signal wire runs - placing and modelling queries

[andrewnummelin]

The first task given to me for a club layout was to make the facing point locks to complement the excellent point rodding runs that had already been installed (including replaceable bits for spanning baseboard joints!). This job is nearing completion but as the Ambis etch I used included signal wire detectors I started to think about the wire runs and how to represent them. Consulting the reference books I have, it was easy to determine how to set out point rodding but I could find almost nothing on signal wire runs.
The club layout represents a Welsh ex-GWR branch terminus set in the mid 1950s and assumed not to have any track circuits. From various bits of information I'm guessing on the following:
- signal wire runs closely follow point rodding;
- wire runs are outside the rodding;
- "bends" in the runs are made using pulleys rather than cranks;
- detectors are fitted for facing points but not trap points.

Here's the schematic station plan with track in black, point labels and rodding in blue, signals in red and signal wire runs in green.

ABB.jpg

Now the guesses and a question:
- switch 1a detector would control home signals B1 & B2 as part of a simple run from the signal box, there would be no detector on S1b;
- switch 2a detector would control home signals B3, B4 & B5, there would be no detector on trap S2b;
- switch 3a detector would control home signals B3 & B4, there would be no detector on trap S3b;
- switch 4b would not have a detector (similar logic as for such a switch not having a facing point lock);
- how far apart would the signal wire posts be? (By every third or fourth rodding stool?)

So the signal wires would run as follows:
- B1 & B2 from signal box, under the tracks, pulleys to turn left, through S1a detector and on to the signals;
- B3 & B4 from signal box, under the tracks, pulleys to turn right, pass S3a, pulleys to turn 180°, through detectors S3a and S2a and on to the signals;
- B5 from signal box, under the tracks, pulleys to turn left, through detector S2a and on to the signals;
- B6 & B7 from signal box, under the tracks, pulleys to turn right, follow tracks, link direct to B7, pulley to go under the track to B6;
- B8 from signal box, under the tracks, pulley to turn left, on to the signal (passing back under the track off scene);
- G1 & G2 from signal box, under siding, pulleys to turn right then pulleys and passing under the tracks, pulleys to turn wire towards the ground discs. Would the pull be to from the rear of the disc or would it have to pass the disc and turn back to pull from the front?
- G3 & G4 from signal box, pullies to turn as appropriate. Then again, pull from front or rear of the discs?

Now to the modelling. I have a Masokits etch so now I have to check whether it includes enough components or will I need more than one?
Then there's the matter of additional bits: the etch suggests one will need 0.4mm rod (I have this), 36SWG (0.19mm) soft brass and 41SWG (0.11mm) phosphor bronze wires. With the loss of Eileen's Emporium I've had to do searches on the internet but I have failed to find either of these wire types - has anyone any idea where I might buy some?

[bécasse]

I am not sure that your assumption that the traps wouldn't be detected is correct. I have certainly seen traps that had detectors but I can't remember now, over half-a-century later, the pertinent circumstances. I would certainly expect any trap passed over in a facing direction by passenger trains to be detected.

[Noel]

A few thoughts on the diagram, Andrew, which I hope you will excuse. I think, assuming B1 and B2 are distants, B2 wouldn't exist [just leave a working B1 'on'], and quite possibly not B1 in that form either, having been replaced by a fixed distant in the 1930s.
Is the line behind the platform a siding or a bay? If the latter, the trap is very unlikely, if the former the signal is perhaps more likely to be a ground signal, and B3 as well.
The box is is in a difficult place; somewhere around S1a seems much more likely, and much more convenient for dealing with collection and delivery of staff, tablet or token.
I think S4b might have an FPL, as passenger carriages will probably be pushed back in front of the station buildings rather than make passengers walk up the platform.
G3 could be a yellow disc, but I suspect that a double disc is much more likely.

Ground signal operating arms can be in front of, or behind, the disc, depending on who designed and installed it, and when... The manner of applying detection, if present, will vary accordingly.

[Neil Smith]

From memory of LMS practice, I would certainly expect the ground signals protecting facing points to have detection - in part because as Noel says you could have a single lever to pull off one of two discs, selected by the route set.

I have a copy of Adrian Vaughan's history of GW signalling - I could have a look for you later if that helps?

All the best

Neil

[andrewnummelin]

Gents,
Many thanks for the rapid replies. As you may have guessed my knowledge of signalling is limited.

The inspiration for the club layout was Dowlais Central but this had a double track approach whereas the club layout assumes single track.

Signals B1 & B2 are (outer?) home signals and control the junction between the main line and sidings. (There was something similar approaching Dowlais.)

It looks as though the trap point S3b (bay) should have a detector (unless a GWR expert confirms otherwise). (The bay in Dowlais was controlled by ground discs and I've seen no evidence that it was used for passenger services: on the model it is used for passenger trains so there are normal signals.) The bay platform line also has a cattle dock so I guess the trap point is justified.

There was variation on the prototype regarding facing point locks for the engine release (discussion elsewhere when I was counting how many fpls I'd have to make). Dowlais did not have a fpl on the engine release crossover so I guessed that normal practice would have been to leave the coaches where they were during run round so the train did not depart over the crossover: indeed I've seen a photo of a train so placed at Dowlais. However I have also seen photos of trains pushed right back to the stop blocks there and I've been told that this was done sometimes elsewhere even when not necessary because of the length of the train. It was assumed that the signalman would have been on the ball and would not have any reason to reverse the points once a loco had run round.

I suspect the location of the signal box on the model was a result of the design of the scenery but it looks to be not inconvenient.

DSC05055.jpg

I thought that during the period modelled it would have been unlikely for G3 to be yellow or a double disc. Personally I wonder if G4 is really necessary to protect leaving the headshunt - in the rare occasions that a movement would need to be stopped at it, I think a flag out of the signal box window would have been more likely.

Adrian's book doesn't have a section on signal wire runs but I have not yet looked at all the pictures for clues rather than definitive information.

[Noel]

I thought it looked vaguely familiar... At Central the point rodding for S3a & b and S2b ran along behind the line into the bay, with the signal wires for B7 (a ground signal), B6 and G2 behind them [i.e. further from the track]. The rodding and signal wire to S4a & b and G1 ran along the face of the platform, with no FPL for 4b so far as I can see http://www.alangeorge.co.uk/dowlaiscentralstation.htm. S5 has a hand lever. The "bay" is definitely a siding with a trap; it seems to have been a coal siding in later days, although the cattle pens were still there. There was a fence along the back of the platform, although its maintenance seems to have not continued past closure to passengers [apart from workmen's trains] in 1952.

The bay platform line also has a cattle dock so I guess the trap point is justified.

I'm not sure that the BoT would ever have permitted that, because of the risk of collision with standing vehicles? If they did, there would presumably have to be some way of warning the driver of an approaching train that the bay was already partially occupied.

[Stuartp]

There was variation on the prototype regarding facing point locks for the engine release (discussion elsewhere when I was counting how many fpls I'd have to make). Dowlais did not have a fpl on the engine release crossover so I guessed that normal practice would have been to leave the coaches where they were during run round so the train did not depart over the crossover.

Obviously not GW practice, but FPLs on the engine release crossover seem to have been very much the exception on ex-LMS lines* in Scotland, whereas pushing the coaches back to the buffers and departing over the unocked points was common if not actually the norm. At Dalmellington (G&SWR), despite the station remaining fully signalled to the end, the loco release points were not only unlocked but were worked by hand levers.

[grovenor-2685]

Facing point locks are required to prevent serious train wrecks when a train splits points which are not fully home when running at speed. A release crossover at a terminus can at the worst drop a couple of bogies onto the ballast at very slow speed and the guard can reasonably be expected to take rapid action, hence not really much of a risk. From what I have seen fpls in these cases are a relatively recent thing, mostly on preserved lines where the 'requirements' are not questioned. Back in the day they would just be the exceptions that proved the rule.

[Neil Smith]

Now the topic has slightly broadened, it's worth also noting, going back to detectors, that they do act as a sort of poor man's FPL. The detector is anchored to the stock rail, and detects whether the blades are fully home before the signal can be pulled off. Once said signal is off, the points cannot be moved, either deliberately irrespective of what any interlocking would require, or equally through other factors be these changing temperature causing expansion effects, broken bolts on the rodding, etc.

All the best

Neil

[grovenor-2685]

Detector slides have a bit more tolerance than fpls and are also made of much weaker material. They don't do the same job, the detectors are there to stop the signal coming off if the turnout is not set correctly, they will only stop the point moving if the signal is off provided not much force is applied.

Detection of points for traps and for shunt moves was covered by the MOT 1950 requirements as:

Detection by the relevant signals of the closed switch of facing trap points
in sidings is desirable, also of trailing points in passenger lines which are used
in the facing direction for shunting

Ie. desirable, so no hard and fast rule. I think that detection was provided in most cases, at least for the point where the shunt signal was sited.
NB. In the 1902 version detection is only mentioned in relation to facing points that need fpls.

You might find this page useful,
http://www.norgrove.me.uk/resources/rodding.htm

[andrewnummelin]

Gents,

Many thanks again for the very valuable replies. I confess to having forgotten about Keith's exceptionally useful website.

I think I now have enough info to make a good go at modelling at least an impression of the signalling wire runs apart from one thing that no one has picked up yet: following Eileen's demise where can I now get the 41swg phosphor bronze and 36swg brass wire mentioned in the Masokits instructions?

[martin goodall]

There was variation on the prototype regarding facing point locks for the engine release (discussion elsewhere when I was counting how many fpls I'd have to make). Dowlais did not have a fpl on the engine release crossover so I guessed that normal practice would have been to leave the coaches where they were during run round so the train did not depart over the crossover.

Obviously not GW practice, but FPLs on the engine release crossover seem to have been very much the exception on ex-LMS lines* in Scotland, whereas pushing the coaches back to the buffers and departing over the unocked points was common if not actually the norm. At Dalmellington (G&SWR), despite the station remaining fully signalled to the end, the loco release points were not only unlocked but were worked by hand levers.

Actually, it was also standard practice on the GWR not to lock the engine release crossover with an FPL (although the normal locking of the signal box lever frame would prevent the starting signal being pulled off unless the engine release crossover was set Normal, and pulling off the starter would back-lock the lever for the engine release crossover in the Normal position).

[Horsetan]

...You might find this page useful,
http://www.norgrove.me.uk/resources/rodding.htm

Some of the photos don't show up when viewed on a tablet.