Distants on platforms

Yes, most green/yellow signals along platforms were for guards use. Termed 'platform repeaters' they simply repeat the starting signal.
Mounted at headwalls are 'headwall repeaters' which are the same thing.

Some station starters also embody a repeater for the advance starter.

So the starter can display

Red (no second aspect as you don't need to know whats on the advance starter).
Green Yellow: Starter green, advance starter red.
Green Green: Starter green, advance starter green.

Eg. Mansion House e/b

Some even more descriptive, running, signals exist on the w/b approach to East Ham where the post contains the signal now, repeater for the next two/three signals and a fog repeater for the next signal.

Red
Green Yellow Yellow (next signal red)
Green Yellow Green (next signal green, signal after that may be red (if next post display is G Y Y), third signal will be red)
Green Green Green (next two/three signals green).

Reading these signals correctly enables a faster approach to East Ham (the signals are on a bend so you can't see the whole sequence before you'd need to otherwise apply brakes)

NR has a different braking distance, so the Drivers need to know more signals in advance = single yellow next one at red, double yellow next one at yellow, green next one at double yellow. There were thoughts of flashing green for the fifth aspect.

Now, unlike LUL NR has a standard overlap - 183m, LUL tailor the overlaps to the signal and almost every train has the same braking characteristics, train speeds are much lower than NR and braking is guaranteed by the trainstop/trip cock. Two aspect signals cover the speed/braking/headway combo more than adequately on the UndergrounD. NR with its wide variety of stock/braking characteristics and no trips need warning much, much further out.

Now for electric colour lights - what would you like to know:

1. electrically lit, mechanically operated?
2. oil lit, electro-pneumatically operated?
3. electrically lit, electro-pneumatically operated?

or:

4. Moving lenses or multi-lens units?

Depends exactly what you mean by electric colour light!! ;D

EDIT; just thought of another combination or two:

Gas lit, EP operated?
Electrically lit, hydraulically operated?
Oil lit, hydraulically operated?
Electrically lit, electrically operated?

The simple answer to your question is we don't!

At least not in the context you're most likely thinking of. Where there is a double green, you are looking at two separate signals (the bottom one is a repeater for the next signal) ;D ;D

I've knocked up a quick diagram which I hope explains it:

Well, this is why I enquired exactly what form of electric colourlight signals - I could have been mean and asked for a further subdivision: internally or externally pivoted lens units, which can them be further subdivided into rotating or linear movement!

It is somewhat disingenuous to refer to tunnel signals as armless semaphores only the early colour lights that were not mounted in tunnel can be definitively stated to have used semaphore parts as these were manufactured in-house by the larger pre-grouping companies. A number of the pre-1923 companies had some 'armless semaphores' used for restricted clearances and not in tunnel. Pure mechanical colour lights were used in a few circumstances by the LNER and they were indeed armless semaphores in the truest sense, some were battery operated too (and motor worked off the accumulators)!

Tunnel signals were designed and patented as a complete departure from the semaphore ethos; there was a considerable variety in designs between the signalling contractors all attempting to avoid patent litigation.

I think too, the Hall signals of pre-1908 vintage also merit a mention; although the principal daytime aspect was produced by a red cashmere cloth, the upper light was illuminated continuously.

I'm also surprised that you quote 1919 for the LOR signals; it was due to accidents in 1919 that Col. Pringle recommended the overhaul of the Timmis' semaphores. The tender for the resignalling of the Overhead wasn't issued until 12/3/20 after inspection of a trial section. The full daylight colour signalling system was brought into use on 27/7/21 and inspected by Maj. Hall on 3/8/21.

Interestingly, there is also a precedent for the PRR colour lights - using your yardstick of 'electric lights on a stick' then the laurels must go to the US Electric C^oy of Pittsburgh, who had magnetically operated colour lights in use from 1902, in Britain the equivalent (magnetically operated) was a green-aspect-only system fitted to the Swindon Tramways in 1908. Although I'm pretty sure the Brecknell, Munro and Rodgers system of two red bulbs and a green was fitted to the Burton and Ashby Light Railway in 1906 and the Weymss Light Railway in the same year.

For tunnel signals, the story isn't quite so clear-cut: the good old Interboro' has a hybrid semaphore/colour light installation dating from 1904, and I have a US&S EP interlocking technical manual from 1914 that shews a pure colour light signal in use at Pennsylvania Station on the Hudson and Manhattan (Hudson Tubes, NY) and that installation was commissioned in 1910.

British equivalent colour-light tunnel signals would be the all-electric moving lens installation on the Waterloo and City dating from 1898 and the grand-daddy of multi-lens units would be the Great Northern and City with the automatic signalling controlled by track circuit and treadle dating from 1904.

Ah. I've wandered. I'll leave fresnel doublets for another time.

I admire the Japanese practice of getting the Driver to point and speak out aloud every signal aspect and speed restriction - there is a subtitled video somewhere on Youtube of the expert ? Nozomi driver doing exactly that and being worried about running 5 seconds early. Can't find the link though. Pah!

You know I have often wondered what the logic was behind the EP Colour light in LU tunnels. Someone went to a lot of trouble to use an EP valve and air motor to move the red and green filter up and down in front of an electrically illuminated light. I was amazed when I saw a photo of one in a book of photos of the Underground in Foyles bookshop many years ago.
The mechanism was not dissimilar to the CLR, DR and LER 'long tom' trainstops that were still in use when I began my service with LT. The whole concept appeared to defy logic as if the technology existed to operate the air motor and prove it then the same technology could operate red and green lights directly. I am left wondering if electric lights were either distrusted or unreliable whereas there was no doubt as to the position of the air worked aspect. Presumably the red aspect would've been above the green, the opposite to normal, as the piston pushed the coloured aspects upwards.
For those who are wondering or may not know why, the red signal aspect in a modern LU colour light is at the bottom of the signal head in order for it not to be easily obscured e.g. by a build up of snow.

Depends which design of tunnel signal - if the bob weight was above the light source or not. IIRC the problem with electric lights was a mix of the short life of light bulbs (the Livermore light notwithstanding) and the difficulty of proving return indication locking via reliable ECRs - EP operation was a straight mechanical contact, but it took a while for lamp proving relays to migrate across the pond.

Funny you should say that; as the fresnel 'non-doublets' installed by the Pennsy between Paoli and Overbrook on 14/2/15 were always referred to as 'amber'. However, every single aspect was amber, the interpretation was based on the alignment of the three signal lights (which broadly replicated the position of a semaphore arm) and are the basis for the horizontal/45°/vertical bars used in signalling plans to this very day!

Track circuits on LU would have been with the Ealing and South Harrow. The pain was perfecting indication locking, which required a lot more ECRs than at present.

Indication locking, rather than track locking was seen in the early days as the the way to go. AIUI all the early installations for the Yerkes group tubes were track circuited, or were certainly tracked by 1906/7 in the case of the Central London, though I'd need to do a bit of digging to prove the CLR.

Remember, vane relays were not on the scene until slightly later, early track circuits in this country were DC. Hmm; traction spikes!!
Theoretically, if you are just using indication locking then there is no need for track circuits.

Yes I had always thought that the Yerkes lines were track circuited when built. I do know that not only DC track circuits were used but also electric trainstops, when I worked in the signal school in the 1980s I came across some old blue 'prints' dating from the 1950s IIRC which showed them still in service. I have never seen an electric trainstop or a photograph of one whereas a few of those early dc type relays are still in service. The last 'D' type I saw was one that I moved from Queen's Park signal cabin relay room to the IMR in south shed and that must be almost 20 years ago but of course 'C' types are still in use in many T/T relay rooms in the battery charging circuits.

I have a note somewhere about electric trainstops in service until the early 50s; I think it could be the Ealing Broadway area. I also think the extensions in the 20s at the top end of the Northern were all-electric (rather than EP), so you have probably seen a picture of an electric train stop.

Small world!

If we are talking about the same relay, I think it is the only one in that IMR, 'D' type blue and white and next to it a 'C' type also blue and white. ISTR that's 12v or for a dc relay, I have the colour codes somewhere but not readily to hand!

The traction sections from Kilburn Park to Queen's Park only have a T/T room at Kilburn Pk SS as there is no Queens Park SS. The two relays hang on the Queen's Pk end of the tunnel wires and provide indications to the T/T panel on the Bakerloo Line Controller's desk. These indications, unbelievably, are routed on common or garden telephone pairs in the main telephone cable jumpered at all station MDFs back to Baker St.

The original T/T indications appeared as white lights on the signalling diagram in QPK cabin.

I don't know if the T/T prints for Kilburn Park were updated, certainly the circuits are shown on the Bakerloo Controller's T/T panel prints which show the schematic wiring of all the T/T sections on that panel. At the time I submitted a sketch of what we did locally which was a simple drawing showing the relays and outgoing fuses etc but the comms DO was (I'm sure you know this!) notorious for dragging its heels on updates especially for small alterations.

This relocation of the two relays was part of a larger project in which I changed all the existing T/T phones on the Bakerloo from the early Bakelite/Ceramic 1920s types in wooden boxes to the new standard Redatron CB T/T instruments and new FP200 tails connected to red and black singles onto the tunnel wires. AFAIK this was the only line rewired to this standard, almost everything before was 2core LC 1/064 and most afterward was 2L1.5 and 4L1.5 pyro.
I was involved in most of the comms installation works for the then new Bakerloo Control room.

Tubeprune, that's a great shot, I have never seen one at that angle before, but no it is not the type I was thinking of. Interesting to note the small air motor which appears to be more the size of the Davis motors that were fitted to H,J & K trainstops until recently.
Also the EP valve with separately attached armature, there were/are still some of those around, Baron's Court and South Harrow spring immediately to mind. In fact the one on the viaduct at South Harrow is precariously placed for those who have never worked on the type, it is not unknown for the armatures to fly!

The type I was thinking of had the motor mounted vertically directly beneath the aspects and may well have been in a sub surface tunnel rather than a tube tunnel when I think about it. The shot I saw was taken directly in front of it and so perhaps the whole apparatus was not in view. Clearly I can see a bob weight in your image but if one existed in the image I saw then it was certainly not obvious.