passengers self-evacuating at North Pole Junction, 15 July 2019

[whole diary]

[Trowres]

I don't recall seeing mention of this incident before.

RAIB^▸ (Rail Accident Investigation Branch) have published a safety digest:

https://www.gov.uk/government/publications/safety-digest-092019-north-pole-junction

At around 16:54 hrs, train 2L97, the 16:36 hrs London Overground service from Clapham Junction to Stratford, formed of a 5-car class 378 unit with between 500 and 600 passengers on board, stopped at North Pole Junction, a short distance beyond Shepherd’s Bush station. The train, which was operated by Arriva Rail London (ARL), became unable to draw power during the switch over from the third rail 750V DC^▸ (Direct Current) supply to the 25kV AC overhead supply, and consequently was unable to move.

At 17:22 hrs some systems on the train shut down and the first carriage lost all its lighting and forced ventilation. The lighting and ventilation on the rest of the train shut down at 18:06 hrs.

Around 17:53 hrs, 57 minutes after the train became stranded, passengers began to self-evacuate onto tracks where both the third rail and the overhead line traction supply current were still energised and trains were running. From social media messages, it is clear some passengers on the train were feeling the effects of heat, and some were becoming stressed and fearful for their own safety.

Due to difficulties in communicating with the correct signal box (the driver should have been speaking with Victoria signalling centre but was unable to contact it using the GSM-R^▸ (Global System for Mobile communications - Railway.) radio system)*, and conflicting messages between ARL and Network Rail, there was a significant delay in agreeing a course of action

* Driver was actually connected to Wembley Mainline signal control centre

The cause of the extended delay was the absence of a decision as to the most appropriate course of action. No single party took the lead.

The driver of the train spent over 27 minutes of the first hour after the train came to a halt, on the phone or radio. In at least two conversations, both lengthy, the driver was speaking to the ARL control room on one device and to the signaller using another. The driver had to repeatedly remind each person he spoke to that the train was not connected to the traction power supply.

At 17:16 hrs the driver walked through the train to reset passenger emergency alarms and emergency door egress handles, to open windows throughout the train, to explain to the passengers what was happening, and to see if he could regain traction power from the cab at the back of the train. This took 11 minutes.

The driver was shouting from the window of his cab to passengers, who were leaning out of the doors, to ‘stay on the train’, at the same time as he was speaking with both ARL control and the signallers.

This incident demonstrates the importance of:

• good and accurate communications between staff on the ground, and railway infrastructure and train service controllers
• effectively supporting the drivers of trains involved in stranding incidents, minimising the time needed to explain things to control centre staff, so that the driver can focus on their passengers and fault finding where practicable (see recommendation 1 of RAIB report 16/2018)
• undertaking regular and accurate assessments of the situation to understand if it is escalating
• ensuring that any limitations of equipment, such as GSM-R boundaries, are known and understood by operators with appropriate mitigations in place, or put in place as soon as possible once identified during an incident
• understanding the risks of passengers remaining in stranded trains for prolonged periods in hot weather without adequate facilities or ventilation
• regularly updating passengers on a developing situation

[ChrisB]

These trains don't have toilets I believe - and thus the maximum time that should be allowed before allowing evacuation has to be shorter than for those trains with toilet facilities.

I dson't believe this factor isa taken into account in these situations.

[Witham Bobby]

I was stuck on a Central Line tube train between stations one hot August day, many years ago.  The train was well filled, but not, thank God, rammed.  I doubt that we were stuck for any more than 20 minutes, but it seemed like an eternity.  The sense of panic that set in among many of the passengers was hard to contend with.  You do really wonder if there's going to be enough air, and if you're ever going to be rescued.  When we got to Tottenham Court Road, several passengers had to be assisted - London Ambulance people were there, at the ready.

Since that day I've avoided the tube on the occasions when I'm in London at busy times, and I'm aware of sensations of claustrophobia in any crowded situation.

I do have considerable sympathy with the passengers on this stranded train.  It doesn't matter to the passengers which organisation was responsible for sorting this failure out for them, they just needed someone to grasp what was going on, very quickly, and to come-up with an appropriate and fast remedy.

[broadgage]

Another case of a stranded train and no effective action being taken to facilitate prompt rescue.
"keep them on the train no matter what" Some escapees were actually ordered back onto the train.

These incidents continue to occur, despite the reviews, studies, and recommendations after each such stranding. I expect more such events for two reasons.
Firstly new trains are generally worse than old in terms of on board conditions (no toilets, no or very limited opening windows, minimal seating)
Secondly new trains are hugely complex with far more features to break or trip and cause the stranding. (consider for example all the new trains that got stuck after a drop in national grid frequency, older trains would not even have known about the frequency)

I repeat my earlier statement that all new electric trains should have either a small diesel engine or a battery pack sufficient to run at very much reduced speed for at least 20 miles, or to run lighting, ventilation and other crucial services for several hours.

I also believe that all new trains must have toilets, for situations like this.

Meanwhile, the stranding of a busy train with poor on board conditions needs to be regarded as an EMERGENCY and not just as another delay.
Evacuation should be considered after 30 minutes, and be mandatory after 60 minutes in the case of poor conditions. Longer times are acceptable if toilets, seats for all, lighting, and ventilation are available.