Diverge/Converge/Approach signalling

[dstelley]

I know the difference between a converge and a Diverge Junction but from a signaling aspect (US), are they handled different? If I'm approach a Diverging junction and have to reverse into it once past, to me that would be a diverge, but how would it prototypically be signaled?

'04's ? 06's ?


(C)--------\
(A)---------\------------(B)

From (A) through (C)

Thanks

Side note, any word about when TaneSP1 scripted signals will be "fixed" again?

 

[Eboy87]

First of all, the train wouldn't see a special indication telling it to go through a switch, then back up and take the siding.



Let me assume a few things here.

F------------E--\
A------------B---\---C----------D

<-- WEST EAST-->

So, a train going from A to F, assuming a stop signal at F, would see CLEAR on B, then stop with its tail end just past C. They'd have to call the dispatcher for the signal, and see a Medium Approach or Slow Approach depending on the switch speed (for NORAC) on C. For Western roads, it would be Diverging Approach. They'd have to back the train into the siding that way. Of course, in Trainz, you can eliminate the "Call the dispatcher" part and just throw the switch. Now, if the siding is unsignalled, C will show a RESTRICTING instead of any more favorable indication. If a train was headed westbound from beyond D, and was about to take the siding, D would show APPROACH DIVERGING or APPROACH MEDIUM/SLOW, again depending on the turnout speed.

I've been using justinroth and RRSignal's signals rather than built in ones, but in the above, A, B, D, E, and F would all be '04s, and C would be a 06.

This is of course if I'm understanding your question correctly. And for reference, here are some well done signal charts:

Union Pacific
NORAC

 

[JCitron]

I would use a Type 06 at C, and Type 04s at B and E respectively.

Same as Eboy87...

John

 

[dstelley]

Eboy87,
Thanks, that was almost exactly what I was looking for. All to often I see in routes where people use 1 or 2 or 3 head signals used all over. I thought it would likely be a pair of 04's and an 06 as you described.

I have tried and tried to get justinroth's scripted signals to work since SP1 and they just don't. I can program them well enough so that 2 out of 3 times (but not all, and not even on the same signal) the aspect is correctly shown (Approaching diverge or whatever) but more often than not the HUD on the right shows a completely different setup, or the HUD is right and the signal shown is dead wrong. I have a good screenshot showing this foolishness. Looking out the cab window the JMR... signal will show a R/R/Y which in this instance is correct. The HUD says "red" and the AI controlled driver is accelerating right up to line speed.
I'm longing for JR's 08 Safetran interlocking to work like it did so well in TS12....

 

[JCitron]

There seems to be a disconnect between what the HUD is saying versus what the signal aspects are showing. This could be a script timing issue, but I'm not sure so I don't quite know how to explain it well enough to report this as a error.

 

[Eboy87]

I have the problem with the HUD, and sometimes the cab signals in JR locos (RRS only on that particular error), when using the scripted signals as well in Mac TS2. My understanding is Trainz doesn't recognize any of the aspects other than a CLEAR, APPROACH, and STOP. So a red over flashing yellow MEDIUM APPROACH will just be registered as a yellow APPROACH, a red over green MEDIUM CLEAR would register as a green CLEAR. The extra aspects are just shown thrown scripting while those base indications are what control the trains.

At least, that's the way I've understood it. I'm sure their respective creators can elaborate a bit more.

 

[martinvk]

...
Let me assume a few things here.

F------------E--\
A------------B---\---C----------D

<-- WEST EAST-->

...

Don't know about US 04 or 06 but I find that when using Dutch signals and putting an absolute at B facing West and a Diverging Right at C facing East and E Facing West works very well. Perhaps not prototypical but Trainz AI seems to understand.

 

[JCitron]

Don't know about US 04 or 06 but I find that when using Dutch signals and putting an absolute at B facing West and a Diverging Right at C facing East and E Facing West works very well. Perhaps not prototypical but Trainz AI seems to understand.

That is exactly what I use and this has worked 100% of the time for me except recently in T:ANE SP1. Something has changed here which affects the signals no matter who makes them.

 

[cascaderailroad]

US RH Running: /-S-TM-----------------------------<-\
---S-----------S-/->-------------------------------TM-S-\-S----------S---

S = Signal
< > = Track Directional Marker
/ \ = Switch
TM = Track Marker

I find placing too many signals confuses AI

You need not name signals, nor junctions

AI works just fine with all junctions named: ` (the lowercase of ~)

Track direction markers will only let a train in on one direction

Drive Via a named trackmarker

 

[martinvk]

That is exactly what I use and this has worked 100% of the time for me except recently in T:ANE SP1. Something has changed here which affects the signals no matter who makes them.

Just made a small test route using my normal signal logic and is still works in SP1 80345.

All facing points have a diverging left or right signal, depending on which track is the mainline. All mainline trailing points have an absolute signal. All converging trailing points have the same signal that is on the facing point side.

=====>div right=====\\
=====>abs=========\\=======div right<==


=====>abs=========//=======div left<==
=====>div left======//

> signal facing left
< signal facing right

abs 49267:24006:2
div right 49267:24014
div left 49267:24016
there are a series of div left and right for various junction speeds as well as left and right absolute signals