Studies on the release of high-pressure hydrogen gas in the case of vehicle fire

Toshiya Hirose, Kazuo Matsushima, Kazuyuki Narusawa, Tetsuo Taniguchi

Research output: Contribution to journalConference articlepeer-review


At the time of a vehicle fire, high pressure hydrogen gas in a tank (a high pressure hydrogen gas cylinder) of a fuel cell vehicle (FCV), which is a passenger vehicle, must exhaust through a pressure relief device (PRD) as quickly as possible in order to prevent any accidental bursts by a temperature rise of hydrogen gas in the cylinder. The high temperature region surrounding a vehicle develops when the hydrogen gas is released through a small nozzle to the air directly. Therefore, to suppress the high temperature region, the effectiveness of a diffusion box is considered further. A pressure relief device (PRD) detects differences in temperature of the environment surrounding an FCV on fire and releases hydrogen gas in a tank to the air by which the valve opens when the temperature in the environment becomes high. The PRD also releases hydrogen gas through a nozzle, e.g. installed upward or downward, to the outside of the vehicle. The PRD is required to be installed in an FCV. A diffusion box covers a hydrogen cylinder and can release hydrogen gas through holes in the bottom plate. Without releasing hydrogen gas to the air directly, to use the diffusion box is an effective method to limit to a smaller area where the temperature becomes high. In this study, tests are carried out from the perspective of safety upon the activation of the PRD and the vehicle fire caused by a leakage of fuel hydrogen gas. Two types of bonfire tests are carried out. One is an ordinary bonfire test and another is a bonfire test with a diffusion box. The bonfire test with a diffusion box is carried out not by an FCV, but by using a testing device assuming a diffusion box in an FCV. Furthermore, the effects of flammable conditions which are defined by both of hydrogen concentration and volume of the air-hydrogen mixture on the deflagration are studied. The conclusions are as follows; 1When the hydrogen gas is released by PRD activation, the diffusion box reduces the maximum distance of the high temperature region from the release point. 2When the hydrogen gas is released with a diffusion box by PRD activation, a shock by explosion is smaller if the hydrogen gas release is vented into the opening section of the diffusion box, because a time delay for ignition is shorter. 3The type-4 hydrogen cylinder used in this study shows that the backfire to the remained hydrogen gas in a cylinder is not confirmed after continuing 60 minutes. 4When the air-hydrogen gas mixture has 6% in hydrogen concentration, the deflagration is not confirmed, but it is observed when a hydrogen concentration is higher than 12%. The explosion phenomenon is affected by the volume of high hydrogen concentration plume of mixture by its leakage.

Original languageEnglish
JournalSAE Technical Papers
Publication statusPublished - 2010 Jan 1
Externally publishedYes
EventSAE 2010 World Congress and Exhibition - Detroit, MI, United States
Duration: 2010 Apr 132010 Apr 13

ASJC Scopus subject areas

  • Automotive Engineering
  • Safety, Risk, Reliability and Quality
  • Pollution
  • Industrial and Manufacturing Engineering


Dive into the research topics of 'Studies on the release of high-pressure hydrogen gas in the case of vehicle fire'. Together they form a unique fingerprint.

Cite this