H21- Phase 1 Technical Summary Report

The UK Government signed legislation on 27th June 2019, committing the UK to a legally binding target of Net Zero emissions by 2050. Climate change is one of the most significant technical, economic, social and business challenges facing the world today.
The H21 NIC Phase 1 project delivered an optimally designed experimentation and testing programme, supported by the HSE Science Division and DNV GL, with the aim to collect quantifiable evidence to support that the UK distribution network of 2032 will be comparably as safe operating on 100% hydrogen as it currently is on
natural gas. This innovative project begins to fill critical safety evidence gaps surrounding the conversion of the UK gas network to 100% hydrogen. This will facilitate progression towards H21 Phase 2 Operational Safety Demonstrations and the H21 Phase 3 Live Trials, to promote customer acceptability and ultimately aid progress towards a government policy decision on heat.
DNV GL and HSE Science Division were engaged to undertake the experimentation, testing and QRA update programme of work. DNV GL and HSE Science Division also peer reviewed each other’s programme of work at various stages throughout the project, undertaking a challenge and review of the experimental data and results to provide confidence in the conclusions.
A strategic set of tests was designed to cover the range of assets represented across the Great Britain gas distribution networks. The assets used in the testing were mostly recovered from the distribution network as part of the ongoing Iron Mains Risk Reduction Replacement Programme. Controlled testing against a well-defined master testing plan, with both natural gas and 100% hydrogen, was then undertaken to provide the quantitative evidence to forecast any change to background leakage levels in a 100% hydrogen network.
Key Findings from Phase 1a:

• Of the 215 assets tested, 41 of them were found to leak, 19 of them provided sufficient data to be able to compare hydrogen and methane leak rates.
• The tests showed that assets that were gas tight on methane were also gas tight on hydrogen. Assets that leaked on hydrogen also leaked
• on methane, including repaired assets.
• The ratio of the hydrogen to methane volumetric leak rates varied between 1.1 and 2.2, which is largely consistent with the bounding values expected for laminar and turbulent (or inertial) flow, which gave ratios of 1.2 and 2.8, respectively.
• None of the PE assets leaked; cast, ductile and spun iron leaked to a similar degree (around 26-29% of all iron assets leaked) and the proportion of leaking steel assets was slightly less (14%).
• Four types of joint were responsible for most of the leaks on joints: screwed, lead yarn, bolted gland, and hook bolts.
• All of the repairs that sealed methane leaks also were effective when tested with hydrogen.