Route cancel behind trains

London Transport's OCS panels, by contrast, used pushbuttons rather than switches so TORR functionality was easily achieved. It wasn't TORR circuitry though; those panels functioned by remote-controlling lever frames (Westinghouse Style N or N2, which were conventional miniature levers, or Westinghouse Style V which used vertical rotating shafts with handles directly attached for emergency use). The circuitry was designed so that signal levers were returned to Normal as soon as the requisite signal aspect and TC occupation conditions were met, thus releasing the route for the next move.

Solid State Interlocking was, as I understand it, the first interlocking design to incorporate TORR by default. Nowadays, of course, TORR functionality can be incorporated in the remote control circuits for conventional relay interlockings where these have been recontrolled from a ROC workstation.

May be worth adding that TORR was unknown before the commissioning of London Bridge Powerbox (1971 or thereabouts?)

May be worth adding that TORR was unknown before the commissioning of London Bridge Powerbox (1971 or thereabouts?) In a relay interlocking it needs a lot more relays and all the design and testing that goes into that, so it costs quite a lot of £££ and there needed to be a good financial case for it.

This went spectacularly wrong at Aldersgate (Barbican) one morning in January 1955 when a train was signalled out of the sidings. The system specifically allowed you to preset the next route, which the signalman did on his push button desk. As the train left the sidings, rusty rails caused the track circuits to bob, the route released, the preselected route was set and the points moved under the train that was coming out of the sidings, derailing it.

kbarber in post 150616 said:

Solid State Interlocking was, as I understand it, the first interlocking design to incorporate TORR by default. Nowadays, of course, TORR functionality can be incorporated in the remote control circuits for conventional relay interlockings where these have been recontrolled from a ROC workstation.

I believe the later versions of geographical relay sets had TORR, though I'm not sure on the timing vs SSI.

The Leeds area re-controlled a lot of remote relay interlockings with broadly only the station being resignalled and re-locked. None, or very few, of those remote interlockings had TORR but it was needed for ARS operation - not to mention desirable for the signallers! Rather than modify the interlockings, which is an expensive and, in some cases, impossible anyway, the interfaces between the interlockings and the control centres had TORR added in the interface. They could do this because it wasn't considered a safety-critical feature, just safety-related. Correct sequence of occupying and clearing tracks detected, then "pull" the entrance button.

TUT in post 150618 said:

This went spectacularly wrong at Aldersgate (Barbican) one morning in January 1955 when a train was signalled out of the sidings. The system specifically allowed you to preset the next route, which the signalman did on his push button desk. As the train left the sidings, rusty rails caused the track circuits to bob, the route released, the preselected route was set and the points moved under the train that was coming out of the sidings, derailing it.

Ouch. Were the points not track-locked by the train being on the track circuit containing them? Or was that the one that released because of the rusty rail?

kbarber in post 150616 said:

May be worth adding that TORR was unknown before the commissioning of London Bridge Powerbox (1971 or thereabouts?) In a relay interlocking it needs a lot more relays and all the design and testing that goes into that, so it costs quite a lot of £££ and there needed to be a good financial case for it.

The cost justification is reducing the workload on the signallers, meaning you don't need so many.

According to the circuits in the Green Book, it's only one or two extra relays per signal in a freewired system.

It also notes that there were two reasons UK signal engineers didn't like TORR. One was light vehicles, which could make the track seem clear. The other was that a momentary short in an IBJ could make it look like a train had moved from one track circuit to the next and therefore the route could be released. This is why the standard TORR logic involves three track circuits being occupied and cleared in the correct order.

Cambridge PSB is perhaps 10 years after London Bridge, but it doesn't have TORR. I don't know if that was a deliberate decision or was because the geographic interlocking didn't have it included.

TUT in post 150618 said:

This went spectacularly wrong at Aldersgate (Barbican) one morning in January 1955 when a train was signalled out of the sidings. The system specifically allowed you to preset the next route, which the signalman did on his push button desk. As the train left the sidings, rusty rails caused the track circuits to bob, the route released, the preselected route was set and the points moved under the train that was coming out of the sidings, derailing it.

Ouch. Were the points not track-locked by the train being on the track circuit containing them? Or was that the one that released because of the rusty rail?