What causes signal failures in the UK?

[button_boxer]

What happens if travelling at high speed and a blank signal is shown due to a bulb failing? Say that all the signals in the area (including the one with a blown-out bulb) are 4-aspect signals, the previous signal is green (thus giving no warning or need to slow down), then the driver sees a blank signal? I think he would not be able to stop in time.

If I read the earlier replies correctly this could only happen if the signal failed during the brief window after the train had passed the previous signal. Once a particular signal has failed presumably the interlocking would force the immediately preceding signal to show yellow rather than green.

 

[EM2]

What happens if travelling at high speed and a blank signal is shown due to a bulb failing? Say that all the signals in the area (including the one with a blown-out bulb) are 4-aspect signals, the previous signal is green (thus giving no warning or need to slow down), then the driver sees a blank signal? I think he would not be able to stop in time.

Would he then slam on the emergency brakes? What about the AWS (i.e. if the associated AWS magnet gives a 'clear' indication, could the driver assume the signal is green?)

As far as I remember (and it was over 20 years ago I was an S&T tech), if a there is a complete aspect failure, which is almost unknown because all lamps have two filaments, then the preceding signal will failsafe to danger.
However, that not may be a correct recollection!

 

[moggie]

Still true, the previous signal will revert to red assuming it is capable of showing a red aspect. Otherwise it reverts to it's most restrictive aspect. For example if the failed lamp (irrespective of what aspect was displayed) is on a signal which has a Y/G repeater, the repeating signal goes to Y and the signal on the approach to that repeater (assuming a stop signal) is placed to red. I hasten to add this is for TCB lines, not AB lines.

Now - if in the unlikely event a signal lamp fails after a train has passed the previous signal then clearly the driver will not have any forewarning. This is why route knowledge is so important as it is down to the driver knowing where the next signal is and what aspect they are expecting to receive. Of course they should be driving at a speed consistent with the aspect they last passed. If the failed signal has points beyond it there is no major safety implication as the route remains secured by the route locking which remains locked irrespective of a signal lamp fault. I suppose the worst case scenario is when the driver passes a single yellow in rear approaching a signal which then suffers a failure of the red aspect where a preceeding train could be standing ahead in the next signal section beyond the failed signal. In practice a rare event due to infrequent combination of train positions, total lamp fault (mitigated by double filament bulb) and the timing of the fault occurence but neverthe less highly reliant on the driver's knowledge of where the signal displaying the red aspect should be. Any drivers out there with experince of this scenario?

 

[Old Timer]

Still true, the previous signal will revert to red assuming it is capable of showing a red aspect. Otherwise it reverts to it's most restrictive aspect. For example if the failed lamp (irrespective of what aspect was displayed) is on a signal which has a Y/G repeater, the repeating signal goes to Y and the signal on the approach to that repeater (assuming a stop signal) is placed to red. I hasten to add this is for TCB lines, not AB lines.

Now - if in the unlikely event a signal lamp fails after a train has passed the previous signal then clearly the driver will not have any forewarning. This is why route knowledge is so important as it is down to the driver knowing where the next signal is and what aspect they are expecting to receive. Of course they should be driving at a speed consistent with the aspect they last passed. If the failed signal has points beyond it there is no major safety implication as the route remains secured by the route locking which remains locked irrespective of a signal lamp fault. I suppose the worst case scenario is when the driver passes a single yellow in rear approaching a signal which then suffers a failure of the red aspect where a preceeding train could be standing ahead in the next signal section beyond the failed signal. In practice a rare event due to infrequent combination of train positions, total lamp fault (mitigated by double filament bulb) and the timing of the fault occurence but neverthe less highly reliant on the driver's knowledge of where the signal displaying the red aspect should be. Any drivers out there with experince of this scenario?

Two points here, firstly the AWS helps to locate the black signal and secondly in any case the Driver will be braking for the red aspect.

An accident has happened on the Brighton Main Line at Patcham Tunnel in the sort of circumstances outlined above.

The introduction to the Accident Report reads :

"Shortly before 23.20, the 21.50 London (Victoria) to Brighton electric passenger train, formed of 12 coaches, was brought to a stand at Signal CA.4 some 506 yards north of Patcham Tunnel. At 23.22 it was struck violently in the rear by the 21.40 Victoria to Littlehampton electric passenger train, formed of 8 coaches. The driver of this train had received a single yellow (caution) aspect at signal CA.164, just over 1½ miles in rear of Signal CA.4, but the train's speed had not been significantly reduced by the time it passed the next signal, CA.6. This signal, which should have been showing a red aspect protecting the rear of the 21.50 train, was almost certainly unlit as the 21.40 train passed it. The train continued at speed until the driver made an emergency brake application on sighting the stationary train ahead. The brake application had little time to take effect and the train was still travelling at between 45 and 50 mile/h when it ran into the back of the one in front. The collision was very destructive and I regret to report that one passenger, together with the driver of the 21.40 train and another railwayman, who was on duty but travelling as a passenger, lost their lives. One passenger was seriously hurt, 6 were taken to hospital but were allowed home after treatment: and a further 50 received minor injuries."

 

[Marklund]

What about the AWS (i.e. if the associated AWS magnet gives a 'clear' indication, could the driver assume the signal is green?)

The AWS circuit has a Lamp Proving Relay in it, so if the lamp were out, the AWS electro-magnet would not energise.

 

[moggie]

Yes of course, interestingly I believe the presence of AWS at all colour light main signals (in TCB areas) and the rarety of such an event occurring is why it was decided it was not cost effective for the safety benefit obtained to fit TPWS to what would be Auto / passable signals.

History had passed me by on the Patcham Tunnel event. I'll have a read of the report - a couple of years after I started my rail career!

 

[Old Timer]

Yes of course, interestingly I believe the presence of AWS at all colour light main signals (in TCB areas) and the rarety of such an event occurring is why it was decided it was not cost effective for the safety benefit obtained to fit TPWS to what would be Auto / passable signals.

History had passed me by on the Patcham Tunnel event. I'll have a read of the report - a couple of years after I started my rail career!

Partly that and partly the fact that a Driver through his route knowledge should know broadly how far away a signal is from its predecessor - this confirmed by traction inspectors.

The whole argument is that if a Driver correctly applies the brakes at the Y signal then he should be travelling at a considerably reduced speed, so that even should he miss it because there is no AWS he should still be able to stop if he sees a train ahead. In the case of the accident in question the Driver continued without braking until he saw the rear of the train in front.

There is little difference between that and a distant signal in mechanical territory being out.

 

[miikey]

We have had signal failures at London Liverpool Street due to the track circuit being dropped. This can happen due to a number of causes. Fillings from rail grinding trains bridging IBJ's and causing the track cuircuit to drop. This can also happen when two rail fastenings are close to each other. If something bridges the gap the track circuit can drop causing a signalling failure. We have insulated all the sleepers and fastenings to prevent this at Liverpool Street now.

My friend bridged an IBJ with a jack before, that dropped the track circuit and delayed trains... :P