As somebody very aware about EMC and its affect on normal radio I am a bit puzzled by your explanation. How can a few percentage change in the 50Hz frequency affect the accuracy of things like track circuits (presumably working at dc)? Or does the signalling system itself use similar low frequency data which can clash from the close proximity of magnetic radiation from the traction motors? And at 50Hz all the coupling will be by magnetic fields rather than normal RF frequency coupling.
Apologies if I didn't make it clear, but the issue isn't the signalling equipment varying it's frequencies, but all the other equipment that produce interference frequencies producing an even wider spectrum of interference frequencies, if the grid were to be allowed to deviate its frequencies even further.
Only very simple track circuits work on DC. Any used in a DC traction area can't use DC. And any track circuit in an AC traction area can't use 50Hz AC. And a track circuit in a dual traction area, or where there might be interference from an adjacent source, can't use either DC or 50hz ac. Early electrification systems used frequencies specially generated, the Southern used 75hz for example, and many early ac electrification schemes used 83.3 hz. However, generating and distributing these special frequencies was expensive, and so other solutions were sought.
The RT type track circuit used a single frequency for each track circuit, with neighbouring track circuits using different frequencies. The frequencies are produced by tuned reeds, which are very stable and inherently immune to fluctuations in the supply frequency. However, the system is obviously reliant on the frequencies it uses not being generated elsewhere, as such interference could result in false operation of the track circuit. Later types of track circuit, such as the TI21, use frequency modulation techniques. If such track circuits are subject to interference at the critical frequencies, they will still fail, but at least they will fail in a safe mode. However, that could still be highly disruptive, particularly if significant numbers were to subject to interference.
Track circuits and the traction supply are directly connected, as they both use the running rails. However, the issue is not just track circuits. Point detection circuits in dual traction areas also have to be immune to both traction supplies, so also used similar immunisation methods, including equipment based on the RR/RT system.
The RR reed system had dozens of frequency channels. As I recall, there was a hierarchy of frequencies, with safety critical functions having a restricted frequency set. The RT track circuits, for example, used only the 8 (as I recall) most immune frequencies. However, non safety critical functions in frequency division multiplex FDM systems could use a much wider range of frequencies.
Both GEC and Westinghouse (the two main UK signalling suppliers) produced FDM systems, but I think only GEC's RR/RT systems could be used for safety critical functions. The Westinghouse system only saw limited use in the UK, but the GEC equipment was widely used, especially for track circuits. For a long time, it was one of only two permitted types of jointless track circuit.
Whilst track circuits are the biggest issue, signalling schemes in the 1980s and later used FDM equipment quite extensively. FDM was useful as a multi drop system, used for dropping off individual controls such as automatic signals replacements along long sections of plain line, and picking up individual track circuit indications. Doing this with direct wire circuits could get very expensive, particularly in ac traction areas where signalling circuits are limited to a maximum of 2km length, to avoid the possibility of large voltages being induced on them.
Many signalling schemes also used FDM as a backup means of sending override controls in case the main TDM remote control failed. And as it used voice frequencies, FDM was also useful for sending circuits via direct BT and BRT lines, especially as BT started to move away from through copper circuits.
So there is potentially a LOT of legacy signalling equipment around that could be affected by any move to allow a wider variation in the grid frequency, and the consequent wider spread in harmonic interference frequencies that could lead to.
