So here I am on a warm February 4C day creating blocks on my NEW C track layout.

As the basic requirement for a block is; it must hold my longest train. Which in a sense can be a locomotive or a train with 100 cars?

Confusing but interesting so....

Time for a break so I decided to come upstairs for coffee and while I'm enjoying my coffee I come across this :

http://www.mckinleyrailw.../item/19-block-detection

What I find very interesting is this paragraph;

We have put occupancy detection on all sections of track including points and track sections too small to be a block. The computer software has a safety feature that will not release a path through a junction until it has become unoccupied. To take advantage of this, we have glued small 10K Ohm resistors (surface mount technology) bridging one axle of every wagon and unlit carriage. The small flow of current across each axle ensures that the occupancy sensors indicate to the computer when a complete train has left a block, and not just the location of the engine at the front of the train.

Yes I know about the 10K Ohm resistors but I never though of detecting all sections. If you have created blocks on your layout what are your thought on this?

Regards.

This limitation is true for analog. It is not true for software control of a digital system.



The addition of resistors to axles applies to 2-rail operations. Märklin axles are not isolated from either wheel so adding a resistor will achieve nothing.

If you use software, detection is only required for the leading edge of the train if the length of the train is known. All occupancy of tracks (or 'blocks') can be deduced by the software from that information. If the length of the train is unknown, then any form of detection (s88 points or occupancy) at the rear of the train can be used by the software, but it will be slightly less reliable than using train length IMHO.

Yes, the addition of resistors to axles applies to 2-rail operations not Marklin. What surprised me was to put occupancy detectors on all sections.

Before I commenced on wiring the new “C track” layout I realized that it would be very easy to detect all the track but is it necessary? Not from what I have read in RR&Co Traincontroller. From knowing what constitutes a block the answer is no but it is intriguing that his layout is done so.

Regards.

Hi Michel

Yes, in most cases, a block must be longer than the longest train (in my case 1.7 m on main lines, .4m on branch /secundary lines). This is the case on my lay-out. On Rocrail there is a possibility -I am pretty sure- to have trains occupying more than one block but I didn't try it.

In normal operation (i.e;: not with super long trains) with Rocrail, it computes the train length and compares with available block length less the minimum block length (a parameter to set the computation precision- 10 mm in my case) to decide if the train can be sent to the block in question

Yes, the most tracks including detection provide safety of operation but in my case using C-track I did not modify the switches to include this detection. My present lay-out is in operation for 4 years and this was not a problem
Cheers

Jean

If I followed the "all blocks must be longer than the longest train" I wouldn't be able to run my ICE3 or Railjet passenger trains. Perhaps that requirement was a constraint under a non-computer controlled layout, but it certainly doesn't apply to software control. Out of my 32 blocks on my layout, probably only 10 blocks are longer than my ICE3 or Railjet. However, these trains are not restricted to those 10 blocks due to software definitions of "crossing blocks" in Rocrail when a route is established, thus allowing them to run almost anywhere on the layout.

I have blocks varying anywhere from ~2,000 mm to ~500 mm allowing the longer trains to cross through the shorter blocks and the shorter blocks allowing a railbus or locomotive to stop in those blocks if desired.