GWR Ground signal usage

[GWR_Express]

I confess to being no expert but I’ve just completed a project installing a 4 phase block signalling system using bells, indicators and working semaphore signals and I’m in the process of installing ground signals on my GWR O gauge railway. But, there seems to be a lack of understanding both on my part and of others as to how they should function. I am thinking they are instrumental in controlling back shunting and as such they would do the following:
Be to the left of a diverging road at a point to indicate a clear or blocked road. As such they could not be linked to the point they live next to as they would be indicating either the points are set to ahead and not diverging as well as whether it is clear or not to proceed into the diverging road in my book I don't see that they could do both.
I am thinking this now as my original plan was to link the signal to the point blades which I have been told is wrong! I read somewhere that they would be separately controlled from the signal box according to the situations mentioned above.
If this is the case and you can confirm that my thinking is correct where in relation to the point would the signal be sited?
I have seen many photos showing these signals apparently connected in some way to an adjacent point either by a rod and/or wire as well.
My weblog where all the information about my railway is hosted is at: http://www.bpodmore.co.uk
I would be grateful for your opinion of the above.

[Danny252]

I am thinking this now as my original plan was to link the signal to the point blades which I have been told is wrong! I read somewhere that they would be separately controlled from the signal box according to the situations mentioned above.

Correct - a signal operating with the points would generally be considered as a points indicator, and would provide no information regarding the state of the line ahead, or any information about whether a train was permitted to proceed. Whilst such things did exist, I don't believe they were ever common on the GWR.

I have seen many photos showing these signals apparently connected in some way to an adjacent point either by a rod and/or wire as well.

This would likely be for detection. A set of apparatus is attached to the points (and, if applicable, facing point lock) to prevent the signalman clearing the signal unless the points are physically proven to be in the correct position (and locked, if applicable).

I am thinking they are instrumental in controlling back shunting and as such they would do the following:
Be to the left of a diverging road at a point to indicate a clear or blocked road. As such they could not be linked to the point they live next to as they would be indicating either the points are set to ahead and not diverging as well as whether it is clear or not to proceed into the diverging road in my book I don't see that they could do both.

They are not necessarily linked to shunting backwards, or do they generally indicate whether the road is clear or blocked. A ground signal being cleared indicates that the route has been set (and usually confirmed as such by detection), perhaps which route is set (if the signal(s) have that capability), and that the train has permission to proceed as far as the line is clear. However, it does not state that it is clear to the next signal -- there could be a train in the way, which it is the driver's responsibility not to hit!

Furthermore, not all shunting moves were controlled by discs. Especially at earlier dates, railway companies kept signalling to a minimum; this would often mean that discs were only provided to control trains leaving sidings (a legal requirement from a certain date), but trains into the siding were controlled by hand signals.

A good number of signalling diagrams of GWR layouts are available on the main site, which show everything from very minimal layouts (e.g. Highley) to comprehensively signalled locations (e.g. Wednesbury).

[davidwoodcock]

Points indicators were once very common on the GWR as they were the first (and only at the period) type of ground signal used by the company, and were installed at the points governing the exit from sidings on to running lines in accordance with the then requirements (post-mid-1870s?) of the BoT. Obviously, any movement that had to continue past the point when it wasn't set for movements on to the running road had to pass the point indicator "on" but this was once normal even with worked ground signals, being the main reason for the introduction of yellow ground signals by three of the four post-grouping companies in the mid/late-1920s. (The GWR remained the odd-man-out and yellow ground signals didn't appear on the Western Region until the mid-1950s.) Most GWR points indicators were gradually replaced by worked ground signals (especially in the early part of the 20th century), but some (e.g. at Highworth) remained until the bitter end (1962 in Highworth's case).

[StevieG]

Perhaps some recollections of my experience of about 18 ex-GWR boxes on BR Western Region in the 1960s might be helpful. These included ground shunt disc arrangements of both probable pure GWR origin, and of the BR era incorporating some later ideas.

It appeared to me that latter GW practice was for each disc located at a point end to normally only be operable for movements diverging through those points.
'Yellow' discs have been mentioned : These were one solution to the logically problematic situation of movements legitimately needing to pass over disc-protected points while they were in their Normal position (usually the 'straight' or non-divergent) while the red exit disc was still in the 'On' (Danger, or stop) position. This was most usually in sidings where the points were an exit to the 'main' line but the siding also continued beyond the points.
As said, 'yellow' discs (yellow bar on the face and yellow 'On' night indication, instead of red) AFAIK, on the Western Region, were a later BR-era development.

Where different box-worked point ends were located near each other (e.g. along a running line, a crossover to the other line and points into a siding being closely adjacent), this could result in each point end having its own disc.
As 'yellow' discs were not normally found on running lines, this could mean that a movement to diverge at the second point had to pass the first disc in the 'On' position (i.e. at Danger). On BR(WR) in some cases such a 'first' disc was not fitted with a red glass, so it's night-time 'On' indication was white instead, but in daylight there was nothing to differentiate a 'white' red disc from one which ought never to be passed while 'On', and traincrew's route knowledge was the only way they knew which red discs could be passed while 'On' and which should not.

But where the need to shunt through running line points -
- to two or more routes, or
- it was important for staff to know to which route they were being signalled, or
- was over points in either position (Reverse or Normal),
was frequent, then by the BR period (& before?) many places had multiple-disc signals as a single unit, for as many routes as were needed; the discs being mounted vertically above one another, often up to three in number, possibly more.

Latterly, almost certainly during the BR period, at some layouts (e.g. Weston-Super-Mare, where signalling was modified in the 1950s) where necessary single discs were made able to be cleared for two or more routes, with the nearest points in either position.

So, if you want to take this kind of detail into account for your disc installations GWR_Express, you might want to decide approximately what time period that you're modelling.

Incidentally, regarding -

" .... confirm that my thinking is correct where in relation to the point would the signal be sited? .... "

, if you're asking, apart from 'left' or 'right' (it was not uncommon for siding exit discs to be on the right, by the way), how near their points that discs should be sited, those with the older type of detector of a rod coming directly from the point's switch toe ('blade' end) across into the base of the disc signal, then of course the disc was right beside the point toes. but the later type of detector mechanism with slides at right angles, did not need the disc to be particularly close, and they were often then something like 10 feet on the approach side of the points.

[Mike Hodgson]

You seem to see backing movements as important. As you are modelling GWR, you should be aware that they had a special signal for this, although they often used ground signals like other companies, illustration on this page of the main site:

https://signalbox.org/signals/semaphore9.htm

If you want to see an actual example where this might be used, there's one at Bewdley on the Severn Valley Railway. An advantage of this type is that it could be seen more easily from the footplate than a ground signal when propelling stock where platforms, bridges etc may obstruct visibility. However the driver would still need assistance from the Guard or Shunter as the signal does not mean the line is necessarily clear.

[GWR_Express]

Thanks to all for your input here. Together with this I have several pictures showing ground signals used to both show entrance and exit routes but not to any standard rule. With this in mind I will endeavour to use the signals appropriately, certainly not as I previously thought!!
Many thanks again for the useful info