开云体育

Problems with an SML Pair #signalmasts #sml


 

I cannot get a single mast pair T7 to T2 via B7 to work as expected.
T7 is a ghost block with a detection block pre and post. T1, T2 and T3 all share a detection block. B17 Shown below is a single block with detection sensors.
I have two paths (B7, B8) coming in from T7 which I've shown below that share same entrance to T7 and exit from T2.?
The SML was auto-generated. The diverging path works as expected. That is, it knows that T7 must be thrown, T1 must be closed (invert) and T2 to its right must be thrown (invert) to pass traffic. The mast on the right (T2) never changes aspect although the T7 mast changes to approach as I'd expect.
?
When auto-generated, jmri paired the T2 mast with another mast in the yard block (BYE) for unknown logic. Complete layout follows for completeness. Note: All start at B17 and counter-clockwise on B1 or B9. B9, B11 are up and over B8,B7.
I copied the logic from T1 mast to T2 mast in the xml but it produced the same results.
I have done grand resets (deleting all paths and SML in this case) several times and I've run out of ideas.
File is uploaded problems and a convenience link below.
I hope somebody here has some suggestion.
//stevemac


 

If you set the block routes to bi-directional, the SML generates properly.

Dave Sand



----- Original message -----
From: stevemac <steveo@...>
Subject: [jmriusers] Problems with an SML Pair #signalmasts #sml
Date: Tuesday, April 01, 2025 2:02 PM

I cannot get a single mast pair T7 to T2 via B7 to work as expected.
T7 is a ghost block with a detection block pre and post. T1, T2 and T3 all share a detection block. B17 Shown below is a single block with detection sensors.
I have two paths (B7, B8) coming in from T7 which I've shown below that share same entrance to T7 and exit from T2.?

The SML was auto-generated. The diverging path works as expected. That is, it knows that T7 must be thrown, T1 must be closed (invert) and T2 to its right must be thrown (invert) to pass traffic. The mast on the right (T2) never changes aspect although the T7 mast changes to approach as I'd expect.
?
When auto-generated, jmri paired the T2 mast with another mast in the yard block (BYE) for unknown logic. Complete layout follows for completeness. Note: All start at B17 and counter-clockwise on B1 or B9. B9, B11 are up and over B8,B7.

I copied the logic from T1 mast to T2 mast in the xml but it produced the same results.
I have done grand resets (deleting all paths and SML in this case) several times and I've run out of ideas.
File is uploaded problems and a convenience link below.
I hope somebody here has some suggestion.
//stevemac


 

If you set the block routes to bi-directional, the SML generates properly.

Thanks Dave. I originally did not have deniedBlocks and several SML failed. Since the layout is designed for only one direction I (painstakingly) right-clicked each block, edit block, edit routing, select send/receive appropriately for every block.
?
All other pairs?display?aspect correctly with?deniedBlocks. Why would this single path/pair fail? I certainly will attempt this change but not knowing the reason doesn’t help my understanding. Can you tell me why?


 

I deleted the SML, set the block bi-directional and now Discover fails to find any mast pair. Since the logic behind SML isn't intuitively obvious to me I subsequently added a T7 trailing main mast and tried again but it fails to Discover any pairs.
?
I've likely tried every combination and I keep getting the same result.


 

It appears that you have created an unusual condition. ?Based on looking at the Route Table, BD7 -> BD17 -> BD8 is being treated as a reverse loop. ?This is probably caused by three turnouts in BD17.

Dave Sand


----- Original message -----
From: stevemac <steveo@...>
Subject: Re: [jmriusers] Problems with an SML Pair #signalmasts #sml
Date: Tuesday, April 01, 2025 3:37 PM

I deleted the SML, set the block bi-directional and now Discover fails to find any mast pair. Since the logic behind SML isn't intuitively obvious to me I subsequently added a T7 trailing main mast and tried again but it fails to Discover any pairs.
?
I've likely tried every combination and I keep getting the same result.


 

>This is probably caused by three turnouts in BD17
Last night I deleted paths and SML, removed/replaced masts and turnouts. After a rebuild it worked correctly. Nothing is obvious comparing the two files but as you imply there was something making the LayoutBlockManager think the two masts were unconnected. Must have been a hidden track segment zig-zag. Thanks!