Welcome to the forum   
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Share
Options
View
Go to last post in this topic Go to first unread post in this topic
Offline applor  
#1 Posted : 09 June 2016 06:17:32(UTC)
applor

Australia   
Joined: 21/05/2004(UTC)
Posts: 1,653
Location: Brisbane, Queensland
Hey guys,

Just after some advice on RocRail and if it is capable of working with the below scenarios, since this affects my layout design.

1)

I have a branch line terminus station with a runaround loop.
Can the software order a vehicle to:

1 A train is coming from (1) and stops at the station on (2)
2 activate a de-coupler to separate the lok from consist
3 drive the lok forward past the turnout and stop
4 activate the rear points
5 reverse drive the lok to take the loop (3) past the consists
6 stop when past the entry points
7 reverse the lok again and drive into the consist to re-attach at (2)
8 leave the station and continue with standard route on branch line (1)

Assuming of course that the track is setup appropriately with S88 detection and having the de-coupler connected to an appropriate digital control module.

You can see the plan here:

shunt2.jpg

2)

A train is again coming from (1) and takes the loop around the branch station.
The track loop is bi-directional and a train has entered the loop from the bottom side (4) and another train arrives from the main route (1), does RocRail keep track of the direction the train in the loop is travelling and hence knows to send the newly arriving train into the loop from direction (4) and not in the wrong direction (5) which is where the first train needs to exit from?

I can make the loop uni-directional of course which would avoid the problem but I would prefer to keep it bi-directional.
That would mean changing my signalling however.

Thanks guys!
modelling era IIIa (1951-1955) Germany
Offline French_Fabrice  
#2 Posted : 09 June 2016 14:08:21(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,475
Location: Lyon, France
Hi Eric,

I think both of your questions are related to "automatic mode"...

For 1) I've not experimented until so far such a sequence. What I'm thinking is a set of "Schedules" complemented by "Actions" should do it.
But I'm seriously doubting it will run correctly in an automated way, because of the uncoupler. Uncouplers never work as expected and most of the time, some couplers stay ...coupled...
In my opinion, the sequence you are describing should be done in manual mode, then when the loco has swapped you can reinsert it in the automatic trafic.
I think some parts of the sequence can be easily automated, but the uncoupling and the coupling are the most sensitive parts and subject to random events...

For 2) yes, it's feasible; have a look at the "Critical sections" in the wiki. You must also carefully plan the number of blocks leading to the loop as section (1) need also to be set in a critical section...

Cheers
Fabrice

Edited by user 09 June 2016 19:00:51(UTC)  | Reason: Not specified

Offline applor  
#3 Posted : 10 June 2016 01:27:46(UTC)
applor

Australia   
Joined: 21/05/2004(UTC)
Posts: 1,653
Location: Brisbane, Queensland
Thankyou for your reply Fabrice and yes this is for automatic operation.

I understand the de-coupler may be troublesome.
I was thinking under digital control that you should be able to set the activation period for the de-coupler - So I would set it to activate for a few seconds and have the lok start driving away whilst it is still active which should give a reliable de-couple.
If that sort of operation has not been done before then it is something I would need to experiment with. The issue with that is I would need to plan and install a number of S88 detection areas in order to attempt this operation, rather if I was to do it purely manually then I would not need any at all.
Trying to avoid more work if it is not going to work:)

Thanks for confirming 2) - I have had a read of critical sections in the wiki. This is exactly what I need for my branch line (1) to the main station.
In the case of the loop though I do not think it is appropriate because I can have multiple loks in that loop (block group) - they just all need to enter the block group in the same direction as the lok already inside the block group to avoid a deadlock.
I may need to ask on RocRail forums.

I plan to have all my track equipped with S88 detection (except maybe not my branch station or the BW). I will use the diode trick on detection rails to reduce any contact issues.
modelling era IIIa (1951-1955) Germany
Offline applor  
#4 Posted : 10 June 2016 03:42:42(UTC)
applor

Australia   
Joined: 21/05/2004(UTC)
Posts: 1,653
Location: Brisbane, Queensland
One other question for you Fabrice - do you think it is beneficial to have S88 detection of points position for RocRail?

I am using Marklin 75491 motors with the stop limits removed which can then be used to detect the points position:

https://translate.google...iVepIk8lw9myXdR7Z0eFj0PA

I understand it is not required for operation (the software simply activates the point as required by the route) so I guess the benefit would be it prevents the software activating a point when it is already in the correct position, maybe reducing wear and tear in the long term?
modelling era IIIa (1951-1955) Germany
Offline French_Fabrice  
#5 Posted : 10 June 2016 19:20:10(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,475
Location: Lyon, France
Hi Eric,

Some thoughts...

When you say your loop is done with "block group", thus allowing many locos stored in the loop (one in each block of the loop, of course), I'm not sure allowing bidirectional traffic in such a block group is a good idea. You will encounter very quickly some deadlock situations, in my opinion...
You should allow only unidirectional traffic in the loop... An alternative instead of using block group should be the use of a "staging block", which is also unidirectional...
To avoid some long wait time is this area, you should double the line (1) with an other track (add a siding). It will allow more smooth operation for entering/exiting this zone.

About the detection of point position, I think it's not worth the effort.
My opinion is only a mechanical feedback of the blade position will provide perfect feedback (see https://www.marklin-user...-II--The-town#post467342 and later posts for some thoughts about that issue). As it is very difficult to do, the acceptable compromise is to let it as is... Of course, remove the micro switches, ...or use servos...

About a possible increase of the pulse length when decoupling, take extreme care for not frying the motor/solenoid. In Rocrail, each switch/decoupler can have a specific pulse duration time. I don't know the Amps eaten by a decoupler... I hope it's not too much... As a side effect, you may dedicate a power block for "amps hungry" devices, to avoid some voltage drops...
But I'm still not convinced you have a reliable situation when decoupling in automatic mode... Decouplers (and couplers) are too much unpredictable...

Good luck, and don't hesitate to ask more questions if needed.

Cheers
Fabrice


thanks 1 user liked this useful post by French_Fabrice
Offline applor  
#6 Posted : 12 June 2016 06:09:27(UTC)
applor

Australia   
Joined: 21/05/2004(UTC)
Posts: 1,653
Location: Brisbane, Queensland
Originally Posted by: French_Fabrice Go to Quoted Post
Hi Eric,

Some thoughts...

When you say your loop is done with "block group", thus allowing many locos stored in the loop (one in each block of the loop, of course), I'm not sure allowing bidirectional traffic in such a block group is a good idea. You will encounter very quickly some deadlock situations, in my opinion...


This is exactly the point I was trying to make and what my question relates to - if the block group is empty, it is bi-directional. However if the block group has occupancy can it change to uni-directional based on the direction that the loco in the block group is travelling. This would avoid any deadlock situation.

Originally Posted by: French_Fabrice Go to Quoted Post

About the detection of point position, I think it's not worth the effort.
My opinion is only a mechanical feedback of the blade position will provide perfect feedback (see https://www.marklin-user...-II--The-town#post467342 and later posts for some thoughts about that issue). As it is very difficult to do, the acceptable compromise is to let it as is... Of course, remove the micro switches, ...or use servos...



Thankyou, I will save myself the hassle then:) Didn't want to bother with servos and having to do all that custom work and programming to fit and look good.

Originally Posted by: French_Fabrice Go to Quoted Post

About a possible increase of the pulse length when decoupling, take extreme care for not frying the motor/solenoid. In Rocrail, each switch/decoupler can have a specific pulse duration time. I don't know the Amps eaten by a decoupler... I hope it's not too much... As a side effect, you may dedicate a power block for "amps hungry" devices, to avoid some voltage drops...
But I'm still not convinced you have a reliable situation when decoupling in automatic mode... Decouplers (and couplers) are too much unpredictable...

Good luck, and don't hesitate to ask more questions if needed.

Cheers
Fabrice




Thanks and will do!
modelling era IIIa (1951-1955) Germany
Users browsing this topic
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.

| Powered by YAF.NET | YAF.NET © 2003-2024, Yet Another Forum.NET
This page was generated in 0.387 seconds.