Technical study · Engineering & Design

Thermal Runaway Vent Gas in a Battery Enclosure

CFD modelling in COMSOL to test whether a battery enclosure's ventilation keeps flammable vent gas below safe limits if a cell goes into thermal runaway, and the simplest fix if it doesn't.

Technician inspecting rows of industrial battery cells

What the client needed

A battery integrator delivering containerised battery systems in Australia came to us after a fire engineer reviewing the design asked whether the enclosure ventilation would keep flammable vent gas below safe limits if a single cell went into thermal runaway. They needed an answer quickly, without an expensive product redesign.

The question they needed answered: If a cell vents, does the ventilation keep gas below 25% of the lower flammability limit (LFL)? If not, what is the simplest change that does?

  • 25% LFLdesign limit met with the revised layout
  • 0extra fan capacity needed
  • 1 paneldrawing changed, across the whole product line

What we did

  • Defined the vent gas release rate and composition from UL 9540A cell test data. Vent gas is typically rich in hydrogen, carbon monoxide and light hydrocarbons.
  • Built a CFD model of the enclosure in COMSOL Multiphysics, including racks, inlet and exhaust, buoyancy and gas mixing.
  • Ran the release for the original ventilation layout, then tested other inlet and exhaust positions and flow rates.
  • Checked results against NFPA 69 and NFPA 68 criteria and the requirements of NFPA 855.

What we found

With the inlet and exhaust on the same wall, fresh air short-circuits straight to the exhaust. Buoyant vent gas forms a layer under the roof, above 25% LFL, in the part of the enclosure furthest from the fan.

Moving the inlet low on the opposite wall sets up a sweeping flow along the ceiling to a high-level exhaust. Gas above 25% LFL is then limited to a small region near the release and at the exhaust itself.

Doubling the fan rate in the original layout helps far less than moving the inlet, and adds cost, noise and auxiliary load for the life of the asset.

Deflagration vent panels are still needed as a second layer of protection. The model shows where gas collects, which sets where the panels go.

Two CFD sections of a battery enclosure comparing vent gas concentration for the original and revised ventilation layouts
Vent gas 60 seconds after release, original layout (top) and revised layout (bottom). Representative 2D section; the study model is 3D and transient.

What we recommended

  • Move the inlet louvre low on the opposite end wall and keep the exhaust high. No increase in fan capacity.
  • Start the exhaust on gas detection (hydrogen and CO) as well as temperature, so ventilation starts at the first sign of off-gassing.
  • Put deflagration vent panels in the roof area where the model shows gas collecting.
  • Keep the validated model as a design tool for future enclosure variants, instead of repeating physical tests for every change.

Key assumptions

Vent gas composition and release rate based on representative UL 9540A test data for LFP cells. Standards referenced: NFPA 855, NFPA 68 and NFPA 69.

Related work

Talk to our engineers

Tell us what you're working on, and one of the people below will get back to you.

Get in touch+61 2 8034 7980[email protected]

Talk directly to

  • Andrew LelliottOperations Director
  • Alistair WardropeTechnical Director

Want to join us? We're always keen to hear from engineers who value strong fundamentals and real delivery experience.

Careers at Amarna