Long Block Sections

Vern

Trainz Maverick
Posted this over on UK Trainz but thus far no response so thought I would test the knowledge base here in case someone has the answer...

Here's an oddity.

Fed up of planting endless trees, gorse bushes and the occasional bungalow on the Mallaig route I decided to make a start on the signalling, using the LMS sempahores from the DLS.

Last night I signalled Mallaig to Arisaig and on a test run successfully got the AI driver to change the points at Arisaig to proceed in the Up direction and the starter/section signal came "off". Tonight I placed the signals at Glenfinnan. On setting up a run in the Down direction, on attempting to depart Glenfinnan the points changed over successfully but the signal would not clear with a "Track terminates or closed" message. I have already proved the route from Glenfinnan to Arisaig and the track is complete with no breaks (maybe the odd wiggly bit to sort out).

The only thing I can think of is that the block section is 17 miles long from Glenfinnan to Arisaig and this has somehow confused the signal AI.

Any suggestions or fixes appreciated.
 
Yes it seems Trainz does not recognise block signalling where sections are over a certain length as putting in a new signalled crossing loop at Lochailort (where there was one until the early 1960's) cured the problem. However the adjacent section from Mallaig Jcn to Glenfinnan is nearly as long (16 miles) and there is no historical precedent for a crossing loop at any of the intermediate stations.

This does seem to have revealed a limitation of the Trainz AI. There are quite a few railway sections in the world and particularly UK/Ireland where there may be a lengthy run between blockposts. Putting intermediate signals in is not the answer as unlike the US, single line sections certainly those worked by token or train-staff are treated as one block and do not have intermediate signals.

Unfortunately this does rather limit the use of Trainz for prototypes with long single track sections - most of the more remote UK lines are ruled out and I had been contemplating something based on Ireland, but Clonmel to Tipperray on the Waterford & Limerick, for example, is nearly 25 miles between block posts.
 
I suspect there may be a cunning workaround using rules and triggers, but it hasn't occurred to me yet....

I'll post if anything comes to mind...

Paul
 
I would suggest you place Invisible signals to break the section into smaller pieces - or write a "Rule" for Key Token working...:o
 
Invisible signals as suggested and Wulf 9's "Static Block Rule"; KUID2:104722:255001:3 I think is the one ess1 is thinking of (if not then I don't know).
 
I would suggest you place Invisible signals to break the section into smaller pieces - or write a "Rule" for Key Token working...:o

The problem with putting these signals inside blocks is that now the block is subdiveded and could let a driver enter the block already having another unit in it ?
 
That's what I thought, unless you overlap the bounds of the signal, i.e. in a 15 mile section the intermediate signal for A to B is 8 miles from A/7 miles to B while in the reverse the intermediate signal from B to A is 8 miles from B/7 miles to A.

I'm going to try that but it's still an inelegant solution to a problem which shouldn't be there in the first place. I know TRS was originally conceived for collectibles/model railway style operation but from TRS2004 onward it has been marketed as a prototype simulator, so it needs to work like one. They've had 7 years to get it right - I'm sure in that time I'm not the only one who has tried to build this style of route.

This is nearly as bad as the MSTS coupler breaks after running more than 10 miles with no points or switches to divide the track nodes.
 
Vern did you see Pommie's 2nd reply in the thread on "length between signals". I've copied the relevant part of it below:
Thats over 45k and the AI can see the signal and return a green so the AI driver will do the correct line speed, what I did was to place the short multi-industry half way (I left every setting at the default zero)
It seems that the AI can see the industry and then looks for another 25~30k for a signal :)
Using this work around is easy and cannot be seen in driver, I don't know if it will work using two industries for an even longer run but if not at least it doubles the block length.

Cheers David
Seems to be a possible solution to your problem.

Bob Pearson
 
Yes thanks for pointing that out, I may give it a try.

Still a situation which shouldn't require a workaround though.
 
...However the adjacent section from Mallaig Jcn to Glenfinnan is nearly as long (16 miles) and there is no historical precedent for a crossing loop at any of the intermediate stations.

...

Yes there is - nearly. According to John Thomas's "The West Highland Railway", p131
On the outbreak of war... A naval base was established at Corpach and new loops and sidings were put in. The capacity of the line was increased by the placing of a new signal box at Camus-na-ha, 4 1/2 miles from Mallaig Junction.
The naval base is, I think, where the Linnhe Caravan Park is now. Don't know if there was actually a loop on the main line, but it certainly sounds from that as if there was at least a block post, similar to the RETB token point at Loch Eil Outward Bound Station today. I think you'd be justified in putting a loop in there Vern. I'll ask around some of the older railway staff here and find out if it was an actual passing loop. Problem solved?
 
Thanks David!

Unfortunately I still don't think that will solve the signalling issue as the limit seems to be under 10 miles. The trick of placing a passive interactive industry didn't work either, so I'm now going to look at repositioning the invisible signals about halfway between Locheilside and L.O.B.
 
Solved It!

Well sort of. I took out the semaphores and replaced them with default (TRS2004) UK colour lights. Everything now functions as it should, no need for interactive industries, intermediate signals or anything else. Next step is to remove the un-necessary crossing loop at Lochailort, resignal Arisaig/Mallaig then it's back to 3D placement.

I daresay the purists will object but IMHO better to have a Mallaig route with colour light signals than not at all - which is how it was beginning to look.
 
Why not wait a bit to do the signals
I understand that the TC Edition3 (British Line)
will come with a new semaphore signal system that should solve the block issue, You will probably have to buy the prog but should be worth it
 
Camus-na-ha loop was just west of the bridge to what is now the Linnhe Caravan Park, on the falling gradient (westbound) straight stretch and it was a full passing loop. My contact said "I remember being sent there with (name of driver) on a J36 to fetch a single wagon that had been left in the loop. He was raging at being sent all that way just for a single wagon." The loop was taken out in 1961 or 62.

I was also wondering about TC3 Settle & Carlisle Edition as having a re-write of the block code. This (Bloodnok's influence now he's down-under?) could well give us proper working semaphores.

I tried a quick test layout, 2 stations with passing loops on a single line. I used PathControl. Everything seemed fine, the path was set, the train started to move, then as it came up to the points at the end of the loop I got "Waiting for acess to track beyound junction", even though the path was set, the signals (starter distant and home) were all off and the points correctly set and there was no train on the route...
 
I may look at redoing the route once TC3 comes out but it is using a fair bit of built in content for TRS2004.

I'm happy enough now things are working with the colour lights, as I say it's not prototypical but the route has been built for people to have a bit of fun driving through a semblance of the Highlands not a slavish copy of the real thing.

With the traction pack included in TC3 though I've already got a couple of ideas forming in my head for a new project!
 
That's what I thought, unless you overlap the bounds of the signal, i.e. in a 15 mile section the intermediate signal for A to B is 8 miles from A/7 miles to B while in the reverse the intermediate signal from B to A is 8 miles from B/7 miles to A.

I'm going to try that but it's still an inelegant solution to a problem which shouldn't be there in the first place. I know TRS was originally conceived for collectibles/model railway style operation but from TRS2004 onward it has been marketed as a prototype simulator, so it needs to work like one. They've had 7 years to get it right - I'm sure in that time I'm not the only one who has tried to build this style of route.

This is nearly as bad as the MSTS coupler breaks after running more than 10 miles with no points or switches to divide the track nodes.

Dido,dido, dido!
 
Back
Top