Ventilation System Influence on Hydrogen Explosion Hazards in Industrial Lead-Acid Battery Rooms

Overview

Lead-acid batteries give off hydrogen during charging, so a battery room whose ventilation fails can build up an explosive atmosphere. Hydrogen is flammable in air from 4% to 74% by volume, and battery rooms must stay below the 4% vol lower explosive limit (LEL). The study first ran full-scale hydrogen release tests in a 20 m³ room with a 1.0 × 0.5 m battery-imitating box; at the slowest, most realistic release rate hydrogen filled the room fairly uniformly instead of collecting under the ceiling, so when the LEL is reached almost the whole room volume can be flammable.

FDS simulations then modelled a 20 m³ room with twenty 2100 Ah open lead cells, with hydrogen emission estimated per BS EN 62485 at 9.7 × 10^-5 m³/s. Without ventilation, theoretical times to reach 10% LEL (0.4% vol), 40% LEL (1.6% vol) and 100% LEL (4% vol) were 840 s, 4260 s and 8520 s; the simulated rise was slightly slower because the lowest layer stayed hydrogen-free. BS EN 62485-based design gave 42.3 m³/h mechanical extraction or a 0.12 m² natural-ventilation opening for this room.

All five ventilation scenarios kept hydrogen below 20% LEL. Natural ventilation with a ceiling exhaust was most effective, mechanical ceiling extraction was less effective than natural ventilation, and short extraction hoods could leak hydrogen into unventilated zones. Detection and ventilation systems were confirmed as the key to eliminating fire or explosion risk.

Limits: one room size (20 m³) and one battery type (open lead cells); although the event tree analysis estimated scenarios, real rooms can have very different geometries. The finding that hydrogen does not accumulate under the ceiling at low release rates still matters for detector placement.

Key points

1. Battery rooms must keep hydrogen below its 4% vol LEL; hydrogen is flammable in air from 4% to 74% by volume.
2. Full-scale tests used a 20 m³ room and a 1.0 × 0.5 m battery-imitating box; at the slowest, most realistic release rate hydrogen filled the room fairly uniformly instead of concentrating below the ceiling.
3. Modelled case: 20 m³ room with twenty open lead batteries of 2100 Ah; hydrogen emission per BS EN 62485 of 9.7 × 10^-5 m³/s.
4. Without ventilation, theoretical times to reach 10% LEL (0.4% vol), 40% LEL (1.6% vol) and 100% LEL (4% vol) were 840 s, 4260 s and 8520 s; the simulated rise was slightly slower because the lowest layer stayed hydrogen-free.
5. BS EN 62485-based design gave 42.3 m³/h mechanical extraction or a 0.12 m² natural-ventilation opening for this room.
6. All five ventilation scenarios kept hydrogen below 20% LEL; natural ventilation with a ceiling exhaust was most effective, mechanical ceiling extraction less so, and short extraction hoods could leak hydrogen into unventilated zones.

Using it in design

• Calculate battery room ventilation from hydrogen emission per BS EN 62485 (here 42.3 m³/h or 0.12 m² for a 20 m³ room) and separately show the time to reach LEL if ventilation fails.
• Prefer natural ventilation with a ceiling exhaust opening where possible; if using mechanical extraction, check for hydrogen leaking into zones outside hood coverage.
• Place hydrogen detectors considering that gas at low release rates mixes fairly uniformly through the room height rather than only collecting at the ceiling, and link detection to the ventilation-failure alarm.

Who should read it

Electrical and fire-safety engineers designing ventilation and explosion protection for lead-acid battery rooms.

Document contents

1. 1. Introduction
2. 2. Battery Rooms Failure Scenarios
3. 3. Conducted Real Scale Experiment (3.1 Experiment Layout, 3.2 Experiment Results)
4. 4. CFD Analysis of Different Ventilation Systems Effectiveness (4.1 Explosive Hazards in Battery Rooms without Ventilation, 4.2 Ventilation Systems in Battery Rooms)
5. 5. Conclusions

Related standards

• IEC 62485-2
• EN 62485-2
• NFPA 1
• IEEE 1635

Read the document

Publication details

Authors

Dorota Brzezińska

Publisher

Energies (MDPI AG)

Year

2018

Volume

11

Issue

8

Pages

2086

Document language

English

Source type

Scientific

Access

Open licence — downloadable

DOI

10.3390/en11082086

Licence

CC BY 4.0 — share with attribution

Licence evidence

www.mdpi.com

Citation