Many distants on main lines were replaced by colour lights even before WW2, which was partly to allow them to be further out for higher speeds but also to make them more visible to drivers.
The LMS had a policy of converting 'Isolated Distants' from mechanical to colour-light signals for some years prior to WW2. I understand the main reason was to avoid trains being delayed in foggy/smoggy weather while 'Fogmen' were called out to attend such mechanical distant signals and had to be paid overtime, of course! The secondary benefit was the elimination of all the operating wire and dozens of support pulleys. A third benefit was the easing of the load on the signalman as he now only 'operated' an electric light switch - but I am sure that benefit was not foremost in the LMS's mind!
Were there problems in practice with the temperature compensation? Obviously the expansion to be made to balance out is greater on a long wire, but so are the possibilities for other factors to upset the balance. The main one (I think) was that some parts were heated by direct sunshine, other parts were not. So did those long wires often hit the stops either way?
I don't see that as a problem with a wire, as it's the expansion over the whole length that matters. And presumably there was enough play in the adjustments to account for that, and the signaller could get it set correctly by reference to the aspect repeater if the signal wasn't visible from the box.
There were various mechanical ways of compensating over a long distance - this includes a 'lost motion' lever to increase the length of pull on the last part of the wire to allow for motion lost in the first part of the wire.
The 'Compensator' is a pair of opposed cranks linked together at approximately the mid-point of the rodding run. Books such as "Railway Signal Engineering" by LP Lewis/JH Fraser and "Mechanical Railway Signalling" by H. Raynar Wilson go into some detail about calculating the location of such compensation.
"Isolated Distants" were regarded by the LMS S&T Engineer, A F Bound, as an important marker and he introduced the practice of painting the (tubular) post in alternate 2 foot wide black and white bands. This applied to both semaphore and colour light signals.
I was under the impression that these levers were more about ensuring the signal returned to danger?
The compensator basically changes push motion into pull, or v.v. The aim being to have the same amount of rodding in push as in pull. However, you don't necessarily need a compensator, you can achieve the same thing where cranks are used to change the direction of the rodding, eg to pass under the track, you can also change push to pull.
