Fighting Fires Without the Water
by Leigh Soutter and Paul Schreier
Figure 1: United Technologies Research Center UTRC of East Hartford (Connecticut) says the only thing that makes fire worse for computers, documents, and valuable art is adding water to extinguish the flames. UTRC designs fire- suppression devices that suffocate flames with non-combustible gas instead of dousing them with water. Orchestrating complex suppression systems like this has required multiple simulation programs along with controls software using conventional methods. However, Dr. Mike Dorobantu from UTRC and Dr. Jesper Oppelstrup from the Royal Institute of Technology KTH of Stockholm (Sweden) set up the physics and the control algorithms in one COMSOL model.
United Technologies Corporation (UTC) designs the knobs we turn everyday - the ones that control air conditioners, power and electronic equipment, satellites, and airplanes produced by companies like OTIS, Carrier, UTC Fire & Security, UTC Power, Pratt-Whitney, Sikorsky, and Hamilton-Sundstrand. Almost everyone reading this page probably has used something touched by UTC. UTRC is the innovation center of UTC and is responsible for developing novel technologies and design methodologies.
Fire-supression system
Figure 2. The inert gas fire suppression system designed by UTRC and protections against overpressuring the room with gas. Click image to enlarge.
UTRC brings this vast experience to bear on designing ways to snuff out fires with inert gas (Figure 2). Sensors check the room for high temperatures and chemical signs of fire. When the sensor detects a fire, it triggers the release of a non-combustible non-toxic gas from a tank. The gas passes through a nozzle, expanding to as low as -78 °C, with the flood of the gas forcing oxygen from the room to drop the oxygen concentrations to or below 12%. The inert gas is non-destructive, reduces temperatures, and also fights fires in ventilation systems or behind walls, where sprinklers cannot touch.
If your room is on fire, getting as much inert gas to the flames as fast as possible might seem tempting. In reality the overkill can produce a spiral of problems that actually adds to the fire. "Releasing the gas too quickly pressurizes the room. The windows can blow out and the fire spreads. The classical remedy is a large relief vent in the room, but that makes the whole system unattractive to the tenants" explains Dr. Dorobantu.
COMSOL Modeling
Figure 3. COMSOL models of the inert-gas fire suppression combine multiple physics, systems controls ODE, and validation all in a single model file. Click image to enlarge.
To design the fire-suppression systems, Dr. Dorobantu teamed up with Dr. Oppelstrup, passing COMSOL Multiphysics models back and forth overseas. For their work, a single COMSOL file contains a number of different physics models which they connect and control with ODEs (Figure 3). The physics account for the gas discharge, the room pressures, chemical reactions, and heat transfer. The control ODEs are equations entered directly in the GUI that switch the various physics simulations on and off.
To test out different room and suppressor designs, they move the sensors, the vents, and the gas release nozzles. They also change the shape of the room or add furniture to it. For example, the animations below (Figure 4) show temperatures, concentrations, and gas velocities for a room with one vent and two gas nozzles.
 Play movie |
 Play movie |
Figure 4. COMSOL animations for the inert-gas fire suppression model: (A) temperatures, and (B) oxygen concentrations and velocities.
About Dr. Dorobantu
Dr. Mike Dorobantu works at UTRC within the Modeling, Analysis, Simulation, & Computation (MASC) initiative. Their goal is to make United Technologies a world leader in modeling and analysis for products and processes, with respect to development time and costs, system performance and robustness, and product quality and reliability. The three key elements to achieve this goal include: (1) developing rigorous system-level models of products and processes, (2) implementing these models through engineering standard work in UTC product development, and (3) creating a robust product development process which substantially reduces use of rigs and prototyping.
The way he sees it, COMSOL modeling supports the goals of the MASC initiative. According to Dr. Dorobantu, "My work with Dr. Oppelstrup was mostly overseas exchanges that would have been almost impossible had we been writing our own finite element code. But using a common tool, COMSOL Multiphysics, made it easy to exchange ideas about the modeling and how the physics interact."