Use of Computer Models to Predict Temperature and Smoke Movement in High Bay Spaces.
Use of Computer Models to Predict Temperature and Smoke
Movement in High Bay Spaces.
(4935 K)
Notarianni, K. A.; Davis, W. D.
NISTIR 5304; 64 p. December 1993.
Sponsor:
General Services Administration, Washington, DC
Available from:
National Technical Information Service
Order number: PB94-145976
Keywords:
computer models; temperature; smoke movement; clean
rooms; computational fluid dynamics; detector response;
field modeling; fire detection; fire models; forced air
flow; fire plumes; fire tests; high bays; response time;
sprinkler response
Abstract:
The Building and Fire Research Laboratory (BFRL) was
given the opportunity to make measurements during fire
calibration tests of the heat detection system in an
aircraft hangar with a nominal 30.4 m (100 ft) ceiling
height near Dallas, TX. Fire gas temperatures resulting
from an approximately 8250 kW isopropyl alchol pool fire
were measured above the fire and along the ceiling. The
results of the experiments were then compared to
predictions from the computer fire models DETACT-QS,
FPETOOL and LAVENT. In section A of the analysis
conducted, DETACT-QS and FPETOOL significantly
underpredicted the gas temperatures. LAVENT at the
position below the ceiling corresponding to maximum
temperature and velocity provided better agreement with
the data. For large spaces, hot gas transport time and
an improved fire plume dynamics model should be
incorporated into the computer fire model activation
routines. A computational fluid dynamics (CFD) model,
HARWELL FLOW3D, was then used to model the hot gas
movement in the space. Reasonable agreement was found
between the temperatures predicted from the CFD
calculations and the temperatures measured in the
aircraft hangar. In section B, an existing NASA high
bay space was modelled using the computational fluid
dynamics model. The NASA space was a clean room, 27.5 m
(90 ft) high with forced horizontal laminar flow. The
purpose of this analysis is to determine how the
existing fire detection devices would respond to various
size fires in the space. The analysis was conducted for
32 MW, 400 kW, and 40 kW fires.
Building and Fire Research Laboratory
National Institute of Standards and Technology
Gaithersburg, MD 20899