.
I'm not a radio engineer, but I'm sure there are plenty on HN who can correct and clarify what I've written.
My suspicion is that this means an exciter and a stack of amps per service, which then go through a two stage combiner and out to the antenna. There might even be a pair of exciters and amps per service depending on redundancy.
The combiners (certainly for FM/DAB/TV services) also cause cumulative attenuation as the signal gets combined each time, so even if all 3 are radiating at the same power, the first in the chain might need twice as much amplification to make up for losses.
edit: MB21 (of course) has some fantastic technical info about Droitwitch: https://tx.mb21.co.uk/gallery/gallerypage.php?txid=1454&page... and there's some great pics here, too: https://www.radiorewind.co.uk/radio1/droitwich.htm
I believe they're still using a pair of Marconi B6042 transmitters (250kW each, in parallel) to provide at least one of the services.
I was very much getting myself confused with some of their other transmission sites where they take multiple DAB or DTV services, modulate, amplify and combine them and then broadcast through the same antenna.
Going by [1], emitted power.
[1] https://www.bbceng.info/Operations/transmitter_ops/Reminisce...
Droitwitch LW's antenna uses a T-aerial suspended between two 210m steel masts acting as massive capacitive top-loaded vertical monopole. The signal isn't beamed or shaped, it propagates omnidirectionally and this style of antenna offers _0 dB_ of ERP increase.
Even worse, they're transmitting AM, so the power output dynamically increases with the volume of the analogue audio being transmitted. If you cut off the input to Droitwitch, it'd still be putting out a 500kW carrier wave. When audio is applied the amplitude of the carrier is modulated, so for peak loudness (someone shouting or the loudest spike in music) it can take an extra 50% power to create the upper and lower sidebands - at peak, the Vapotron tubes could be putting out a combined 750kW.
The amplification stage is only ~70% efficient as well, so at peak power it's possible that the site is pulling nearly 1MW from the grid.
--
Compared to a modern UHF DTV transmitter station the differences are wild. The big transmitter near me is putting out 6* DTV MUX's at 174kW ERP each, but that's through a 15dBd UHF array at the top of the mast which gives an obscene amount of gain.
- Mains draw at the wall ~150kW (including cooling and ancillary systems).
- Total TPO (RF energy leaving the cabs) from each of the six transmitters is only ~52kW combined (8.7kW each)
- Output of the combiners after losses of ~0.5dB is ~46kW. We can expect another ~1.5dB of attenuation after forcing it up 300m of waveguide to the top of the tower so we're now sat at a "mere" ~33kW of RF energy going into the bottom of our antenna.
- 33kW with a +15dBd gain gets us to an ERP from the antenna of 1.044 MW.
--------------
Note: Numbers compiled from public sources. All mistakes and misunderstandings are mine. Whilst I do work in a tangentially related industry this is completely out of my area of expertise - in the same way that working as a cleaner at an aeroplane does not mean one knows how to fly or maintain a plane.
Some people save energy by turning their appliances off at the plug rather than leaving them on standby. Clearly Radio 4 chose to avoid having Brian Blessed on too frequently instead ;)