Signalling single-track.

nicky9499

SSoTW Bot
diag.jpg


How would I signal the section from A to G as one block in which only a single train can be in it anytime? Divergences B, C, D, and E are industrial sidings and AI is usually not programmed to be take them. However, they are still fully signaled at the junction connecting to the mainline. F leads to a six-track yard of which the 3 easternmost ones are part of the mainline and is used by freight AI.

diag2-1.jpg


Both freight and passenger tracks converge at Y a short distance down south. Occasionally freights will be broken up into the 3 western tracks and brought to the local sidings B, C, D, and E by a road switcher.

Given diverge signals, absolute signals and any triggers or trackmarks that may be required, how do I go about signalling A to G/Z such that:

a) the AI treats A to G/Z as a single block,
b) no AI train coming from the north or south will enter the block when a switcher is within A to Z?

Cheerio,
Nicholas.
 
Signals

I am no genius with signals, but there are, I beleive, simple rules with single-line working.
A siglan leaving the A station and another immediately after the point to protect the tracks within the station itselef. The outward signal should control the whole line to G. All sidings should have a signal of some sort before the track enters the main single line to protect the single line and hold and train in the siding until the main is clear to enter.
The same thing applies at G, a single signal to protect the line right through to A, and one of the single track to protect the sidings to Z. All sidings should again be individually signalled.
Basically you are treating all sidings as adjoining lines so should be signalled to protect the main single.

Hope that helps.

Angela
 
Hold on. Does this mean I do not place any diverging signals (namely either the USA02 or Searchlight 06 LD/RD) from A through G/Z?

I have created a scaled down representation to test such a setup with only 04/06 trailing signals, and it has worked to a certain extent. The only problems I face are trains in opposing directions forcing junctions into positions that is not favorable to either party, resulting in both signals at A and G/Z showing track direction marker pointing towards the train. This can only be solved by either temporarily taking over manually, using STW drive rule, or ReleaseJunction rule.

STW drive, if used, will cause problems elsewhere on the route because when the train runs under this rule it would not automatically switch junctions. ReleaseJunction is troublesome because I would have to automatically add this command to every train approaching the single-track block in question. Even if I were to define specific junctions in the ReleaseJunction rule during session setup, there are too many junctions that an opposing train can control.

Also, when a train leaves from Z, on top of the problem mentioned above, it will also go undetected by any train that has passed the signal at A or G. Manually checking for trains passing these signals before leaving the yard is tedious as the block is a couple of miles in length.

Any help will be much appreciated.

Cheerio,
Nicholas.
 
What Angelah said is correct in my opinion.

diag.jpg

I have put a green dot where I believe you should put the signals. You only need to signal the main line at A and G and then a signal on each diverging line to protect the main line.

Nathan
 
A suggestion

Here it is, and it's just a suggestion....

Download my Hayling Island branch, it practically all single line and works well. It's a small'ish file too...

Angela
 
I will, thank you. Does your route require many dependencies on the download station?

Napalm, I will try out that configuration when I get home and post back, thank you.

Cheerio,
Nicholas.
 
Probably

I will, thank you. Does your route require many dependencies on the download station?

Napalm, I will try out that configuration when I get home and post back, thank you.

Cheerio,
Nicholas.

I expect so, it's almost impossible to build a route without using something from the DS. It was made in 2004 so you should find it simple enough.

Best of luck,

Angela
 
Not to hijack this thread, but has anyone had a chance to try the AI traffic with signalled single track in TS2009?

Auran said they made improvements with the AI so I was wondering if they made ones that would prevent cornfield meets and Mexican Hat dances.

John
 
It seems that their level of intelligence has somewhat remained the same. As stated above, the signalling setup Angelah and Napalm works for one train on the mainline and one (me) working the local switcher. Two trains at opposite ends of the mainline will result in either

a) both trains will not release the junctions immediately ahead of them (the ones where double track mainline converge to enter the city) so both cannot move, or

b) train/trains will hold a junction and I will be unable to get out of a siding. If there are many trains passing through, I'm just stuck there for the entire day (or hijack the trains in question holding my junction and ReleaseJunction :hehe:)

Sigh.

Cheerio,
Nicholas.
 
Hi, I am just posting as I would like to follow this thread. I am working on a similar type layout/signal system but I am not at the Signal stage yet.

Thanks,

Craig
:):):)
 
This could be a problem going from double to single with yard access in between. I've had very little success with this when working with AI traffic.

If you signal it prototypically the AI have no idea how to work the line. Napalm's suggestion would work, but it's not prototypical. It really depends on how realistic you're trying to make it.

I've been told that the Path rule helps in this respect, but I'm yet to determine whether this is the case.

Best of luck. :)
 
My take on signalling has been to always just signal ONE of the converging tracks. When faced with a diverging track, the train will always take one or the other, so that facing point signal is not needed (in Trainz). I only put signals where they are absolutely needed to prevent a collision.

Using trackmarks will solve the train's issue of not releasing the junction ahead of them.
 
Hello Euphod,
You mentioned trackmarks. Care to elaborate more on how I should go about using these?

Jamez, it is a freelance route so there is no pressing need for 100% prototypical signaling. After all, what's the use of having prototypical signals when nothing will work after that? I'm also interested in your suggestion regarding paths.

Cheerio,
Nicholas.
 
diag.jpg


Assuming right side running, Northbound trains will stop at signal at "G", then will travel to Trackmark A if given the "Drive Via...Trackmark A" command. In a similar fashion, Southbound trains will stop at signal at "A", and then proceed to Trackmark B if given the "Drive Via...Trackmark B" command.
 
Whilst I agree wholheartedly with my learned friend, I would add a signal on the trailing side only of the junctions B through F inclusive. Which brings us full circle to Napalm's solution, except he (being a gentleman of the Old Country) has signalled it for left-hand running....

Andy :)
 
Whilst I agree wholheartedly with my learned friend

Well, I don't know to whom you are referring, but I only was attempting to show the trackmarks. If you are suggesting sidings B through F be signaled to protect the main, then I concur with your brilliance!
 
Single Track Working

Whilst I do not have the know how to provide you with a sketch or marked up sketch, I would use the route controls provided by js-TRF to control the signals and routes with the signals positioned by Angelah. This is a scheme that is considerably better than Path Control, and would suit the situation.

You would need to do some "programming" with signals numbered and suitable trackmarks and triggers, but the trains would sort themselves out in the order of triggering their position and route.

May I refer you to their website for further info and downloads:-

www.js-home.org/trainz

I have used their set of downloads to control both N.American and English junctions/station throats.

They really work too!!!
 
Thank you guys. I shall be out of the country until Saturday. Will post back on the results. Initial testing shows that Euphod's method will work with a divergence signal on both junctions A and G. A local switcher out of the yard is still undetected though.

Cheerio,
Nicholas.
 
Back
Top