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 Edmunds  
#1 Posted : 09 May 2011 16:44:40(UTC)
Edmunds


Joined: 12/04/2011(UTC)
Posts: 13
Location: Riga, Latvia
Hi all,

I am now off to workshop to lay the roadbed for a section of my HO railroad with Marklin track. I have followed the very valuable advice from this forum and went for ECoS for command station, which really is great - I have tested both, Marklin and DCC equipment works seamlessly along with each other. This a real revolution, since I no more care about the secret proprietary standard of each manufacturer!

I have also tried a short section of contact track and it seems to work :). Well, as it should. Pretty straight forward thing. What I am worried about and what you, guys could help my with while I'm at my roadbed is how long these contact sections should be.

The great website of Dale S. shows like 7-ties-long. Some other places I have read that it should be at least as long as the longest locomotive. Both, but specifically the latter, makes for pretty long contact sections.

I was previously doing reed contacts and though maybe not 100% reliable at all times, they took 3mm of track length, so I'm a bit confused with doing 2 or three contact tracks, say 150mm each on a 600mm yard track.

Any real life examples of an actual implementation of contact track controlled blocks would be much appreciated.


Thank you all,

/Edmunds

Edited by user 09 May 2011 21:30:17(UTC)  | Reason: Selected "Watch this topic..." checkbox

Offline arconell  
#2 Posted : 10 May 2011 00:01:18(UTC)
arconell


Joined: 27/07/2010(UTC)
Posts: 174
Location: Kreis Kleve, Germany
Hi Edmunds,

It all depends on how you intend to control the block sections. If that is by automatic computer or Ecos control*, it depends on the software that controls the block sections. Usually, once a block section has been defined including its detectors, the control mechanism recognizes when a train hits the first detector in a section and considers that section occupied until the last detector in the direction of travel is free again. More sophisticated software also know the length and speed of the train and the length of the section and take these data into account as well, f.i. by declaring 2 sections occupied when a train is too long for the first section it entered.
Because most control systems can handle both short impulse detection, i.e. reed switches or contact rail, as well as the S88 mass detection which you can make as long as you like, it doesn't really make any difference so long as your control system knows what type of detectors you are using and where they are.

In most control systems however, using S88 mass detection has the added advantage that when long enough, say some 10"/30 cm, you can also use them as a braking/stopping section to bring your train to a smooth stop at precisely the right point.

For me an average length of some 30-40 cm with 2 such feedbacks per block section works out just fine for any length of train including the smallest solo shunter, a Köf.
I don't bother with block sections shorter than 60 cm, simply because it wouldn't make sense to make them into a separate bock section. However, it would make sense in f.i. a small fiddle yard if you want that automated as well. But then I would use the whole length of each track section in the yard as one single occupancy detector.

*I am not at all familiar with the Ecos control system, but I would think in terms of its basic control algorithms it won't be that much different from other systems.

regards, Robert
thanks 1 user liked this useful post by arconell
Offline Edmunds  
#3 Posted : 10 May 2011 08:40:04(UTC)
Edmunds


Joined: 12/04/2011(UTC)
Posts: 13
Location: Riga, Latvia
Robert,

Thank you for your reply. I'm familiar with computer control - I had a pretty sophisticated TrainController layout before with two-rail.

After some thinking and reading your post I deduct:

1) The actual length of the detection section does not matter - what matters is, is where it begins i.e. where the train is, when the system receives a "short" from the detector;
2) The S88 detection sections can be used for both, occupancy and momentary detection.

An example would go like this:

Assume you have a station track, 60cm long. The train is supposed to stop at a signal. For the sake of simplicity, let's say it is unidirectional - trains travel from left to right only. You gap one rail in the beginning of the block(1) and at 55cm from the left(2) and at the end of the block(3). When a train comes, the system senses the train enters the block through(1). Now, since this is a short block, the system starts the respective braking curve for the train to have it at creep speed by 55cm. When the train bridges(2), a stop command for the train is issued and the locomotive stops a few centimeters after, right at the signal. When the train leaves past the last gap (3), the block is marked free and can be allocated for oncoming traffic.

About right?

/Edmunds
Offline arconell  
#4 Posted : 10 May 2011 10:17:39(UTC)
arconell


Joined: 27/07/2010(UTC)
Posts: 174
Location: Kreis Kleve, Germany
Yeap, that's right.

As a variant of your example, using S88 mass detection, you may want to combine gap 2 and gap 3 into one longer (say 25 cm) section. When the train enters this section it slowly reduces its speed to zero and stops exactly at the signal. Now when it leaves the station again, the section will be marked free once the last wheelset of the last coach has freed the outer (rightmost) part of this long detector which should of course extend all the way to the end of the block section. In this variant you need 2 detectors for the station block instead of 3. This works beautifully providing the computer system knows the speed curves and stopping distances of each train. The advantage here is that it saves you one detector decoder position and a bit of wiring. See the little drawing.

UserPostedImage

Best regards, Robert
thanks 1 user liked this useful post by arconell
Offline dntower85  
#5 Posted : 10 May 2011 21:02:02(UTC)
dntower85

United States   
Joined: 08/01/2006(UTC)
Posts: 2,218
Location: Shady Shores, TX - USA
Hi Edmunds,

I have an Ecos and use various lengths of contact tracks. One is very short 5 maybe 6 cm but I only pan to use it to trigger actions on the computer. Most are one track section long.
The circuit track sends a very short signal if the lok is running fast but the Ecos still picks it up
DT
Now powered by ECoS II unit#2, RocRail
era - some time in the future when the space time continuum is disrupted and ICE 3 Trains run on the same rails as the Adler and BR18's.
Offline Edmunds  
#6 Posted : 10 May 2011 22:13:59(UTC)
Edmunds


Joined: 12/04/2011(UTC)
Posts: 13
Location: Riga, Latvia
Hi,

Just to confirm. I only gap the real rails ever. More specifically - one of them.

Two only possible reasons to electrically isolate the "third rail" as well would be for programming track section and sections powered by another boosters. Right?

/Edmunds
Offline arconell  
#7 Posted : 10 May 2011 23:43:38(UTC)
arconell


Joined: 27/07/2010(UTC)
Posts: 174
Location: Kreis Kleve, Germany
Absolutely correct Edmunds!
Schöne Grüsse nach Riga aus KalkarSmile , Robert
thanks 1 user liked this useful post by arconell
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 0.572 seconds.