CS2, Rocrail and End pulse

Märklin Central Station 2

CS2, Rocrail and End pulse

Postby cjeung » 27.10.2014, 01:20

I have been able to set up CS2 controlled track blocking using single sensor blocks. This is because the CS2 allows for edge triggering (i.e. start and end pulse). There are, however, some errors in the stability of reading an edge trigger (I will get back to this later). With Rocrail, it seems that the same track blocking can be implemented. This would involve the use of the "end pulse" event trigger. I am aware of the caveat not to use this because of the same stability issues that can occur in the CS2. For example, dirty wheelsets can often cause track occupation sections to register "off" when indeed the train is still "on". So a train car can register as on-off-on, when the block is still occupied.

In Rocrail, I have come across a different issue. For my block entry, I used the following settings:
block rte settings.png


When I run an auto test. I can see the sensor register as active, but no change to the server status. When the end pulse triggers, my block gets a ghost occupancy warning, with the server warning of a Missing Enter event (see image below). What may have caused the Enter event not to register?
block rte settings.png


As I mentioned earlier on the CS2's instability of end pulses. I can correct for it using forward occupancy conditions (I won't explain here, because it is a CS2 methodology).
But what to do in Rocrail? Is it possible to create a stable state instance, where when an end pulse is triggered, it must read off for a set period of time (say 2 seconds) before the "in" state is activated? Or, as in the CS2, implement some conditonal statement, where the "in" action is only activated if another sensor reads 'occupied'?
Cheers, Curtis.
* Running primarily Marklin digital, with a CS2. Running rocrail in Linux
cjeung
 

Re: CS2, Rocrail and End pulse

Postby rjversluis » 27.10.2014, 16:43

Hi,

I suggest you should user enter2in.
End pulses are of bad quality.

Better is to user two sensors: enter and in. (BBT)
Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBUS - CANGCx ] - [ G: CBUS - CANGCx ]
rjversluis
Site Admin
 

Re: CS2, Rocrail and End pulse

Postby cjeung » 27.10.2014, 17:16

Sure, I understand the stability recommendations, so it's not necessary to have the options unless they can be functional. Otherwise, can the issue be addressed?
Also, there was no answer as to why the "enter" event went missing.

These are the images that failed to show.
[img]block%20rte%20settings.png[/img]

[img]block%20rte%20settings.png[/img]
You do not have the required permissions to view the files attached to this post.
Cheers, Curtis.
* Running primarily Marklin digital, with a CS2. Running rocrail in Linux
cjeung
 

Re: CS2, Rocrail and End pulse

Postby rjversluis » 27.10.2014, 17:45

OK, I can reproduce the issue; I will look into it.
Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBUS - CANGCx ] - [ G: CBUS - CANGCx ]
rjversluis
Site Admin
 

Re: CS2, Rocrail and End pulse

Postby rjversluis » 27.10.2014, 18:28

It was not possible to use one sensor in the same block for two events.
I added it now.
Best Regards, Rob.
:!: PS: Do not forget to attach the usual files.
:!: PS: Nicht vergessen die übliche Dateien an zu hängen.
[ macOS - Linux] - [ N: CBUS - CANGCx ] - [ G: CBUS - CANGCx ]
rjversluis
Site Admin
 

Re: CS2, Rocrail and End pulse

Postby eroncelli » 27.10.2014, 20:40

I wonder about the need of "pulse" trigger: in my layout, I use only normal contacts with the CS2.
Everything works perfectly even with "single sensor" blocks: delay between the ENTER and IN functions can be set (IN is simply a delayed action)
HO Maerklin, CS2+MS2, PC with Win10, Android phone, electronics by IEK, decoder by Maerklin-ESU-TAMS
".. and let your dog enjoy Rocrail"
eroncelli
 

Re: CS2, Rocrail and End pulse

Postby cjeung » 29.10.2014, 00:31

@Rob

Excellent! Can't wait to test it out (as soon as I figure on when to update, I picked last stable build I believe)


@eroncelli,

I wrote an article a couple of years ago for the Marklin Digital Newsletter, which uses pulses. Regardless of how you trigger the s88, the CS2 only reads a "pulse". The advantage of start/end pulse triggering is that there is no dependence on timing. The pulses allow triggering of proper events for any length train. The downside is a false triggering, which with some creative thinking can be overcome.

Relating to Rocrail, I've started to think about the differences in control, compared to what is done with CS2. Rocrail, is much about Loc control, and very proactive. (i.e. a train sensor is activated, rocrail controls speed of train to stop). CS2 blocks are somewhat reactive. example, a locomotive passes a sensor block, and CS2 sends signal to track to activate a brake module, CAUSING a trailing loc to stop. Quite a different approach in logic.

Curtis

send me a private message if you wish to know more
Last edited by cjeung on 29.10.2014, 00:33, edited 1 time in total.
Cheers, Curtis.
* Running primarily Marklin digital, with a CS2. Running rocrail in Linux
cjeung
 

Re: CS2, Rocrail and End pulse

Postby eroncelli » 29.10.2014, 20:09

I know the difference between the two systems (Maerklin and RR); a Finite State Automate (FSA) logic, as used by RR, is more effective and safer
The logic is not "stop the train" but, better, "don't start the train" if it can't go to the destination block.
With this logic, no matter if more pulses are received: the first one triggers the event.
Triggering on rise/fall is useful only if the "signal" has no "bounces" and it's very difficult to control, as you know.
HO Maerklin, CS2+MS2, PC with Win10, Android phone, electronics by IEK, decoder by Maerklin-ESU-TAMS
".. and let your dog enjoy Rocrail"
eroncelli
 


Return to CS2