INiG-PIB in the SHIMMER project: existing leak detection methods can be used effectively in networks carrying natural gas–hydrogen mixtures, especially with up to 20% hydrogen — portable detectors showed the highest overall effectiveness, complemented by OGI cameras and TDLAS lasers
On 14 September 2026 the Oil and Gas Institute – National Research Institute (INiG-PIB) published conclusions from laboratory tests of leak detection methods for natural gas–hydrogen mixtures (NG-H₂), carried out in the European SHIMMER project. According to the Institute, injecting hydrogen into the existing gas network poses new challenges for the tightness of the infrastructure; detailed results are given in the project report D3.5.
- The tests were carried out on INiG’s laboratory test bench (GU_84), which allows controlled leaks of methane and its mixtures with hydrogen to be generated and is equipped with devices simulating atmospheric conditions that may interfere with measurements. Leaks of 0.1–4.75 l/min — typical small leaks in gas networks — were generated at a nozzle, a threaded connection and a flange; the mixtures contained 0–20% hydrogen, and 50% and 80% were also tested to assess the lower detection limit.
- Four methods were evaluated with eight devices — portable detectors (sniffers), optical gas imaging (OGI), acoustic gas imaging (AGI) and tunable diode laser absorption spectroscopy (TDLAS) — in series of 10 independent measurements. A leak was considered detected when the concentration of methane or hydrogen in air was at least 50 ppm; sniffers with methane sensors detected all leaks (100% sensitivity), as did those with hydrogen sensors, and at 0–20% hydrogen no changes to procedures or equipment are needed.
- OGI cameras detected all leaks of mixtures with 0–50% hydrogen; at 80% the performance drops significantly and depends on the camera. TDLAS detected all leaks with a sensitivity above the assumed 75%. The lower detection limit of the acoustic method varies with, among other things, the shape of the leak, the hydrogen content and the device — according to the report, this method appears to be the least suitable for NG-H₂ mixtures.
- In an assessment against eight criteria (including the lower detection limit, resistance to changes in gas composition, availability of devices and measurement standards, costs and personnel qualifications) on a scale of 1–5, the sniffer method scored highest. There is no universal method: according to INiG-PIB, a multi-method approach remains key — sniffers and OGI complement each other, and TDLAS provides remote detection in hard-to-reach locations. The project is funded by the Clean Hydrogen Partnership (grant agreement No 101111888).