Fire Protection Systems for Rail Transportation of Class A Explosives. Interim Report.
Fire Protection Systems for Rail Transportation of Class
A Explosives. Interim Report.
Bukowski, R. W.
NBSIR 80-2170; 30 p. November 1980.
Sponsor:Department of Transportation, Washington, DC
Available from: National Technical Information Service
(NTIS), Technology Administration, U.S. Department of
Commerce, Springfield, VA 22161.
1-800-553-6847 or 703-605-6000;
Fax: 703-605-6900; Rush
Service (Telephone Orders Only) 800-553-6847;
Order number: PB81-153975
bombs (ordnance); computer models; fire detection
systems; fire suppression; large scale fire tests; heat
transfer; railroad accidents; small scale fire tests;
As a result of several accidents involving fire induced
detonation of military explosive during rail shipment, a
research project, funded by the Federal Railroad
Administration (FRA), was initiated at the Center for
Fire Research (CFR) at the National Bureau of Standards
(NBS). This project was initiated to evaluate various
methods of protection of Class A explosives from fire,
and to identify one or more cost-effective approaches
which could be explored in greater detail in later
studies. Active systems (detection, notification, and
extinguishment) and passive systems (thermal insulating
barriers) were evaluated regarding cost, feasibility and
level of protection provided for the major hazard
scenarios involved in rail shipment of explosives. The
passive, thermal barrier approach was selected as the
most reliable and less costly of the options studied
while providing an acceptable level of protection.
Small-scale and full-scale tests were conducted to
obtain performance data on one specific thermal barrier
material. Based on this data, a computer model was
developed which can predict temperatures of the boxcar
floor, top surface temperature of a thermal barrier, and
casing/explosive interface temperature of a wood-pallet
mounted bomb for a range of fire sizes. The model
predications compare favorably with measured results
from a limited number of experiments. Further
experimental data are needed to refine the model and
establish an acceptable confidence level in the
predicted values. The proposed work necessary to
provide this refinement and verification is described.