Automated direction finders are only reliable in relative close proximity to NDBs, due to various adverse effects. In particular, the night effect limits their useful range for navigation independently of the NDB transmission power.
However, there was at least one historical navigation system, Omega, that has offered global coverage through high-powered VLF beacons.
In contrast to NDBs and ADFs, Omega was based on hyperbolic navigation, and in particular compared the phase difference between signals received from different beacon stations. These stations were transmitting in the VLF band, with their transmissions reaching over ten thousand kilometers due to the unique propagation characteristics of that band.
What I dont understand: If there is a capability to transmit a radio signal that far, why can't the same principle be used to determine the direction to the transmitter using an ADF, using radio direction finding principles, and then determining one's position by intersecting the determined bearings to the various transmitting stations as is possible using NDBs?
Is there something unique to the VLF band that makes it less prone to the distortions affecting NDBs, or is it subject to them as well, but the phase comparison mechanism is somehow less prone to errors and measurement ambiguities than radio direction finding would be over the distances involved?
Has Omega just solved for these problems by modeling VLF propagations sufficiently?
Or is it something else entirely?
Note: This question was originally part of Was radio navigation used for oceanic crossings before GPS? If not, why not?, but I've spun it off of that for clarity.