Upcoming Games

(UTC times)


Full list
Add a game

Upcoming Events

No events to display

Who's Online

Person82, waucott (2 users seen recently)

Kemble Reversing Bug

You are here: Home > Forum > Simulations > Released > Swindon A and B > Kemble Reversing Bug

Page 1 of 1

Kemble Reversing Bug 20/02/2010 at 16:28 #742
TRC666
Avatar
135 posts
Whilst experimenting with a fictional closure I came up with the following problem.

I have created train 5Z35 which enters at Kemble Siding. I route it from shunt signal 1349 to signal 157 as usual, and when the train arrives at Kemble Tunnel to reverse, route it from shunt signal 1350 to signal 160, again in the usual manner. The train will then proceed to Kemble platform 2 and become 2Z35. Now, when the train departs from Kemble, it appears to depart on its own authority (and without abandoning timetable or manual reversing) as it has no signals to obey until signal 157 (and its associated repeater) which is near Loco Yard Junction. It will still depart Kemble and run to signal 157 even if the route from signal 150 to 158 (for down trains) is set, this has potential for a head-on collision if a down train is in the area. The only way to hold the train is to set the points against it at Kemble, the Train List will show that it has run out of valid track, but it will continue again if the points are set again for it. I would have thought that the train would refuse to reverse or proceed without player/host intervention if it is an unsignalled move. I have been to the real Kemble station myself several times and I am pretty certain there is no aspect or shunt signal allowing trains to reverse in platform 2 (nor is there one for down trains to enter platform 1).

I will also add that trains leaving Kemble platform 2 in the up direction always leave their TDs behind and 'flood' the track circuits on the first set of points.

Below is a link to a ZIP file containing a saved game and small test timetable for anyone who wants to replicate the bug easily.

http://www.filefactory.com/file/b037e3h/n/kemble.zip

EDIT: This timetable should work with either mode (platform 4 or parcels), but has only been tested properly in platform 4 mode.

Log in to reply
Kemble Reversing Bug 20/02/2010 at 16:28 #6837
TRC666
Avatar
135 posts
Whilst experimenting with a fictional closure I came up with the following problem.

I have created train 5Z35 which enters at Kemble Siding. I route it from shunt signal 1349 to signal 157 as usual, and when the train arrives at Kemble Tunnel to reverse, route it from shunt signal 1350 to signal 160, again in the usual manner. The train will then proceed to Kemble platform 2 and become 2Z35. Now, when the train departs from Kemble, it appears to depart on its own authority (and without abandoning timetable or manual reversing) as it has no signals to obey until signal 157 (and its associated repeater) which is near Loco Yard Junction. It will still depart Kemble and run to signal 157 even if the route from signal 150 to 158 (for down trains) is set, this has potential for a head-on collision if a down train is in the area. The only way to hold the train is to set the points against it at Kemble, the Train List will show that it has run out of valid track, but it will continue again if the points are set again for it. I would have thought that the train would refuse to reverse or proceed without player/host intervention if it is an unsignalled move. I have been to the real Kemble station myself several times and I am pretty certain there is no aspect or shunt signal allowing trains to reverse in platform 2 (nor is there one for down trains to enter platform 1).

I will also add that trains leaving Kemble platform 2 in the up direction always leave their TDs behind and 'flood' the track circuits on the first set of points.

Below is a link to a ZIP file containing a saved game and small test timetable for anyone who wants to replicate the bug easily.

http://www.filefactory.com/file/b037e3h/n/kemble.zip

EDIT: This timetable should work with either mode (platform 4 or parcels), but has only been tested properly in platform 4 mode.

Log in to reply
Kemble Reversing Bug 20/02/2010 at 17:03 #6839
Peter Bennet
Avatar
5402 posts
The matter of authorisation has since been added as a core-code feature so the train should not depart on an unsignaled move without authorisation. On the point about the route from 150 to 158 with the 'authorisation' code is resolved in that the train will not depart without it. However, there is nothing to stop you authorising it in the wrong situation- but then that's aready the case in that you can instruct trains to pass red signals etc.

Hope that will help.

Peter

I identify as half man half biscuit - crumbs!
Log in to reply
Kemble Reversing Bug 20/02/2010 at 23:08 #6843
GeoffM
Avatar
6376 posts
TRC666 said:
The train will then proceed to Kemble platform 2 and become 2Z35. Now, when the train departs from Kemble, it appears to depart on its own authority (and without abandoning timetable or manual reversing) as it has no signals to obey until signal 157 (and its associated repeater) which is near Loco Yard Junction. It will still depart Kemble and run to signal 157 even if the route from signal 150 to 158 (for down trains) is set, this has potential for a head-on collision if a down train is in the area. The only way to hold the train is to set the points against it at Kemble, the Train List will show that it has run out of valid track, but it will continue again if the points are set again for it.
The train has no way of knowing that a route has been set against it. In reality for such an unsignalled move you would require additional protection such as pilotman working or flagging on the ground, depending on the circumstances. You'd certainly want to collar 150 and 158, and turn off ARS in the area, when performing such moves.

TRC666 said:
I would have thought that the train would refuse to reverse or proceed without player/host intervention if it is an unsignalled move. I have been to the real Kemble station myself several times and I am pretty certain there is no aspect or shunt signal allowing trains to reverse in platform 2 (nor is there one for down trains to enter platform 1).
It doesn't know it's an unsignalled move. Think of a branch line with no signals for an arriving train to depart from. It has to continue on its own authority until it encounters a signal - in your case the next signal is 157R and 157. There are no reversing signals at Kemble so trains cannot reverse there logically. In SimSig the train is obeying what you told it to do in the timetable, and it will probably quite happily do it at Chippenham too (in the down direction) if you told it to do it.

TRC666 said:
I will also add that trains leaving Kemble platform 2 in the up direction always leave their TDs behind and 'flood' the track circuits on the first set of points.
Of course, since it is not a signalled move. So the operation is correct here.

SimSig Boss
Log in to reply
Kemble Reversing Bug 22/02/2010 at 16:33 #6875
TRC666
Avatar
135 posts
Thanks to both of you for the replies, much appreciated!

GeoffM said:
It (the train) doesn't know it's an unsignalled move. Think of a branch line with no signals for an arriving train to depart from. It has to continue on its own authority until it encounters a signal - in your case the next signal is 157R and 157. There are no reversing signals at Kemble so trains cannot reverse there logically. In SimSig the train is obeying what you told it to do in the timetable, and it will probably quite happily do it at Chippenham too (in the down direction) if you told it to do it.
There is a similar situation, on the Bristol sim, where trains can reverse down to up unsignalled at Clifton Down (platform 1 I believe), but they phone in to request authority to proceed onto the single line. Just out of interest was this feature also supposed to be in Swindon A+B or was it perhaps left out accidentally or intentionally?

Log in to reply
Kemble Reversing Bug 22/02/2010 at 18:45 #6880
Peter Bennet
Avatar
5402 posts
The version of SwinDid currently available predated that feature.

Peter

I identify as half man half biscuit - crumbs!
Log in to reply