Signalling Settings In Advance?

[BR Class 1.618]

Hello all!

I have a question:

How far ahead of time are signals set by the signalmen? For example, if there is a train due to enter a signalman's "patch" at 1537 and the time is currently 1515, and there aren't any other trains due for a while. Would the signalman most likely set the route at that time, or would he wait until later?

My immediate thought is: to minimise the time that the train spends to slow down at a signal, set the route immediately; however, I have heard...not sure as to the validity of this, but...that the signalman will not set the route until the last minute, because he wants to leave his signals red for safety...? Again, I'm not entirely sure that this has any basis at all.

Thanks.

 

[Llanigraham]

Absolute Block - signals and route are set when the train is offered by the preceding Box (generally) but there can be many local variations.
For example:
At Bromfield on the DOWN I will drop the barriers and set the signals when I get "the approach" from Onibury, who give it when they get "entering section" from Craven Arms, and on the UP I will drop the barriers and set the signals when the annunciator sounds as the train leaves Ludlow Station. Both are about 2 minutes.

 

[causton]

Another reason for not clearing until the last minute is wondering whether the next train is actually going to turn up on time and in the order you expect!

If the train gets to the preceding station at 1530 and then reports a fault so does not make it there, or another train runs early and overtakes it, or another train running late in the other direction (or early) has a conflicting route with the one you have just set, it is wasting a bit of time. Don't want to set a route all the way along for this train and then an early running freight train pops up! Especially if it has a different route, you don't want another train to turn up and accidentally route it the wrong way cause the route was set for the first train!

 

[edwin_m]

Automatic route setting systems are normally configured to keep two green signals ahead of a train, unless there is a conflict ahead where a different train is given priority. Human signallers will in general clear signals further ahead than that, if they are pretty sure a train approaching a junction will have a clear run through, so that the train can keep running while they are looking after something else.

 

[Crossover]

I was thinking about this only the other day whilst at Brough awaiting a Leeds bound service - the semaphore only cleared as the train approached. Although not high time for it, I did wonder if for winter there may be a safety aspect of not risking the signal freezing, or otherwise becoming stuck, on a proceed aspect and thus not setting the route ages before? Is there any grounding in my thought?

 

[BR Class 1.618]

I suspect that that is a prime concern with a semaphore style signalling system. On a multi-aspect colour system, I doubt it...

 

[carriageline]

If it's maintained properly it should never stick! It's weighted so should fall back to danger


Sent from my iPhone using Tapatalk

 

[Tomnick]

I think it's clear by now that there's no single answer to the question! Indeed, the method of working varies greatly between different areas of the one box that I'm working today. At one end of the panel (TCB), I've had a route set for a good 15 minutes before the train's expected to step into the train describer - I can see that the adjacent box has already put the route in, and I'd have to set the same route even if a different train subsequently turns up! It's easy enough to pull a route out if necessary, in most cases, if the order of trains does change - it's equally easy to forget to set a route if you leave it until the last minute, then get distracted by something else more important.

At the other end of the panel (mainly AB), there's a level crossing to consider - lower the barriers and pull off (right through) for Down trains upon receipt of TES from the box in rear; pull off as far as the section signal for Up trains upon receipt of TES, but only lower the barriers and clear the section signal about a minute after the train's actually passed the box (maybe someone can spot the potential for an easy oopsie there ). Elsewhere on the patch, a couple of boxes also use the 'train approaching' bell signal (good to hear that it's still in proper use elsewhere too!!), whilst others have to wait two or three minutes from receipt of TES (or an annunciator, in one case) before doing anything.

Absolute Block - signals and route are set when the train is offered by the preceding Box (generally) but there can be many local variations.

Offering forward and pulling off upon receipt of TES from the box in rear is the 'standard' mode of signalling, not when the train is offered - is this what you meant?

 

[High Dyke]

At the other end of the panel (mainly AB), there's a level crossing to consider - lower the barriers and pull off (right through) for Down trains upon receipt of TES from the box in rear; pull off as far as the section signal for Up trains upon receipt of TES, but only lower the barriers and clear the section signal about a minute after the train's actually passed the box (maybe someone can spot the potential for an easy oopsie there )

Maybe...but not for open discussion on here.

As Tomnick says though there is no simple answer to this question due to variants in signalling operation.

Today i'm working a 'fringe' box - i.e. works Track Circuit Block (TCB) one way and Absolute Block (AB) the opposite side. I too have a level crossing to take into consideration - one situated on a busy 'A' road, and a station. For trains approaching from the AB side it takes four minutes from Train Entering Section (TES) to arriving - most signallers that work here wait about 1 minute before lowering the barriers and clearing signals.

For trains approaching from the TCB side it gets a bit more complicated. Any approaching train also affects the ECML - that's hints where i am. Train annunciation is received and i can pull the lever for my first controlled signal; however the signal works on approach control, so for a stopping train i can leave it to clear as it is intended, but for a non-stopping train i also need to lower the barriers and clear the section signal (it protects the level crossing) to permit the first signal to clear straightaway. The challenge we face here is that you don't know for certain when the train will cross the ECML, and yes i've been caught out before by operating the level crossing and clearing signals about 10 minutes before the train actually passes (not popular with motorists...).

 

[edwin_m]

I was thinking about this only the other day whilst at Brough awaiting a Leeds bound service - the semaphore only cleared as the train approached. Although not high time for it, I did wonder if for winter there may be a safety aspect of not risking the signal freezing, or otherwise becoming stuck, on a proceed aspect and thus not setting the route ages before? Is there any grounding in my thought?

Originally semaphore signals were left at clear except when a train had recently passed, but that changed in Victorian times following an accident at Abbots Ripton where the signals froze in the clear position in a blizzard - not helped by being slotted post signals where the arm just sat in the slot when clear.

Since then they have only been cleared when a train is expected, and signalmen are expected to check that the arm falls back to danger when they replace the signal lever, either visually or using electric arm repeaters. It's still possible for a semaphore to freeze "off" and there is an accident at Chinley on Railways Archive from this cause.

Seeing semaphores clearing with the train close to them is often to do with the rules on what to do if a train arrives but hasn't been accepted by the next box. Most boxes control more than one stop signal on each line, for example there will often be separate signals for arrival and departure at a station. However if the train hasn't been accepted all of these signals must be kept at danger, each one in turn being cleared once the signalman is satisfied the train is running at a safe low speed, until the last one where the train will have to wait to be accepted. This reduces the risk that the driver will see the first signal(s) clear ahead, and accelerate away forgetting that the last signal might be at danger.

 

[Tomnick]

Maybe...but not for open discussion on here.

All legitimate, don't worry! TES isn't given when the train passes the box (as you won't normally have cleared the section signal by then), but rather at a later point specified in the SBSIs. It takes a definite effort not to send it when the train passes the box though!

Train annunciation is received

...

The challenge we face here is that you don't know for certain when the train will cross the ECML, and yes i've been caught out before by operating the level crossing and clearing signals about 10 minutes before the train actually passes (not popular with motorists...).

Do you get the annunciator even with no route set across the ECML then? We've got a similar situation a little way further south, albeit with no reason for trains to be held at the last signal at the box in rear - here, the description won't step into our 'APPR' berth until it's occupying the berth at that last signal and the signal is cleared for the appropriate route.

 

[High Dyke]

All legitimate, don't worry! TES isn't given when the train passes the box (as you won't normally have cleared the section signal by then), but rather at a later point specified in the SBSIs. It takes a definite effort not to send it when the train passes the box though!

No worries. I know how excited the senior ranks get about discussion forums... Back to differing ways of working. At this location you transmit the description upon receipt of TES from the box in rear. A different location, since altered, you used to transmit the description when the train passed the box. Nowadays at the location we usually enter the description into the system upon TES, but it only steps over and annunciates at the neighbouring PSB when it passes the box.

Do you get the annunciator even with no route set across the ECML then? We've got a similar situation a little way further south, albeit with no reason for trains to be held at the last signal at the box in rear - here, the description won't step into our 'APPR' berth until it's occupying the berth at that last signal and the signal is cleared for the appropriate route.

You can do. Sometimes trains appear on the diagram / in the train describer at the same time, but others you get the TD annunciation in the 'APPR' berth before it appears on the box diagram - which has varying degrees of time.

 

[Tomnick]

No worries. I know how excited the senior ranks get about discussion forums... Back to differing ways of working. At this location you transmit the description upon receipt of TES from the box in rear. A different location, since altered, you used to transmit the description when the train passed the box. Nowadays at the location we usually enter the description into the system upon TES, but it only steps over and annunciates at the neighbouring PSB when it passes the box.

Interesting stuff. Here, we have to send the description (to their 'APPR' berth, if I'm not mistaken) in order to request the slot over the single line, then it steps from our signal berths (we have a small train describer installation at that end of our panel) as normal when the train passes our last signal. At another box, with a more conventional fringe to the adjacent powerbox, we put it in the 'APPR' berth at some point (usually when the signals have been cleared, but that's not specified) and it steps from there into their first signal berth when it passes the home signal. Variety is the spice of life!

You can do. Sometimes trains appear on the diagram / in the train describer at the same time, but others you get the TD annunciation in the 'APPR' berth before it appears on the box diagram - which has varying degrees of time.

Sounds typically inconsistent! At least we know where we are with ours. It's not uncommon for a second train to annunciate immediately as the first clears the overlap on the first signal, as the route's already in for the second one and it's already approaching their last signal, so it steps as soon as the signal clears. Do they annunciate to you for the other route, round the anonymous curve, beyond the PSB's last signal too then?

 

[High Dyke]

Sounds typically inconsistent! At least we know where we are with ours. It's not uncommon for a second train to annunciate immediately as the first clears the overlap on the first signal, as the route's already in for the second one and it's already approaching their last signal, so it steps as soon as the signal clears. Do they annunciate to you for the other route, round the anonymous curve, beyond the PSB's last signal too then?

That happens here. I can have one stood in the station on the Up road and the PSB can signal another train over the Flat Xing towards me. No we don't get annunciations for the ECML.

It gets more complicated for station working on the Down line. There is a short t/c from my stop signal through the level crossing and then one long one t/c from there (at my end of the platform) towards the signal protecting the Flat Xing. For stopping trains we have an inhibit switch that berths the description on our berth (replicated in the PSB), but won't step over until we activate the switch. Some trains are booked to stop for 6 minutes, but may be even longer.

 

[Marklund]

If it's maintained properly it should never stick! It's weighted so should fall back to danger

Nope, and nope.

Rain followed by a heavy frost often causes problems.
You can have as much special kilfrost grease in the spindle, but if the whole of the mechanism at the top is soaked, it can stick off if the box has been switched out overnight.

And not all signals have balance weights due to the length of the signal wire run.

 

[carriageline]

Nope, and nope.



Rain followed by a heavy frost often causes problems.

You can have as much special kilfrost grease in the spindle, but if the whole of the mechanism at the top is soaked, it can stick off if the box has been switched out overnight.



And not all signals have balance weights due to the length of the signal wire run.

And you learn something new everyday!


Sent from my iPhone using Tapatalk

 

[Marklund]

And you learn something new everyday!

The signals I had (they're long gone now) without weights were peculiar.
You could maybe expect the one less than 10 yards from the box not to have one, but the other was about 150 yards away. :-?

 

[BR Class 1.618]

To be quite honest, anything that possibly can go wrong, most likely will go wrong, unfortunately.

As for the signal that was 150 yards away... that I cannot reason with. :?

(I may not have experience in the world of rail, which I would like to gain, but I do have decent experience in the world of Engineering--not with locomotives, though. )

 

[Railsigns]

It ought to be clarified that the purpose of the weight on a semaphore signal is to pull the wire back when the controlling lever is returned to the normal position the frame, and not to return the signal arm to 'danger' (which should, in any case, return to 'danger' by gravity alone).

If a standard upper quadrant signal arm is stuck up in the 'clear' position, it doesn't matter how heavy the weight is, it won't help return the arm to 'danger'. Lower quadrant signals are a different matter, because there is a solid rod linking the weight lever to the arm.

 

[nottinghamcity]

One way of getting a view of the variables that have to be considered is to have a go with simsig, you quite quickly work out where to and how far to set routes to, and the impact of overdoing it!

 

[BR Class 1.618]

One way of getting a view of the variables that have to be considered is to have a go with simsig, you quite quickly work out where to and how far to set routes to, and the impact of overdoing it!

Surprisingly enough, that is where my question originated. I have no real experience signalling, so my knowledge of the subject, and, indeed, some of the replies here, is not great (my understanding of them, not the actual replies, which are wonderful). I am slowly becoming more knowledgeable as I 'play' the simulator, but I cannot help feeling that there are huge bits that one cannot experience in the simulator that happen in the real world, which is inevitable.

 

[High Dyke]

I am slowly becoming more knowledgeable as I 'play' the simulator, but I cannot help feeling that there are huge bits that one cannot experience in the simulator that happen in the real world, which is inevitable.

Which...oddly enough seems to be the way Network Rail are conducting their selection process. I understand that they are now asking about 'Non-Technical Skills' as they've realised that experience can never be replicated in a question and answer session or an online recruitment assessment. However they are now more inclined to assess you on how you think and not what experience you have.

 

[BR Class 1.618]

However they are now more inclined to assess you on how you think and not what experience you have.

I think that seems reasonable. I reserve, though, that the best way to learn about a craft, or a skill, is to be an apprentice, and learn under your "master." Unfortunately, it doesn't seem like anyone has time these days, and, admittedly, it's not necessary in many cases, but there are obvious advantages to learning from an expert.

Oh well.