Welcome Guest ( Log In | Register )

2 Pages V   1 2 >  
Reply to this topicStart new topic
> Siting and use of GWR bracket signals
geoffwba
post 21 Feb 2020, 10:37
Post #1


Fireman
Group Icon

Group: Plus+
Posts: 201
Joined: 18-November 09
From: Birmingham UK
Member No.: 6,764



On my layout I have two routes converging at a junction. The left route is from a loop and sidings (which could be used for a passenger train) whilst the right route is the main line. To protect the junction with home signals, I am wondering whether it was normal GWR practice to have two separate signals or have a bracket signal. As far as I know bracket signals were really intended to control access to two diverging routes.
Go to the top of the page
 
+
John Webb
post 21 Feb 2020, 15:40
Post #2


Railways Czar
*********

Group: Members
Posts: 4,298
Joined: 20-August 05
From: St Albans, Herts.
Member No.: 144



Some clarification is required for a sensible answer. Are we talking about (1) two single lines joining or (2) a single line into a double line or (3) a double line into a double line?

If (1) and the lines ran parallel for a short distance, then a bracket signal might be used sited between the two lines.
If they converge at an angle, and in cases (2) and (3), it is most likely that separate home signals, set back a little from the junction, would be used.

Bracket signals were not confined to indicating the diverging route at junctions. They were popular too at terminal stations, where platforms had a line each side, as 'starters' for each line, as the bracket signal took up less space.

In addition, the GWR had a policy of marking signals used generally only by freight trains by putting a ring on the end of the arm centred on the white stripe.

Regards,
John
Go to the top of the page
 
+
Julian2011
post 21 Feb 2020, 16:32
Post #3


C55
*******


Group: Members
Posts: 2,308
Joined: 29-January 11
From: South Wiltshire - OO and an intended toy in N
Member No.: 9,766



Also, if it's 1, and the line is run in both directions, you will need a pair on a bracket before the toe end, for traffic working the opposite way.

Julian


--------------------
Go to the top of the page
 
+
Bear 1923
post 21 Feb 2020, 19:31
Post #4


Regional Controller
*******


Group: Members
Posts: 2,612
Joined: 31-March 11
Member No.: 10,215



Sorry John... huh.gif I'm going to strongly disagree with you on item 1... Unless you can come up with an example somewhere in the UK - that is - outside the very early period when some truly weird things were done (by later standards) as regards signal sighting... However, even then, I'd be extremely surprised to see a bracket signal in the V between two converging lines.
Post 1889 (if not earlier) the influence of the Board of Trade meant that signals were placed to the left of the line to which they applied or above the centre line... Anything to the right of the centre line was "wrong side" and had to be individually justified. If there was even the slightest risk of a signal being misread it would not be allowed wrong side - where necessary the signal would be right side with a bad sighting which was compensated for by a banner repeater - and, in an extreme case, by a severe speed restriction on the approach - also - even with both of these additions - it would be likely for such a signal as a junction home signal to be an inner home protected by an outer home. This protecting by a signal in rear could be shuffled along the station limits so that a home protected a starter and a starter protected an advanced starter - although the last is the least likely.

Semaphore signals protecting any convergence - even on fast lines - could be located right up to about a yard in rear of the fouling point. Unlike todays "defensive" thinking where an overlap will be included between the signal and the fouling point wherever possible the older thinking was that train crews were instructed to stop dead in rear of a stop signal that was against them ("on"). That was it - no discussion. Train crews were expected/required to have their trains under control. If an over-run occurred all the crew could lose their jobs immediately. They could, of course, also lose their lives - and some of the ways train crew lost their lives were excruciating - cooked by being buried by the coal from their tender igniting with the coal in the firebox... Kind of persuasive to not getting the job wrong. Something to keep in mind is that train speeds, especially unfitted freights, were much slower. However, this said - the signal would normally be to the left of the line it applied to.

There is a slight case for a bracket signal though... this would be where available space caused the signal for the right hand line to be bracketed out over the left hand line so that the doll, arm and light got into the location it would be in if a post could be fitted in on the left of the right hand line... The main post of this bracket would be on the left of the left hand line. However, this would more than likely be a single doll bracket - and the left hand line's signal would be on a separate post (which would stand on the approach side of the brackets structure - possibly set some way back.

Brackets for the same direction on island and other double sided platforms - this was not so much to save space (fit into space) but so that the guard when on the platform and platform staff could see the signal - which might otherwise be hidden by the train. This particularly applied to signals out of bay platforms.

cool.gif

Go to the top of the page
 
+
John Webb
post 21 Feb 2020, 21:17
Post #5


Railways Czar
*********

Group: Members
Posts: 4,298
Joined: 20-August 05
From: St Albans, Herts.
Member No.: 144



QUOTE (Bear 1923 @ 21 Feb 2020, 19:31) *
Sorry John... huh.gif I'm going to strongly disagree with you on item 1... Unless you can come up with an example somewhere in the UK - that is - outside the very early period when some truly weird things were done (by later standards) as regards signal sighting... However, even then, I'd be extremely surprised to see a bracket signal in the V between two converging lines...

I'm sure somewhere I've seen a photo of two single lines that ran parallel for half a mile or so before joining near a station - presumably to avoid having to put in an extra signal box at the point where they could have joined! But for the life of me I can't recall either where it was nor in which book or website I saw it. I'll admit my memory may be at fault on this one, however.

Regards,
John
Go to the top of the page
 
+
Bear 1923
post 21 Feb 2020, 21:19
Post #6


Regional Controller
*******


Group: Members
Posts: 2,612
Joined: 31-March 11
Member No.: 10,215



The OP has got me thinking... ohmy.gif

The siting of a signal (and adjacent signals) - that is the location of it - is a direct result of the sighting requirements - which are determined by curvature, gradient, line speed, environmental conditions (eg structures, embankments/cuttings. bridges and a whole load more - including risk of smoke from factories) and - eventually - track layout.
"
"Normally" signals (ie running signals) on adjacent lines are located on or close to a line perpendicular across the tracks. Generally this assists sighting and minimises risk of confusion. As an alternative one, usually on the lesser line, may be pushed back to the rear ("on the approach to"). I'm assuming that the OP knows that a signal for a lesser line will be lower in height than the adjacent signal for a senior line...

Meanwhile... one option would be to put the lesser signal on a straight post on the left of its line with the arm at 12 to 15 feet above rail head.and then bracket the senior signal out over the centre line of the senior line with its main post between the two lines. This is less of a structural issue than bracketing over the lesser line. (In fact going over the lesser line might mean a small gantry across all lines). Bracketing the senior signal up and to the right would make a good differentiation between the two signals and create good sighting for higher speed on the main.

It is possible that a bracketed and high main signal would have a repeater arm on the main post. The GWR did a very nice/weird repeater arm. I figured it out from the Swindon S&T parts catalogue that this stubby arm used the "SB" (Special Boss) arm of the SB signal - commonly called a "Centre Balanced" arm by enthusiasts. On the repeater the Special Boss spectacle plate is inverted and the arm, wholly on the left side of the post, is pivoted at its right hand extreme end. (There were lots of these but finding a modern picture is impossible so far - does anyone know of an extant example please?)

The lesser line signal is very unlikely to be bracketed from the left up and over the centre line - unless it is set very well to the rear and away from any misreading by crews on the main line.

John is correct about the GWR using ringed arms - although exactly what their practice was where and when is an arcane subject - that I cannot fathom. I suspect that over the decades they were not consistent ( ohmy.gif GWR not consistent?! ohmy.gif ) That or they may have had different practices in different areas - possibly as well as at different dates.

The GWR also used 3ft arms for some lesser lines (standard arm being 4foot). A ringed arm would be non-passenger. As far as I can tell a 3ft arm would also be non-passenger - except (typical GWR to have an "except") where a 3ft arm was used due to serious lack of space problems - usually a wall getting in the way.

Back at the OP - sidings would be non-passenger - by definition. If the route serves sidings and a possibly passenger loop the sidings would be trapped from the loop and signalled (in and out) separately - usually with GWR independent discs (aka "dummies") Outbound they might have a ringed arm. The loop would need signal protection from the connections with the sidings - In semaphore form this would probably be additional to the signal protecting the convergence with the main - unless the sidings connected very close to the convergence. Passenger lines would normally be all 4ft arms.

Of course the 4ft arms could be SB arms... dribble.gif

Malvern has a nice example of a standard 4ft arm on the main and a 4ft SB - separate posts - for exit from the loop. That example has the posts stepped along the line by a few yard rather than directly across from each other.

cool.gif
Go to the top of the page
 
+
Bear 1923
post 21 Feb 2020, 21:24
Post #7


Regional Controller
*******


Group: Members
Posts: 2,612
Joined: 31-March 11
Member No.: 10,215



QUOTE (John Webb @ 21 Feb 2020, 21:17) *
I'm sure somewhere I've seen a photo of two single lines that ran parallel for half a mile or so before joining near a station - presumably to avoid having to put in an extra signal box at the point where they could have joined! But for the life of me I can't recall either where it was nor in which book or website I saw it. I'll admit my memory may be at fault on this one, however.

Regards,
John


Absolutely true John biggrin.gif

Due to the BoT/MoT restrictions on FPLs distance from the lever many junctions were arranged in a way that resulted in both single and double line pairs running parallel for long distances - several miles sometimes. However - I would be astounded to see the signals for the two adjacent (single or single next to a double) lines being put on a bracket between the two lines - one signal would inevitably be wrong side and too close to the other line. I don't think that an ASLEF delegate on the sighting committee would have contemplated such an arrangement for one millisecond.

Running lines parallel like this not only avoided having to provide two boxes close together - with all the maintenance and staffing costs - but also avoided the complications of having to provide slotted Distants and related issues. (including lots of signal lamps that had to be lamped - an additional staff cost plus the paraffin).

B)

PS Actually, it occurs to me, it might have happened on the broad gauge in Ireland... They did some totally freaky things. huh.gif

B)
Go to the top of the page
 
+
John Webb
post 22 Feb 2020, 09:23
Post #8


Railways Czar
*********

Group: Members
Posts: 4,298
Joined: 20-August 05
From: St Albans, Herts.
Member No.: 144



I have seen a bracket signal - not on the GWR - used to protect the junction for two converging lines, although it was not two parallel lines converging. It was at Stratford Low Level on the line to North Woolwich where the Stratford East curve from the GE main line came in to join the N Woolwich line. A bracket signal stood at the end of the Down Woolwich platform with the main post taking the signal for the Woolwich line, and the left-hand post, bracketed out to the left of the main post, and twisted to get the arm at right-angles to the track as the link curved towards the Woolwich line; I assume this was done to give the best sighting for a train using the link. Each post also carried a "WHISTLE" board below the signal!

I knew little about signals when I regularly trainspotted at Stratford and used the N Woolwich line to get there, and never took my own photo of what I now realise to be an unusual signal. It does appear in the Middleton Press book "Branch lines around North Woolwich" published in 2001 and the more recent book "London's East End Railways - Part 2 Branch lines to the Docks", published by Book Law Publications in 2013.

John
Go to the top of the page
 
+
Bear 1923
post 22 Feb 2020, 19:24
Post #9


Regional Controller
*******


Group: Members
Posts: 2,612
Joined: 31-March 11
Member No.: 10,215



ermm.gif Well, if you will go looking around Stratford... What would you expect?

Actually - one might expect things to be done conventionally. However, the arrangement you describe achieves two thing for the link line (1) it aims a signal and aspect at approaching traffic on the link (2) it enables staff on the platform to see the signal - to some extent at least (depending on how far out of alignment the adjusted arm was). Without having seen a picture I might guess that, as a doll on a bracket, the signals were also raised fairly high - at least above the normal 15 to 18 foot (vague/broad) minimum standard for lower quadrant signals.

(The standard height for the most restrictive aspect of a colour light signal - multiple aspect or searchlight - is 12 ft above rail head "Driver's Eye Level". The lamp shining through the aspects on an Upper Quadrant semaphore would be a bit higher than this while a Lower Quadrant would be about the same or a bit higher again. This simply relates to the effect of the arc of the semaphore arm. One reason for adopting UQ was that, broadly, it gave simpler options for height positioning than LQ - and was significantly better in overhead electric areas (among the stanchions etc).

There may have been other exceptional examples such as this Stratford one. There would always be a specific case for each one.

As far as I have seen rotating a whole doll (or dolls) - or rotating a post - to align a signal for sighting on a tight curve is at least unusual, if not rare. On the other hand both the lamp cases of semaphores and the whole heads of colour lights and the brackets they sit on are built so that the cases and heads can be rotated to shine their beam to an optimum location on the approach side of the signal. Both can also be adjusted from a vertical position (forward or back) to allow the beam to shine up or down a gradient as appropriate. This is extremely difficult to illustrate with a photograph. The vertical adjustment is achieved by adding washers or shims between the base of the lamp/head and the support at the front or back as necessary.

A conundrum for you...

Two double track running lines with no track connection between them running more-or less parallel several hundreds of yards apart. One (we'll call it the south line) has some company building next to it. The (north) other has three tall lower quadrant semaphore signals - but the arms of the signals are parallel to the track not perpendicular to it. unsure.gif Why would this be? What is going on? dribble.gif
Go to the top of the page
 
+
Graham Plowman
post 23 Feb 2020, 08:08
Post #10


Regional Controller
*******

Group: Members
Posts: 2,407
Joined: 5-September 06
From: Sydney, Australia
Member No.: 869



QUOTE (geoffwba @ 21 Feb 2020, 21:37) *
On my layout I have two routes converging at a junction. The left route is from a loop and sidings (which could be used for a passenger train) whilst the right route is the main line. To protect the junction with home signals, I am wondering whether it was normal GWR practice to have two separate signals or have a bracket signal. As far as I know bracket signals were really intended to control access to two diverging routes.


Geoff,

If it helps, I have a comprehensive article about how to signal a layout here: http://www.mrol.com.au/Pages/Vu/SignallingAndInterlocking

Brackets tended to be used for diverging routes. Separate posts would be used for converging routes. An example of this would be on the left end of the platforms in the first picture in the above article.

Having said that, on the eastern end of Minehead's platforms there is an example of a double bracket (both left and right) where two running lines leave opposite sides of an island platform. Note however, that both brackets are extended from the centre post over to the side of the platform to which they apply.


--------------------
Graham Plowman
(British outline 00 - NCE PH PRO-R, Lenz 100 - DCC Sound and computer controlled signalling/interlocking of Ashprington Road with SSI software)
Go to the top of the page
 
+
John Webb
post 23 Feb 2020, 08:29
Post #11


Railways Czar
*********

Group: Members
Posts: 4,298
Joined: 20-August 05
From: St Albans, Herts.
Member No.: 144



QUOTE (Bear 1923 @ 22 Feb 2020, 19:24) *
A conundrum for you...

Two double track running lines with no track connection between them running more-or less parallel several hundreds of yards apart. One (we'll call it the south line) has some company building next to it. The (north) other has three tall lower quadrant semaphore signals - but the arms of the signals are parallel to the track not perpendicular to it. unsure.gif Why would this be? What is going on? dribble.gif

Driver/Fireman sight testing, perhaps?

John
Go to the top of the page
 
+
Bear 1923
post 23 Feb 2020, 19:41
Post #12


Regional Controller
*******


Group: Members
Posts: 2,612
Joined: 31-March 11
Member No.: 10,215



sad.gif mad.gif sad.gif Foiled!!! You got it in one clap.gif thumbsup.gif clap.gif

dribble.gif
Go to the top of the page
 
+
geoffwba
post 24 Feb 2020, 09:59
Post #13


Fireman
Group Icon

Group: Plus+
Posts: 201
Joined: 18-November 09
From: Birmingham UK
Member No.: 6,764



Thanks to all for the comprehensive replies ! To clarify matters, and bearing in mind this is an 00 gauge layout with of course restricted space, there is one single "branch/loop" line converging with one track on a a double track main line.Both tracks are on a curve. Access to the loop from the main line is protected by a ground shunt signal since train movements would on my layout normally only be in one direction. It is a bit difficult to explain without a diagram so I have attached my layout diagram and the converging lines are shown in yellow. I think to be prototypical one semaphore for each line would be correct.
Attached Image
Go to the top of the page
 
+
Bear 1923
post 24 Feb 2020, 20:38
Post #14


Regional Controller
*******


Group: Members
Posts: 2,612
Joined: 31-March 11
Member No.: 10,215



I enjpy pontificating! biggrin.gif So thankyou smile.gif

I'm afraid that at the size I can reproduce your diagram I can't figure out what's where. This doesn't matter because you don't need (or want!) correcting. It's your railway! I like that you look for a reasoning rather than just plonk in whatever signal you have/want.
One trick that anyone can use to figure out what signals (and even what track arrangement) would have been used for a layout is to sketch it out straight and roughly according to the length proportions that you think would apply. From there you work on what you want to/ can curl into the space you have available. I hope this makes some sense.
cool.gif
Go to the top of the page
 
+
Peter Armond
post 25 Feb 2020, 11:47
Post #15


Footplate Inspector
*****

Group: Members
Posts: 725
Joined: 1-February 06
From: Lusaka, Zambia
Member No.: 560



I was wondering how much space this takes up? Do you have access to the centre of the layout?
Go to the top of the page
 
+

2 Pages V   1 2 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS    Lo-Fi Version Time is now: 16th July 2020 - 18:20