Upcoming Games

No games to display

Full list
Add a game

Upcoming Events

No events to display

No overlap at Parcels platform

You are here: Home > Forum > Simulations > Released > Bristol > No overlap at Parcels platform

Page 1 of 2

No overlap at Parcels platform 02/12/2011 at 12:25 #24283
maxand
Avatar
1637 posts
My task is to set a route through Bristol Temple Meads for 1C801, currently inbound on the Down Main and due to stop at the Parcels platform before heading out to Worle Jn.

As it's a passenger train, it seems natural to set the route from 20 to 30 direct, like this:



This sets the route but now I can't set a route between 30 and 32, getting the message "No overlap available".

After much angst, I discovered that setting the route 20-558-574-30 also gets 1C801 to the Parcels platform, but now it becomes possible to set a further route from 30 to 32 and onwards.



Can anyone explain why the first method doesn't work but the second does? I would have thought the subsidiary shunt signals on 20 and 30 would have obviated the need to use 558 and 574. Thanks.

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 12:35 #24284
headshot119
Avatar
4869 posts
Have you tried swinging the points in advance of signal 30 reverse, then try setting from 20 to 30.

Can't give anymore help as I'm not at home.

"Passengers for New Lane, should be seated in the rear coach of the train " - Opinions are my own and not those of my employer
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 12:40 #24285
Late Turn
Avatar
696 posts
Max,

The main route into the parcels platform (i.e. the route from 20 to 30) is indeed the one that you should set. The only overlap that appears to be available for that route is straight ahead into the short spur (whether that was correct for the layout at the time, I don't know) - you'll see this if you set the route from 30 to 32 first and then attempt to set the route from 20 to 30. In other locations, you'll find that the overlap can 'swing' to an alternative overlap if one is available - try setting a route from 154 signal (on the Down Filton Main approaching Bristol East) to 24 signal, and then (assuming the overlap does run straight ahead towards plt 7 rather than across the road to the Down Through!) either set a route onwards from 24 to 30 or try a move that conflicts with the overlap such as 45 (plt 7) to 57. Since an alternative overlap is available in this case, you'll see it swing across to that alternative overlap.

When you try and set the successive shunt routes (using the intermediate shunt signals), you correctly note that there is no overlap beyond 30 signal - this is quite common for shunt routes, as the risk from a low-speed, non-passenger movement overrunning is obviously lower than that from an arriving passenger train full of punters! This is part of the reason why passenger trains are not normally allowed to proceed on the authority of a shunt signal.

Hope that helps!

Tom

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 12:42 #24286
maxand
Avatar
1637 posts
I think I may have answered my own question.

Choosing the first route, between 20 and 30, when the train arrives, the overlap disappears (is released) as soon as it stops:



At this point it's possible to set the next route from 30 to 32 and onwards.



However, if one looks at the first pic in this series one sees that the points between 30 and 32 were locked normal by the route from 20 to 30. In that case, shouldn't I have got the message "Points locked normal by another route" when trying to set 30-32, instead of "No overlap available"?

Never a dull moment with this sim.

Last edited: 02/12/2011 at 12:48 by maxand
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 12:47 #24287
maxand
Avatar
1637 posts
Tom, thanks for your ultra-quick response which arrived as I was doing my second one. :)

Good point about less overlap needed for shunts. Must try out that "swing" overlap you talk about. So the answer here seems to be to use 20-30 (particularly as it's a passenger, not a freight, train) and WAIT for the overlap to release after it stops moving. Thanks for your help.

(Added) Thanks Headshot, missed your post too. Just tried out your suggestion:



As you can see the points were manually swung and locked. Unfortunately (you guessed it!) when I try setting 20-30 I now get "Points locked in reverse by another route". So this means that the overlap from 20-30 has to be along the "normal" direction (of the points), not the "reverse". Which makes sense - an overshooting train runs a much greater risk of derailment if being forced to diverge rather than run straight ahead.

Now this makes me stop and think. The route from 20 to 30 is by no means straight; it passes through two sets of points (each with its own shunt signal). Therefore, whether the set route is 20-30 or 20-558-574-30, its speed on entering the Parcels platform would have to be the same - I'd say 15mph going through the shunts. So why was it necessary to specify such a long overlay for the Parcels platform using controlled signals (20 and 30) if the only access to the platform is via a shunt?

Last edited: 02/12/2011 at 13:15 by maxand
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 12:51 #24288
Forest Pines
Avatar
525 posts
Routes between running signals on passenger lines have an "overlap" which extends beyond the signal itself. On your first screenshot, for example, the route set into Platform 12 extends past the end of the platform and onto the points, and the route set into the parcels platform extends past the signal and into the short dead-end neck.

Sometimes, where there are facing points in the overlap, the overlap can be "swung" - in other words, you can move the points even when they have an overlap route set over them, and the route of the overlap will change. However, some overlaps can't be swung. This is, I think, what is happening here: as long as the overlap route past 30 is set, you can't move the points.

(some other overlaps can be swung unless the train is just approaching, but that isn't the situation here)

The solution to this is: overlaps with facing points in them do have a timeout. Set the route from 20 to 30, and wait for the train to arrive. After the platform track has been occupied for the timeout period, the overlap route will drop - the track beyond 30 will change back to grey - and you can set the route onwards.

(if the overlap can be swung unless the train is approaching, the timeout still happens but the overlap might not appear to drop - in some locations you have to keep retrying the route to see if it's allowed yet).

Like many things, this is location-dependant, and I don't have Bristol on this PC, so apologies if the above is slightly wrong. Routes from shunt signals don't have overlaps, which is why your solution worked (neither do routes on freight-only lines, incidentally, or some routes in low-speed areas). If you want to see where the overlaps are, go to the display options and tick "Show track circuit breaks" - as well as showing the track circuit breaks, this will also show most of the overlap extents, apart from overlaps that end on "diagonal" stretches of track, with a symbol rather like the one for buffer stops.

Edit: hah, while I was writing all that you discovered the answer!

Last edited: 02/12/2011 at 12:53 by Forest Pines
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 13:12 #24290
maxand
Avatar
1637 posts
Thanks Forest Pines - our posts crossed. Seems we're both agreed on what to do here.

I don't know how the penalty system works here, but since the train is expected to make the same speed to the platform no matter which waypoints are used to set the routes, I would hope no extra penalty is incurred on setting it through the shunt signals.

Even though we are told not to use shunt signals to set routes for passenger trains, I can't see that this would make any difference from getting a train from 20 to 30.

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 13:22 #24292
Late Turn
Avatar
696 posts
I suppose the 'penalty' should be that the Driver phones in and queries the signal at 20 (I'm not sure what route indication he'd get, but it wouldn't be a platform number as the route's not initially set into a platform!). I do wonder whether there are shunt routes from the main signals approaching the station in reality - they don't seem to serve any purpose in reality (why would you want to run an arriving train up to one of these shunt signals?).
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 13:27 #24293
maxand
Avatar
1637 posts
Even though we are told not to use shunt signals to set routes for passenger trains, I can't see that this would make any difference from getting a train from 20 to 30.

From the driver's point of view, all he sees are the shunt signals 558 and 574 telling him which way to go. He doesn't know which signals were clicked to set the route, so all he can do is abide by the speed (15mph) till he reaches signal 30. Therefore there doesn't seem to be any logic in making the overlap at 30 longer if it were set from 20 than if it were set from 574. Therefore, might this overlap discrepancy be seen as a rule built into the SimSig core code and made to apply inflexibly to all routes set between controlled signals? :)

Late Turn said:
Quote:
why would you want to run an arriving train up to one of these shunt signals?

To take advantage of the shorter overlap length. Only joking.

Some sims (can't remember which) seem to actually prevent me from setting routes using shunt signals which happen to lie between two controlled signals, or between a controlled and an automatic signal. Can't see the logic here, though I'd expect there to be a reason for this.

Last edited: 02/12/2011 at 13:42 by maxand
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 13:48 #24295
Firefly
Avatar
521 posts
Two things.

Firstly I don't believe the parcels platform should be used for passenger trains. It's effectively a goods line. Looking at your pictures (I'm not near my computer so can't check the sim), but looking at your photo it looks like you can only exit the platform on a shunt signal, not a move you can do with a passenger train.

Quote:
Even though we are told not to use shunt signals to set routes for passenger trains, I can't see that this would make any difference from getting a train from 20 to 30.
Shunt routes cannot be used for passenger moves because they don't have overlaps and they often don't require track circuits to be clear.

I suspect the parcels siding was signalled for parcel trains terminating from London, therefore no need to provide a swinging overlap at Country end.

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 13:53 #24296
Late Turn
Avatar
696 posts
Just a quick reply for now.

Max - the big difference (at the moment that the approaching train potentially overruns signal 30) is that a train signalled via the shunt route(s) shouldn't be a passenger train, so the consequences of an overrun should be less severe.

Firefly - I've just quickly checked, and yes, signal 30 appears to be fixed at red with only a shunt route out towards the south.

Tom

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 15:28 #24299
Firefly
Avatar
521 posts
Quote:
Firefly - I've just quickly checked, and yes, signal 30 appears to be fixed at red with only a shunt route out towards the south.
Thought as much. It was probably only ever used to allow the engine to run-round.

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 15:43 #24300
Firefly
Avatar
521 posts
Quote:
Now this makes me stop and think. The route from 20 to 30 is by no means straight; it passes through two sets of points (each with its own shunt signal). Therefore, whether the set route is 20-30 or 20-558-574-30, its speed on entering the Parcels platform would have to be the same - I'd say 15mph going through the shunts. So why was it necessary to specify such a long overlay for the Parcels platform using controlled signals (20 and 30) if the only access to the platform is via a shunt?
Afraid Not. When a Main Route "presets" a shunt signal the 15mph rule does not apply. Obviously in this example the speed will be slow because of the point work, however you could have a preset shunt on a piece of 125mph railway and clearly you wouldn't expect an express train to slow to 15mph just to pass it.

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 16:32 #24302
clive
Avatar
2738 posts
" said:
From the driver's point of view, all he sees are the shunt signals 558 and 574 telling him which way to go. He doesn't know which signals were clicked to set the route,
Not so. If 20-30 was clicked then he gets a yellow. If 20-558-574-30 was clicked, he gets a red plus subsidiary. They have different meanings.

Quote:
Therefore, might this overlap discrepancy be seen as a rule built into the SimSig core code and made to apply inflexibly to all routes set between controlled signals? :)
I don't know what's happening in this particular case, but overlaps are configured individually when you write the sim. There might not even be an overlap in some cases.

Quote:
Some sims (can't remember which) seem to actually prevent me from setting routes using shunt signals which happen to lie between two controlled signals, or between a controlled and an automatic signal. Can't see the logic here, though I'd expect there to be a reason for this.
The reason is probably because that how it is in real life.

For example, in Euston there's two shunt signals on each route between the last inbound signals and the platforms. You can't set a route from the running signals to either of those shunt signals, because they were provided for moving engines between platforms. An inbound train should be going to a platform, not stopping in the middle of the station throat!

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 16:53 #24305
Firefly
Avatar
521 posts
Quote:
Some sims (can't remember which) seem to actually prevent me from setting routes using shunt signals which happen to lie between two controlled signals, or between a controlled and an automatic signal. Can't see the logic here, though I'd expect there to be a reason for this.
To see the logic you have to question and work out why the signals are there in the first place.

Signals are placed specifically in accordance with the operators requirements.

Take Newton Abbot on the Exeter sim as an example.



Signal 9 has a Main Class Route, a Warner Class Route and a Call On Class Route.

It does not have a shunt class route, therefore you cannot route from 9 to 677. (Any move that ends at a shunt signal has to be a Shunt Class route)

Why doesn't it have a shunt class route?

Basically there is no point in providing a shunt class route from No.9. Firstly any train standing at signal 9 must have come from Totnes so would not be a shunting move. But primarily there's no reason to provide a shunt route because you can get to every Platform on a Main, Warner or Call-On and there's no reason for any train routing from 9 signal to stop at 677 signal. What would be the point?

677 signal itself is only there to allow turn back moves from the platforms.

Therefore a train or engine can come out of one platform, stand behind 677 shunt signal and then route back into a different platform.

679 is there so that the same turn back moves can be achieved without blocking the main line.


Bristol on the other hand gains lots of flexibility by allowing shunt moves from the main signals. I've drawn some of the possible options. Blue is a train entering the picture and yellow is a train going away.




Make sense?

Last edited: 02/12/2011 at 17:00 by Firefly
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 17:09 #24307
Steamer
Avatar
3922 posts
A good reason why passenger trains should not be routed using shunt signals is because they are generally less visible than main signals, and also don't have any train protection devices (AWS, TPWS etc.) fitted, so there is more chance of the train passing at danger. Also, since the train driver has to assume each shunt signal is at danger until he sees it clear (which may only be when he's very close to it), he can't drive very fast, whereas if he gets a single yellow aspect, he knows the line is clear to the end of the platform.

Having said that, the train in question won't be passenger if it's using the parcels siding (for some reason, they almost always run as Class 1 instead of Class 3), however it's good practice to route main signal to main signal unless the move specifically requires main signal to shunt signal to sunt signal to main signal.

"Don't stress/ relax/ let life roll off your backs./ Except for death and paying taxes/ everything in life.../ is only for now." (Avenue Q)
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 17:40 #24312
ralphjwchadkirk
Avatar
275 posts
The reason you mustn't set shunt routes for trains is simply that a cleared GPL or subsidiary does not guarantee that the track ahead is unoccupied. You cannot run a passenger train under those conditions, so we use main aspects for passenger trains (and they should be used wherever possible anyway) because the clearance of a main aspect does come with the guarantee that the track ahead is clear. The only time you can signal a passenger train with subsidiary signals is when permissive working is authorised.

Module TS 4.11:
Quote:
You must always use the main aspect when there is an associated subsidiary or position-light signal, except when:


  • shunting

  • attaching or detaching

  • the line is occupied and permissive working is authorised for that line and the type of train concerned

  • it is necessary during single line working or working by pilotman

  • the main aspect is locked at danger, as shown in section 6.3 of module T1B Movement of trains during failure of, or when working on, signalling equipment.



Last edited: 02/12/2011 at 17:46 by ralphjwchadkirk
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 21:13 #24317
Forest Pines
Avatar
525 posts
" said:
I do wonder whether there are shunt routes from the main signals approaching the station in reality - they don't seem to serve any purpose in reality (why would you want to run an arriving train up to one of these shunt signals?).
According to the signalling notice issued when the box was commissioned, those routes did exist originally - see http://www.signallingnotices.org.uk/scans/49/1969%20-%20S%202639%20Britol%20TM.pdf

I am slightly puzzled by one aspect of that diagram - it seems to show that for each of the signals at that location, a subsidiary clear aspect with no route indication is a shunt route set to the next shunt signal - 20 to 558 for example. However, this also applies to signal 26 - it lists it as having routes to 560, 562 and 570, all of them with no route indication given. Unless there's something I'm missing somewhere.

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 21:35 #24318
postal
Avatar
5190 posts
" said:
Having said that, the train in question won't be passenger if it's using the parcels siding (for some reason, they almost always run as Class 1 instead of Class 3).
Most of the trains like that were trains running for Mail traffic. Because of the requirements of the Royal Mail indoor operation, nearly every mail train is carrying time critical traffic at some part of its journey so time keeping throughout the journey is important. For that reason, in the late 1980's, Royal Mail negotiated a premium with the railway to have nearly all of their trains run as Class 1 (and be regulated accordingly).

“In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe
Log in to reply
The following user said thank you: Steamer
Re: No overlap at Parcels platform 02/12/2011 at 21:52 #24322
Firefly
Avatar
521 posts
Quote:
I am slightly puzzled by one aspect of that diagram - it seems to show that for each of the signals at that location, a subsidiary clear aspect with no route indication is a shunt route set to the next shunt signal - 20 to 558 for example. However, this also applies to signal 26 - it lists it as having routes to 560, 562 and 570, all of them with no route indication given. Unless there's something I'm missing somewhere.
It was never a requirement to provide a route indication for shunt signals. The idea was that the train would be going slow enough to see which direction the points are set.

Route indicators on shunt signals is a modern thing (80's era and onwards)

FF

Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 22:03 #24323
Forest Pines
Avatar
525 posts
I think what puzzled me mostly is the inconsistancy - there are some shunt routes which do have indications, such as from 28 into the goods yard.
Log in to reply
Re: No overlap at Parcels platform 02/12/2011 at 22:35 #24328
Firefly
Avatar
521 posts
Same with routes to the in road,

They obviously provided a route indication for all destinations from a signal, but no indication for the intermediate shunts.

Log in to reply
Re: No overlap at Parcels platform 03/12/2011 at 14:03 #24349
maxand
Avatar
1637 posts
Thanks everyone for your comments and help. I had no idea this could get so complicated.

ralphjwchadkirk wrote:
Quote:
The reason you mustn't set shunt routes for trains is simply that a cleared GPL or subsidiary does not guarantee that the track ahead is unoccupied.
I understand the point you are trying to make, but what you say seems to be at variance with the fact that in some sims I'm sure I can set shunt routes. Were you referring specifically to passenger trains? Did you mean "from GPL to GPL"?

Had to look up that GPL acronym. For those like me who have not encountered this term it stands for Ground Position Light (signal), i.e., the standard light signal (as opposed to semaphore signal) used for shunting. Good example here in Wikipedia. (Must add to Wiki glossary)

What comes across to me from reading all your comments, particularly Firefly's post #15, is that sims might have inbuilt rules (not visible to us) preventing some specific routes from being set, which would explain my inability to set routes even when following guidelines in the Wiki and sim manual. In other words, it may come down to trial and error to find a route from A to B that SimSig will accept, a bit like a Skinner box for trainee signallers. Even though the necessary waypoints are specified in the TT this information can be very terse. Another way to put it is that without us realizing it, a sim may insist that we accomplish a move in a certain way only. Those with a good grasp of the finer details of signalling would not have any problems here, but for the rest of us it could be very challenging (to put it politely).

The amount of in-depth knowledge required to follow protocol in setting routes appears to be quite staggering. You guys have given me a lot to re-read, which I don't mind.

Some of you have said that I shouldn't be routing a Class 1 passenger train through a Parcels platform. If you check 1C801's timetable, you will see it specifically mentions the PCL platform at Bristol Temple Meads.

While on the subject of shunt signals, I stumbled across this interesting RSSB 2002 document - Shunting Signals - Special Topic Report . Probably no surprise to you pros, but others might like to check it out.

Log in to reply
Re: No overlap at Parcels platform 03/12/2011 at 14:13 #24351
maxand
Avatar
1637 posts
It just occurred to me that if SimSig aims to simulate real IECC panels designed for fully trained signallers, the number of error messages they are likely to be presented with should be far fewer than those we beginners see, hence this might explain the relative lack of error messages compared with those seen in "games" that need to cater for rank beginners in order to sell.

Perhaps there's a need for more detailed explanation of error messages, equivalent to "spoon-feeding". I realize this would mean a lot of extra programming for each sim (unless they were part of the core code), but maybe the answer is as here, with elaboration of sticky points in the forum accompanied by suitable walk-throughs.

Log in to reply
Re: No overlap at Parcels platform 03/12/2011 at 14:37 #24357
clive
Avatar
2738 posts
" said:
It was never a requirement to provide a route indication for shunt signals. The idea was that the train would be going slow enough to see which direction the points are set.
Right. The original convention was that shunt indications from a running signal normally didn't carry a route, while calling-on aspects did carry one even if there was only one possible route. Routes were only included on shunt aspects when it was necessary to avoid confusion.

So, for example, if a signal read into a yard or into a single running platform ahead, "catseyes" on their own read into the yard and catseyes with a number was calling-on into the platform.

Log in to reply
The following user said thank you: maxand