RailWest

Time to visit the 'Dorset' again :-) I’m trying to sort out the thorny matter of the ground-frame at Shillingstone. It was not in the 1878 layout, but existed by 1901. Was it perhaps the case that the loop had been lengthened at the south end by that time, putting the facing points beyond the contemporary limit for direct mechanical linkage? The exact date of its abolition is not known, but circumstantial evidence suggests 1915. The 1901 signal diagram marks it on the Up platform at the south end, but maps suggest that it was off the platform (probably just past the ramp). Photographic evidence appears almost non-existent, except that there is a hut of some sort in the background of one view looking south. That hut has similarity with the later greenhouse, but it would be an odd place for the latter so near to the track. Conversely the actual greenhouse appeared to be rather un-greenhouse-like to me, so I wonder if perhaps the greenhouse in fact was the old GF hut modified for a new use? But when was the greenhouse first recorded at Shillingstone? Can anyone add/clarify anything please?

Correct. Regulation 1 specifies that the token must be replaced in the instrument before the signalman (at B) sends 2-1 to A. Regulation 7(a)(ii) requires the signalman at A to send 3-3 to B if the line is still occupied outside the Home signal after receiving that 2-1.

Do not forget that GWR key token machines had only manual indicators. Also IIRC the same principles applied to electric tablet and electric staff working. Unlike 3-position block instruments for double-line working, with token/staff/tablet machines you can't simply turn an indicator to 'Train in Section' in order to record a 'Block Back' as they don't work like that. This is where good booking practice in the Train Register comes into play :-)

I would disagree. Although there were a lot of loops on the GWR with traps, in general they tended to be either for speeding up the services (eg the Barnstaple and Minehead branches), or for safety reasons such as severe gradients etc (and then more as a protection against 'break aways' on unfitted goods trains). The Stationmaster may have other views? :-)

Correct on both counts :-) Indeed, I would suggest that trap-less version is/was by far the more common arrangement - not only cheaper to install and maintain, but facing points in passenger lines were usually avoided as far as possible. Conversely, in a 'light railway' scenario it was not uncommon to find passing-loops with only Homes, the authority to start away into the next section being given verbally by the signalman (or whoever was doing the equivalent job) when handing over the staff/tablet/token for the next section.

Actually, my comment was meant in admiration, but I'm not sure of the right smiley for that ! Although our 2900s have long since gone to wherever such things end up, I still have the odd reel of magnetic tape for them somewhere in the garden shed - it's quite good for keeping the birds off the fruit bushes, though eventually the black oxide wears off in the rain......:-)

I remember the 2900 series :-) Mind you, I had a boss once who could remember LEO machine code! I worked for many years in maritime IT - I like to think that by the end I understand at least some of what the sailors were telling me, and that hopefully they understand at least some of what I told them. At least I never sank anything, but they did manage to wreck quite a few pieces of our kit :-( Never tried to ask them about block working though.....

If you look at the bottom left, it says 'Boston, Mass'. More obviously, all the signals are the wrong way round for the UK, as they point to the right not the left !

Ironically, I had to look this up just recently, but I forget the Regulation number now :-( Don't know whether things have changed in recent years, but..... IIRC:- 1. Provided no tokens were out, you could shunt into a section by Blocking Back without needing a token. This could be done at both ends of the same section at the same time if required. 2. If there was a token out for a train already in the section going away from you, then you could shunt into the section behind it without Blocking Back, but you had to BB immediately if you got Train Out of Section before your shunt had been withdrawn. 3. If there was a token withdrawn at the other end for a train to come towards you - keep out!

>>>The average modeller with an average fictitious layout , simply doesn't have to understand the specifics you mention... It all depends what the 'average modeller' (and I put myself firmly in that category!) wants to do/achieve. If he just wants the 'right sort of signals' in 'roughly the right places', then maybe OK. But if - as you say when talking about interlocking' - he wants to avoid obvious howlers, then I would argue that he does need to have a modicum of knowledge of some of the associated issues. For example - I once saw a very nice layout equipped with all the right signals in all the right places, but.....the layout was for a pre-grouping period, but it used signals of a later type and installed in accordance with practices not introduced until nationalisation. Not a bit of knowledge that the builder might have gleaned simply from reading a 'simple guide to signalling' and looking at a few photographs.

>>>But a few post after it, it was poo-pooed because the GWR didn't always leave enough room for the their trains between the last potential place of a collision and the starter signal so needed permission of the box in advance to enter their block section to shunt, an unnecessary complication which some of the professionals seem hell bent on throwing into this thread..... With respect, such an unwarranted jibe IMHO shows a lack of understanding of exactly why there are some of use who caution that "it is not as simple as...". There were many places (on the GWR, if you want to be specific) where, if you placed the 'section' signal sufficiently far enough away for the longest train to be able to draw clear of a siding or crossover for shunting purposes, the front of the train would have been out-of-sight of the signalman. This might be because of the curvature of the line or an intervening overbridge etc. Now consider the case of a light engine, or perhaps a very short goods train, drawing up to the 'section signal' to await 'LIne Clear' from the box ahead. In the days before the widespread use of track-circuits there would be nothing to remind the signalman of the presence of that engine/train, other than what he could see from his box. Sadly, the inevitable results of occasional human failings was an accident when a second train was signalled into the back of the first. Consequently it became the practice in earlier days to place the 'section signal' in such a location that any engine stood at it could be seen from the signal-box - which is why, in many cases, a shunt-ahead arm became necessary to facilitate shunting. This was not an "unnecessary complication", but an example of adopting what today would be called "safe working practice". The modeller needs to understand that there is more to the correct placing of such a signal than simply the length of the longest train to be shunted.

Pardon? The only 'block' in signalling terms would be a 'block section' and I would suggest that most layouts would lie outside of a block section - to be more precise, outside of one block section if a terminus, outside/between two block sections if a through station.

Ah, but.....a signal faces the oncoming traffic to which it is intended to apply. Therefore the 'front' and 'back' of a signal are viewed from the opposite perspective than that of the driver of the appoaching train. So there is no inconsistency in terminology really

>>> It's a long way short of being full knowledge but is a step in the right direction, and a model with such signals is going to be more accurate than one without,....... I would not disagree with that view. As a 'starting point', then no doubt perhaps it does achieve what the OP was seeking. However, I do not think that negates my point that it depends how far you want to go towards being 'correct'. If a modeller is just happy to have signals that look the right sort and are in the right place, then fine. But if he also wants to operate his layout with prototypical correctness (is that PC in another form - eek!), then I would argue that he does need to understand at least the basics of block working etc as well. Clearly there are many differing views on this matter - I'm not suggesting that anyone is better or worse than any other, merely offering my own perspective.

>>>It shows the junction, it shows the two stop signals needed to safeguard it, and it shows without any words why they're needed. To people who just want to get things a bit more right, that's all they need. That would depend upon what it was that people wanted to get right, and how far they wanted to go towards being correct. For example, the picture as shown IMHO would not stop a modeller committing the faux pas of having trains approaching the signals on both lines at the same time, even if both arms were 'on'. (S)he would need to understand about accepting trains under Regulation 4, not fouling the Clearing Points, possible provision of Outer Homes etc.

Nice to see someone making an effort with proper rodding etc. Should be a bit easier for Spetisbury...:-)

A good point, with which it is hard to disagree. However, a contrary view.... I would suggest that many modellers 'learn' from what they see at exhibitions in much the same way as they 'learn' by visiting heritage railways and/or railway 'historical exhibits' (which sometimes may be less-than-accurate). I have always held the view that, if you are going to 'educate the public' then there is a moral - if not legal - obligation to educate them correctly. If, for practical reasons (eg lack of relevant information, objects etc), you can not do so with a high degree of accuracy, then you should at least make clear the known discrepancies. I do not believe that we should (appear to ) condone 'bad habits'. Meanwhile, it's back to my 16.5mm gauge track and old Tri-ang wheels.....

This talk about Outer and Inner Homes and Clearing Points etc reminds me..... There was one particular location with both Outer and Inner Homes where, because of the particular nature of the location, the interlocking required the Inner Home to be pulled first to release the Outer Home. This meant that, in the event of a conflicting move in advance of the Inner Home, then any approaching train was always held at the Outer Home. On a particular model of this location the Outer Home was 'off scene', so the owner-operator (unaware initially of the prototype working) always allowed approaching trains to draw up to the Inner Home while awaiting the conflicting movement, as he felt that it added 'interest' to see both trains in the visual part of the layout at the same time. So - visually the signalling looked OK, but operational the layout was incorrect. As the owner - having been made aware of the prototype - chose to continue with his chosen practise, to what extent (if any) can we criticise him for doing what he likes with his own model?

...no, those were off-scene :-) I doubt they had green levers in the box interior though......

>>>A simple way around this is to model a prototype then from photographs and diagrams install the signals..... With some caution :-) Some examples from personal experience:- 1. A book of photographs, track-plans and signal-diagrams for a particular railway. You will be surprised (or maybe not) by the number of times puzzled modellers have queried why they can't seem to get information for their chosen station to 'match up', either oblivious to - or ignoring - the unfortunate fact that in many cases the photos, plans and diagrams all relate to different periods. 2. A junction station which once had two signal boxes, later replaced by one. At the same time the junction layout was re-modelled and the signalling altered accordingly. A well-known model of that station on the exhibition circuit had the signalling of one era superimposed on the track-layout of the other era! 3. A model of a particular station in the early 1920s, signalled entirely with UQ signals.

To mis-quote... "Proper signalling of a layout is not a matter of life-and-death - it's more important than that"

Looking at signal-box diagrams "in isolation" is also fraught with problems. Many layouts were altered over the years, at different times with different standards/requirements in place, so it was quite common to find somewhere with shunt signals at some points and none at others. To 'understand' the signalling of a prototype you need to know more about its history. This is why two or three seemingly identical layouts on the same railway may appear to have been signalled in different ways.

>>>I also note that often the home signal is in rear of the station platform (unlike the up home diagram in post #34), so it would be unusual to be able to bring a passenger train to a stand at the platform and then shunt in front of it. I don't see why. If there was a Home in rear of the one shown on the diagram as No 2 (ie before the train arrived at the platform), then you could apply Rule 39(a) and allow the train to draw past it up to No 2 (which would be kept 'on'). Once the train was a stand there, no reason not to shunt across 7/9 if necessary. >>>The passenger train doing its own shunting - e.g. detaching a horsebox - is a different matter, as it's within station limits. In the diagram as shown, once the train was past 2 then it would indeed be within station limits. Whether it shunts, or another engine does the shunting, I don't see why you consider it a 'different matter'. >>>In fact the layout in post #34 would be inconvenient, as to detach a vehicle and put it in either the up or down siding, the up home would have to be pulled off. So? Why would that be inconvenient? >>>My point is, even if the signals are off-stage, one should operate believing in their existence. I agree completely !! :-):-)

To quote from some preceding posts..... "I use the term " stop " signal specifically, because in reality starters and homes etc have no role on the majority of model layouts. There is no " section " or " limits "...." "Starting signals are the most common found on a model as they control starting from a station ..." Sorry chaps, but no. Once again, it depends upon the railway in question, but if you look at the GWR or (to a lesser extent) the SR, then the terms 'Home' and 'Starter' - and the variations thereof - relate to the position of the signal relative to the controlling signal-box. So, in the example diagram 2 is a Home, even though in effect it controls 'starting' from the station. Most models do have 'station limits' which exist within the modelled area, with the start/end of the 'block sections' often - but not always - being within the visible modelled area. >>>It was being suggested that a shunting move could take place from the Up yard to the Down line as controlled by ground signal 10 whilst this train was held at 3. You said this would be OK BUT that signal 3 would be locked by that move, and would be unable to be released until the shunt move was complete (and points/signals reset). Now I can understand that 2 would be locked since any approaching train would otherwise run into the shunting move (or move off from the platform into the shunting move). But why 3? It would only be cleared when the next block was clear" Quite simple...one of the basic rules of interlocking would require that clearance of signal 3 would lock points 7 (and hence points 9, which would be released by 7) in which ever position they were at the time - either normal or reverse (known as 'both ways' locking) to ensure that the points are not moved under any train that moves off once 3 is cleared. Therefore, because mechanical interlocking is reciprocal, reversing 7 would lock 3. Furthermore, 3 and 10 would be required to lock each other because they control opposing moves - otherwise, if you could (accidentally) pull off both 3 and 10 at the same time, which way is the train supposed to go ??? :-) It's nothing at all to do with the state of the block section in advance of 3.

>>>As noted by other posters the home signals are placed as to protect the points and any potential movement over them. The starter signals are placed at the longest train length to allow for any shunting without encroaching into the next block section...... With respect, that sort of statement illustrates exactly the sort of problem that occurs when trying to give a 'simple' explanation. For example....many 'Home' signals do not protect points at all, but simply the 'signal section' between that stop signal and the next stop signal. Although there were indeed many stations without a (Home) stop signal in rear of the platform, as on the Up line in the example diagram, I would argue that it was a less common arrangement and therefore not the best illustration. There were many examples where Starting signals were NOT sufficiently far out to enable a train to shunt without going into the block section - hence the common use of Shunt Ahead arms by the GWR for example. What the diagram by itself, and even the accompanying comments, does not make clear - for example - is that you could not allow a Down train to approach No 12 if you had a train in the platform shunting the sidings (fouling the Clearing Point). You need to understand the basics of block working if you are going to operate the layout in any sort of accurate fashion. Sadly, one of the biggest problems that I encounter with layouts at exhibitions etc is where the builder has gone to a heritage railway for inspiration. He sees something and replicates it, unaware of why the prototype has done that and/or ignorant of the fact that what he sees is a modern interpretation of the rules using older style equipment. Sometimes he gets well-meant but misguided 'advice' from a volunteer there, who knows - or thinks he knows - what that signal does, but not why it is used as or where it is.