Could Virtual Block be created from mini-blocks?

Trains Magazine General Discussion
1

Ignoring the cost side of it, could a Virtual Block system using mini-blocks be technically possible? …and would such a system provide greater efficiency of track usage and more flexable train control of varying trains (i.e size, speed, weight, etc)??

In theory Moving Block is supposed to be one of the ‘ultimates’ in efficiency because the block ‘wraps’ around the train and moves with it. My thought was that if the blocks were smaller, and computing technology existed, the benefits of Moving Block, Fixed Block, and Varying Block could all be delivered via these smaller blocks with an abundance of differing trains running the line. The computers would process the weight, length, braking, and real-time speed, and automatically ‘bundle’ X number of these mini-blocks into the needed space (Virtual Block) for a particular train at that moment in time. The train slows down, or or mini-blocks are removed, and as the train goes faster the system adds mini-blocks to its headway. As the train proceeds down the line, the Virtual Block moves down the line by dropping/adding a mini-block. Each mini-block is signalled, and the signal at this mini-block display the appropriate aspect for the Virtual Block.

I hope my diagram works:

Fixed (current) Block
|===============|===============|===============|===============|===============|

Mini-Block
|=====|=====|=====|=====|=====|=====|=====|=====|=====|=====|=====|=====|=====|=====|

Virtual Block indicated by (), for fast heavy train, :
|==T==|(=====|=====|=====|=====)|=====|=====|=====|=====|=====|=====|=====|=====|=====|

|=====|==T==|(=====|=====|=====|=====)|=====|=====|=====|=====|=====|=====|=====|=====|

|=====|=====|==T==|(=====|=====|=====|=====)|=====|=====|=====|=====|=====|=====|=====|

|=====|=====|=====|==T==|(=====|=====|=====|=====)|=====|=====|=====|=====|=====|=====|

Virtual Block indicated by (),for slow light train:
|==T

2

I’d like to see that incorporated into an automatic train control system–cab signals, no wayside signals (except absolute signals where required).

3

Sure, cab signaling makes perfect sense!

4

Normally the block trails the train, so in Virtual Block it would need to be in front of the train so this could be an unique challege.

5

Would Virtual Block be Failsafe.

6

…as failsafe as any RR signaling system out there today. [;)]

7

Point about modern systems is that they don’t NEED formal blocks at all – if you look at some of the modern PTC systems, they determine continuously from train speed, gradients and track condition, etc. what the correct ‘following distance’ plus safety margin should be, and control speed and braking appropriately.

The big problem I see with ‘mini-blocks’ is that you need an architecture of sensors to define where the blocks are relative to the track, and all the sensors have to be operating correctly for the system to work. Any ‘failed’ sensor will automatically have to ‘fail safe’ to a restrictive indication; anything more ‘adaptive’ would involve quite a bit of logic circuitry for something that’s almost certainly done better as continuous PTC anyway…

You might also note that train-order format for a miniblock system is going to involve significant added complexity if it follows existing railroad practice. I’ll be among the first to agree that some form of ‘continuous train order’ (following the likely kinds of format that will wind up being used with a system like NAJPTC) will be a future development… but again, I think it makes better sense to have movement authority, etc. reside with the train as an entity, tracked against something continuous like absolute track milepoint, than to involve an intermediate logic layer of lots of discrete logical blocks…

8

Can you send me some www links where I can read more about this?