Fire on a hydrogen bus

The incident occurred in a workshop of a bus company. Hydrogen leaked from a tank on a new hydrogen and fuel cell bus parked there. The cause of the leak is unknown.
From conversations with the persons directly involved, the hydrogen bus had recently been delivered to the bus company, and few days before the fire, a maintenance training had been given engineers, followed days later, the trained people plugged the bus to a fast charger. As soon as the bus was plugged they heard three loud bangs and they run away in panic. On the contrary, the fire fighters post-accident investigation reports that were few seconds between a first puff and the explosion. The released gas ignited before the hydrogen detection system could detect hydrogen. Multiple jet fires followed; one of the jet fires was horizontal instead of vertical because the outlet tube was bent. The fire extended to a diesel bus and to workshop materials. The whole workshop and the two buses went lost.
It is unclear how much hydrogen was in the bus on board tanks and which safety measures kicked in. The workshop had sensors to detect hydrogen and to open roof windows if hydrogen was a specific concentration level in air.
Nobody was injured, but the fire brigades met several challenges in their fire-fighting strategy. The intervention last for several hours.

The INITATING CAUSE has been reported as a leak of hydrogen, which ignited triggering the opening of the TPRD’s. Nothing is known on the cause of the leak. Theoretically, the starting point could have been the accumulation of hydrogen under the bus bonnet and its ignition from an electrical spark; or, alternatively it could also have been a unplanned opening of one TPRD.
Due to the confined location and a horizontal jet fire, a domino effect followed involving all flammable materials in the workshop.
The ROOT CAUSE is of difficult identification. The bent vent pipe creating a horizontal flame, the possible presence of ignition sources and the vicinity of other buses suggest shortcoming risk assessment and procedures. Also, according to the fire fighters, the workshop configuration created a air circulation feeding the fire.

on-board H2 storage,
hydrogen bus,
bus workshop

the hydrogen bus had only arrived in the location one week before. From conversations with the persons directly involved, the hydrogen bus had recently been delivered to the bus company, and few days before the fire, a maintenance training had been given to the bus company personnel by the manufacturer's engineers. On the day of the incident, the just-trained personnel, under the supervision of the plugged the bus to a fast charger, probably to charge the on-board battery system. As soon as the bus was plugged they heard three loud bangs and they run away in panic.

No injury, but a high material cost, far above 200000 €. The whole workshop was lost.

These returns of experience are focussed on the first responders’ strategy and its improvements.
No public report is available on the technical aspects of the event, such as the cause of the first release, the operations ongoing at that moment in the workshop, how many buses were present and how many were hydrogen buses. To know the role and behaviour of the thermal pressure relief devices would be very important.
(1) In hydrogen buses, it is mandatory to have the outflow direction of the TPRD upwards. However, during an incident, it cannot be always assumed that a jet fire will be directed upwards, because vent pipes can be bent by impacts or by wrong installation.
(2) The jet fires triggered a domino effect because the bus was inside, facilitated by flammable materials and by the air circulation. The supply of cool air and the discharge of warm air through the broken skylight contributed to size of the fire. In these cases, a recommendation for first responder would be to open everything that can be opened.
(3) If one hydrogen tank has been blown off, this does not mean that the other hydrogen tanks have also been blown off. It was difficult to determine if there was no longer a hydrogen hazard.
(4) The sound of a hydrogen tank being blown off is deafening and should not be underestimated when . The flaring off of the hydrogen tanks caused so much noise that the fire brigade could not communicate. Phone messaging was used during the period that the hydrogen tanks were blowing off.
(5) Detection of hydrogen can be intended as a warning, for example as an acoustic/visual as a signal to switch off the hydrogen system and/or to enable escape. It can also be part of an (automatic) prevention/mitigating system. However, detection does not necessarily prevent an incident: the speed at which the incident developed was faster than the detection speed. Moreover, location of detectors is based on accidental scenarios and risk assessments assumptions. During a real incident, they may result placed in positions, which are not directly affected by the first hydrogen release; alternatively, ventilation can cause that hydrogen is not detected. Therefore, the presence of detection can give a false sense of security.