Signals

THE SIGNAL BOX


Railway signalling discussion

Port Dinllaen

For railway modellers to discuss and share ideas on realistic signalling arrangements.

Port Dinllaen

Unread postby tynewydd » Sun May 11, 2014 11:05 am

A scheme late in planning (very) that has 4 stations centered around this big one -> Port Dinllaen. Advice/brickbats please!

I have been working on this whole plan for over 18 months so far, but I have begun to notice that this particular "railway that could have been" has started to become quite popular of late - see, for example, the latest RM. Where my ideas differ is perhaps that I have taken things further and imagined PD as a terminus junction between the GWR/LNER and LMS - all drawn by the lure of a second route to Dublin. This leads to a boundary terminus ala Bath Green Park but MR/ER/GWR in the period modelled - circa 1960. The imaginary "faction" shenanigans to get that to happen are set forth in this document which also contains ideas about typical train movements at PD.

The track diagram - as far as I have got - is here. I took a late vintage Holyhead, foreshortened it, welded on a soupçon of Bath Green Park, completely altered the MPD and the local goods yard which is more Buckingham GCR... My abject apologies for the myriad mistakes of omission and commission I have made - all of which will certainly revolt you. :oops:

Questions -
  1. I'm having trouble with call-on signals - which will be needed for the station pilot to grab carriages from Pl1 and move them to Pl2 or Pl3 and for new locos to back down onto trains that are in Pl4. It seems likely that I will want to be able to call those engines out of MPD Out and hold them before they enter the Platforms - so that implies calling on over the Up and Down - which I suppose means a "normal" signal as well above each? In my current plan I have used a ground signal and a ground calling on subsidiary. But I have never seen one of the latter - although I have seen a picture of a ground distant. Do they always need to be post-mounted miniatures instead?
  2. The detailed trackplan of Holyhead I have is of a late vintage - since it was re-modelled in between about 1970 and 1980. There are a reasonable number of theater/indicators in that plan - but I don't know what is typical for a 1960 period, and when to use a set of miniature arms and when a stack of ground signals instead.
  3. The necessary compression to get the station into 15' length including a rapidly diverging "branch" has pushed the throat together quite a bit and led to several slips (although nothing as dense as Newcastle!) - but it also means that most signalling will have to be at the edge of the throat, I think. I do hanker after a gantry as marked - but it would mean that for some routes the signal would be over the center of the points or slip it controls - is that ever OK - or would you have to move it either further out (left) or further in (right) or have two gantries (or none)?
  4. Gantry sighting - the current gantry is maybe 140 scale feet from each of two road bridges. What would the plan be there - co-acting under-slung repeaters, perhaps? What is the standard on when they are required?
  5. When you have a boundary station - does the signalling type change abruptly as control of the metals changes over? I envisage that at one time there were two signalboxes, one of which controlled the GWR branch (still could be modelled as there, in fact). But even suppose they were rationalized into one - would we see LQ arms on the ex Cambria/GWR as whereas the rest is UQ with perhaps as an odd survivor or two of LQ MR days - like the famous advanced starter for Pl2 at Holyhead ?

Anyway, thanks so much for looking and I welcome any pointers you can give - however radical. The entire folder of stuff including the above am working on is here.

Adam Richards

PS You are all welcome to reuse any of the signal parts I have in the Illustrator file inside that folder. If there's interest I can put them into a separate template file.
tynewydd
Crossing box
Crossing box
 
Posts: 49
Joined: Tue Feb 15, 2011 2:04 pm
Location: Les Etats-Unis, Cote Ouest

Re: Port Dinllaen

Unread postby Danny252 » Sun May 11, 2014 12:34 pm

tynewydd wrote:I'm having trouble with call-on signals - which will be needed for the station pilot to grab carriages from Pl1 and move them to Pl2 or Pl3 and for new locos to back down onto trains that are in Pl4. It seems likely that I will want to be able to call those engines out of MPD Out and hold them before they enter the Platforms - so that implies calling on over the Up and Down - which I suppose means a "normal" signal as well above each? In my current plan I have used a ground signal and a ground calling on subsidiary. But I have never seen one of the latter - although I have seen a picture of a ground distant. Do they always need to be post-mounted miniatures instead?


For your "ground distant", all I can think of is that you've seen a yellow disc, which has an entirely different meaning to "distant"! However, you have used one correctly for the quarry headshunt.

Regarding shunting, ground/disc/shunting signals always act "calling on" - the same disc will clear whether the platform is empty or occupied, and it is the driver's job to act accordingly. Only main running signals would require calling on arms, but even they can be dispensed with if Station Yard Working is in force.

The main site has a page on shunting signals, which you might find useful

When you have a boundary station - does the signalling type change abruptly as control of the metals changes over? I envisage that at one time there were two signalboxes, one of which controlled the GWR branch (still could be modelled as there, in fact). But even suppose they were rationalized into one - would we see LQ arms on the ex Cambria/GWR as whereas the rest is UQ with perhaps as an odd survivor or two of LQ MR days - like the famous advanced starter for Pl2 at Holyhead ?


I assume you mean a signalbox controlling the quarry access on the GWR single line? In that case, I would expect its signals to be LQ - after all, it's the GWR who built and maintain them! In the case of rationalisation, I'd lean towards them having been replaced as part of an overall resignalling of the station in this example, but boxes inheriting signals of the "wrong" type wasn't unheard of. Another consideration if you did this would be that each box would need its own distant signals (situated under the previous box's starters for distances this short), which might get a bit messy in modelling terms!

The gantry seems fairly reasonable, but the space does sound quite limited!
Danny252
Rest-day relief
Rest-day relief
 
Posts: 693
Joined: Thu Aug 26, 2010 4:39 pm
Location: As close to the levers as they'll let me be

Re: Port Dinllaen

Unread postby Richard Lemon » Sun May 11, 2014 2:38 pm

Adam,

Sorry to be a pedant, but should it not be called Porth Dinllaen or to be strictly correct "Porthdinllaen"?

Diolch!

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: Port Dinllaen

Unread postby Mike Hodgson » Sun May 11, 2014 4:18 pm

My first comment would be that I don't see the signalbox locations on the plan. With mechanical point operation, it has to be a fairly short distance from the points - there is a limitation on the length of rodding. By the 1970s/80s, power operation of the points would not be an issue, but if your box is the "original" one from an earlier date it would have been so constrained.

Given your hypothetical history, I agree I would have expected there to be at least two boxes, with the GWR tending to use the platforms on that side. This is unlikely to have been rationalised between the wars, but conceivable after nationalisation. I would expect that history to have set a precedent for platforms to have established a company bias to platform allocation and usage. The GWR box would be the smaller, and probably not large enough to handle the whole station; a single structure is therefore likely to be either a BR design or a resignalled LMS/LNER one depending which company got the job of signalling that joint operation.

I would also expect two goods sheds and engine sheds originally, a GWR one and a larger joint one. Perhaps your former GWR shed can be deemed to have been off-scene a little way off along that branch. The position of your yard is inappropriate for it to have been the LMS/LNE yard, as they would have to cross the GWR to reach it - it would have been on the other side or further back along that line.

The next thing that strikes me is that you have two roads crossing the station throat diagonally from bottom left to top right with double slips on the left hand road. The right hand of these roads does not serve the main platforms 2/3/4. Most modellers would have gone for only one road, both to minimise cost of pointwork and to save space - the trade-off might be against an extra carriage length in the platforms. Traditionally single slips were preferred, though double slips would not be unusual at a terminal station. Given that you have these two roads, I would expect those diamonds/slips to be arranged to permit parallel movements eg arrive platform 1 while departing Pl 3 on the main lines. This would involve more single slips but fewer double slips/diamonds (use the left hand diagonal for arrivals, the right hand one for departures). You have a third diagonal parallel to these two between Platforms 1/2. I don't see what purpose this road serves - it does not seem to allow any more flexibility.

If I am reading it correctly, the highest signal on the 3-doll branch Home signal reads to Pl 4, implying this is the principal route. However as I said earlier I don't think this would be the most usual main route from the Branch. I would have thought both originating and terminating branch trains would generally use Platform No 1 via the goods road you have labelled Gds In, certainly while you had a GWR box there. So I think your ringed arms are probably inappropriate, and also the trap, whilst correct for a goods road, would also not be there if it is a passenger route.

I would expect any trains from the branch requiring to work forward onto the main or vice versa would typically reverse in Pl 2/3. Terminating/originating trains on the Main would typically use 4/2/3 in order to avoid unnecessary traffic conflicts. If the Gds In road is only for goods use, any branch passenger working conflicts with any other movement except departure from Pl4 to Main.

The running signal and call-on protecting the crossover at the end of Pl4 seems unlikely (and not mirrored on Pl 1b for the run-round point there). I assume this crossover is for engine release (though you can't use it while you have a goods on the Goods Out road). I would not normally expect a signal there. Instead I would expect the crossover to be locked normal when a train is arriving/departing Pl4, or perhaps a release to a ground frame working the crossover.

The quarry siding off the branch would probably be worked by a ground frame by the 70s/80s. Is the big X a public level crossing? This would probably have had a GWR gate keeper's hut, as I doubt the GWR box could be right beside it and be within the limits for point rodding nearer the platforms. It might possibly have been be combined with Danny's Quarry sidings box; your (demolished) former GWR engine shed might also be deemed to be close by, so it could even be a medium sized box with a goodly number of spare levers. By the 70s/80s this would have been reduced to a non-block post; the crossing might well have been converted to barriers.

I take it the two bridges are road over-bridges with their supporting pillars? These would undoubtedly give rise to sighting problems (especially all the pillars supporting the one across the platform ends), so you will need to consider signal heights and whether any banner repeaters are necessary. Look at it from a driver's eye point of view - I doubt they would see signals on the gantry unless underslung.

Your gantry does seem to be in the right place for signalling purposes. Unfortunately for adequate sighting those signals probably want to be mounted on or right up against the road bridges, but unfortunately that would put them on the wrong side of the points they apply to. The usual way round this problem is to move the junction signals further back. In this case that would mean the platform starters would all be junction signals clearing only when the route is set right through the station throat. It makes the interlocking more difficult but I suppose you can accept that! As a modeller I suppose your decision is whether you want the bridge or a gantry :)

The cattle dock would have become redundant by the 1970s, though the siding could remain useful for eg engineer's stock.
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2475
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Port Dinllaen

Unread postby Mike Hodgson » Sun May 11, 2014 4:31 pm

Richard Lemon wrote:Adam,

Sorry to be a pedant, but should it not be called Porth Dinllaen or to be strictly correct "Porthdinllaen"?

Diolch!

Richard


Nonsense! It is well known that the railways were built and run by Englishmen who could neither spell nor pronounce Welsh names, (and who weren't that hot with English names either).

This message was posted from Drenewydd Gelli Farch, a place with two corrections to its spelling on the village name-board! GWR Hall class 4967 only ever carried the English name of the local "big house".
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2475
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Port Dinllaen

Unread postby Richard Lemon » Sun May 11, 2014 5:19 pm

Ah, but Porth Dinllaen is the English version!!!!!!!

Anyway, it is a lovely place, great for Seals - but that is not relevant to the signalling!

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 375
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: Port Dinllaen

Unread postby tynewydd » Mon May 12, 2014 12:00 am

Danny252 wrote:For your "ground distant", all I can think of is that you've seen a yellow disc, which has an entirely different meaning to "distant"! However, you have used one correctly for the quarry headshunt.


This is the one I meant - from this sites S&T forum - http://forum.signalbox.org/viewtopic.php?f=6&t=3699#p43554 - I assume the lower disc is supposed to be a distant because of the black field and the yellow banner with fishtail and black chevron.

Your note about ground signals reading as calling-on themselves is very helpful and explains why I never saw such a picture of those.

Danny252 wrote:I assume you mean a signalbox controlling the quarry access on the GWR single line? In that case, I would expect its signals to be LQ - after all, it's the GWR who built and maintain them! In the case of rationalisation, I'd lean towards them having been replaced as part of an overall resignalling of the station in this example, but boxes inheriting signals of the "wrong" type wasn't unheard of. Another consideration if you did this would be that each box would need its own distant signals (situated under the previous box's starters for distances this short), which might get a bit messy in modelling terms!


I actually meant a box positioned right at the end of the GWR line because I assume it would be handy for token exchange - needed for a single line - and give visibility around the tight curve.

As a prototype note - in Holyhead before rationalization, according to "Chester to HolyHead Railway" by Anderson and Fox there were at least two boxes - No 1 (40 levers) controlled the end of the throat and No 2 (74 levers) the Platforms. I say at least because think there was a No 3 that controlled the mail pier at one point as well. Anyway, arriving trains passed 3 distants, one for No 1 and two positioned under homes both of which had levers from No 2 box. The rear one was an "outer distant" perhaps? I know there was an accident at HH where one train rear-ended another in Platform 1 and excessive speed down the falling gradient was seen to be a contributory cause - having multiple distants that could be pulled off leading to terminal platforms might not have helped either....

Departing trains, however, did not see any distant as they passed between No 2 and No 1 control. I imagine this implies that the Up Main block post was owned by No 1 and the platform starters only gave clearance to it.

Danny522 wrote:The gantry seems fairly reasonable, but the space does sound quite limited!

I'll wait to build consensus. I am half-thinking that moving that gantry to just behind the left-hand bridge might clear things up a bit, with Up direction signals above and mounting under it any signals needed in the Down direction so they are visible under that bridge.

Thanks again,
Adam
tynewydd
Crossing box
Crossing box
 
Posts: 49
Joined: Tue Feb 15, 2011 2:04 pm
Location: Les Etats-Unis, Cote Ouest

Re: Port Dinllaen

Unread postby Pete2320 » Mon May 12, 2014 3:50 am

tynewydd wrote:This is the one I meant - from this sites S&T forum - viewtopic.php?f=6&t=3699#p43554 - I assume the lower disc is supposed to be a distant because of the black field and the yellow banner with fishtail and black chevron.

Ah, that one almost certainly is a distant but is also a red herring! Note that it stands next to a miniature railway so is probably acting as a running signal.

Pete
Pete
Pete2320
Main line box
Main line box
 
Posts: 2113
Joined: Wed Feb 20, 2008 10:50 pm
Location: Barton on Humber

Re: Port Dinllaen

Unread postby Danny252 » Mon May 12, 2014 12:35 pm

Pete2320 wrote:
tynewydd wrote:This is the one I meant - from this sites S&T forum - viewtopic.php?f=6&t=3699#p43554 - I assume the lower disc is supposed to be a distant because of the black field and the yellow banner with fishtail and black chevron.

Ah, that one almost certainly is a distant but is also a red herring! Note that it stands next to a miniature railway so is probably acting as a running signal.

Pete


I think it's worth perusing this forum with a bit of caution - a lot of the things you'll find illustrated are here because they're the exact opposite of normal practise!
Danny252
Rest-day relief
Rest-day relief
 
Posts: 693
Joined: Thu Aug 26, 2010 4:39 pm
Location: As close to the levers as they'll let me be

Re: Port Dinllaen

Unread postby JRB » Mon May 12, 2014 1:49 pm

At least Port Dinllaen is getting more thoughtful planning than a lot of full size installations.
JRB
Double-manned box
Double-manned box
 
Posts: 3206
Joined: Sun Dec 9, 2007 10:12 pm
Location: GWR

Re: Port Dinllaen

Unread postby tynewydd » Mon May 12, 2014 3:00 pm

Mike Hodgson wrote:My first comment would be that I don't see the signalbox locations on the plan.

Right because I was wrestling with it. The easiest place (space-wise) for a single rationalized box would be between the two bridges on the MPD side.

Mike Hodgson wrote:I would also expect two goods sheds and engine sheds originally, a GWR one and a larger joint one.

Well I have the two sheds in the MPD - ala BGP. I might be able to squeeze a second shed into the good yard, but I can't get another yard into the other side of the tracks. In my notes you may see that I have provisioned a LMS bypass for freight that avoids PD all together. Perhaps I can plead that the width of the PD site is restricted by geography and the Admiralty to a narrow strip and so all goods facilities had to be shared. In general, I wonder if we should see the GWR as a late-comer, forced to put up with whatever arrangements the existing owners can be bothered to provide.

Mike Hodgson wrote:The next thing that strikes me is that you have two roads crossing the station throat diagonally from bottom left to top right with double slips on the left hand road. The right hand of these roads does not serve the main platforms 2/3/4.

Right - the HH prototype has semi-dedicated engine release road. It doesn't serve 2/3 because they are departure only. The reason for that is that the boats arrive adjacent to Pl 2; discharge and then move around to Pl1 to onboard. It's other function was that it has once had an additional slip coming from the adjacent goods line which led across to the Gds Out side. This is a place where Peco's range of points is deficient!

Mike Hodgson wrote:Given that you have these two roads, I would expect those diamonds/slips to be arranged to permit parallel movements eg arrive platform 1 while departing Pl 3 on the main lines.

Yes - I see the reason is that I have managed to attach the Up and Down shifted one over from HH somehow. The Gds In was the Dwn, the Dwn was the Up and the current Dwn was a Up/Down Passenger (once a Down Gds). Hmm - time for a rethink on that alignment.

Mike Hodgson wrote:If I am reading it correctly, the highest signal on the 3-doll branch Home signal reads to Pl 4, implying this is the principal route.

Yes - I agree basically - I had intended for Pl1 to be the termination point of most inbound regardless of branch or mainline. I was thinking of the "through" trains that would be say Cardiff to Manchester. Those have to "cross-over" in one direction or other regardless of which way is "normal". Currently only Pl4 has the capacity for double-headed trains to arrive and have a change of motive power as well.

Mike Hodgson wrote:I would expect any trains from the branch requiring to work forward onto the main or vice versa would typically reverse in Pl 2/3. Terminating/originating trains on the Main would typically use 4/2/3 in order to avoid unnecessary traffic conflicts. If the Gds In road is only for goods use, any branch passenger working conflicts with any other movement except departure from Pl4 to Main.

Right - consequence of above shift of exit tracks. But we also have to consider the effect of the boats. Pl4 would be the least favored of the platforms for arriving OR departing boat passengers (further to walk, no overhead roof - important in North West Wales!). Therefore I can see an argument that when there are competing services for London and the owners of the station are LMS/LNE, they might consign the GW trains to that spot out of spite.

Mike Hodgson wrote:The running signal and call-on protecting the crossover at the end of Pl4 seems unlikely (and not mirrored on Pl 1b for the run-round point there). I assume this crossover is for engine release (though you can't use it while you have a goods on the Goods Out road). I would not normally expect a signal there. Instead I would expect the crossover to be locked normal when a train is arriving/departing Pl4, or perhaps a release to a ground frame working the crossover.

It is odd, but it's a feature of the HH plan post re-org. The main signal seems to have no lever (fixed?) and the calling-on is operated by lever 3 of a GF. The other two leavers are #1 "Release" and #2 the crossover points. I guess Release was actually to lock the associated signals in the box?

Mike Hodgson wrote:The quarry siding off the branch would probably be worked by a ground frame by the 70s/80s. Is the big X a public level crossing?

No sorry - that was my ham-fisted way of acceding to the strictures of the plan posting rules - its supposed to be two opposing arrows showing bidirectional working. Then the track got extended...

Mike Hodgson wrote:I take it the two bridges are road over-bridges with their supporting pillars?

Yes. I agree about the sighting. I may be able to nudge the LHS bridge left, however.

Mike Hodgson wrote:The cattle dock would have become redundant by the 1970s, though the siding could remain useful for eg engineer's stock.

Yup - that's another reason why I'm early 60s. HH lost all its cattle traffic only after the Menai Bridge fire in 1970.

Thanks again for the help, I'll cogitate on the alignment of the entry tracks and update.

Regards,
Adam
tynewydd
Crossing box
Crossing box
 
Posts: 49
Joined: Tue Feb 15, 2011 2:04 pm
Location: Les Etats-Unis, Cote Ouest

Re: Port Dinllaen

Unread postby Mike Hodgson » Mon May 12, 2014 5:40 pm

Pete2320 wrote: Ah, that one almost certainly is a distant but is also a red herring! Note that it stands next to a miniature railway so is probably acting as a running signal.

Pete


Correct.
Dave posted it to illustrate his slotting arrangement.
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2475
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Port Dinllaen

Unread postby Mike Hodgson » Mon May 12, 2014 7:50 pm

tynewydd wrote:It is odd, but it's a feature of the HH plan post re-org. The main signal seems to have no lever (fixed?) and the calling-on is operated by lever 3 of a GF. The other two leavers are #1 "Release" and #2 the crossover points. I guess Release was actually to lock the associated signals in the box?

Yes, the main signal would have been fixed at danger, as there is no running movement beyond it. You have to have a main signal there because it's not allowed to have the previous running signal leading to only a shunt signal. The GF and the signal box would lock each other out, ie the points must be normal whilst a train is arriving or departing, and the signal levers for such moves would have to be normal before the GF could be worked. The distance from the box (point rodding run) and lack of view would make it much easier to work this way than trying to work the points from the box.

The 1965 Holyhead accident was put down to a misjudgment on the driver's part, because the call-on authorises him to enter the occupied platform, prepared to stop short of any obstruction, and this he patently failed to do. He fully expected to find the train there, but evidently expected it to be shorter/nearer the stops and his speed was only slightly excessive.

The signalmen didn't escape comment either though - they were regularly failing to apply rule 39(a), clearing the outer home signal prematurely when the inner home was at danger and 44(b) clearing the call-on before the train had come completely to a stand, although in mitigation they were not in a position to know whether the train had stopped because they did not have a proper view. I don't think the distants came into it. It used to be normal practice to clear distants into terminal platforms, but only when the road was clear to the stops (it wasn't, and the distant was at caution)
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2475
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Port Dinllaen

Unread postby davidwoodcock » Mon May 12, 2014 9:55 pm

Mike Hodgson wrote: Yes, the main signal would have been fixed at danger, as there is no running movement beyond it. You have to have a main signal there because it's not allowed to have the previous running signal leading to only a shunt signal.


Wasn't the use of fixed semaphore stop signals a comparatively recent (say 1960s) feature that grew out of the introduction of fixed red+subsidiary colour light signals (which were definitely in use by 1960)?

I would have thought that earlier just a small-arm subsidiary signal would have been used, being cleared to permit the shunt movement instead of the dolly later provided. I seem to remember that Holyhead had somewhere in its layout the shortest arm LQ LNWR signal that I have ever seen which had just this sort of function.
davidwoodcock
Rest-day relief
Rest-day relief
 
Posts: 947
Joined: Tue Jul 12, 2011 7:58 am
Location: Champlon, Belgique

Re: Port Dinllaen

Unread postby tynewydd » Tue May 13, 2014 7:16 am

Mike Hodgson wrote: Yes, the main signal would have been fixed at danger, as there is no running movement beyond it. You have to have a main signal there because it's not allowed to have the previous running signal leading to only a shunt signal.


The signal in question surrounded by an very ornate fence!

Seems the photo is relatively recent based on that illuminated sign. This also gives you an idea of the perils of passengers on this platform in inclement weather with an easterly wind.

Adam
tynewydd
Crossing box
Crossing box
 
Posts: 49
Joined: Tue Feb 15, 2011 2:04 pm
Location: Les Etats-Unis, Cote Ouest

Next

Return to Signalling - model railways/simulators

Who is online

Users browsing this forum: No registered users and 1 guest