Signalling on ECML around Welwyn

[crazystripe]

Having incurred yet another delay around Welwyn on the northbound ECML this evening (just like most evenings) I was wondering whether additional signal blocks might be a cheap(er) way of increasing capacity without needing to address the obvious "real" issue of the 2-track section over the viaducts and through the tunnels.

Opentraintimes map shows that a train following a delayed service ahead is forced to slow just past Hatfield, for the most part, and doesn't really get moving quickly again till the service in front is well on the way towards getting back onto the 4-track section approaching Knebworth. Why couldn't additional signalling blocks be added between Hatfield and Knebworth to let the following train get closer to the service ahead, rather than today's situation which causes such a "lag" on the following service speeding up again? The problem today is that a delay on one service almost invariably leads to multiple services being delayed as well. Surely this could be avoided with tighter signalling?

 

[yorkie]

Yes, you can put the signals closer together, but it would reduce line speeds.

 

[Darandio]

Having incurred yet another delay around Welwyn on the northbound ECML this evening (just like most evenings) I was wondering whether additional signal blocks might be a cheap(er) way of increasing capacity without needing to address the obvious "real" issue of the 2-track section over the viaducts and through the tunnels.

Opentraintimes map shows that a train following a delayed service ahead is forced to slow just past Hatfield, for the most part, and doesn't really get moving quickly again till the service in front is well on the way towards getting back onto the 4-track section approaching Knebworth. Why couldn't additional signalling blocks be added between Hatfield and Knebworth to let the following train get closer to the service ahead, rather than today's situation which causes such a "lag" on the following service speeding up again? The problem today is that a delay on one service almost invariably leads to multiple services being delayed as well. Surely this could be avoided with tighter signalling?

Surely adding more blocks means a reduction in the speed due to braking distances? This therefore completely negates any advantage you were trying to gain.

 

[Railsigns]

You can put the signals closer together without reducing the permissible speed if you have consecutive double yellows in the aspect sequence.

 

[D1009]

Having incurred yet another delay around Welwyn on the northbound ECML this evening (just like most evenings) I was wondering whether additional signal blocks might be a cheap(er) way of increasing capacity without needing to address the obvious "real" issue of the 2-track section over the viaducts and through the tunnels.

Opentraintimes map shows that a train following a delayed service ahead is forced to slow just past Hatfield, for the most part, and doesn't really get moving quickly again till the service in front is well on the way towards getting back onto the 4-track section approaching Knebworth. Why couldn't additional signalling blocks be added between Hatfield and Knebworth to let the following train get closer to the service ahead, rather than today's situation which causes such a "lag" on the following service speeding up again? The problem today is that a delay on one service almost invariably leads to multiple services being delayed as well. Surely this could be avoided with tighter signalling?

I believe the distance between the signal protecting Digswell Jn on the down slow and the previous one does not meet modern standards, so the system known as "double reds" has to be employed. This means that the previous signal is approach controlled rather than displaying a single yellow as normal.

 

[The Planner]

You can put the signals closer together without reducing the permissible speed if you have consecutive double yellows in the aspect sequence.

We try and remove those sequences rather than add them in.

 

[najaB]

Having incurred yet another delay around Welwyn on the northbound ECML this evening (just like most evenings) I was wondering whether additional signal blocks might be a cheap(er) way of increasing capacity without needing to address the obvious "real" issue of the 2-track section over the viaducts and through the tunnels.

I think that things will get a bit better with ETCS control rather than lights on sticks.

 

[crazystripe]

We try and remove those sequences rather than add them in.

What's the reason why consecutive signals at double yellow are undesirable?

 

[Bald Rick]

What's the reason why consecutive signals at double yellow are undesirable?

It reduces capacity. Trains are timetabled on green aspects. An extra YY means extra headway.

 

[Railsigns]

It reduces capacity. Trains are timetabled on green aspects. An extra YY means extra headway.

It shouldn't reduce capacity, since consecutive double yellow aspects are used where the signals are spaced more closely than ususal.

 

[The Planner]

It does though as a driver will back off on a restrictive aspect and as Bald Rick says, we calculate headways on greens. New St resignalling is taking a double double yellow sequence out.

 

[D1009]

It does though as a driver will back off on a restrictive aspect and as Bald Rick says, we calculate headways on greens. New St resignalling is taking a double double yellow sequence out.

It's a few years since I've been a planner, but I can't remember double double yellow sequences affecting the Train Planning Rules.

 

[Railsigns]

Consecutive double yellows shouldn't make any difference to headways because where they're used, the signals are more closely spaced, but the distance from the first caution aspect to the red shouldn't be any different - it should still be a little bit more than the required braking distance. It doesn't matter how many signals the train passes between first braking and reaching the red.

 

[The Planner]

It's a few years since I've been a planner, but I can't remember double double yellow sequences affecting the Train Planning Rules.

You are assuming people knew about them!
--- old post above --- --- new post below ---

Consecutive double yellows shouldn't make any difference to headways because where they're used, the signals are more closely spaced, but the distance from the first caution aspect to the red shouldn't be any different - it should still be a little bit more than the required braking distance. It doesn't matter how many signals the train passes between first braking and reaching the red.

If the distance of a R Y YY YY G is no longer than a R Y YY G then no it shouldnt but I would be surprised if the distance is the same.

 

[edwin_m]

Consecutive double yellows shouldn't make any difference to headways because where they're used, the signals are more closely spaced, but the distance from the first caution aspect to the red shouldn't be any different - it should still be a little bit more than the required braking distance. It doesn't matter how many signals the train passes between first braking and reaching the red.

Doesn't this mean that if the following train is being checked by the double yellow today, it will still be checked in the same place by the first of several double yellows if more of them are added? Therefore it doesn't appear to be a solution to the problem noted by the OP, although it might make a marginal improvement depending what happens after the initial check.

I think ERTMS level 2 would allow shorter blocks to be created in areas such as this. In the version without lineside signals (as I think is planned for ECML) there would be no complication of repeated aspects. This would slightly reduce the headway locally, because headway has to be calculated based on the worst case position of trains relative to blocks and for shorter blocks this won't be quite as bad. However if shorter blocks are used over the entire route then the trains could be following more closely anyway and the second one will still be checked.

ERTMS level 3 would be a bit better still, as the minimum spacing between trains is calculated according to the speed and braking performance of the second train (plus a safety margin) so in effect the blocks are infinitely small. This system doesn't of course exist yet.