Learning Signal Usage - Confirmation?

LAST EDITED ON Feb-10-02 AT 09:22 PM (EST)I think I've got a handle on a few things and would like some confirmation from some of the more experienced signal gurus out there.I've been setting up an Automatic Block System using Searchlight signals exclusively. The single track mainline has several passing sidings that are "protected" at the entrance by double head absolute's and at the exits by single head abs. "Permissive (interval)" signals are used along the single track sections I previously posted an issue regarding what I thought at the time was an erroneous "medium approach (flashing yellow)" indication that I had been getting with a couple of the permissive signals. Several respondants had suggested that this may be related to the number of clear blocks ahead that the script defined, but increasing this parameter from 4 to 6 as suggested made no difference.I finally realized that this indication was perhaps not erroneous at all, but was instead indicating that the switch ahead was set to the diverging siding (in this case a reduction in speed is often required to safely pass the points) as I was getting this indication only on the signals preceding the double head absolute signals that protect the siding entrance - and only when this signal was indicating the divergence (red over green), which is of course correct. However, if the heads of the siding enterance signals are linked in reverse to the switch, the preceding permissive signal will show a "medium approach (flashing yellow)" even though you are clear through the straight section.My study of the signal scripts suggest that this is caused by the fact that the lower head of the dblheadabs signal script is written to indicate "clear_1" and the upper head indicates "clear_2", so I suspect this is the trigger for the preceding signals indications.To clear it all up on my route, I went back and made sure that all dblheadabs signals protecting the siding entrances were linked properly - top head to straight through, bottom head to diverging track. I then replaced the snglheadper signals imeadiately preceding the above with dblheadpermissives.With this setup, I now get the proper color/flashing/dbl yellow indications for any given situation and not what otherwise appeared to be an erroneus "medium approach" indication when I had a straight through switch setting ahead.I still now however seem to have a problem with a block not being released to me when there is an oncoming train but when I should have plenty of time to get to the next passing siding (waiting 45 minutes for an oncoming train to pass when I can make the distance in 5min!). I supose this could be related to AI timing but I wonder if the "NumSigClearAhead" parameter for the absolute signals protecting the block are looking too far ahead -they are currenlty set to 6 - do they not need only to look to the next absolute signal? I have the opposing paths routed correctly through the sidings with no convergence.I hope that I am correct in my assesment thus far and would appreciate some confirmation or feedback.
Thanks!

Printer-friendly page | Top

Learning Signal Usage - Confirmation? [View All], Daylighter, 09:17 PM, Feb-10-02, (0) RE: Learning Signal Usage - Confirmation?, Daylighter, Feb-10-02, 09:47 PM, (1) RE: Learning Signal Usage - Confirmation?, sniper297, Feb-10-02, 10:19 PM, (2) RE: Learning Signal Usage - Confirmation?, Daylighter, Feb-11-02, 12:58 PM, (3) RE: Learning Signal Usage - Confirmation?, bobs, Feb-11-02, 01:22 PM, (4) RE: Learning Signal Usage - Confirmation?, Daylighter, Feb-11-02, 01:41 PM, (5) RE: Learning Signal Usage - Confirmation?, Daylighter, Feb-11-02, 09:22 PM, (6) RE: Learning Signal Usage - Confirmation?, bobs, Feb-12-02, 04:41 PM, (7) RE: Learning Signal Usage - Confirmation?, Signalmaster, Feb-12-02, 07:14 PM, (8) RE: Learning Signal Usage - Confirmation?, crow357, Feb-12-02, 08:29 PM, (9) RE: Learning Signal Usage - Confirmation?, bobs, Feb-13-02, 01:57 AM, (10) RE: Learning Signal Usage - Confirmation?, crow357, Feb-13-02, 09:37 AM, (11) RE: Learning Signal Usage - Confirmation?, Daylighter, Feb-13-02, 10:15 AM, (12) RE: Learning Signal Usage - Confirmation?, bobs, Feb-13-02, 12:56 PM, (13)

Conferences | Topics | Previous Topic | Next Topic

LAST EDITED ON Feb-10-02 AT 10:08 PM (EST)To further clarify I thought I would list my current signal usage and placement.This is for a single track mainline with passing sidings using Searchlight Signals exclusively.1. Single Head Permissive - used along single track mainline for train seperation2. Double Head Absolute - used to protect the siding entrance and is placed just prior to the switch - top head linked to straight, bottom head to diverging siding.3. Double Head Permissive - used in place of signal #1 above if it is the last or only permissive signal preceding a #24. Single Head Absolute - used at the siding exit to protect the single track block ahead and is placed just prior to the switch that converges the sidings back to a single track mainline.
Note that for MSTS, AI trains will overun permissive signals indicating red at the exits of sidings and stop on the switch, so you must use an absolute signal at these locations even though some RR's, like the SP, only used permissive signals "APS".

Printer-friendly page | Top

The way the switches are set up ahead of the AI train and which train takes priority are dependent on a number of factors, mostly the signumclearahead combined with the spacing of the signals (sometimes you need to add additional signals within a block to space out the AI dispatcher setting the path) and the speed of the AI train, which is controlled by "expected performance" in the service editor. Also, in the activity editor, the "player drivable path?" checkbox has to be checked for the AI path so that HAL9000 (AI dispatcher) don't set all the switches for the AI path at the moment it spawns.

Printer-friendly page | Top

>The way the switches are set
>up ahead of the AI
>train and which train takes
>priority are dependent on a
>number of factors, mostly the
>signumclearahead combined with the spacing
>of the signals (sometimes you
>need to add additional signals
>within a block to space
>out the AI dispatcher setting
>the path) Sniper, If a section of single track mainline between passing sidings is short, like 2 miles or less, are you saying that you must still install several (more than one?) interval signals? I generally have single track mainline signals ranging in spacing from 1 - 2 miles, and at least have one dbleheadperm on short sections to indicate the condition of the upcoming passing siding. In this case, should the signumclearahead parameter be reduced?Thanks!

Printer-friendly page | Top

Your assessment of your earlier problem and the Clear_1/2 thing is correct. At least for the scripts I wrote for the MP3 signals, Clear 1 is used for divergence and clear 2 for normal. This was done for the reason you indicated, so that the previous signal can warn of the divergence, even if it is a diverging clear. The same is true for Approach 1 and Approach 3.There is one exception to this in my scripts. When you approach a merge (two tracks to one), the signal protecting the merge has one head. In order to receive warning of this merge, I had this signal show Clear 1 instead of Clear 2 by creating a checkbox in the signal properties that you need to set. The previous signal then sees Clear 1, thinks it is a divergence, and shows double yellow or flashing yellow, depending on which signal it is. This matches BNSF practice and GCOR rules. NORAC rules, and the east coast RRs, seem to use a double signal at the merge, with the top always being red. Accomplishes the same thing.

Bob Siederer
Co-author of Marias Pass 3.0

Printer-friendly page | Top

>There is one exception to this
>in my scripts. When
>you approach a merge (two
>tracks to one), the signal
>protecting the merge has one
>head. In order to
>receive warning of this merge,
>I had this signal show
>Clear 1 instead of Clear
>2 by creating a checkbox
>in the signal properties that
>you need to set. That is an interesting option I had not considered.I currently do not have your MP3 signal scripts installed. Since I have many searchlight signals in the route already, will I run into problems if I copy the MP3 scripts into my route and load up the MP3 Signals?Thanks for the confirmation on the previous issues as well!

Printer-friendly page | Top

LAST EDITED ON Feb-11-02 AT 09:24 PM (EST)>>There is one exception to this
>>in my scripts. When
>>you approach a merge (two
>>tracks to one), the signal
>>protecting the merge has one
>>head. In order to
>>receive warning of this merge,
>>I had this signal show
>>Clear 1 instead of Clear
>>2 by creating a checkbox
>>in the signal properties that
>>you need to set.
>
>That is an interesting option I
>had not considered.
>
>I currently do not have your
>MP3 signal scripts installed.
>Since I have many searchlight
>signals in the route already,
>will I run into problems
>if I copy the MP3
>scripts into my route and
>load up the MP3 Signals?
>
>
>Thanks for the confirmation on the
>previous issues as well! Hello Bob,After considering this for while, I started wondering if the two head absolute signal protecting the siding entrance should not also display CLEAR_1 for the divergence instead of CLEAR_2 as well.OOPS! As I was writing this I checked your signal files and I see where you have aready done this too.Hmmm. Me thinks It is time to use MP signal scripts. Should have listened to you a while back. So tell me, can I just copy your sigscr.dat and sigcfg.dat files into my route and then replace the original search light signals where necessary with your MP signals?I am also using distant semiphores and will need to paste this script in.

Printer-friendly page | Top

As put in MP3, the scripts and config files contain all three sets of signals...the original BNSF 3 lights, the Searchlights 2.2, and my version of the searchlights. There are no common shape file names among the three. You could copy the script and config files over your existing ones and everything should continue to work as is. But you would have to delete any existing SL signal and replace it with the appropriate MP signal to get it to work. And, you can't mix the two for operations as the scripts are different enough that the results would be unpredictable. You can edit just fine, and can even run for testing, just don't expect all the sigs to show the right thing until you have all of them from the same set. You could also add your semaphores, but would have to look at how they would interact with the existing scripts.

Bob Siederer
Co-author of Marias Pass 3.0

Printer-friendly page | Top

I'm essentially abandoning the semaphore detection used in the UK/Aust signals and going with the colorlight methods. This way they will be compatable with any of the current signals but most closely the SL22 and MP3 sets.
Hank

Printer-friendly page | Top

Well, I must have missed the correct combination of signals and parameters.Could someone please tell me the correct signals and checkboxes (I'm using the SL signals from Marias 3.0) so that a signal will turn green so I can move from the divirging route back onto the main line. I could use permissive signals but I understand that the AI trains will ignore this.I've also tried the backing up and the creeping forward procedure.Thanks.

Printer-friendly page | Top

If you mean the MP searchlights, you would use a single absolute on each leg of the switch in the trailing point direction. The one coming from the diverging route would have the box checked to cause CLEAR_1 instead of CLEAR_2, but that shouldn't affect whether or not it turns green to let you onto the mainline, just what the signal before it says.Coming from the points direction, you would have a double headed absolute, top head linked to the straight path, bottom to the diverging. If the diverging had a restricted speed limit, you would check the box to force a restricted indication.The exception to the above (there HAS to be an exception, it just works that way!) would be if your NEXT signal (in the trailing direction) also allowed a divergence. In that case, instead of a single head, you would want a double head and would link the paths such that this signal could show you that you would diverge at the next opportunity after joining the mainline.It would be easier to draw a picture...I've got to finish the tutorial I am writing as that surely would help clear up some of this.

Bob Siederer
Co-author of Marias Pass 3.0

Printer-friendly page | Top

Ok, I've done exactly that and I think I've just figured out why it won't turn green.After that last abs signal protecting the end of the passing siding, there are nothing but permissives until the end of the track. And I read in another post, that the abs signals don't look at permissives, so, that last abs signal is saying there isn't any more track.I'll have to test it out.Man, I sure appreciate the help you and others have given me. Hopefully I'll know something some day be able to help you... Art

Printer-friendly page | Top

>After that last abs signal protecting
>the end of the passing
>siding, there are nothing but
>permissives until the end of
>the track. And I
>read in another post, that
>the abs signals don't look
>at permissives, so, that last
>abs signal is saying there
>isn't any more track.
That would certainly cause the problem you are experiencing.There is supposed to be an end of track marker signal available but I have not found it and don't know if it would resolve such a situation.

Printer-friendly page | Top

If your signal is the last absolute, then it should turn yellow, but not stay red. End of track acts as an absolute stop signal for script purposes.

Bob Siederer
Co-author of Marias Pass 3.0

Printer-friendly page | Top

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2002 by Nels Anderson. All Rights Reserved.