Jump to content

Signalling Swan Hill

kitpw

453 views

Having made a few posts about the general arrangement and a few details about Swan Hill, I'm just getting around to signalling so it might be a good moment to post the provisional signal diagram "for comment and suggestions" and see what comes up! 

 

Swan Hill is GWR and dates to 1927 (or thereabouts). The track layout is loosely based on a reduced version of Uxbridge Vine Street with the goods yard accessed on a reversal (from road No 3) similar to the arrangement at Windsor central station.  Traffic is predominantly passenger and the branch is double track throughout (for a double track branch with nothing much at the end of it, see GWR Uxbridge High Street). As at Uxbridge Vine Street, engines can only run round by reversing stock out of the arrival road (No 2) and running round on the crossover (points 1a and 1b). Parcels, horses, milk and other perishables are unloaded in Siding 1 and the loading dock.  Other freight - coal, timber, construction materials and so on - goes to the goods yard via Road No 3.  Road No 3 can be used by a push-pull service in addition to providing goods yard access: all three roads have direct access to the Up Branch line.

 

The layout is wired for DC operation with 4 controller "zones" - the Up sidings (siding 1, loading dock and headshunt) are on controller 1 (C1: coded yellow for wiring). The Up branch is C2 (red), the Down branch C3 (blue) and Road 3 and the goods yard are C4 (green). Switches, relays and a Megapoints servo driver reverse the points, change frog polarity and allocate controllers and track feeds. Thus, reversing point 11 on the diagram allows C2 to take over control from C1 and, similarly, reversing points 10, 23 and 24 allocates C2 to take over from C4. Reversing point 23 causes C3 to take over from C4 (except for the line to the right of point 24) so that access can be obtained from Road 3 to the Down Branch.  The same actions remove power feeds to parts of the track so that conflicting moves are prevented. When all points are set normal, the 4 controllers can operate the different areas independently. 

 

What is needed now is signals...
 

 

 

SwanHill-signalling01.pdf

SwanHill-signalling01.jpg

  • Like 3
  • Informative/Useful 1
  • Craftsmanship/clever 1


13 Comments


Recommended Comments

Jeremy C

Posted (edited)

Isn't signal 3 (down home) on the wrong side of the crossover? Even allowing for the crossover being within the clearing point (and so not actually being protected by that signal), the position would surely be inconvenient since it would be in the path of any locomotive running round.

 

I also wonder about 10, 11 and 12 in quick succession, all of which would need to be cleared for a train backing into 1 road after running round.

 

However, I am no signalling guru (far from it!) and know nothing of GW practice, and would be happy to be corrected.

Edited by Jeremy C
  • Agree 1
  • Thanks 1

Share this comment


Link to comment
3 hours ago, Jeremy C said:

Isn't signal 3 (down home) on the wrong side of the crossover? Even allowing for the crossover being within the clearing point (and so not actually being protected by that signal), the position would surely be inconvenient since it would be in the path of any locomotive running round.

Jeremy - thanks for this observation - amazing what one doesn't notice in planning/drawing these things.  I've moved the home signal 'Up' to the other side of the cross-over together with the clearing point, both now opposite point 1b - and a braking distance from the arrivals signal (or inner home). Thankfully, easy to do on paper (digital paper!). 

 

You'll have gathered that I'm no signalling guru either!  Thanks for taking the time to look and comment.

Kit PW

  • Like 1

Share this comment


Link to comment

Just like 1a and 1b some other switches would be pulled together by the same lever.  The following list of examples being non exhaustive*: 

 

6 and 24,

7 and 12,

10 and 23. 

 

*It's not exhaustive because I haven't spent sufficient time to look at how 11, 17-20 might be linked up.

 

From a signalling pov.

 

Would there be starting signals for the up siding and bay? I think only the siding. 

It seems there is nothing protecting the crossover 1 in the down direction. 

I don't think 22 would be a start signal. Probably a disc. (I'm assuming this is not a running line)

In early times 10-12 could be the same signal, sited as 10, depending how close they are. You could alternatively have two or more? stacked discs at 10 for these routes to aid sighting.

4 and 5 might be the same height if the platforms have equal status. 

  • Agree 1
  • Thanks 1

Share this comment


Link to comment

Rich - many thanks for your very helpful comments.  

 

On the actuating of points, the numbers refer to servos as much as to switches (so that they can be adjusted individually). At the moment, each servo has its own switch/lever which is generally a 2 or 3PDT type switch: they do the frog polarity as well as the Megapoints servo controller and the third way is used for relays allocating controllers to the 4 areas - Up sidings (loading bay and siding 1), Up branch, Down branch and Road 3/Goods Yard. All rather DC I'm afraid (if not positively steam age tech). Once I've got the signalling sorted out on paper, I'll have another look at the switching (see picture below:  thankfully there's a wiring diagram and what isn't obvious is labelled!) - it could do with the simplification you suggest!

 

On the placing/types of signals:. 
Siding and bay - a single signal would be preferable: I assume that this would be in the 'Y' between the two.  
I have set back the down home signal (in the drawing) to protect the crossover 1a/1b: it got moved when I drew it out as a signal diagram.  
Signal 22 is interesting: the signal diagram I have for Windsor (which has the reversal access to the low level Goods Yard) is a low-res pdf but seems to show a start signal with route indicator for the Up, Down and Goods Yard directions - might this be a better option altogether?
10 - 12 are very close together: I agree that a signal at 10 (only) would be a better bet - possibly 3 stacked discs.
4 and 5 - should be same height: corrected!

 

Many thanks for your help on this - much appreciated: I'll post an updated diagram in due course.

Kit PW
 

100723.jpg

  • Like 2

Share this comment


Link to comment

On top of Rich's crossovers, I would pair 11 with 19&20 on one lever and have 17&18 on a second lever.

The distant signal has to be braking distance (100s or 1000s of metres depending on speed/gradient) from the first stop signal so doesn't appear on most layouts.

If you really want one, have it as a worked distant on the same post as 26 controlled by the next signal box (for trains leaving the terminus).

Hope that helps

Will

  • Thanks 2

Share this comment


Link to comment

Will - thanks for your suggestions/comment.  11, 19 and 20 are linked to the extent that when point 11 is reversed, track power is taken off both siding and loading dock and will only be restored by setting signals 27 or 28 off - in line with Rich's comment I will reduce these two signals to one.  The wiring is arranged so that when signals are installed, these interventions can be made. I'm not sure how the Megapoints board would cope with several servos switched from a single switch - I haven't tried it - but I will get in touch with David Fenton and ask the question.    

The Down distant is somewhat "notional" and fixed.  It ought to have a dotted line and a distance on the diagram which, as you say, would be a kilometer or more Up branch.  I haven't looked at "Box B" yet - if I ever get there, there is scope to extend the layout so a working distant on signal 26 would be appropriate, even if it stands at the entrance to a fiddle yard!

Many thanks - all helpful!

Kit PW 

  • Like 1

Share this comment


Link to comment

I can see how my earlier comments are ambiguous. What I meant is that I would keep the signal from siding 1 as a starter but have a disc instead of 27. I don't think 27 and 28 would be combined. This assumes that a train will not depart directly from the loading bay i.e. any traffic from here is moved to a road with a starter before departing. If it does then 27 remains as a starter. Maybe neither 27 or 28 is a starter and traffic from this area is marshalled into a platform before it sets off.

  • Agree 1
  • Thanks 1

Share this comment


Link to comment
kitpw

Posted (edited)

Rich - thanks for the clarification: I think a start signal at 28 would be appropriate as the siding is long enough (in modelling terms) for vehicles to be marshalled for exit onto the Up branch, leaving a disk on the short bay at 27.  

Kit PW

 

 

Edited by kitpw
  • Like 2

Share this comment


Link to comment
WillCav

Posted (edited)

I've been thinking about signal 3 and the crossover.

 

On the real railway, you would want to be able to shunt the crossover whilst a train is heading towards the terminus.

 

There would likely be absolute block in place between the signal boxes and the last signal at box B would remain 'on' until the block section and 440yds beyond the first home signal are confirmed clear through the block instrument. So as well as your offstage distant, you need an offstage home signal 440yds before the crossover.

 

The only thing this will influence on your model is the number/colour of levers and the diagram in your signal box (if visible).

 

Regards

 

Will

Edited by WillCav
  • Informative/Useful 1

Share this comment


Link to comment

 

Will - thanks for those thoughts and taking time to study it!  I will look at introducing these 'virtual' signals into the way that Swan Hill is operated.  For an Outer Home signal to be pulled off, points 1a/1b (and probably 6 and 7) would need to be normal so there is clearly an interlocking requirement there.  In terms of operating the layout, that interlocking requirement can be met by taking off the Down line track feed on the Up side of point 1a when the crossover is in use (points reversed). When 1a is normal, actuating the virtual Outer Home signal would restore the track feed - whether the signal is real or imagined, it would have a lever (switch) in the signal box and since it would be out of sight of the SB, an indicator that the signal had answered the lever.

 

The signal diagram is looking a good deal more sensible - it needs to be tidied up before posting a revised version - thanks to all for comment and ideas.

 

Regards - Kit PW

 

 

  • Like 1

Share this comment


Link to comment

Kit,

 

There are a few important things to think about with semaphore signal boxes.

 

Location: it needs to be in a position to be able to see the tail lamp of the trains as they go past before giving line clear to following trains.  It also needs to be next to the road if there is a level crossing and close to the points it controls. Modern design is 350m max for manual operation.  Signals can be considerably further as they are lighter to pull.

 

Size: if it looks too small for the location, it will look wrong. It can be too big as the company would design in some spare levers - or the track could have been rationalized leading to spare levers.

 

Heritage: the architecture should match the company or its predecessor- same with the signalling. 

 

Operation: 2 identical layouts could be signalled in different ways due to the operational requirements.  If you need to share platforms (say, pilot onto back of train) then you need calling on signals whereas if it were units only, you don't.   Think of all the moves you need - including run rounds and engine releases. You need signals for all of these ideally.

 

In 7mm, you can do the point rodding and signal wires. Don't forget expansion compensators. 

 

Good luck with the signalling - it is hard to get right as there aren't many resources about the subject.

 

Hope this helps a bit

 

Will

  • Like 1
  • Thanks 1

Share this comment


Link to comment
17 hours ago, WillCav said:

it is hard to get right as there aren't many resources about the subject.

 

It is well written but I think the Vaughan book wouldn't help too much.

 

The GWSG published a book last year, GWR signalling practice, which they strived to make the definite work on this subject. 

  • Thanks 2

Share this comment


Link to comment
kitpw

Posted (edited)

Will - thanks for your thoughts and ideas.  It helps enormously.

 

Signal box location - pretty much as per the diagram:  arriving trains will pass the SB before entering the platform(s). I drew on GWR terminus layouts for prototypes - typically Barnstaple Victoria Road but also Birmingham Moor Street: there are quite a few others which are similar.

 

I've relied on the Institution of Railway Signal Engineers publication 'Mechanically operated points' for guidance on layout and compensation. I assumed that the crossover 1a/1b would be Westinghouse type motors with hand generator(s) in the signal box - ok, I think, for 1927 although more common after 1940: in phase 1 of this project, points 1a and 1b will, in any case, be 'off scene'.

 

The size of a signal box is quite difficult to ascertain from published data: some sources give the number of levers with no overall dimension or a dimension without the number of levers: Historic England managed a whole publication about signal boxes without either. However, with levers at 5.5" centres (from Vaughan 'Great Western Signalling') and a requirement for at least 35 levers (to include notional detonators, FPL etc), the frame would be not less than 16'0". With end clearance and an internal staircase, I reckon double that, so about 32' in length. The box will be a GWR 'type 7' with 3 over 2 sash windows, gable roof and internal stair (the Uxbridge box fits that description except has an external stair). The signals will all be pre 1927 GWR types (rather than Saxby & Farmer) on timber posts with timber arms including the "O"and 'S' types rather than the plain small arm subsidiary style. Point rodding is largey installed but with some still to do. As for signal wires, I'm looking forward to those incredibly fiddly looking wire pulleys!

 

I have a copy of GWR Journal No 13 which describes the working of the Uxbridge branch - I need to study it in more detail and develop something along those lines for Swan Hill.

 

Rich - thanks for your comment - I have Vaughan, which is useful and will see if the GWSG book can be got via the library or s/h - It's £45 on the bookstands!  One of the things that's striking about GWR stations/signal diagrams is that some seem to have almost no signals where a similar layout in another place has many - quite baffling but, as Will says, I guess it's all operationally dependent.

 

Progress has stalled a bit recently but I've included a recent picture below... the lack of signals at the platform end is more than apparent!

 

Regards

Kit PW
 

162042.jpg

Edited by kitpw
  • Like 1

Share this comment


Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.