Is Signaling one of the most overlooked or at least intentionally simplified parts of the layout?

[OnTheBranchline]

I think it would make sense if it was because some people in a sense want to 'plop and play' - i.e. "here's the signal, that's the signaling done, now I can play trains".  Sure, the layout may have a signal, but do most people have the appropriated signal wires to go alongside?

 

In building my diorama, I've done some reading and there's a lot more to signaling than just installing a signal. But a lot of signaling equipment isn't RTR or in straight forward kit form so that's probably why people don't bother.

 

 

[BachelorBoy]

Probably. 

[Hibelroad]

Signalling is very much a neglected feature on many otherwise highly detailed layouts. Part of the problem is that the whole area is somewhat mysterious, partly due to the fact that there are few magazine articles or books explaining what it’s all about and how to model signals and associated equipment correctly. I’m am guilty of neglecting signalling mainly due to ignorance and would agree that it’s a subject deserving wider discussion. 

[Jeremy Cumberland]

Track layout comes higher up my own list than signalling, with facing points and double slips in abundance. I don't particularly mind signals being totally absent - my eyesight isn't good enough to see ground signals in 4mm at a normal viewing distance anyway, and few layouts require much in the way of main signals - often just a starter in each direction is enough. Operators can avoid conflicting moves whether they have got signalling or not, although correctly interlocked and working signalling - that drivers obey - does help reinforce correct operation.

 

But working signalling has a curse. One layout I saw recently featured a long stretch of track with meticulously-modelled signalling all operated correctly as far as I could tell. Sections were controlled by block bells (I don't know about instruments), and out of section wasn't given till trains were past the clearing point. From a signalling perspective, it was magnificent. But it meant waiting about 5 minutes between trains, which at an exhibition is too much, I think, unless you can contrive to have something going on away from the running lines. which this layout did not.

 

There is a balance to be struck, but there is surely more than one way of striking it.

[melmoth]

One consideration when modelling signalling is the distances involved. IIRC, the distance between a  distant signal and a home signal, for example, should exceed the braking distance of a train - which would/could be hundreds of yards. Given the compression most of us need to adopt to fit a layout into the space available, there might not be much signalling required, if modelled accurately.

[whart57]

None of us have room to do signalling justice. Just a simple Distant  - Home - Starter sequence needs about twenty foot of track in 00 and then there are the lengths of blocks. It is possible to build a layout which is fully signalled and operated as such but that requires severe compromises on pointwork (no long flowing points) and train length (hello, three coach express and six wagon heavy freight). Even then most of the signalling activity is small stuff you can barely see, like ground signals, or can't see at all like unlocking point locks.

 

My personal opinion is that unless you are really building a signalling demo layout, that the aim regards signalling is to make signal operation consistent with reality. No running through red lights, semaphores reset to danger after a train has passed. Where appropriate the use of calling on signals and the like adds interest. But pinging bells is rarely more than unwanted noise.

[Vecchio]

I think it is also difficult to make any interlocking between signals and points. I have signals in my layout Donnersbachkogel, there are 3 tracks in the station, they should be interlocked so only one (the one where also the points are right) should be able to show green. When you are at a show it is difficult to work them (ok, if you have some 100 feet of mainline you can use automatic block software, for instance the one from Megapoints controller. Now everybody knows what layout I mean). My layout fills a van but still distances between signals are rubbish. May be computer control could help, but then the feeling that you drive a train is gone - it is the computer doing it, and you are just struggling in getting the software right....

[DY444]

Simple answer: yes.  I've lost count of the number of otherwise excellent layouts where signals appear to have been put in random places as an after thought.  Many people get hugely exercised by the tiniest detail on locomotives etc and then run them on layouts where the "signalling" is unlike anything seen anywhere ever.

 

If you've got a roundy-roundy tail chaser with no fiddle yard then that's likely to be impossible to signal in any meaningful way unless it's enormous.  Otherwise my view is any layout which is FY to scenic and back to FY can be signalled properly and all the comments about not enough space to do it right fall away.  You've just got to consider the signalling as you design it rather than forget about it until you've built it. 

[cctransuk]

1 minute ago, DY444 said:

Simple answer: yes.  I've lost count of the number of otherwise excellent layouts where signals appear to have been put in random places as an after thought.  Many people get hugely exercised by the tiniest detail on locomotives etc and then run them on layouts where the "signalling" is unlike anything seen anywhere ever.

 

If you've got a roundy-roundy tail chaser with no fiddle yard then that's likely to be impossible to signal in any meaningful way unless it's enormous.  Otherwise my view is any layout which is FY to scenic and back to FY can be signalled properly and all the comments about not enough space to do it right fall away.  You've just got to consider the signalling as you design it rather than forget about it until you've built it. 

 

Every word true - but, and it's a big BUT - sadly, the vast majority of viewers, including myself, will be totally unaware of the deficiencies - or otherwise!

 

Signalling and operation are - and will remain - two of the least understood aspects of the real railway, past or present.

 

CJI.

[Flying Pig]

57 minutes ago, Jeremy C said:

But working signalling has a curse. One layout I saw recently featured a long stretch of track with meticulously-modelled signalling all operated correctly as far as I could tell. Sections were controlled by block bells (I don't know about instruments), and out of section wasn't given till trains were past the clearing point. From a signalling perspective, it was magnificent. But it meant waiting about 5 minutes between trains, which at an exhibition is too much, I think, unless you can contrive to have something going on away from the running lines. which this layout did not.

 

 

There are very few layouts indeed where it takes several minutes for a train to pass through a completely modelled section.  The vast majority represent station limits and a bit of track either side.  The gap between trains can be greatly compresssed, even with correct working of the signals - it's just a matter of when the next one is released from the fiddle yard after the first has disappeared.

[toby_tl10]

Lack of knowledge in signalling? For a straight piece of track, it's easy enough I think. In modern days, it goes red, single yellow, double yellow, and green. With semaphores, it's either stop or go, sometimes it has a caution.

 

But when it involves points, I have absolutely no idea what is happening, and I believe others might feel the same. What signal do I put here? Where do I put a ground signal? Do I put a signal gantry here?

[Phil Bullock]

 

It can be done! Here’s point rodding and signals on the new layout all built using MSE components from Wizard Models. Signal wires to follow once I have determined exact positions for all signals. 
 

Interlocking can also be achieved depending on your operating system … we are on DCC and programming routes in to ECoS means that all operations should be hassle free…. 
 

Operating individual points/signals isn’t an option … about 20 to get back to where you started in the fiddle yard, one is bound to get missed. We won’t be using bell codes other than the odd call attention ding for distracted operators. 
 

Also key to have a reset route to return everything to normal once a train had passed and before pegging the next one.

 

I don’t for one moment kid myself it’s perfect but It worked on the last layout as long as operators drive to the signals…. Failure to do so results in a form 1 requiring the transgressor to buy a round In The pub…. 

[The Stationmaster]

The biggest problems when it comes to signalling model railways areas follows - in descending order of influence -

1. Track layouts which are impossible to signal because they ignore the layout planning 'rules' which govern the real thing.  I frequently find when looking at questions about adding signals that it becomes too difficult purely because the track layout can't be signalled correctly.

 

2. A lack of understanding of what signalling is and how it works.  This needs a wider knowledge, still quite straightforward, about the purpose of signals and what they do beyond the simplistic 'put one here' approach.  Sometimes down to inadequate or even misguided research.

 

3. Limited understanding - to the extent it will help or be useful in operating a model of a railway rather than just running trains.  This also feeds back at the planning stage into Item 1 on this list.

 

4.  Critical once you get this far and know how to do it.  The difficulties, or perceived difficulties, of making the correct signals and making them work.

 

5.  A limited understanding of how you need to apply selective compression to tje planning of a model railway and how to relate that to signalling.  For example, already mentioned above, apart from lower arm examples an ordinary 'lone' distant signal in 4mm scale is going to be at least 17feet from the first stop signal to whch it applies although shorter distances could be found in Pre-Group days.   So unless. you are modellinga stratch of open country you don't need a distant signal on many model railways.

 

6. Limited or erroneous understanding of the subject from reading poorly researched books - and nowadays online articles etc - written by people who themselves didn't properly understand what they were writing about.

[DY444]

1 hour ago, toby_tl10 said:

Lack of knowledge in signalling? For a straight piece of track, it's easy enough I think. In modern days, it goes red, single yellow, double yellow, and green. With semaphores, it's either stop or go, sometimes it has a caution.

 

But when it involves points, I have absolutely no idea what is happening, and I believe others might feel the same. What signal do I put here? Where do I put a ground signal? Do I put a signal gantry here?

 

Then you have a choice.  Don't signal it at all, guess and get it wrong or do some research and ask questions.  It's not like there's a shortage of reference material.  At the end of the day it's up to you but imo no signalling is better than totally wrong signalling. 

 

It does seem to me though that "I don't understand signalling" is becoming some sort of badge of honour up there with "I don't do electrics" ( the two are obviously related!).  As an aside you rarely hear "I don't do scenery" or "I don't do wagons".  The lengths people go to with things like Class 47 boiler ports, which to be honest is a pretty niche specialism, knows no bounds and yet some of the same people will blithely plonk any old signal anywhere. 

 

They say the beauty of railway modelling is that it requires a collection of practical and artistic skills, an eye for detail and an understanding of the base subject matter.  I've never understood why for some reason that frequently doesn't stretch to cover electrics and signalling.  

Edited February 27, 2023 by DY444

[Legend]

Must admit my approach has been to stick signals where they look nice . I know ......cringe !

 

I understand the fundamentals and over the years did initially try to get it right . Distant and home junction signals etc , but given compressed distances on model railways often that looked ridiculous , so finally gave up any pretence 

[Nick C]

It's certainly something that needs to be considered at the planning stage - all too often there's a thread in this section - "Here's my layout, track all laid etc, where should I put signals" - which as @The Stationmaster says, is sometimes impossible! 

 

For most layouts, working ground signals are un-necessary, unless you're a particular signalling or operation afficionado, but it's fairly easy to get decent stop signals, at least for grouping era onwards - Dapol and Ratio both have suitable semaphore examples either RTR or kit - then MSE for earlier or more complicated stuff (though it'd be great if there was more information available as to what bits you need to create different kinds of signals).

 

Working colour-lights are even easier, with several makers, and Simon Paley's comprehensive book on the subject.

 

I can't claim to be an expert by any means, but as a signaller on a heritage railway, I want the signalling on my own layouts to be right - or at least as close as I can manage!

[t-b-g]

Signalling is, sadly in my view, either neglected or not considered important enough to warrant time on research and installation.

 

There is one very highly regarded layout, widely praised as a superb representation of a real place, that has the wrong types of signals and some of them in the wrong places.

 

I have told the layout owner/builder but he doesn't seem bothered enough to do anything about it. Of course modelling the signals at a real location should be easier than at a fictional one but even then, things changed over the years and to get things spot on you need lots of information correct for the period you are modelling.

 

I do like to see a layout with correct signals, in the right places, being used properly. They are rather rarer than perhaps they should be. I have been spending many hours working out signalling arrangements for a new layout and I have consulted a "real life" signalman as to whether certain shunt signals would be provided or whether the moves were likely to be flagged from a signal box. I am still looking for a nice clear photo of an LD&ECR pattern wooden post bracket signal. They are not easy to find but I would like to get it as accurate as I can. 

 

I have built quite a few signals for other layouts over the years and you can certainly tell whether a track layout represents prototype practice well or not from the number and complexity of the signal required. When a fairly small station requires a forest of signals to protect everything properly, there is a good chance that it doesn't really represent how the real railways would have done things. 

[Talltim]

My problem is that although I understand the basics of UK route based signalling, I just can’t get my head around US speed based signalling, especially as there seem to be so many variations. So much easier to model a ‘dark’ branch line

[Pete the Elaner]

58 minutes ago, The Stationmaster said:

The biggest problems when it comes to signalling model railways areas follows - in descending order of influence -

1. Track layouts which are impossible to signal because they ignore the layout planning 'rules' which govern the real thing.  I frequently find when looking at questions about adding signals that it becomes too difficult purely because the track layout can't be signalled correctly.

 

2. A lack of understanding of what signalling is and how it works.  This needs a wider knowledge, still quite straightforward, about the purpose of signals and what they do beyond the simplistic 'put one here' approach.  Sometimes down to inadequate or even misguided research.

 

3. Limited understanding - to the extent it will help or be useful in operating a model of a railway rather than just running trains.  This also feeds back at the planning stage into Item 1 on this list.

 

4.  Critical once you get this far and know how to do it.  The difficulties, or perceived difficulties, of making the correct signals and making them work.

 

5.  A limited understanding of how you need to apply selective compression to the planning of a model railway and how to relate that to signalling.  For example, already mentioned above, apart from lower arm examples an ordinary 'lone' distant signal in 4mm scale is going to be at least 17feet from the first stop signal to which it applies although shorter distances could be found in Pre-Group days.   So unless. you are modelling a stretch of open country you don't need a distant signal on many model railways.

 

6. Limited or erroneous understanding of the subject from reading poorly researched books - and nowadays online articles etc - written by people who themselves didn't properly understand what they were writing about.

 

Some very good points there.
Signalling seems to be a bit neglected because it is less well understood.

Many see no further than trains running on rails.
Signals are also often treated as just a part of the scenery, so frequently not operational.
It is easier to replicate signals for the few of us building scale replicas of real locations, but even then, they are less well photographed than rolling stock or station buildings. When we do see good photos of them, it is not always possible to see where they are placed in order to understand why they are placed there.

 

Many of the other surrounding structures are often mis-represented on layouts too: Stations without nameboards, poor access (passengers need a way to get there), bridges & tunnel portals the wrong shape, wrong brick bond used. I may sound a little pedantic now but some are just as easy to get right as wrong.

[Phil Bullock]

Unfortunately even with the 27ft of the new layout to play with space still forces compromises.

 

We have two boxes …. Abbotswood on the B&G and Norton Junction on the OWW. Signalling should be ok on those main lines although closer than on the prototype…. But the length of trains we want to operate will force compromise on the chord connecting the two. 
 

The biggest train will be a pair of 37s on 10 100 ton tank cars … this needs to traverse the chord to proceed via Worcester to Albion in Brum. 11ft long. I haven’t measured it yet but I suspect that if it gets checked at Norton Junctions home signal the rear of it will not be past Abbotswoods home and it may even foul the junction at Abbotswood. What do I do? Options I can see are :

 

1. Shorten the train so it fits between the two signals 

2. Omit one of the signals … Abbotswood starter 

3. Invent a local rule …. Either the train cannot leave Abbotswood until Norton has given line clear (and cleared all its signals) or if the end of the train stands foul inside Abbotswoods starter then no traffic can be passed over the junction to Birmingham …. Abbotswood cannot accept anything northbound from Pirton Sidings. 
 

I really don’t want to apply 1 as the prospect of a pair of 37s thrashing to restart their heavy train won’t look the same with fewer cars. Nor 2 as most trains will be clear of Abbotswoods starter if checked at Norton. Presumably @The Stationmaster Abbotswood could clear back to accept a train from Pirton Sidings to go to Birmingham as long as the train going up to Norton is in the section and clear of the starter? 
 

Complicated ? I did think about these issues when planning the layout but shortened distances don’t go away…

[russ p]

One other problem is lack of suitable signals . My layout is fully signalled but semaphores are only posable and not working.  Whilst it is possible to make ratio signals work they aren't very robust so they probably wouldn't last long. I admire people who like Phil can make MSE products to work,  I'm not great at soldering and they are very fiddly.  I've managed to make shunting discs and a banner repeater posable 

I don't know how or if it could be done but a modular signalling system possibly based on BR standard  equipment would be fantastic 

[Jeremy Cumberland]

2 hours ago, Legend said:

Must admit my approach has been to stick signals where they look nice . I know ......cringe !

 

I understand the fundamentals and over the years did initially try to get it right . Distant and home junction signals etc , but given compressed distances on model railways often that looked ridiculous , so finally gave up any pretence 

I think this is a perfectly valid approach for those who don't have the inclination to put much energy into signalling. What looks "nice" is often what we expect things to look like, and we expect things to look that way because that is what we have previously observed, perhaps unconsciously. I'll put a signal at the end of the platform because it looks right. I might not know what it does, but it seems to belong there.

 

It might not be a perfect representation of the prototype, but it isn't likely to be a long way off.

[Michael Hodgson]

2 hours ago, The Stationmaster said:

The biggest problems when it comes to signalling model railways areas follows - in descending order of influence -

1. Track layouts which are impossible to signal because they ignore the layout planning 'rules' which govern the real thing.  I frequently find when looking at questions about adding signals that it becomes too difficult purely because the track layout can't be signalled correctly.

 

2. A lack of understanding of what signalling is and how it works.  This needs a wider knowledge, still quite straightforward, about the purpose of signals and what they do beyond the simplistic 'put one here' approach.  Sometimes down to inadequate or even misguided research.

 

3. Limited understanding - to the extent it will help or be useful in operating a model of a railway rather than just running trains.  This also feeds back at the planning stage into Item 1 on this list.

 

 

Spot on, particularly the order of impact.

 

Most layouts are freelance, not based on some given location but a question of what will fit in the limited space we are all constrained by.  If you do model a real place, at least there is a good chance that you can find out from photos etc what signals there were, and even if no record is available, there were enough rules of thumb that meant you could have a plausible stab at it.  But when the layout is fictitious and if you don't understand the principles of signalling, it's hardly suprising that you treat signals as mere "scenery", the same as the bus on an overbridge, to be plonked wherever takes your fancy.

 

I suspect the technical challenge of getting semaphores to work is a big consideration for many modellers, who put it in the "too difficult" category; but I'd rather see a layout correctly signalled even if the signals don't work - the same as one accepts that the passengers on the platform don't board the train when it stops.

 

If you want to know more about signalling, this link should help give a basic understanding

https://signalbox.org/block-system/

 

4 hours ago, Flying Pig said:

 

There are very few layouts indeed where it takes several minutes for a train to pass through a completely modelled section.  The vast majority represent station limits and a bit of track either side.  The gap between trains can be greatly compresssed, even with correct working of the signals - it's just a matter of when the next one is released from the fiddle yard after the first has disappeared.

 

That's true, and even if plain track is modelled it's more heaviliy compressed that the station itself, which is why a speeded up clock is a good idea if you want to run a timetable.

 

 

5 hours ago, Jeremy C said:

 

But working signalling has a curse. One layout I saw recently featured a long stretch of track with meticulously-modelled signalling all operated correctly as far as I could tell. Sections were controlled by block bells (I don't know about instruments), and out of section wasn't given till trains were past the clearing point. From a signalling perspective, it was magnificent. But it meant waiting about 5 minutes between trains, which at an exhibition is too much, I think, unless you can contrive to have something going on away from the running lines. which this layout did not.

 

 

That's true - it's not just most modellers' ignorance of signalling practice, the general public has even less idea about signalling, and they haven't a clue about whether you're doing it right. 

 

The reason "trainspotters" tend to be viewed as boring nerds, is that if you stand at the lineside in most places for any length of time, nothing happens and you'll see very few trains.  An exhibition is not about realism and correct operating - the punters expect to see toy trains moving - and if not, they will soon walk away muttering very pretty but it doesn't work.  To be fair, commuters get just as bored standing on the platform grumbling about how long they've got to wait and why are the trains are so infrequent.

 

However the real railway works in a standardised way using a fleet of identical rolling stock on any one route - you will not see a procession of every class of loco the company ever made (even if you can tell GWR classes apart) especially at a branch line terminus.  To that end, just one train circling a roundy-roundy layout is actually more relaistic as the trains do all look the same !  But agin the punters want to see variety.  If you want realsim with more variety of rollling stock, you need to model a major main line which has a lot of different services just passing through. 

 

 

[OnTheBranchline]

7 hours ago, Phil Bullock said:

 

It can be done! Here’s point rodding and signals on the new layout all built using MSE components from Wizard Models. Signal wires to follow once I have determined exact positions for all signals. 
 

Interlocking can also be achieved depending on your operating system … we are on DCC and programming routes in to ECoS means that all operations should be hassle free…. 
 

Operating individual points/signals isn’t an option … about 20 to get back to where you started in the fiddle yard, one is bound to get missed. We won’t be using bell codes other than the odd call attention ding for distracted operators. 
 

Also key to have a reset route to return everything to normal once a train had passed and before pegging the next one.

 

I don’t for one moment kid myself it’s perfect but It worked on the last layout as long as operators drive to the signals…. Failure to do so results in a form 1 requiring the transgressor to buy a round In The pub…. 

 

Where do you get the signal ground wire holders and what do you use for signal wire?

[Phil Bullock]

26 minutes ago, OnTheBranchline said:

 

Where do you get the signal ground wire holders and what do you use for signal wire?

Planning to use Wizard models signal wire and Signal wire posts … several versions of the latter available. Plenty of pulleys for 90 degree turns on the point crank etched.

 

Will post more pictures once done , may be a while.