Locked switches

hosebag

New member
How are the radius of a locked switch controlled?
Some switches can be turned with a car 3 cars away from the switch, and some require you to be 5, 6, or up to 10 cars away.
 
Yes the radius can be set.
In Surveyor, click on the track tab.
Change to the Trackmark mode.
Click on advanced tab.
Click on the 20.00 (default radius number).
Change the 20.00 to something else. 5 is a good number in meters, which is roughly 17 feet.
Click on Set trigger radius
Click on the junctions you want to change.

You can also use the adjust trigger radius which will allow you to mouse up and down to adjust the radius.

Ignore the yellow-thing that appears on the junction. It's showing the junction adjustment.
 
I presume you know that if Driving the locking depends on train speed. The faster the train is moving the further away the lock occurs. I imagine you already know that and post above is what you are after.
 
The nifty trick I outlined is for yards and sidings where you're switching in tight quarters, but generally the default settings are fine for the mainline. As 1611mac says, the speed determines how far out that the junctions will be locked and set. If you think about it, it makes sense. You don't want the junction set the wrong way if you're speeding down the tracks.

The peeve I have with this is that AI ignore user-driven trains and will snap the junctions for themselves before we have a chance to get them ourselves. There have been many occasions where I find myself stuck while pulling out on to a mainline from a branch line because the AI some distance away decides to lock the junction in its favor just as I approach the switch even though I have already set the junction for my path.

I see this as a bug and not a feature because all consists should be treated equally regardless of the driving method. Prior to T:ANE, this wasn't so much of a problem as it is now so something has changed since that version came out.
 
I've had this and think I got a solution- of sorts !

The nifty trick I outlined is for yards and sidings where you're switching in tight quarters, but generally the default settings are fine for the mainline. As 1611mac says, the speed determines how far out that the junctions will be locked and set. If you think about it, it makes sense. You don't want the junction set the wrong way if you're speeding down the tracks.

The peeve I have with this is that AI ignore user-driven trains and will snap the junctions for themselves before we have a chance to get them ourselves. There have been many occasions where I find myself stuck while pulling out on to a mainline from a branch line because the AI some distance away decides to lock the junction in its favor just as I approach the switch even though I have already set the junction for my path.

I see this as a bug and not a feature because all consists should be treated equally regardless of the driving method. Prior to T:ANE, this wasn't so much of a problem as it is now so something has changed since that version came out.

This happens to me:

"There have been many occasions where I find myself stuck while pulling out on to a mainline from a branch line because the AI some distance away decides to lock the junction in its favor just as I approach the switch even though I have already set the junction for my path. "

if I notice in time, I pause the game, go to the AI train, insert a " halt train " a " wait 1 min. " and usually the switch releases. When that doesn't work, some of my AI only have one command like navigate to a portal or industry, I take note of that and put in a " abandon schedule command " and that for sure releases it and after my consist is in the passing track I re-enter the original portal command.

The feature that we can just change or add commands to a schedule on the fly, is really helpful. In all my sessions I have AI trains full of commodities I use and when I need to I will re-direct to an industry instead of letting them exit at a portal on the other end of the route.

Willy
 
I had this problem and found a solution that worked:
At start of session, all was normal. Switch Z is the problem, being locked by a train (Train A) far away, making Train B wait.

Then Train A passed a jct X, but A was still a long way from switch Z.
Then Train B passed a jct Y, but B was less than .25 km away from switch Z.
Yet Train B got stuck just before switch Z and had to wait for Train A to pass thru switch Z. No way to run a railroad.


Solution: Somewhere along the stretch of track Train A would travel between X and Z, I added a siding. This makes switch Q. No plans to use the siding.
So now Train A is between X and Q instead of X and Z, which means Z will now be locked for use by Train B and all is well.
So in short, the AI must consider a section of track between switches as the controlling factor regardless of length.
 
Last edited:
I had this problem and found a solution that worked:
At start of session, all was normal. Switch Z is the problem, being locked by a train (Train A) far away, making Train B wait.

Then Train A passed a jct X, but A was still a long way from switch Z.
Then Train B passed a jct Y, but B was less than .25 km away from switch Z.
Yet Train B got stuck just before switch Z and had to wait for Train A to pass thru switch Z. No way to run a railroad.


Solution: Somewhere along the stretch of track Train A would travel between X and Z, I added a siding. This makes switch Q. No plans to use the siding.
So now Train A is between X and Q instead of X and Z, which means Z will now be locked for use by Train B and all is well.
So in short, the AI must consider a section of track between switches as the controlling factor regardless of length.

I must try this. A stub sent into the bushes might help here as well. The route I'm running into this problem is quite tight already being a commuter and switching operation.
 
This happens to me:

"There have been many occasions where I find myself stuck while pulling out on to a mainline from a branch line because the AI some distance away decides to lock the junction in its favor just as I approach the switch even though I have already set the junction for my path. "

if I notice in time, I pause the game, go to the AI train, insert a " halt train " a " wait 1 min. " and usually the switch releases. When that doesn't work, some of my AI only have one command like navigate to a portal or industry, I take note of that and put in a " abandon schedule command " and that for sure releases it and after my consist is in the passing track I re-enter the original portal command.

The feature that we can just change or add commands to a schedule on the fly, is really helpful. In all my sessions I have AI trains full of commodities I use and when I need to I will re-direct to an industry instead of letting them exit at a portal on the other end of the route.

Willy

I've manually stopped and started the AI without needing a halt command or abandoning the schedule. The only time I've run into totally stuck junctions is when the AI is already quite close to the junction to begin with. When that's the case, I will let the AI pass by before I go about my switching duties.
 
A litle different situation

I've manually stopped and started the AI without needing a halt command or abandoning the schedule. The only time I've run into totally stuck junctions is when the AI is already quite close to the junction to begin with. When that's the case, I will let the AI pass by before I go about my switching duties.

On the Rocky Mountains Montana track:

there are long stretches of single track that have signals along the way and I guess that is good if one is following an AI but where the problem comes in is when you are close to the end of the single track and the AI steals the switch and changes it to create a head-on and you can't change it cause he has it locked. The only good thing is that he usually sits at the other side of his signal waiting for me but he still won't release the switch.

As you say, if I were in his position and he was approaching on the single track I would definitely wait till he passed.

But, how do you:

" manually stopped and started the AI without needing a halt command or abandoning the schedule "

To do it " manually " would that not entail dropping his AI status and turning him into a player driven train by clicking on the bottom right of the screen where you can change to DCC, Realistic, or AI?

Thanks

Willy
 
On the Rocky Mountains Montana track:

there are long stretches of single track that have signals along the way and I guess that is good if one is following an AI but where the problem comes in is when you are close to the end of the single track and the AI steals the switch and changes it to create a head-on and you can't change it cause he has it locked. The only good thing is that he usually sits at the other side of his signal waiting for me but he still won't release the switch.

As you say, if I were in his position and he was approaching on the single track I would definitely wait till he passed.

But, how do you:

" manually stopped and started the AI without needing a halt command or abandoning the schedule "

To do it " manually " would that not entail dropping his AI status and turning him into a player driven train by clicking on the bottom right of the screen where you can change to DCC, Realistic, or AI?

Thanks

Willy

To stop the AI driver, click on the 3-dots located in the far right of his display. Change from AI driving to either cab-control or DCC. As soon as you change to manual control, the AI will stop driving.

When you want to start him up again, click on the >>-thing where you put in commands and click on Continue Schedule located at the top part of the menu.

What you are seeing on the Rocky Mountain route is prototypical for single-track with sidings. I've see that in real life will traveling and around my area when a freight will sit there for quite sometime until the opposing traffic passes. The difference between real life and AI driving is the real drivers don't keep kvetching about waiting at a signal.
 
CTRL + right click on the train, and select 'Stop Train' to stop the AI train.
CTRL + right click on the train, and select 'Continue Schedule' to let the AI train continue its schedule.

Peter
 
On the Rocky Mountains Montana track:

there are long stretches of single track that have signals along the way and I guess that is good if one is following an AI but where the problem comes in is when you are close to the end of the single track and the AI steals the switch and changes it to create a head-on and you can't change it cause he has it locked. The only good thing is that he usually sits at the other side of his signal waiting for me but he still won't release the switch.

As you say, if I were in his position and he was approaching on the single track I would definitely wait till he passed.

But, how do you:

" manually stopped and started the AI without needing a halt command or abandoning the schedule "

To do it " manually " would that not entail dropping his AI status and turning him into a player driven train by clicking on the bottom right of the screen where you can change to DCC, Realistic, or AI?

Thanks

Willy

To prevent an AI train from simply grabbing one of your junctions you could try using <kuid2:160293:101022:3> Token Management 2. This recently uploaded driver command and its companion rule are really designed for AI sessions but with a bit of adaptation could be used by player-driven trains as well.

The principle is that any train wanting to use a section of track (no matter how long or short) should request a token permitting the train to pass. While one train is the owner of the token, no other train is allowed to use the section.
On approaching a controlled section, players should request the appropriate token and wait until the command clears before proceeding.

Tokens are defined by ensuring that a trackmark name starts with a # symbol, such as "# Section 1". A player-driven train would need to issue a token request before entering section 1 and relinquish it after exiting the section. AI trains would be provided with commands to do the same.

If the token trackmark lies within the protected section then there is an option to automatically release the token once the train leaves that piece of track. The menu item for that is: "Request and wait: Token name". Even though the token ownership might have been transferred, the train will not proceed until the track has been vacated.

There are several other options for advanced use such as clustering several tokens into one GO/NO GO command. The asset details window has a full list of instructions.

Trevor
 
Back
Top