Towards the Enhancement of Aircraft Cargo CompartmentFire Detection System Certification using

Smoke Transport Modeling

Walt Gill and Jill Suo-Anttila

Fire Science and Technology Department

Sandia National LaboratoriesAlbuquerque, NM

David BlakeFire Safety Section

FAA Technical Center

International Fire and Cabin Safety Research ConferenceNovember 2004

Sandia is a multi-program laboratory operated by Sandia Corporation, a Lockheed-Martin Company, for the United States Department of Energy under Contract DE-AC04-94AL85000.

Sandia National Laboratories Team Members

• Experimental – David Blake, Walt Gill, and Jill Suo-Anttila

• Model Development – Jim Nelsen and Stefan Domino

• Graphical User Interface and Code Development– Carlos Gallegos

• Technical Support– Louis Gritzo, manager of the Fire Science and

Technology Department

Modeling Smoke Transport in Aircraft Cargo Compartments

Goal: Develop a CFD-based simulation tool to predict smoke transport in cargo compartments

• Improve the certification process– Identify optimum smoke detector locations– Specify sensor alarm levels – Identify most challenging fire locations– Reduce the number of flight tests

• Fast running • Suitable for non-expert users • Experimental data for source term

characterization from FAA experiments• Validated using FAA full-scale experiments

Built on firm FAA knowledge base

Validated using FAA experiments

Airlines, Air-Framers, Certifiers

Robust and fast running

Software Design

Graphical User Interface

Pre-Processor

Analysis Module

Post-Processor

pmj

jS)u(

xt

Pre-Processor Overview

• Provide models for different aircraft

• Boeing 707, 727, 747, etc.

• User defined

• Capabilities

• Refine mesh

• Enter fire(s) location and type

• Enter ventilation velocities and locations

• Enter compartment temperature and pressure

• Add obstacles and recessed areas

• Instantaneous visual feedback

Running a Simulation Compartment and Mesh Specification

• Execute the Pre-Processor

• Select the type of compartment

– 707

– DC-10

– User Defined

• Input the dimensions

• Enter the mesh size - # of nodes

FAA_PreProcessor.lnk

custom

707 or DC-10

Running a Simulation Created 707 and DC-10 Meshes

• Automatically generated 707 mesh

• Curvature captured by mesh

• Right side of screen shows selected plane

• Automatically generated DC-10 mesh

• Internal view of compartment

Running a Simulation Recessed Area Specification

1. Advance to selected Y-plane

2. Select desired cells

3. Perform operation using buttons

1

2

3

Running a Simulation Obstacle Specification

Obstacle

Recessed Area

Running a Simulation Ventilation and Fire Specification

1. Select cells

2. Enter type of cell (inlet, outlet, fire) – cell colored to denote type

3. Use table to enter ventilation properties

4. Fire properties in file

31

2

Fire

Inlet

Outlet

Running a Simulation Mesh Refinement Specification

1. Select the plane for refinement

2. Use refinement tool

3. Enter level of refinement

12

3

Resulting Grid

Running a Simulation Running the Analysis Code

• Analysis - - - Run Analysis

• Status monitored on screen

Smoke Transport Analysis Code

• Curvature of compartment is resolved on grid

• HRR, MLR are time varying inputs (as measured in FAA experiments)

• Species tracking: presently soot, CO, and CO2 but addition of more or different species possible

• Simulation time = 1 hour per minute of real time

• Validated using FAA full-scale experiments

computational computational grid cell on wallgrid cell on wall

cellmass VMS /

cellener VQS /

180 240120600

12

10

8

6

4

2

0

Co

nce

ntr

atio

n C

O2,

CO

, HC

l, H

2O

(pa

rts

pe

r th

ou

san

d)

Time (seconds)

Co

nce

ntra

tion

All O

the

r Ga

ses

(pa

rts pe

r millio

n)

120

100

80

60

40

20

0

CO2

CO

H20

HCl

HCN

NO

ethylene

acteylene

methane

HCl

Temp (K)

Post-Processor

Allow users to manipulate data in a variety of ways • contour plots

• time history of field variables

• 3D smoke visualization in time

Code Validation Metrics

Insert most recent movie of temperature distribution

• Thermocouple temperature rise

– 0 - 60 seconds

– 0 -120 seconds

– 0 -180 seconds

• Light transmission– 30 and 45 sec (ceiling and vertical)

– 60 sec (vertical - high, mid, low)

– 120 sec (vertical - mid and low)

– 180 sec (vertical - mid and low)

• Gas species concentration rises– 0 - 60 seconds

– 0 -120 seconds

– 0 -180 seconds

Experimental ceiling temperature distribution at 60 sec

Computational temperature distribution at 60 sec

Status of FAA Full-Scale Validation Experiments

Insert most recent movie of temperature distribution

• 707 experiments completed– Baseline – center fire

– Attached – sidewall fire

– Corner – corner fire

– Determined leakage ventilation had no impact on data

– All 707 experiments were conducted without ventilation

• DC-10 experiments– Ventilation validation

 

707 Validation Simulations

Insert most recent movie of temperature distribution

• Interface described used to create mesh and run simulation

• Results and comparisons follow

Internal view (showing fire and recessed areas) of 707 computational domain

707 Baseline computational mesh

Preliminary Validation – Temperature

Insert most recent movie of temperature distribution

• Baseline 707 experiments– center fire– 40 thermocouples– Model including heat transfer to

the ceiling and walls – h=7 W/m2K in model

• Comparison– Trends captured– Magnitudes predicted away

from fire – Magnitudes agree above fire

better at early times

 

Phase 2 - Temperatures at 60 sec

290

295

300

305

310

315

320

0 10 20 30 40

Thermocouple Number

Tem

per

atu

re (

K)

Exp_60s (K)

Model

Phase 2 - Temperatures at 120 sec

290

295

300

305

310

315

320

325

330

335

0 10 20 30 40

Thermocouple Number

Tem

per

atu

re (

K)

Exp_120s (K)

Model

Preliminary Validation – Light Transmission

Insert most recent movie of temperature distribution

Phase 2 - Smokemeters at 30 sec

70

75

80

85

90

95

100

0 1 2 3 4 5 6 7

Smokemeter (Fwd, Mid, Aft, High, Mid, Low)

% L

igh

t T

ran

smis

sio

n

EXP_30s

Model

Phase 2 - Smokemeters at 45 sec

70

75

80

85

90

95

100

0 1 2 3 4 5 6 7

Smokemeter (Fwd, Mid, Aft, High, Mid, Low)

% L

igh

t T

ran

smis

sio

n

EXP_45s

Model

Phase 2 - Smokemeters at 60 sec

70

75

80

85

90

95

100

0 1 2 3 4 5 6 7

Smokemeter (Fwd, Mid, Aft, High, Mid, Low)

% L

igh

t T

ran

smis

sio

n

EXP_60s

Model

• Baseline 707 experiments– center fire– 6 smoke meters

• Comparison– Good agreement in trends and

magnitudes

• Continue validation of the smoke transport code– Finish code modifications– 707 validation comparisons– DC-10 validation comparison

• Release of code to small user community– Includes theory and users manual– Tutorial at FCS conference

• Revisions and final release of code (Summer ’05)

Future Activities

The Fourth Triennial The Fourth Triennial International Aircraft Fire and Cabin Safety International Aircraft Fire and Cabin Safety Research ConferenceResearch Conference