Learnings

Outcomes

Outcomes from the project are that risks are better understood for some specific scenarios with hydrogen. 

Also training will be facilitated from this project as better understanding of how to fight a hydrogen ignited escape have been realized.

The project has also enabled a better understanding of how a hydrogen gas escape in a downstream scenario will be detected.

Lessons Learnt

Lessons learnt from the project are as follows: 

WP1 and WP2: 

•             A continuously fuelled gas fire was the major hazard present when these incidents occurred. Such fires could cause damage and injury within the hazardous area. 
•                      Third-party fire incidents with hydrogen and natural gas result in similar consequences 
•                     Comparable hole sizes result in similar flame size and flame shape. Height of flame and hazardous area was more strongly affected by wind direction and excavation shape than gas type. 
•                     Radiant heat output measurements of surfaces close to the flame were comparable. 
•                   The gas fires in meter box installations had comparable flame sizes, leak points and direction. 
•                     Buried pipe fires showed comparable flames size and shape on the soil surface. 
•                    A simple device able to generate and propel a salt aerosol could be used to detect the extent of a hydrogen flame and the resulting hazardous area, thereby reducing this risk for hydrogen gas fires. 

WP3: 

•                    Hydrogen and natural gas escapes appear to leak through soil into ducts in similar ways. No scenarios detected hydrogen tracking where natural gas did not track and vice versa; 
•           Due to the higher volumetric flows of hydrogen for a comparable hole size, the volumes of hydrogen tracked, and the concentrations reached were higher in most scenarios; 
•                   Buoyancy does appear to have an impact on the flow rates of gas tracking in ducts. This effect was seen for both hydrogen and natural gas in inclined ducts; 
•                   Hydrogen was found to track faster in the telecoms duct than methane by a factor of 2; 
•                 Weather conditions had a significant impact on both tracking of gases into ducts and the concentrations built up in telecoms chambers; 
•                  Accumulated concentrations in telecom chambers were above the LEL for both hydrogen and natural gas. These chambers naturally ventilated to below LEL once gas flow was stopped over a period of a few minutes; 
•                Gas escapes which track into ducting present an increased risk with hydrogen due to the potential for higher gas accumulations. 

WP4: 

•                     The commercially available detector and automatic shutoff valve worked as expected. They provide an off the shelf system that can be used to protect large domestic and commercial premises. 
•                    In all tests gas was reported in the adjacent room by the detector. 
•                 Time taken to detect hydrogen in adjacent rooms is impacted by ventilation. Variances in the time at which conventional detection at 10% LEL (4000 ppm) and 20% LEL (8000 ppm) occur are cause for concern. 
•                   Both diffusion and convection were found to be important in hydrogen moving between adjacent rooms. Internal ventilation between the rooms is important. However, this effect is maximised with the addition of external ventilation, which increases convection. 
•                   At high concentration (above 4000 ppm) the difference in the concentrations between the rooms was dependent on ventilation. Higher ratios were seen for spaces with low levels of ventilation, which may be an issue for larger leaks. Typically, at a concentration of 10% of LEL (4000 ppm) in the adjacent room, the concentration in the leak room was between 2.5 and 4 times greater (10,000 ppm (25% LEL) –  (40% LEL)). For larger leaks, it is expected that these ratios may increase which may produce flammable atmospheres before detection takes place. 
•            Boundary layer buildup in the rooms is impacted by the placement of internal ventilation. Higher level ventilation between rooms will result in faster detection. 
•             The time to 1000 ppm was similar for all the tests conducted. This suggests that detection is possible and (for the construction of partition used) not strongly dependent on ventilation, but rather diffusion/convection between rooms. 
•            Additional internal ventilation may not be required for domestic properties and door cuts may provide enough ventilation; however further work is required to understand the impact the materials used in the partition walls construction has on the rate of diffusion. 
•           An alarm set point at 1000 ppm has been shown to be a good option. It is suggested that this number is more appropriate for detecting leaks in adjacent rooms.