A question for pguy

Hi Pierre
Thanks for the help earlier in the year all is working fine now. Just have one question about the EIT? is it correct that if I have two trains heading to entry signal A, the first going to platform 3 and the second to platform 1, and train 1 crosses over track that has an exclusive label, that the entry signal will stay red until train 1 has cancelled the path. Train 2 waits at the entry signal and has a clear route to platform 1, all the junctions and signals after the entry signal show the path is available but the entry signal stays red. If a train comes from entry signal B he will get a green to go even if he crosses the exclusive label as his start signal was different.
Many thanks for your work and time in helping others.

Hi Stagecoach.

The answer to your question depends on one important option in EIT paths which is the clear method option. This option defines at what moment the ressources needed to secure access to the path are released.

An EIT path is in fact a collection of path objects (signals, junctions, crossings, … ) ressources and some optional extra ressources which are exclusive set name and tcbs (track circuit blocks).
the underlying EIT system manages in fact a collection of locks with one lock for each ressource. The lock may be available, locked by a path for a train and can have a list of paths and trains waiting for the lock avaibility. When a path is requested, the system checks if all the locks for the needed ressources for the path are available ; if it is the case it gives the lock ownership to the path and requesting train for all these needed ressources and if not it enters the requesting path and train in the waiting list for the ressources currently unavailable. Then when all the locks have been locked (which blocks any conflicting path and train activation) it will move the junction directions and the signals state to their target state, will wait 5 seconds to simulate that these things have taken some time to be done and then will clear the path entry signal so that the requesting train can enter the path. As soon as the owning train has entered the path at the entry signal, the entry signal is forced at EX_STOP state to block any other following train to enter any path before the adequate locks are granted.

The question is now when all these ressources locks will be released and the waiting paths and trains be warned that they can retry path activation request.

It depends essentialy of the clear method choosen :

- with clear on cancel, the locks are released only when the train currently owning the path will leave the exit signal. This will release at the same time all the locks.
- with clear on drive, individual signals (including entry signal) , junctions, crossings locks are released as soon as the train currently owning the path will leave these objects. Exclusive set will still be released only when the exit signal is left as it is a global ressource not located to any position along the path. tcb locks are also released when the train leaves its exit signal.
- with clear after last junction, all ressources are released when the last junction is left : the process is similar to clear on cancel but is fired earlier when the train leaves the last junction without the need to reach the exit signal (this enables at a terminus to use the end of track buffer signal as an exit signal and to release the path after the last junction though the exit signal will never be reached … )

So to come back to your initial question, if the path used by your first train is set with clear on drive option, another path at the same entry signal can be activated for another train as soon as all the ressources needed for the second path have been released by the first path and train. This should occur when all the common junctions between the two path will have been left by the first train. Of course, this works only if the second path does not need any exclusive set lock still currently owned by the first path (exclusive set locks and tcbs are released only after the exit signal or last junction).

if the first path is using clear on cancel or clear on last junction option, all the ressources (including the entry signal lock) are only released when the exit signal or the last junction is left.

EIT has currently the same default value clear on cancel than standard IT for paths, but in most cases it is more efficient to use clear on drive for EIT paths.

If you use the IT enhanced manager rule at run time in driver mode, you can see for a path the current status of all the locks, which are displayed on the path details page. Just do a test : with clear on cancel, you will see that all the path objects locks are kept until the train leaves the exit signal, and you will see that with clear on drive the locks are released following the train progression along the path, and can be reallocated by the system to another path needing the same ressource.

Hope this helps understanding how EITs work.
Regards.
Pierre.

stagecoach said:

Snip...
If a train comes from entry signal B he will get a green to go even if he crosses the exclusive label as his start signal was different.
Many thanks for your work and time in helping others.

Click to expand...

Just to address that specific point, if the EIT path defined from entry signal B also specifies the same exclusive set label, then the entry signal should remain on red until the path from signal A has been fully cleared. That is why the method can be used to protect diamond crossings, for example.

If, however, the path definition from B does not specify the exclusive set label, then the entry signal should go green as soon as the required junctions / signals have been released by the path from A.

I hope this clarifies rather than confuses.

John

Thanks for the replies, I do realise that the exclusive sets can only be used by one train. In my case the second train starts at the same entry signal but does not use the path or exclusive set that the first train does. It only follows the first trains path to the third junction and does not pass the exclusive set. The second train has the path set for its destination but entry signal remained red because it was still under control of the first train. I all ways use the release along the path method.

Train 1 path Sig A, jct 1, jct 2, jct 3, es ?, jct 4, destination.
Train 2 path Sig A, jct 1, jct 2, jct 3, jct 5, jct 6, destination.
The route works fine it was just a confirmation question that the entry signal stayed red.

Hi

With Interlocking Towers (either standard one or EITS), the entry signal is fully managed by the tower, which forces the signal to EX_STOP to block any train to enter any path at this entry signal before the path has been fully granted and activated. When the tower has granted a path for the first incoming train, it will switch the entry signal to automatic or to proceed or other state (depending on your choice) to let the incoming train enter its activated path. As soon as the incoming train has entered the entry signal and path, the signal is switched back again by the tower to EX_STOP state to block any following train to enter any path at the entry signal before the tower grants some path access to the new following incoming train.

With clear on drive, the entry signal lock is released as soon as the rear of the first train has left the entry signal, and the tower should grant access for a following train and path as soon a the first train has left all the common junctions between the two train paths (if there is no other conflict on tcb or exclusive sets).

If you do not observe that on your route, that may be a bug, and if you can send me a cdp with a copy of your route and session I will certainly have a look at it to understand why it does not work in your case and if necessary make a fix.

The IT enhanced manager rule can help to diagnose why a path is not granted by displaying at run time the path details page for the not granted path. the path details page displays for a path all the path options, its current state, which train currently owns the path if the path is active and which trains are queued waiting for the path. It display also all the ressources needed by the path to become active (signals, junctions, crossings, tcbs, exclusive set) with their current lock state and if locked which path owns the lock.
When a path available (in path not set state) is not granted, there should be at least one ressource in the list which should be currently locked by another conflicting path and the display should help to identify it. You then only need to display this conflicting path details page to see which train is currently owning this conflicting path and is in fact the cause of your path activation delay.

When people send me their route and session for diagnosis that is how I identify the cause of the problem, and with some little experience with the run time monitor anybody can make easily this diagnosis.
If looking at the monitor you have some path activation delayed with all the ressources displayed as available, it is then probably a remaining bug and I am then interested for a cdp with a copy of your route and session (with also missing dependencies not available on DLS) to repro the problem and make a fix to overcome the remaining bug for all the community.

Hope this helps.

Regards.
Pierre.

Here is one for you. I have tried everything I can think offto fix this. The signals and junctions that are listed in yellow are not in thepath they are all after the end of the yellow line. The two signals that startthe yellow list are the same ones in red at the bottom. I have added the firsttwo yellows and clicked end here. All the rest deleted and everything lookedgood. Saved and came back in and hit edit and it was all back just like I donenothing.

Don't add but click end here at signal 1537 and then extend path again to the end point.

stagecoach said:

Don't add but click end here at signal 1537 and then extend path again to the end point.

Click to expand...

No sorry that did not work. I even deleted the the path and added it new and it still did the same thing.

This happened to me on a route once and I found the cause to be a duplicate junction and / or signal name. Try doing a search for the first 'yellow' signal and see if there is more than one instance. Although there is nothing to stop route builders creating more than one signal with the same name (beyond a warning message), interlocking towers cannot cope with duplicates.

Have you got the latest version for the tower?

Vostrail said:

This happened to me on a route once and I found the cause to be a duplicate junction and / or signal name. Try doing a search for the first 'yellow' signal and see if there is more than one instance. Although there is nothing to stop route builders creating more than one signal with the same name (beyond a warning message), interlocking towers cannot cope with duplicates.

Click to expand...

Thank you very much. One of the signals in red at the bottom of the list had the same name as another one. I know that you can duplicate names but these where done by auto assignment so I was not looking for duplicates. Thanks again.

Vostrail said:

Just to address that specific point, if the EIT path defined from entry signal B also specifies the same exclusive set label, then the entry signal should remain on red until the path from signal A has been fully cleared. That is why the method can be used to protect diamond crossings, for example.

If, however, the path definition from B does not specify the exclusive set label, then the entry signal should go green as soon as the required junctions / signals have been released by the path from A.

I hope this clarifies rather than confuses.

John

Click to expand...

There was a bug in the problem and Pierre has sorted it out. The entry signal did not change to green when new path was needed. He has said it will be uploaded after final checks have been done.