~ibliography ofThermal tion Data for FireJohn R. Howell, K~Daun.and Hakan Erturk Prepared by Sandia National Laboratories Albuquerque, NewMexico 87185 and Uvennore, Califomia 94550

SAND REPORTSAND2001-3320Unlimited ReleasePrinted November, 2001

....~ibliography of Thermaltion Data for Fire

John R. Howell, K~Daun. and Hakan Erturk

Prepared bySandia National LaboratoriesAlbuquerque, New Mexico 87185 and Uvennore, Califomia 94550

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

Approved for publtc release; further dissemination unlimited.

(....) Sandia National Laboratories

Report Documentation Page Form ApprovedOMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.

1. REPORT DATE NOV 2001 2. REPORT TYPE

3. DATES COVERED 00-00-2001 to 00-00-2001

4. TITLE AND SUBTITLE An Annotated Bibliography of Thermal Radiation Validation Data forFire Applications

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Department of Mechanical Engineering,The University of Texas at Austin,Austin,TX,78712

8. PERFORMING ORGANIZATIONREPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES

14. ABSTRACT

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT Same as

Report (SAR)

18. NUMBEROF PAGES

96

19a. NAME OFRESPONSIBLE PERSON

a. REPORT unclassified

b. ABSTRACT unclassified

c. THIS PAGE unclassified

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

Issued by Sandia National Laboratories, operated for the United StatesDepartment of Energy by Sandia Corporation.

NOTICE: This report was prepared as an account of work sponsored by anagency of the United States Government. Neither the United States Government,nor any agency thereof, nor any of their employees, nor any of their contractors,subcontractors, or their employees, make any warranty, express or implied, orassume any legal liability or responsibility for the accuracy, completeness, orusefulness of any information, apparatus, product, or process disclosed, orrepresent that its use would not infringe privately owned rights. Reference hereinto any specific commercial product, process, or service by trade name,trademark, manufacturer, or otherwise, does not necessarily constitute or implyits endorsement, recommendation, or favoring by the United States Government,any agency thereof, or any of their contractors or subcontractors. The views andopinions expressed herein do not necessarily state or reflect those of the UnitedStates Government, any agency thereof, or any of their contractors.

Printed in the United States of America. This report has been reproduced directlyfrom the best available copy.

Available to DOE and DOE contractors fromU.S. Department of EnergyOffice of Scientific and Technical InformationP.O. Box 62Oak Ridge, TN 37831

Telephone: (8651576-8401Facsimile: (865)576-5728E-Mail: [email protected] ordering: http://www.doe.gov/bridge

Available to the public fromU.S. Department of CommerceNational Technical Information Service5285 Port Royal RdSpringfield, VA 22161

Telephone: (800)553-6847Facsimile: (703)605-6900E-Mail: orderslamtls.fedworld.govOnline order: http://www.ntis.gov/ordering.htm

2

SAND2001-3320Unlimited Release

Printed October 200 1

An Annotated Bibliography of Thermal RadiationValidation Data for Fire Applications*

John R. Howell, Kyle Daun, and Hakan Erturk

Department of Mechanical EngineeringThe University of Texas at Austin

Austin, TX 78712

AbstractThis report details experimental data useful in validating radiative transfer codes

involving participating media, particularly for cases involving combustion. Specialemphasis is on data for pool fires.

Features sought in the references are: Flame geometry and fuel that approximateconditions for a pool fire or a well-defined flame geometry and characteristics that can becompletely modeled; detailed information that could be used as code input data,including species concentration and temperature profiles and associated absorptioncoefficients, soot morphology and concentration profiles, associated scatteringcoefficients and phase functions, specification of system geometry, and system boundaryconditions; detailed information that could be compared against code output predictions,including measured boundary radiative energy flux distributions (preferably spectral)and/or boundary temperature distributions; and a careful experimental error analysis sothat code predictions could be rationally compared with experimental measurements.

Reference data were gathered from more than 35 persons known to be active in thefield of radiative transfer and combustion, particularly in experimental work. A literaturesearch was carried out using key words. Additionally, the reference lists in papers/reportswere pursued for additional leads.

The report presents extended abstracts of the cited references, with comments onavailable and missing data for code validation, and comments on reported error. Agraphic for quick reference is added to each abstract that indicates the completeness ofdata and how well the data mimics a large-scale pool fire. The references are organizedinto Lab-Scale Pool Fires, Large-Scale Pool Fires, Momentum-Driven Diffusion Flames,and Enclosure Fires. As an additional aid to report users, the Tables in Appendix A showthe types of data included in each reference. The organization of the tables follows thatused for the abstracts.

??Sandia Purchase Order #15857

3

Table of Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IO

LAB-SCALE POOL FIRES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Bouhafid, A., Vantelon, J. P., Joulain, P., and Fernandez-Pello, A. C., 1988, “Onthe Flame Structure at the Base of a Pool Fire,” Proc. 22nd Symposium(International) on Combustion, The Combustion Institute, pp. 1291-1298. . . . . . . . . . . . 14

Burgess, D., and Hertzberg, M., 1974, “Radiation from Pool Flames,” from HeatTransfer in Flames, N.H. Afgan and J.M. Beer, Eds., Scripta Book Co.,Washington D. C., Ch. 27, pp. 413-430 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..*.......................... 15

Choi, M. Y., Hamins, A., Rushmeier, H., and Kashiwagi, T., 1994, “SimultaneousOptical Measurement of Soot Volume Fraction, Temperature, and CO* in HeptanePool Fire,” Proc. 25*h Symposium (International) on Combustion, The CombustionInstitute, pp. 1471-1480 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Crauford, N. L., Liew, S. K., and Moss, J. B., 1985, “Experimental and NumericalSimulation of a Buoyant Fire,” Combustion and Flame, Vol. 61, pp. 63-77. . . . . . . . . . . . 17

Fischer, S., Hardouin-Duparc, B., and Grosshandler, W., 1987, “The Structure andRadiation of an Ethanol Pool Fire,” Combustion and Flame, Vol. 70, pp. 291-306.18

Hamins, A., Klassen, M. E., Gore, J. P., Fischer, S. J., and Kashiwagi, T., 1994,“Heat Feedback to the Fuel Surface of Pool Fires,” Combustion Science andTechnology, Vol. 97, pp. 37-62 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Hamins, A., Klassen, M., Gore, J., and Kashiwagi, T., 1991, “Estimate of FlameRadiance via a Single Location Measurement in Liquid Pool Fires,” Combustionand Flame, Vol. 86, pp. 223-228 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Hogben, C. D. A., Young, C. N., Weckman, E. J., and Strong, A. B., 1999,“Radiative Properties of Acetone Pool Fires,” Proc. 1999 National Heat TransferConference, Albuquerque, NM, August. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Inamura, T., Saito, K., and Tagavi, K. A., 1992, “A Study of Boilover in LiquidPool Fires Supported on Water. Part II: Effects of In-depth RadiationAbsorption,” Combustion Science and Technology, Vol. 86, pp. 105-119. . . . . . . . . . . . . . . . 22

Joulain, P., 1996, “Convective and Radiative Transport in Pool and Wall Fires: 20Years of Research in Poitiers,” Fire Safety Journal, Vol. 26, pp. 99-149 . . . . . . . . . . . . . . . . -23

Klassen, M., Gore, J. P., and Sivathanu, Y. R., 1992, “Radiative Heat Feedback ina Toluene Pool Fire,” Proc. 24th Symposium (International) on Combustion, TheCombustion Institute, pp. 1713-1719 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Koseki, H., 1994, “Boilover and Crude Oil Fire,” Journal of Applied Fire Science,Vol. 3, no. 3, pp.243-271 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Koseki, H., and Yumoto, T., 1988b, ‘&Air Entrainment and Thermal Radiationfrom Heptane Pool Fires,” Fire Technology, Vol. 24, pp. 33-47 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

4

Markstein, G. H., 1981, “Scanning-Radiometer Measurements of the RadianceDistribution in PMMA Pool Fires,” Proc. lth Symposium (International) onCombustion, The Combustion Institute, pp. 537-547. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Modak, A. T., and Croce, P. A., 1977, “Plastic Pool Fires,” Combustion and Flame,Vol. 30, pp. 251-265. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Orloff, L., 1981, “Simplified Modeling of Pool Fires,” Proc. lgh Symposium(International) on Combustion, The Combustion Institute, pp. 549-561.................. 29

Shinotake, A., Koda, S., and Akita, K., 1985, “An Experimental Study of RadiativeProperties of Pool Fires of an Intermediate Scale,” Combustion Science andTechnology, Vol. 43, pp. 85-97. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Sivathanu, Y.R. and Gore, J.P., 1992, “Transient Structure and RadiationProperties of Strongly Radiating Buoyant Flames,” Journal of Heat Transfer, Vol.114, pp. 659-665. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Weckman, E.J., and McEwen, C.S., 1991, “The Time Dependent Structure of aMedium-Scale Methanol Pool Fire,” Proceedings of the ASME/JSME JointConference, 1991, Vol. 5, pp. 270-275. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Weckman, E. J., and Strong, A. B, 1996, “Experimental Investigation of theTurbulence Structure of Medium-Scale Pool Fires,” Combustion and Flame, Vol.105, pp. 245-266 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Zhang, X. L., Vantelon, J. P., and Joulain, P., “Thermal Radiation from a Small-Scale Pool Fire: Influence of Externally Applied Radiation,” Combustion andFlame, Vol. 92, pp. 71-84 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Zhang, X. L., Vantelon, J. P., Joulain, P. and Fernandez-Pello, A. C., 1991,“Influence of an External Radiant Flux on a 15-cm-Diameter Kerosene Pool Fire,”Combustion and Flame, Vol. 86, pp. 237-248 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

LARGE-SCALE POOL FIRES . . . . . . . . . . . . . . . ..*....................................*.................... 36

Brosmer, M. A., and Tien, C. L., 1987, “Radiative Energy Blockage in Large PoolFires,” Combustion Science and Technology, Vol. 51, pp. 21-37. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Fire and Blast Engineering Project, Fire and Blast Information Group (FABIG),UK Steel Construction Institute, Final Report, 1997. ..*.............................................38

Gregory, J. J., Keltner, N. R., and Mata, R., 1989, “Thermal Measurements inLarge Pool Fires,” Journal of Heat Transfer, Vol. 111, pp. 446-454. . . . . . . . . . . . . . . . . . . . . . . . 39

Gritzo, L. A., Gill, W., and Nicolette, V. F., 1998, “Estimates of the Extent andCharacter of the Oxygen-Starved Interior in Large Pool Fires,” from Very Large-Scale Fires, Keltner, N. R, Alvares, N. J., and Grayson, S. J. Eds., ASTMSpeciaZTechnical Publication 1336, pp. 84-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

Koseki, H., 1999, CCLarge Scale Pool Fires: Results of Recent Experiments,” FireSafety Science, Vol. 6., pp. 115-132 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Koseki, H., Iwata. Y., Natsume, Y., Takahashi, T. and Hirano, T., 2000,“Tomakomai Large Scale Crude Oil Fire Experiments,” Fire Technology, Vol. 36,no. 1, pp.24-38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Koseki, H., and Yumoto, T., 1988, “Burning Characteristics of Heptane in 2.7 mSquare Dike Fires,” Fire Safety Science - Proceedings of the Second InternationalSymposium, pp. 231-240 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Kramer, M. A., Greiner, M., Koski, J. A., Lopez, C., and Sho-Antila, A., 2001,“Measurement of Heat Transfer to a Massive Cylindrical Object Engulfed in aRegulatory Pool Fire,” Presented at the 2001 ASME National Heat TransferConference, Anaheim CA., June lo-12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

Nakos, J. T., Gill, W., and Keitnerm N. R., 1991, “An Analysis of FlameTemperature Measurements using Sheathed Thermocouples in JP-4 Pool Fires,”Proc, 1991 ASME/JAME Thermal Engineering Conference, March 1991. . . . . . . . . . . . . . . 45

Russell, L. H., and Canfieid, J. A., 1973, “Experimental Measurement of HeatTransfer to a Cylinder Immersed in a Large Aviation-Fuel Fire,” Journal of HeatTransfer, Vol. 95 C, pp. 397-404 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

MOMENTUM-DRIVEN DIFFUSION FLAMES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

Bennett, B. A., McEnally, C. S., Pfefferle, L. D., and Smooke, M. D., 2000,“Computational and Experimental Study of Axisymmetric Coflow PartiallyPremixed Methane/Air Flames,” Combustion and Flame, Vol. 123, pp. 522-546. . . 48

Delichatsios, M. A., and Orloff, L., 1988, “Effects of Turbulence on FlameRadiation from Diffusion Flames,” Proc. 22”d Symposium (International) onCombustion, The Combustion Institute, pp. 1271-1279. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Faeth, G. M., Gore, J. P., Chuech, S. G., and Jeng, S. M., 1989, “Radiation fromTurbulent Diffusion Flames,” from Annual Review of Numerical Fluid Mechanicsand Heat Transfer Vol. 2, Tien, C. L. and Chawla, T. C. Eds., HemispherePublishing Corp., New York, U. S. A., pp. l-37 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Faeth, G. M., Gore, J. P., and Sivathanu, Y. R., 1988, “Radiation from Soot-Containing Flames,” from Proceedings of AGARD Conference No. 422 Combustionand Fuels in Gas Turbine Engines, North Atlantic Treaty Organization, pp. 17-1-17-11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 51

Gore, J. P., and Faeth, G. M., 1986, “Structure and Spectral Radiation Propertiesof Turbulent Ethylene/Air Diffusion Flames,” Proc. 21th Symposium (International)on Combustion, The Combustion Institute, pp. 1521-1531 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Gore, J. P., Faeth, G. M., Evans, D., and Pfenning, D. B., 1986, “Structure andRadiation Properties of Large-scale Natural Gas/Air Diffusion Flames,” Fire andMaterials, Vol. 10, pp. 161-169 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Gore, J. P., Jeng, S. -M., and Faeth, G. M., 1987, “Spectral and Total RadiationProperties in Hydrogen/Air Diffusion Flames,” Journal of Heat Transfer, Vol. 109,pp. 165-171 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

I 6

Gilder, @1992, “Soot Formation in Laminar Diffusion Flames at ElevatedTemperatures,” Combustion and Flame, Vol. 88, pp. 74-82 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Giilder, 0. L., Thompson, K. A., and Snelling, D. R., 2000, “Influence of the FuelNozzle Material on Soot Formation and Temperature Field in Coflow LaminarDiffusion Flames,” Proceedings, Combustion Institute Canadian Section, SpringTechnical Meetings, Ottawa ON. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Jeng, S. M., and Faeth, G. M., 1984, “Species Concentrations and TurbulenceProperties in Buoyant Methane Diffusion Flames,” Journal of Heat Transfer, Vol.106, pp. 721-727 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Jeng, S. M., and Faeth, G. M., 1984, “Radiative Heat Fluxes near TurbulentBuoyant Methane Diffusion Flames,” Journal of Heat Transfer, Vol. 106, pp. 886-888. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........... 58

Jeng, S. M., Chen, L. D., and Faeth, G. M., 1984, “Nonluminous Radiation inTurbulent Buoyant Axisymmetric Flames,” Combustion Science and Technology,Vol. 40, pp. 41-53. . . . ..*.................................................................................................... 59

Jeng, S. M., Chen, L. D., and Faeth, G. M., 1982, “The Structure of BuoyantMethane and Propane Diffusion Flames, “) Proc. lYh Symposium (International) onCombustion, The Combustion Institute, pp. 349-358 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Kennedy, I. M., Yam, C., Rapp, D. C., and Santoro, R. J., 1996, “Modeling andMeasurements of Soot and Species in a Laminar Diffusion Flame,” Combustionand Flame, Vol. 107, pp. 368-382 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Liu, F., Guo. H., Smallwood, G.J. and Giilder, ii., 2001, “Effects of Gas and SootRadiation on Soot Formation in a CoFlow Laminar Ethylene Diffusion Flame,” inRadiation III: Third International Symposium on RadiativeTransfer, M. PinarMengiic and N. Selcuk, eds., Begell House, New York, 2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Markstein, G. H., 1977, “Scaling of Radiative Characteristics of TurbulentDiffusion Flames,” Proc. l6’h Symposium (International) on Combustion, TheCombustion Institute, pp. 1407-1419 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

Mbiock, A., Teerling, J., Roekaerts, D., and Merci, B., “Application of the BEMand Analysis of the Role of Radiation Effects in Labscale Turbulent DiffusiuonFlames,” in Radiation III: Third International Symposium on RadiativeTransfer,M. Pinar Mengiic and N. Selcuk, eds, Begell House, New York, 2001..................... 64

Sandia-Livermore website with Data Archive:http://www.ca.sandia.gov/tdf/Workshop/DatDwnld.html . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Sivathanu, Y. R., Gore, J. P., 1993, “Total Radiative Heat Loss in Jet Flames fromSingle Point Radiative Flux Measurements,” Combustion and Flame, Vol. 94, pp.265-270 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 66

Sivathanu, Y. R., Kounalakis, M. E., and Faeth, G. M., 1990, “Soot and ContinuumRadiation Statistics of Luminous Turbulent Diffusion Flames,” Proc. 23rdSymposium (International) on Combustion, The Combustion Institute, pp. 1543-

I 1550 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......... 67

7

Snelling, D. IX., Thomson, K., Smallwood, G. J., Weckman, E., Fraser, R., andGtilder, 0. L., 1999, “Soot Surface Temperature Measurements in a LaminarDiffusion Flame,,, Proceedings, Combustion Institute Canadian Section, SpringTechnical Meetings, Edmonton AB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Souil, J. M., Joulain, P. and Gengembre, E., 1984, “Experimental and TheoreticalStudy of Thermal Radiation from Turbulent Diffusion Flames to Vertical TargetSurfaces,,, Combustion Science and Technology, Vol. 41, pp. 69-Sl......................... 69

You, H-Z., and Faeth, G. M., 1982, “Buoyant Axisymmetric Turbulent DiffusionFlames in Still Air,,, Combustion and Flame, Vol. 44, pp. 261-275 . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 0

ENCLOSURE FIRES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Brehob, E. G., and Kulkarni, A. K., 1998, “Experimental Measurements ofUpward Flame Spread on a Vertical Wall with External Radiation,,, Fire SafetyJournal, Vol. 31, pp. 181-200 . . . . . . ..**................*..........*.........*..............*.........*...............72

Butler, B.W. and Webb, B.W., 1993 “Measurement of Radiant Heat Flux andLocal Particle and Gas Temperatures in a Pulverized Coal-Fired Utility-ScaleBoiler,,’ Energy and Fuels, Vol. 7, pp. 835-841 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Butler, B. W. and Webb, B. W., 1991, “Local Temperature and Wall Radiant HeatFlux Measurements In an Industrial Scale Coal Fired Burner,,, Fuel, Vol. 70, pp.1457-1464. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Dembsey, N. A., Pagni, P. J., and Williamson, R. B., 1995, “Compartment FireExperiments: Comparison with Models,,, Fire Safety Journal, Vol. 25, pp. 187-227.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................. 75

Foley, M., and Drysdale, D. D., 1995, “Heat Transfer from Flames BetweenVertical Parallel Walls,,, Fire Safety Journal, Vol. 24, pp. 53-73 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Gengembre, E., Cambray, P., Karmed, D., and Bellet, J. C., 1984, “TurbulentDiffusion Flames with Large Buoyancy Effects,,, Combustion and Flame, Vol. 41,pp. 55-67. . . . . . ..*................................................................................................................ 77

Hwang, Yuh-Long and Howell, John R., 2001, “Local Furnace Data and ModelingComparison for a 600MWe Coal-Fired Utility Boiler,,, accepted, Journal of EnergyResources Technology, March. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Ingason, H., 1998, “Modeling of a Two-Dimensional Rack Storage Fire,,’ FireSafety Journal, Vol. 30, pp. 47-69. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

Ingason, H. and de Ris, J., 1998, “Flame Heat Transfer in Storage Geometries,,’Fire Safety Journal, Vol. 31, no. 1, pp. 30-60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Pavlovic, P., JoviC, L., Jovanovic, L., and Afgan, N. H., 1974, “Steady andUnsteady Heat Flux Measurement on the Screen Tube of a Power Boiler Furnace,,,from Heat Transfer in FZames, Afgan, N. H. and Beer J. M. Eds., Scripts Book Co.,Washington, U.S.A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

I 8

Spearpoint, M. J., and Dillon, S. E., 1999, “Flame Spread Model Progress:Enhancements and User Interface,” National Institute of Standards and TechnologyTechnical Report NIST GCR 99-782 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Tree, DR., Black, D.L., Rigby, J.R, McQuay, M.Q. and Webb, B.W., 1998,“Experimental Measurements in the BYU Controlled Proiile Reactor,” Progress inEnergy and Combustion Science, Vol. 24, pp. 355383 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..*............................84

APPENDIX A: DATA CONTENT OF REFERENCES ........................................ 90

Table 1: Summary of Lab-Scale Pool Fires References ......................................... 91

Rankings are from 1 (complete data) to 3 (sparse data) ........................................... 91

Table 2: Summary of Large-Scale Pool Fire References ........................................ 92

Table 3: Summary of Momentum-Driven Fire References ................................... 93

Table 4: Summary of Enclosure Fire References ................................................... 94

APPENDIX B: LIST OF CONTACTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

9

INTRODUCTION

This report details the results of a search for experimental data that can be used to

validate radiative transfer codes involving participating media, particularly for cases

involving combustion. Special emphasis is on data for pool fires.

To be useful for code validation, the following features were sought:

1. Flame geometry and fuel that approximate conditions for a pool fire or, lacking

that, a well-defined flame geometry and characteristics that can be completely

modeled.

2. Detailed information that could be used as code input data. This should include

species concentration and temperature profiles and associated absorption

coefficients, soot morphology and concentration profiles, associated scattering

coefficients and phase functions, specification of system geometry, and system

boundary conditions.

3 . Detailed information that could be compared against code output predictions. This

information would include measured boundary radiative energy flux distributions

(preferably spectral) and/or boundary temperature distributions.

4. A careful experimental error analysis so that code predictions using input data

from 2.) could be rationally compared with experimental measurements of 3.).

Reference data were gathered from two primary sources. First, over 35 persons known to

be active in the field of radiative transfer and combustion, particularly in experimental

work, were contacted with a request for references or reports that meet the above criteria

and/or leads to others who might provide such data. Second, a literature search was

carried out using key words, and additionally the reference lists in papers/reports were

pursued for additional leads.

Coupling between the fluid mechanics and the heat transfer, particularly in pool fires,

presents a challenge to both numerical analysts and experimentalists. It is a challenge for

experimentalists to simultaneously measure local velocities, species and soot

1 0

concentrations, and temperatures. It is unlikely that a complete input data set can be

obtained for a radiative transfer code that requires solution of an energy equation that

incorporates chemical energy generation, soot generation and morphology, advection,

conduction and radiative transfer. In addition, it is necessary to measure or compute all of

the necessary properties required in the governing equations. If only the radiative transfer

equation is to be solved for a given temperature profile and given radiative properties,

then the situation is considerably better, because the radiative transfer solution is then

considered independent of the fluid mechanics. However, even in this case, the local

species and soot concentrations must be known in order to calculate or estimate the

necessary radiative properties.

The material presented in this report consists chiefly of extended abstracts of the cited

references, with comments on available and missing data for code validation, and

comments on reported error. Comments are added on the suitability of each reference for

Sandia’s use in validation.

A graphic for quick reference is added to each abstract, with the vertical scale indicating

the completeness of data available for code input and data on radiative output for

comparison. The horizontal axis indicates how well the data mimics a large-scale pool

fire. These scales should be recognized as being quite qualitative, and that they reflect the

judgment of the authors of this report as to how well the material in the papers meets the

requirements for code validation. They are not a reflection of the quality of work reported

in the references, but only how well the material reported in the reference meets the

criteria for validation of the SIERRA/FUEGO or SIERRA/SYRINX code.

The references are organized into four sections: Lab-Scale Pool Fires, Large-Scale Pool

Fires, Momentum-Driven Diffusion Flames, and Enclosure Fires. Pool fires are all taken

to be buoyancy-driven, and are categorized by whether they experiments were conducted

indoors and were immune to wind effects (Lab-Scale) or outdoors (Large-Scale). The

lab-scale fires tend to be more easily studied in the experiments. Momentum-Driven

Difmsion Flames are even more easily controlled, characterized and studied

experimentally. For this reason, they provide excellent test cases for predictive code

verification even though they do not simulate the more chaotic conditions typical of pool

1 1

fiores. Enclosure fires are difficult to model and to experimentally quantify, and most of

the references do not include complete data for modeling purposes. However, some

references may have value and have been included. Clearly, this is not a perfect categorization,

but it serves the purposes of this report. Within each section, the abstracts are ordered

alphabetically.

As an additional aid to report users, the Tables in Appendix A show the types of data

included in each reference. The organization of the tables follows that used for the

abstracts..

The authors of this report consider the following references to be the most useful for code

validation:

Lab-Scale Pool Fires:

Fischer et al. (1987)

Zhang et al. (1993 and Zhang et al. (1991)

Large-Scale Pool Fires:

Kramer et al. (2001)

Momentum-Driven Diffusion Flames:

Faeth et al. (1989)

Faeth et al. (1988)

Gore et al. (1987)

You and Faeth (1982)

Each of these references has complete input data sets as well as measurements of

radiative flux. Except in one case, they have good experimental error analyses.

1 2

LAB-SCALE POOL FIRES

1 3

Bouhafid, A., Vantelon, J. P., Joulain, P., and Fernandez-Peilo, A. C.,1988, “On the Flame Structure at the Base of a Pool Fire,” Proc. 22ndSymposium (International) on Combustion, The Combustion Institute,pp. 1291-1298.

This paper presents the speciesconcentration, temperature, andmonochromatic absorption coefficient inthe base region of a moderately sizedkerosene pool fire. The experimentalapparatus consists of a 15 cm diameter, 3cm deep fire pan. The pan is filled withkerosene to a depth of 2 mm (1 mmbelow the pan lip) and this fuel level ismaintained throughout the experiment.The flame is confined in a test cell that isvented by natural convection, in order tominimize the effects of ambient drafts.

Temperature measurements are obtainedusing a thermocouple array. Thetemperature measurements are timeaveraged, and the variations in the meanare less than 5%. It is not clear if thethermocouple measurements have beencorrected for radiation effects.Composition measurements are made byfirst extracting samples from the fire,passing them through infrared analyzersthat measure CO and CO;!concentrations, and to a gaschromatograph that measures 02concentrations. The CO and COzmeasurements are averaged, and thevariation is less than 3%. The 02measurements have an uncertainty of6 % . The absorption coefficient ismeasured by laser attenuation, at 633nm. These measurements haveuncertainties of up to 30%.

All measurements are taken in the baseregion of the fire, called the “persistent

zone” because it is characterized by aluminous flame with constant shape andstructure. Data include a contours plotof the temperature distribution, speciesconcentration, and absorption coefficientat 633 nm in this region. (The latter isused to predict soot concentrations.)

This paper will be of limited use for thepresent study. The accuracy of thetemperature data is questionable,however, (experimental uncertainly isbased solely on a statistical analysis ofthe data, and does not account forsystematic error), and velocity andradiant flux measurements were notincluded. Nevertheless, the temperatureand species concentration measurementsare detailed, and may be used to validatethe chemistry m o d u l e o f theSIERRAEUEGO or SIERRA/SYRINXcode.

Applicability toSIERRA

1 4

Burgess, D., and Hertzberg, M., 1974, “Radiation from Pool Flames,”from Heat Transfer in Flames, N.H. Afgan and J.M. Beer, Eds., ScriptaBook Co., Washington D. C., Ch. 27, pp. 413-430.

This reference contains a considerableamount of data concerning the propertiesof pool fires and diffusion flamescollected from different publishedworks. The main purpose of the study isto assess the risk posed by radiationemitted from a pool fire resulting froman accidental spill. Diffusion flamesburning hydrogen, methane, butane, andethylene, and pool fires burningmethanol, LNG, butane, gasoline, andbenzene were studied.

A significant amount of the dataconcerns the burning rate of the pool,which is used to estimate the totalthermal output per unit liquid surface.The flame temperature of a heptane poolfire is plotted as a function of air/fuelweight ratio. Radiation data consists ofplots of radiance measurements andabsorption coefficients as a function ofwavelength, although the experimentalapparatus, measurement technique, andmeasurement location are unspecified.The radiative fi-action for differentgaseous and liquid-supported diffusionflames is also presented in tabular form.

This paper is of limited use to thepresent study. Very little information isprovided concerning the experimentalapparatus, and no uncertainty analysis isprovided. (This information may beavailable in the original publications,however.) Also, all of the data ismacroscopic in nature; no differentialinformation (such as velocity ortemperature profiles) is provided. If thissource can be used, it would be tovalidate the overall performance of theSIERRAKJEGO or SIERRA/SYRINXcode, rather than an individual module.

Applicability toSIERRA Code

1 5

Choi, M. Y., Hamins, A., Rushmeier, H., and Kashiwagi, T., 1994,“Simultaneous Optical Measurement of Soot Volume Fraction,Temperature, and CO2 in Heptane Pool Fire,” Proc. 2fh Symposium(International) on Combustion, The1480.

The goal of the study is to determine therelative importance of the radiationcomponent in the heat feedback to thefuel surface, compared to the othermodes of heat transfer. In theexperiments, the temperature, sootfraction, and CO2 concentration weremeasured for a 10 cm-diameter heptanepool fire. This data was then used todetermine the Hz0 and CO speciesconcentrations from generalized staterelationships, and the radiant fluxincident on the pool surface wascalculated using a reverse Monte-Carlotechnique.

The temperature and soot volumefractions were carried out optically, bymeasuring emissive power. Thetemperature and soot volume fractionmeasurements had an estimated error of50 K and lo%, respectively. The CO;?was measured using a PbSe detectionsystem. These data were then suppliedto the RADCAL inverse Monte-Carloprogram, which solved for the radiantflux incident on the fuel surface. Theerror in the radiant heat flux is estimatedto be 1 W/m2, which is less than 10% ofthe measured value.

Combustion Institute, ppy 1471-

soot volume fraction, CO2, Hz0 and COat one axial location.

The usefulness of this paper is limitedbecause most of the data is presented ina form that would make comparison tonumerical results difficult. Also, theradiation fluxes are not measureddirectly, but are instead foundnumerically based on other data. Thus,the accuracy of this data is verydoubtful. Nevertheless, heat fluxincident on the fuel surface is ofparticular interest to the study, so thispaper may prove somewhat useful forvalidating the radiation module of theSlER.RA/SYlUNX code.

ApplicabilitytoSIERRA Code

Results presented include the MonteCarlo results for the radial distribution ofradiant heat flux incident on the poolsurface and plots showing the timeaverage distributions of temperature,

1 6

Crauford, N. L., Liew, S. K., and Moss, J. B., 1985, “Experimental andNumerical Simulation of a Buoyant Fire,” Combustion and Flame, Vol.61, pp. 63-77.

This paper presents the velocity andtemperature fields from a buoyant,turbulent natural gas fire. Theexperimental apparatus consists of a 25cm diameter circular porous refractoryburner, fueled by natural gas (>95%methane). This apparatus is surroundedby solid walls on three sides, and a finewire mesh on the fourth side to limit theeffects of ambient drafts. Combustionair is entrained from the laboratorythrough the mesh screen.

Temperature is measured using athermocouple array and the velocitymeasurements are made with a laserDoppler anemometer. The experimentaluncertainties of these measurements arenot provided.

Data include temperature, probabilitydensity functions of temperature, and theaxial velocity component, both measuredat the centerline and plotted as a functionof height and measured at one axialstation and plotted as a function of radiallocation.

This paper is of limited use for thepresent study, since no radiationmeasurements are taken, and theexperimental uncertainties are unknown.Also, the complicated burner geometrymay hamper numerical modeling.Nevertheless, the velocity andtemperature measurements in the firemay be useful for validating the fluidmechanics module of theSIERJVJFUEGO or SIERRA/SYRINXcode.

17

Applicability toSIERRA code

Fischer, S., Hardouin-Duparc, B., and Grosshandler, W., 1987, “TheStructure and Radiation of an Ethanol Pool Fire,” Combustion andFlame, Vol. 70, pp. 291-306.

This paper provides a very detaileddescription of the flame above a 0.5 mdiameter ethanol pool fire. Data includethe concentration of the combustionproducts, the temperature distribution,and the velocity distribution from thepool surface to a height of 1.5 m. Thelocal soot absorption coefficient andradiant intensity distributions are alsomeasured.

The species concentration in thecombustion products is measured using acommercial NDIR process analyzer forthe CO and CO2 concentrations, and agas chromatograph for the hydrocarbonconcentrations. The experimental errorin this process is estimated to be 15%.The gas temperature is measured with athermocouple array; this data is adjustedto account for thermal lag and radiationloss. The soot extinction coefficient ismeasured using an argon laser, and thetotal radiation heat flux is determinedusing both wide- and narrow- angleradiometers.

The composition of the combustionproducts is evaluated at 21 differentradial positions and 3 different heightsabove the fire. Temperature data includecontour plots at two heights above thefire, the average radial temperatureprofiles at eight heights above the fire,and vertical temperature profilesevaluated at six radial positions. Theerror in the temperature measurementsbased on the standard deviation of thedata, is approximately 325 K, or about25% at worst. The thermocouple probe

1 8

is also used to estimate the velocityprofile using a cross-correlationtechnique. The radial velocity profiles atthree different heights are provided.

Radiation data consist of the centerlinetotal radiant intensity evaluated at fifteendifferent heights, and the intensityparallel to the surface of the pool,evaluated at ten radial locations. Basedon the error bars on the graphs, the errorin the radiation intensity appears to beapproximately 25%. This paper alsocompares the radiant intensitydistributions obtained experimental totwo numerical simulations.

T h i s r e f e r e n c e w i l l p r o v e t o b eextremely valuable to the presentapplication. The reported data can beused to validate almost every aspect ofthe fire code. Both the velocity andtemperature fields are described indetail, and the radiation emitted from thefire is well documented.

iliRRA Cdde

Hamins, A., Klassen, M. E., Gore, J. P., Fischer, S. J., and Kashiwagi,T., 1994, “Heat Feedback to the Fuel Surface of Pool Fires,” CombustionScience and Technology, Vol. 97, pp. 37-62.

This paper presents measurementsdesigned to investigate the heat feedbackto pool fires burning liquid fuels. Theexperimental apparatus is an annular-ring pool burner with a 0.3 m diameter.The radial variation of the total radiantheat fluxes incident on the pool aremeasured at various angles, using anarrow-view radiometer. Other Datainclude average flame height, averagemass burning rate, heat release fraction,heat loss rates from this experiment andseveral published sources. Fuels includeheptane, toluene, methyl alcohol, andmethyl methacrylate.

The paper discusses the flame characterfor the different fuels, and comparesmany different data to other publishedexperiments and correlations. Theprinciple results presented concern theheat feedback rates and the incidentradiant heat flux distribution over thepool surface. Radiative measurementsinclude the intensity and heat fluxincident on the pool surface, measured at10 radial locations, and the reflectivityof the pool surface, as a function of polarangle. The intensity measurements aretaken in eight directions at each radialstation. Mass burning rates, mass fluxrates, and heat feedback rates at eachradial location are also included.

This paper prove very useful forvalidating the radiation module of theSIERRA/SYRINX c o d e , s i n c e t h eradiation emitted by the fire is wellcharacterized. The radiant heat feedbackto the fire is of particular interest, since


it can be used to approximate the heatflux incident on a low-profile objectimmersed in a pool fire, which is thescenario of direct interest to this study.Unfortunately, this paper lacks anadequate estimate of error, largelybecause the apparatus is unique.(Repetition of the tests indicate anuncertainty of approximately lo%, forthe wide angle total heat flux gauge.)Also, the paper does not include anyvelocity or temperature measurements inthe fire.

1 9

Hamins, A., Kiassen, M., Gore, J., and Kashiwagi, T., 1991, “Estimateof Flame Radiance via a Single Location Measurement in Liquid PoolFires,” Combustion and Flame, Vol. 86, pp. 223-228.

The goal of this paper is to develop acorrelation to estimate the heat loss(radiative) fraction using a radiancemeasurement at a single location. Theheat loss fraction estimates determinedfrom a single measurement arecompared to those calculated usingradiant data obtained at multiplelocations around the fire.

Pools with diameters of 4.6, 7.1, and 30cm were studied. The fuels studied wereheptane, toluene, styrene, methylmethacrylate, methanol, ethanol,hexanol, and a homogeneous azeotropicmixture of 70% toluene and 30% ethanolby volume. The fuel level wasmaintained 4 mm below the rim of theburner for the duration of the test. Thedistribution of radiant energy to thesurroundings was determined by movinga wide-angle radiometer over a “semi-infinite” cylindrical surface surroundingthe fire.

Reported data include the flame height, aplot of the radiant heat flux from a 30-cm diameter heptane flame as a functionof height above the pool surface, andplots of the heat-loss fraction for poolsburning heptane, methanol, and ethanol,for all pool diameters. Tabulated datainclude the heat-loss fraction for all fueltypes and pool diameters. Acomprehensive error analysis ispresented, which suggests that thestandard deviation in repeated heat fluxmeasurements at the same locationranges from 10% to 20%.

This paper shall be particularly usefulfor validating the radiation module of theSIlXlUBYRINX code. A wide rangeof pool diameters and fuel types werestudied, and the .quality of the dataappears to be quite high. It would bemore useful, however, to obtain the localradiation heat flux measurements used tocalculate the heat-loss fraction.

ApplicabilitySIERRA Code

2 0

Hogben, C. D. A., Young, C. N., We&man, E. J., and Strong, A. B.,1999, “Radiative Properties of Acetone Pool Fires,” Proc. 1999 NationalHeat Transfer Conference, Albuquerque, NM, August.

This paper discusses the link betweenthe flame geometry and the radiantoutput of the fire. This is done by firstmeasuring the radiation distribution atvarious locations to determine the spatialvariation of the radiation, using narrow-angle and wide-angle radiometers. Bothsets of results were then usedindependently with the geometriccharacteristics of the fire to approximatethe total radiant heat output. Finally, theradiation distribution is used to evaluatea new in-field technique that uses videoand thermocouple data to estimate heatflux and radiative fraction of the fire.

The experimental apparatus consisted ofa 3 1 cm diameter acetone pool fire. Theentire apparatus was contained within ascreened enclosure to minimize theeffects of ambient drafts. Threedifferent thermopile radiometers (twowide-angle, one narrow-angle) werepositioned at a height of 31 cm above thepool surface and pointed at the centralaxis of the pool. The wide-angledetectors and narrow angle detectorswere positioned 1.5 and 3 m away fromthe fire, respectively. The fire was alsovideotaped, and the resulting image wasdigitally altered to define eight colorregions corresponding to different flametemperatures.

Results presented include the flameheight, the emissive power, radiativefraction, and flame temperature. Thesedata are calculated using the narrow-angle and wide-angle radiometer dataand the thermocouple/video data, and arein good agreement with each other andwith other published data (typicallywithin 1 O%.)

This paper does not reprot thetemperature and velocity fields withinthe tire, which limits its usefulness to thepresent study. Nevertheless, theradiation data is of a high quality so thispaper should prove to be moderatelyuseful for validating the radiationmodule of the SIERRPJSYRINX code.


2 1

Inamura, T., Saito, K., and Tagavi, K. A., 1992, “A Study of Boilover inLiquid Pool Fires Supported on Water. Part II: Effects of In-depthRadiation Absorption,” Combustion Science and Technology, Vol. 86,pp. 105-119.

This paper presents a one-dimensionalmodel that predicts the time required fora sub-layer of water to begin boiling.The model requires radiation absorption,which is experimentally determined.The total radiant heat flux at the fuelsurface was measured for a toluene andn-decane pool fire, and the effectiveaverage absorption coefficient wasmeasured for toluene and Alberta Sweetcrude oil. The total heat fluxmeasurements were carried out on astainless steel pan with a 12 cm diameterand a 2 cm height. A fine wire meshsurrounded the pool fire to minimize theeffects of ambient drafts.

The radiant heat flux was measuredusing a Gardon-gauge type radiometerinserted through a l-cm diameter pipepassing though the center of the pool.The radiometer was aligned with the fuelsurface, and measured the radiantfeedback to the pool surface from thefire.

The absorption coefficient was measuredon a pool fire based in a Pyrex pan witha diameter of 12 cm, and a height of 2cm. The pan was filled with fuel andignited. After 10 s, the radiant heatpassed from the fire through the fuellayer was measured with a Gardon-gagetype radiometer located under the fuelsurface, and aimed at the bottom surfaceof the Pyrex pan.

Data for this experiment includes theradiant heat feedback at the time ofboilover, plots of total heat flux passedthrough the fuel layer as a function offuel layer thickness, fuel surface velocityas a function of fuel layer thickness, theboilover temperature as a function ofinitial fuel temperature, and the boilovertime as a function o f boilovertemperature and fuel layer thickness.

This paper is of limited use to the study.Boilover is not relevant to the presentstudy and very little auxiliary data isprovided. Also, no formal error analysisis included. Nevertheless, the radiativefeedback measurements are relevant tothe accident scenario of interest forvalidating the radiation module of theSIERRASYRINX code, as this datamay be used to estimate the radiationheat transfer from a flame to a low-profile object immersed in a pool fire.


2 2

Joulain, P., 1996, “Convective and Radiative Transport in Pool andWall Fires: 20 Years of Research in Poitiers,” Fire Safety Journal, Vol.26, pp. 99-149

This paper summarizes research carriedout by University of Poitiers, France, inthe last two decades that is focused onconvective and radiative transfer incombustion systems. The studies areclassified in five main topics. These aregas-solid combustion in horizontalconfiguration, gas-solid combustion in avertical configuration, bottom surfacecombustion and soot radiation, tire andgravity and research in progress. Ratherthan presenting details of experimentsand their result, brief conclusions aredrawn referencing the correspondingjournal or conference paper.

Gas-solid combustion in the horizontalconfiguration resembles the horizontalturbulent wall fire and in theexperiments carried out, it was observedthat radiation is not significant andnegligible at laboratory scales. The gas-solid c o m b u s t i o n i n a verticalconfiguration resembles the vertical wallfires and convection is the predominantmode of heat transfer, especially for highconfinement. Radiation becomesimportant as the scale of the burner rises.The research on bot tom surfacecombustion and soot radiation consistsof gathering mass fraction, concentrationand property data to provide necessaryinformation for an accurate radiationcalculation. Pool or natural fire researchinvolves theoretical and experimentalinvestigation of simulated pool fires andliquid pool fires. Souil et al. (1984) isone of the referenced studies thatprovides comparisons of experimentaland theoretical heat flux values of a pool

fire. The details of the study arepresented in the review of the relevantpaper. The liquid fire research focuseson the study of the radiative transferbetween free-burning fires and theirsurroundings. The researchinvestigating the effects of an externallyapplied thermal radiant heat flux on akerosene pool fire is presented in Zhanget al. (1991), w h i c h i n c l u d e smeasurements of temperature andconcentration and how they are effectedby the external radiation applied. Zhanget al. (1993) further investigates theinfluences over the radiative flux fromthe flame on the environment. Moredetailed information about these articlesis presented individually. Other studiessummarized regarding the pool firesinclude the case where the flame is tiltedas in flames resulting in most of the realaccidents due to wind conditions. Thenext main topic of study is gravityeffects, which are outside our researchinterest. 4


The paper references many articles,conference papers and technical reports;therefore this could be good source ofinformation to keep in the database.

2 3

Klassen, M., Gore, J. P., and Sivathanu, Y. R., 1992, “Radiative HeatFeedback in a Toluene Pool Fire,” Proc. 24th Symposium (International)on Combustion, The Combustion Institute, pp. 1713-1719.

This paper gives measurements of theradiative heat flux incident on the fuelsurface of a toluene pool fire, and thesoot volume fractions based on emissionand absorption. This data is then used todevelop a treatment for the equation ofradiative transfer that accounts forturbulent fluctuations of the heatfeedback.

The experimental apparatus consists of a30 cm diameter toluene pool fire. Theradiative intensity is measured using anarrow- angle radiometer positioned 2mm above the liquid surface on a pivot,in order to obtain measurements indifferent directions. The soot volumefractions are measured with a laser.

Results include the radial distributions ofthe mean and RMS temperature and sootvolume fractions based on emission andabsorption, evaluated at three differentheights above the pool surface.Radiative data include the radialdistribution of intensity incident on theliquid surface evaluated in threedifferent directions, and the radialdistribution of the total hemisphericalheat flux. No error analysis is provided.

This paper will be very useful forvalidating the radiation module of theSIERRA/SYRINX code. The radiationdata is in an accessible format, and theexperimental apparatus is described in adetailed way so numerical modelingshould be quite easy. The radiativefeedback measurements are of particularinterest since they can be used to

estimate the radiant flux incident on alow-profile object immersed in the poolfire, which is one of the accidentscenarios to be considered in this study.Unfortunately, the lack of an uncertaintyanalysis and supporting data(temperature and velocitymeasurements) somewhat limits theusefulness of this data.


2 4

Koseki, H., 1994, “Boilover and Crude Oil Fire,” Journal of Applied FireScience, Vol. 3, no. 3, pp.243-271

This article describes the boiloverphenomena in detail and presents a briefsummary of the literature about thetopic. The author uses variousexperimental results that were conductedat FRI (Fire Research Institute), NIST(National Institute of Standards andTechnology) while explaining the detailsof the phenomena. The setups used inthese experiments consist of a pan filledwith water and fuel floating on top.Circular pans with diameters of 0.3 m,0.6 m, 1.0 m, and 2.0 m, as well as a 2.7m x 2.7 m square pan are considered.The measurements are carried out forArabian and Louisiana crude oil,kerosene, and fuel oil.

SIERRWYRINX code, particularly inthe case of sooty flames.

A

z2,

SIERRA Code

A number of thermocouples are placedat certain heights along the fire axis andalong the two other radial locations. Thefuel is ignited and the incident thermalradiation from the flame is measuredusing a radiometer located at variousdistances to the pan axis.

Besides the incident thermal radiation,the distributions of temperature are alsopresented along with the burning time.The input provided for the experimentsare the pan diameter, fuel thickness,average and maximum fuel regressionrate, and heat wave regression rate. Theuncertainty for the measurements isestimated to be 5%.

Although the boilover phenomenon isnot of interest to the present study, theradiation and temperature data couldprove to be useful for validating theradiation module of the

2 5

Koseki, H., and Yumoto, T., 198813, “Air Entrainment and ThermalRadiation from Heptane Pool Fires,” Fire Technology, Vol. 24, pp. 33--47.

This paper describes air entrainment intoheptane pool fires, and characterizes thesmoke-blocking effect on thermalradiation from the fire. Pool fires withdiameters of 0.3 m, 0.6 m, 1 m, 2 m, and6 m were studied. In each case, about 3cm of fuel floated on a pool of water.Once the fuel was ignited, it was notreplenished. The burning rate wasdetermined by measuring. the liquidsurface regression rates. The 0.3, 0.6,and 1 m fires were conducted in a 15 mx 30 m x 18 m enclosed test facility,while the 2 m and 6 m fires wereconducted outdoors.

Temperature was measured using anarray of sheathed K-type thermocoupleslocated throughout the fire. The 2 m and6 m pool fires had arrays of 64 and 58thermocouples respectively, while theremaining experiments wereinstrumented with 80 thermocouples.Gas velocity measurements werecalculated using pressure measurementsfrom an array of nine bi-directionalstainless steel pressure probes, assuminga gas density equal to the density of airat the measured temperature. Totalradiation from the flame was measuredusing five wide-angle (120’) thermopiledetector radiometers. The radiometerswere installed at various radial distancesaway from the fire, level with the pan lipand aligned with the tire axis. Radiationfrom horizontal slices of the flame wasmeasured using five narrow-angleradiometers, each with a differentviewing angle.

Plots are given of burning rate as afunction of tank diameter, time-averagedvelocity along the flame axis as afunction of height, and mean velocity asa function of pan diameter. Further plotsshow temperature profiles along theflame axis for all pan diameters, and acontour plot of temperature over theentire fire domain for the 6 m pool fire.Radiation measurements include radiantheat flux as a function of distance fromthe fire, radiation heat flux as a functionof pan diameter, and radiant emittance asa function of height. The mass airentrainment is also calculated byassuming a “top-hat” velocity profileand that the flame is cylindrical with itsdiameter equal to the pan diameter.

This paper may be useful for codevalidation. The temperature and velocityfields are well characterized, andradiation measurements are included.Also, the experiment closely matches theaccident scenario of interest. No erroranalysis is included, and the accuracy ofthe velocity calculations is suspect. Theorientation of the fuel surface to the panlip strongly influences the tireproperties, but is not well defined.


2 6

Markstein, G. H., 1981, “Scanning-Radiometer Measurements of theRadiance Distribution in PMMA Pool Fires,” Proc. lgh Symposium(International) on Combustion, The Combustion Institute, pp. 537-547.

A scanning radiometer is used toproduce a complete map of the averagedradiance of a 0.73 m diameterpolymethyl methacrylate (PMMA) poolfire. Statistical properties of theradiation fluctuations are also presented.Finally, a simplified analytical modeluses the radiance data to compute theaxisymmetric distribution of a grayemission-absorption coefficient.

The experimental apparatus consists of ascanning radiometer located somedistance away from the tire. Theradiometer beam is directed using twomagnetically actuated mirrors, allowingvertical and horizontal scanning of thecomplete flame region. The results areaveraged over a large number of scans.No experimental uncertainty is provided.

Results presented include the radialprofile of the averaged radianceevaluated at thirteen heights above thepool surface, and the variation ofradiance with height above the poolsurface, measured at seven differentradial stations, the statistical datadescribing the fluctuation of the flameradiance due to turbulence, and theaxisymmetric absorption coefficientdistribution, calculated assuming aconstant flame temperature.

The data presented in this paper will besomewhat useful for validating theradiation module of theSIERRA/SYRINX c o d e s i n c e t h eradiation measurements are qui tecomprehensive. In particular, this data

2 7

would be useful to validate the treatmentof an optically thick (sooty) flame,which results from the combustion ofPMMA. The usefulness of this paper islimited by the lack of an uncertaintyanalysis, and because no temperature orvelocity measurements are included. .Furthermore, the accuracy of theabsorption coefficient data is suspectbecause the calculations are doneassuming a uniform flame temperature,which would seem to be an unreasonableassumption.


Modak, A. T., and Croce, P. A., 1977, “Plastic Pool Fires,” Combustionand Flame, Vol. 30, pp. 251-265.

The purpose of this paper is tocharacterize the influence on radiationfeedback on the burning behavior of 51-mm thick solid, horizontal, square,polymethyl methacrylate (PMMA) poolfires. Eleven pool sizes ranging from25 mm x 25 mm to 1.22 m x 1.22 mwere tested. Two tests were performedfor all but two of the pool sizes. ThePMMA slab was ignited by applying anaccelerant mixture of PMMA chips,paraffin oil, and pentane uniformly overthe slab surface.

The radiative power of the flame wascalculated assuming isotropic flameradiation, and was based a singleirradiance measurement taken with awide-angle radiometer located three tofour pool lengths from the pool centerand at a height equal to one pool length.The radiometer was aimed at the fireaxis. The fuel consumption rate wasdetermined by continuously weighingthe slab as it was burned. Nouncertainty estimates are provided forany of this data.

Data include plots of burning rate andradiative fraction plotted against timeafter ignition and as a function of poolsize, flame emissivity as a function oflocation, and burning rate due to radiantfeedback as a function of location on theslab. Tabulated data include steady-stateradiative fraction and radiative heat flux,steady state and time-averaged burningrate, time-dependant burning rates andheat flux

This paper should prove somewhatuseful for the purposes of this study.Most of the data is provided in bothgraphical and tabular form.Unfortunately, no experimentaluncertainty is included a n d t h eassumption of isotropic flame radiationis somewhat suspect. Also, the fuel is insolid form, which may complicatemodeling.temperatureincluded.

Also, velocity andmeasurements are not


2 8

Orloff, L., 1981, “Simplified Modeling of Pool Fires,” Proc. lfhSymposium (International) on Combustion, The Combustion Institute,pp. 549-561.

This paper develops a simplified modelfor predicting the radiant output of poolfires. A method is developed whereby afire that has non-uniform temperatureand species concentration can berepresented by an equivalent fire withhomogeneous properties. This model istested against experimentalmeasurements o f polymethyl-methacrylate pool fires with diameters of0.38 and 0.73 m. The fuel is in the formof solid plastic pellets. The pellets areplaced on a moveable platform, which inturn rests on inner tubes. The innertubes are inflated during the duration ofthe test to maintain a constant distancebetween the fuel surface and the burnerlip. The entire apparatus is containedwithin a screened enclosure to minimizethe effects of ambient drafts.

The flame shape was determined using amodified 35 mm camera, located sevendiameters from the pool fire and aimedhorizontally at the fire. The film imageswere digitized, and a statistical techniquewas employed to determine the ~llzsflame radius at different heights abovethe pool surface. These results arecompared to data from Marksteinww, consist ing of contours ofconstant radiance over the flame regionobtained from averaged scanningradiometer measurements for the 0.73 mfire. The averaged absorption-emissioncoefficient and the average flametemperature as a function of height arepresented. Also, the average flame

2 9

temperature and absorption coefficientare plotted as a function height above thefuel surface.

The most useful data from this paper isthe radiant flux measured at a target 2.8m from the center of a 381 mm diameterPMMA pool fire, using a wide-angleradiometer. The results are plotted atvarious fuel heights within the burningPan. No error analysis is provided,although the analytical solutions liewithin 12% to 15% of the experimentalmeasurements.

This paper is moderately useful for codevalidation, although most o f t h eexperimental data reported comes Cornother published sources. Unfortunately,the lack of experimental uncertainty andsupplemental data (radial temperatureprofiles and velocity measurements)limits the usefulness of this source.Also, the fuel is initially in solid form,which may complicate the numericalsimulation.


Shinotake, A., Koda, S., and Akita, K., 1985, “An Experimental Studyof Radiative Properties of Pool Fires of an Intermediate Scale,”Combustion Science and Technology, Vol. 43, pp. 85-97.

The goal of the paper is to investigateradiation heat transfer as a heat feedbackmechanism for pool fires. In particular,the transient behavior of radiation heatflux incident on the pool surface at thetime of ignition, and the ratio of internalto external radiation heat flux duringsteady burning are evaluated for severaldifferent pool diameters. Severaldifferent correlations are then developedand evaluated.

The apparatus consists of an iron panpartially filled with water. Heptane isfloated on top of the water, and thenignited. Pool fires with diameters of 0.3m, 0.7 m, and 1.0 m were studied.

The radiant heat flux values weremeasured inside and outside of the .flame, using a Gardon-type radiometerand a wide-angle radiometer,respectively. The extinction coefficientof the flame soot is also measured with achopped He-Ne lase r beam. Noexperimental uncertainty is reported.

Data include the radiation heat fluxincident on the center of the liquidsurface plotted against time while thefire is burning, the ratio of internal andexternal radiative flux measurements,and the radiative heat flux distribution atdifferent radial positions. The paper alsopresents the soot absorbance and volumefraction measured at three differentheights, for three different pooldiameters.

This reference will prove to be valuablein validating the radiation module of theSIERRA/SYRINX code. The radiationdata is very comprehensive. Ofparticular interest are the radiation heatfeedback measurements that can be usedto estimate the heating of a low-profileobject immersed in the pool; this is theaccident scenario of primary interest tothe present study. Unfortunately, noother supporting data (temperature andvelocity distributions) are included, andexperimental uncertainty is not provided.

4


3 0

Sivathanu, Y.R. and Gore, J.P., 1992, “Transient Structure andRadiation Properties of Strongly Radiating Buoyant Flames,” Journalof Heat Transfer, Vol. 114, pp. 659-665.

This paper describes the radiationemitted by buoyant-turbulent acetyleneand propylene diffusion flames. Theexperimental apparatus consists of a 50mm diameter burner tube, cooled by awater jacket. Little information isprovided regarding the experimentalapparatus, although another paper isreferenced for further detail.

All measurements were done using athree-line absorption/emission probe.The intensity leaving the probe volumewas collected divided into three parts bytwo beam-splitters. The intensities at900 and 1000 nm were measured usingcalibrated photomultiplier tubes. Theabsoption coefficient was calculatedbased on the extinction of a 632.8 nmHeNe laser beam. These data were usedto calculate the temperature and sootvolume fraction.

rms spectral intensities for differentflame conditions calculated from thetemperature and soot volume fractionsmeasured through correlations arecompared with the measured intensities.

This paper is of little use for the presentstudy, since measurements are takenonly at one axial location, and thereported experimental uncertainty isquite large. Nevertheless, the spectralintensity and absorption data may proveuseful to validate the radiation code.


The temperature, soot volume fractionand the intensity are measured at a singleaxial point and the radial profiles ofmean and fluctuating temperatures andsoot volume fractions for fires withdifferent Reynolds numbers arepresented. The uncertainties inabsorption and emission soot volumefractions are reported to be 40% and25%, respectively. The conditionalprobability density functions oftemperature conditioned on absorptionand emission soot volume fractions andthe probability density functions ofspectral radiation intensity are developedto display the statistical properties of theturbulent diffusion flame. The mean and

3 1

We&man, E.J., and McEwen, C.S., 1991, “The Time DependentStructure of a Medium-Scale Methanol Pool Fire,” Proceedings of theASME/JSME Joint Conference, 1991, Vol. 5, pp. 270-275.

This paper describes the transient andtime-averaged velocity and temperature

characterized, so this paper may be of

profiles of a 305 mm methanol pool fire.some use in validating turbulencesimulation in pool fires.

The apparatus is contained within ascreened enclosure in order to minimizethe effects of ambient drafts.

The velocity field is calculated usingLDA, and the temperature field isdetermined with a thermocouple array.Measurements of both velocity andtemperature were taken at 20 mmintervals from the pool center to beyondthe rim, and from heights of 20 to 200mm above the pool surface, for a total of121 stations. The IWZS velocity andtemperature measurement errors wereestimated to be 5% based on a statisticalanalysis of the data although there issome concern that systematic errors maybe induced due to instrumentation errorsand the thermal inertia of thethermocouple bead.

.!!=nrE0s


The centerline temperature measured atvarious heights above the pool surface iscompared with published results ti-ompool fires using different fuels (methane,acetone, ethane, ethyl alcohol, methylalcohol, kerosene, and JP-4.) Significantvariation is observed between thetemperature profiles from the differentfires.

This work does not contain any radiationdata, so it cannot be used to validate theradiation module o f theSIERRA/SYRINX code. Nevertheless,the temperature, velocity, and turbulencecharacteristics of the fire are well

3 2

We&man, E. J., and Strong, A. B, 1996, “Experimental Investigation ofthe Turbulence Structure of Medium-Scale Pool Fires,” Combustion andFlame, Vol. 105, pp. 245-266.

The purpose of this paper is to describethe turbulent structure of a medium-scalemethanol pool fire, which drivesentrainment and mixing and thereforecontrol development of the fire flowfield. The experimental apparatusconsists of a 3 l-cm diameter methanolpool fire, burning in quiescentconditions.

Laser Doppler anemometry was used tocalculate the radial and axialcomponents of velocity, while thetemperature field was determined usingthermocouples. Data include graphs ofboth mean and root mean squaredtemperature and velocity measured at 11radial stations and 11 vertical stations(resulting in 12 1 measurements), and avariety of turbulence statistics includingReynolds stresses and turbulent heat fluxterms. The velocity measurements areconsidered accurate to within 5%, at95% confidence. The Reynolds stressmeasurements are accurate to within15% at 95% confidence. The error inthe temperature measurements is notavailable.

This paper will be particularly useful forthe present application because of thelarge number of data points used tocharacterize the velocity and temperatureflow fields. Also, this experimentshould be easy to model, since theexperiment is conducted undercontrolled conditions and the apparatusis described in detail. Also, theexperimental uncertainty is wellcharacterized and quite low. The

turbulent properties of the fire are alsowell documented. Unfortunately, thispaper does not include any radiationmeasurements.


3 3

Zhang, X. L., Vantelon, J. P.,‘and Joulain, P., “Thermal Radiation froma Small-Scale Pool Fire: Influence of Externally Applied Radiation,”Combustion and Flame, Vol. 92, pp. 71-84.

This paper describes the effects ofexternally applied radiation on a small-scale kerosene pool fire. Theexperiments are performed with a small-scale kerosene pool fire, burning in acircular steel pan with a 15 cm diameterand depth of 5 cm. The externalradiative fluxes are generated by radiantheating panels located in the ambientfree stream, but far enough from theflame region to minimize theinterference with flow induced by thefire. Four panels are positioned withrespect to pool axis. Each panelconsisted o f f o u r emitters, eachdelivering 400W nominal power. Fourdifferent imposed flux values arec o n s i d e r e d (SO%, 90%, 1 0 0 % o fnominal value and 100% of nominalvalue located 10 cm below) togetherwith the undisturbed case, where there isno external radiation imposed. Thepaper includes detailed informationabout the setup and measurementtechniques used.

Temperature distributions,concentrations of COz, Hz0 and CO,monochromatic absorption coefficient,soot absorption concentration, netradiative heat fluxes from the flame tothe environment or to the fuel surfaceare all measured. The paper alsoincludes a comparison between theradiative intensities along the axiscalculated accounting for wavelengthdependence and with the equivalent grayproperties. A simple zonal analysis isemployed to calculate the heat fluxesfrom the flame to the environment and to

the fuel surface and compared withmeasurements.

This paper provides a complete set of theinformation necessary to solve theradiation equation for the flame togetherwith its accompanying article Zhang et.al (199 1). The experiment is similar tothe accident scenario likely to beencountered in the present study, exceptat a smaller scale and under controlledconditions. Because the radiationproperties of the flame are described indetail, this paper will be particularlyuseful for validating the radiationmodule of the SIERRA/SYRINX code.

3 4

Zhang, X. L., Vantelon, J. P., Joulain, P. and Fernandez-Pello, A. C.,1991, “Influence of an External Radiant Flux on a 15-cm-DiameterKerosene Pool Fire,” Combustion and Flame, Vol. 86, pp. 237-248.

This article is the complementary articleof Zhang et. al (1993) that includes thedetailed information about the COz, Hz0and CO concentrations, absorptioncoefficient and temperature distributions.The article investigates the effects ofexternal radiation on the maximumtemperature in the flame region, burningrate of the fuel, flame shape andstructure, temperature and concentrationfields.

3 5

LARGE-SCALE POOL FIRES

36

Brosmer, M. A., and Tien, C. L., 1987, “Radiative Energy Blockage inLarge Pool Fires,” Combustion Science and Technology, Vol. 51, pp. 21-37.

This paper discusses the influence of thefuel-rich core region of a pool tire ofradiant heat transfer, and consequentlythe effect on burning behavior. Asimplified model is proposed to accountfor this behavior, and is validated bycomparing predicted results withexperimental results obtained fromPMMA pool fires. All the experimentaldata is taken from other publishedsources.

m o d e l o f t h e SIIXRAEUEGO o rSIERRA/SYRINX code.

Data of interest includes plots of sootvolume fraction as a function of non-dimensionalized height above the flametaken from five different pool fires,spectral absorption coefficients for thecombustion gasses, radiant heat fluxvariation along the fuel surface as afunction of radial distance, and radiativefeedback as a function of pool size.Tabulated data include external radiativefraction, convective, surface re-radiant,and radiant heat flux (all calculated), aswell as measured mass burn rate. Nouncertainty estimates are provided forany of the experimental measurements,but these might be obtained from thereferences.

This paper will not be very useful forcode validation. Although the reportedexperiments are similar to the scenarioof interest, very little information abouteach experimental set-up is provided.Also, no temperature or velocity data isprovided for any of the experiments.Nevertheless, the data may be of someuse for validating the soot-chemistry

3 7i

1

Fire and Blast Engineering Project, Fire and Blast Information Group(FABIG), UK Steel Construction Institute, Final Report, 1997.

The FABIG web site athttn://www.fabin.com providesinformation on large scale fire effects onsteel structures, aimed in particular atsafety requirements for offshorestructures. The Fire and BlastInformation Project sponsored a varietyof experimental projects for measuringthe effects of large confined andunconfined horizontal and verticalturbulent jet and pool fires. This was a&4,400,000 study sponsored by 10 oilcompanies and the UK Health andSafety Executive.

All flames were luminous and smokey,and surface radiative emissive power isreported. Data was collected from manysources, including labs in Norway andthe UK. In addition, a survey of existingdata on pool and jet fires was carried outsimilar to the present study, with an eyeto validation of fire codes.

Data reported for the sponsored studiesincludes fuel type (crude of variousgrades), flow rate, pressure, flamegeometry, external radiation field, flame

surface emissive power and IR emissionspectrum, and meteorologicalconditions.

A mission to Kuwait to investigate thewell fires following the Iraqi warreported data on large jet fires, includingdata on plume geometry and radiativefield.

Detailed data and access to the survey ofavailable references can be obtainedonly through membership in FABIG.The reference is not rated, as theindividual reports may only be accessedby membership in FABIG.

A

>Applicability toSIERRA Code

3 8

Gregory, J. J., Keltner, N. R., and8 Mata, R., 1989, “ThermalMeasurements in Large Pool Fires,” Journal of Heat Transfer, Vol. 111,pp. 446-454.

This paper presents the temperature andheat flux measured at different stationsaround a large cylindrical calorimeterimmersed in a rectangular JP-4 pool fire.The calorimeter is a 6.4 m cylinder witha 1.4 m outer diameter. The dimensionsof the pool fire are 9.1 m x 18.3 m. Thetests were carried out in a light wind (>5m/s) resulting in an asymmetric burn.

Flame temperatures were measuredusing K-type sheathed thermocouples,located at various heights on six water-cooled towers surrounding thecalorimeter and at different positions onthe calorimeter surface. The error in thethermocouple measurements due toradiation loss from the bead wasestimated to range from 4 % to 12 %.Temperature and heat-flux wascalculated on the large calorimeter, andon four smaller calorimeters, wasdetermined using a 1-D inverse thermalanalysis. Errors in the heat-flux dataranged from 2 % to 17 %.

Reported data include the transient flametemperatures and heat flux at each

measurement station, and the averageheat flux values at each measurementsstation for the test series, plotted againstthe calorimeter surface temperature. Theflame temperature measured with thethermocouples is also provided intabular form.

Because the experiment closely matchesthe accident scenario of the presentstudy, and the thorough nature of thedata and error analysis, this referencewould be highly useful for the currentproject except for the lack of speciesconcentration and velocity data.


3 9

Gritzo, L. A., Gill, W., and Nicolette, V. F., 1998, “Estimates of theExtent and Character of the Oxygen-Starved Interior in Large PoolFires,” from Very Large-Scale Fires, Keltner, N. R., Alvares, N. J., andGrayson, S. J. Eds., ASTM Special Technical Publication 1336, pp. 8498.

This paper describes the size, shape, andcharacter of the oxygen-starved interiorof a large-scale pool fire. Thisphenomenon is usually present only inlarge-scale fires, and is not observed inthe majority of small-scale pool fires thatdominate the literature. The emphasis ofthe paper is on the effect this region hason object placement within the pool fireduring survivability tests.

The experimental apparatus consists of a1.8 m diameter JP-4 pool fire in low-wind conditions. The flame temperatureis evaluated at seven radial stations andat five heights. (No error estimate isprovided.) The majority of the data arepresented in the form of a definedvariable, given by the product of themean temperature and the standarddeviation of the temperature at a givenlocation, which indicates the presence ofthe oxygen-starved zone.

A numerical simulation is also carriedout using the VULCAN fire field model,and the predicted temperature and fuelconcentration fields are presented.Thermocouple errors limit the accuracyof the experimental temperature data andthus prevent direct comparison with thenumerical data, but the general trendsobserved in the pool fire are consistentwith the numerical simulation.

this study, especially considering therarity of large-scale pool fire data.Unfortunately, the experimental error inthe thermocouple measurements willlikely prevent direct comparison to theSIFXRAKFUEGO or SIERRA/SYRINXcode results.


The temperature distribution data shouldprove to be useful for the purposes of

4 0

Koseki, H., 1999, “Large Scale Pool Fires: Results of RecentExperiments,” Fire Safety Science, Vol. 6., pp. 115-132.

This survey paper presents results ofrecent pool fire experiments, for thepurpose of modeling an accident at anoil refinery. The experiments rangedfrom lab-scale to full-scale pool fires,and used many different fuels.

The scaling effects of pool fires aredemonstrated by plotting flame height,burning rate, and radiative fraction as afunction of pan diameter. Other datainclude a contour plot of temperaturewithin a 6 m heptane pan fire, a table ofmass burning rates and n-radiancemeasured at a single point for fourdifferent heptane dike fires, smoke yieldas a function of pan diameter for crudeoil and heptane fires, the distribution ofparticle diameters for crude oil pool firesfrom 0.1 m, 1 m, 12 m circular pans, anda 2.7 m square pan, and the axialtemperature distribution as a function oftime for crude oil and kerosene poolfires.

Although maw experiments arereferenced in this survey paper, verylittle of the presented data is useful.Also, no experimental uncertainty isincluded with any of the data.Consequently, this paper will be of littleuse to the present study.

4 1

Applicabil i ty to _SIERRA Code

Koseki, H., Iwata. Y., Natsume, Y., Takahashi, T. and Hirano, T., 2000,“Tomakomai Large Scale Crude Oil Fire Experiments,” FireTechnology, Vol. 36, no. 1, pp.24-38

This paper presents results of large-scalecrude oil pool fire experimentsconducted at Tomakomai, Japan. Theexperiments are carried out by burningcrude oil that is floating over water insteel pans of three different diameters, 5,10 and 20 m in an outdoors facility.Comprehensive information about thefuel is presented in the paper in termskinematic viscosity, density and massfraction of elements, so that acombustion code can provide thenecessary input for a CFD and radiationcode. As the experiments are carried outin an outdoors facility relatively roughinformation is available only in terms ofaverage velocity and direction of thewind.

Irradiance measurements were takenwith 12 to 16 wide-angle radiometerslocated at different radial positions andabout 1.2 m above the ground, aimed atthe fire axis. The distance between theradiometers and the fire axis was set atL/D = 2 and 3, where L is the distanceand D is the pan diameter. Five JRcameras are used to measure theradiative energy from the fire and localradiation and temperature profiles of theflame together with an array of 39thermocouples. The temperature valuesmeasured inside t h e f l a m e w i t hthermocouples are compared with valuesobtained through JR camera data.

In order to observe the effects of such afire on a nearby structure, an adjacentpan is located close to burning pan andthe wall and roof temperatures of theadjacent are measured both bythermocouples and JR camera. Thedistribution of the primary particlediameter of smoke agglomerates is alsoprovided.

This paper will likely be very useful forvalidating the SIERRA/FUEGO orSIERRA/SyRzNX c o d e , s i n c e t h eexperiment is very similar to theaccident scenario likely to beencountered in the present study. Theemphasis of this paper is themacroscopic behavior of the fire, so itwould be best suited to evaluating theoverall performance of the code, ratherthan the validation of individualmodules.

7Applicability toSIERRA code

4 2

Koseki, H., and Yumoto, T., 1988, “Burning Characteristics of Heptanein 2.7 m Square Dike Fires,” Fire Safety Science - Proceedings of theSecond International Symposium, pp. 231-240.

This paper presents the characteristics ofheptane burning in a 2.7 square dike.Two experiments were performed; thefirst had four open tank fires on top ofthe dike fire, while the second wasperformed using only the square pan. Inboth cases, a 2.7 m square steel pan wasused to contain the pool fire. The openfire tanks were modeled using four OS-m diameter cylindrical tanks with aheight of 0.4 m, which were locatedwithin the square pan. A 3-6 cm layer ofheptane was floated over water in thesquare pan and in the cylindrical tanks.All experiments were carried out in a 24mx24mx20mroom.

Temperature was measured using anarray of 60 K-type thermocouples,located throughout the flame region.The total radiation heat flux from theflame to the surroundings was measuredusing themopile-detector meradiometers, located at a fixed radialdistance from the fire axis. Theradiation from horizontal slices of theflame was measured using five narrow-angle radiometers, installed at a fixedradial distance from the flame and at theheight of the pan rim. Vertical velocitieswere measured using seven stainlesssteel bi-directional pressure probes. Thevelocity distribution was calculatedusing the pressure measurements fromthese probes, and the local densityestimated from the temperaturemeasurements. The air entrainment iscalculated from the velocitymeasurements, and by assuming a “top

hat” velocity profile. The compositionof the products of combustion wasdetermined using a gas chromatograph.

Data include a table of mass burningrates, surface regression velocities, andthe total radiation measured at onelocation away from the center of thedike. Plots include the, temperaturecontour plots, as well as air entrainment,and oxygen and carbon dioxideconcentrations as a function of axialheight. Radiation data include plots ofirradiance as a function of distance fromthe flame, and as a function of heightabove the ground.

This paper will be useful for codevalidation. The experimental apparatusis relevant to the accident scenario, andthe temperature data is quitecomprehensive. Unfortunately, only thecenterline velocity measurements areincluded, and the accuracy of thesemeasurements is suspect. Also, noexperimental uncertainty is provided.


4 3

Kramer, M. A., Greiner, M., Koski, J. A., Lopez, C., and Sho-Antila, A.,2001, (‘Measurement of Heat Transfer to a Massive Cylindrical ObjectEngulfed in a Regulatory Pool Fire,” Presented at the 2001 ASMENational Heat Transfer Conference, Anaheim CA., June 10-12.

This paper presents data to validatenumerical simulations of heat transfer toa large object immersed in a pool fire.The experimental conditions conform toUSNRC regulations for testingcontainers used to transport high-levelradioactive waste. The apparatusconsisted of 3800 kg cylindrical carbonsteel calorimeter with a length of 4.6 mand a diameter of 1.2 m. Thecalorimeter was suspended 1 m above acircular JP-8 pool fire, with a 7.16 mdiameter. The fuel floated over a 1 mdeep pool of water, and had an initialdepth of 12.7 cm. No more wasprovided during the test, and the testconcluded when the fire exhausted itselfThe entire apparatus was surrounded bya circle of V-shaped wind fences, whichminimized the effect of wind on the fire.Wind conditions varied throughout thetest, but the ambient wind speed wasalways less than 2.9 m/s.

Temperature was measured on theinterior surface of the calorimeter usingan a r ray o f more than fortythermocouples, located at variousstations on the inner surface of thecalorimeter. The experimental error ofthe temperature measurements isestimated to be the larger of +l .l K or0.4%, with 95% confidence. Heat fluxmeasurements were determined usingthe thermocouple measurements inconjunction with the SODDIT 1-Dinverse conduction code. The estimateduncertainty of the heat fluxmeasurements is within lo%, at 95%

confidence. Eight directional flamethermometers (resulting in 16 directionalmeasurements) are located outside of thecalorimeter, and are used to measure thelocal flame temperatures.

Data include plots of ambient windspeed, flame emissive power, andtemperature distribution as a function oftime. The axial temperature distributionis also plotted at one instant. The heatflux distribution is also plotted as afunction of time.

This paper will prove to be extremelyuseful for the present study, and shouldbe used to validate the accuracy of theSIERIWFUEGO or SIERRA/SYRINXcode when applied to model a full-sizedaccident scenario. The purpose of theexperiment is to verify a fire code, andthe experiment closely matches theaccident scenario of interest. Inaddition, the experimental data iscomprehensive in nature, and a thorougherror analysis is provided.


4 4

Nakos, J. T., Gill, W., and Keltnerm N. R., 1991, “An Analysis of FlameTemperature Measurements using Sheathed Thermocouples in JP-4Pool Fires,” Proc, 1991 ASME/‘AME Thermal Engineering Conference,March 1991.

This paper describes the use of sheathedthermocouples to measure the flametemperature in JP-4 pool fires in a

-variety of test conditions. The pool hasa diameter of 2 m, and a 165 x 130 cmsteel plate was mounted adjacent to thefuel surface.

The experimental data consists ofthermocouple measurements evaluatedat eight different locations in a JP-4 poolfire.


The reported data include thetemperature measurements from onethermocouple, and the differencesbetween the measurements of the otherthermocouples, in order to document theeffects of thermocouple bead emissivity,orientation, and shielding. No erroranalysis is included.

The paper does not intend to describe thetemperature field in the pool fire, andgenerally the data is presented in amanner that renders it unsuitable for thecurrent application.

4 5

Russell, L. H., and Canfield, J. A., 1973, “Experimental Measurement ofHeat Transfer to a Cylinder Immersed in a Large Aviation-Fuel Fire,“’Journal of Heat Transfer, Vol. 95 C, pp. 397-404.

This paper consists of two parts: the firstpresents the temperature and totalradiant flux of a rectangular JP-5 poolfire, and the second presents thetemperature distribution across the outershell of a cylinder immersed in the fire atvarious locations. The pool wasrectangular, with dimensions of 8 ft x 16ft.

The temperatures were measured usingshielded chromel-alumel thermocouples,and the radiant flux was measured usinga narrow-angle Gardon-type gauge. Thefirst test, evaluating the temperature andradiant heat-flux distributions within thefire, was carried out in quiescentconditions. The reported data consists ofcontour maps of temperature and total-radiant heat flux. The error in thetemperature and radiant heat-flux datawas 6 % and 30 %, respectively.

The cylinder used for the second part ofthe test was composed of 304 stainlesssteel, with a soot-coated surface havingan emissivity approaching unity. Thecylinder consisted of a core surroundedby an insulating layer and an outer layer.Four sets of three thermocouples werelocated within the outer layer, atdifferent depths from the outer surface.The four sets were angularly removedfrom each other by 90’. The insulatedlayer resulted in an adiabatic boundarycondition on the inner surface of theouter layer.

The cylinder apparatus was placedroughly in the center of the fire. For this

experiment, a 1 mph wind was blowing,resulting in an asymmetrical flame. Theexperimental results consist of thetemperature profiles through the outerlayer of the cylinder, and the effectiveheat transfer coefficient, calculated bysolving an inverse conduction problem.The error in the temperaturemeasurement is less than 6%.

This paper should prove extremelyuseful for code validation because of thecompleteness of the data, the detailederror analysis, and the relevance of theexperimental apparatus to the presentapplication. However, no concentrationor velocity profile data are given.

ApplicabilitVSIERRA Code

4 6

MOMENTUM-DRIVEN DIFFUSION FLAMES

47.

Bennett, B. A., McEnally, C. S., Pfefferle, L. D., and Smooke, M. D.,2000, “Computational and Experimental Study of Axisymmetric CoflowPartially Premixed Methane/Air Flames,” Combustion and Flame, Vol.123, pp. 522-546.

The purpose of this paper is to study thee f f e c t o f premixing methane/airdiffusion flames. This is done bothcomputationally and experimentally.The experimental apparatus consists ofan inner cylindrical burner jet with aninner radius of 0.555 cm, surrounded byan outer annular jet. The premixed fueland primary air passes upward throughthe inner jet, where it is simultaneouslyignited and exposed to the secondary air,traveling upwards through the outerannular jet. Six different flamed arestudied, having equivalence rationsranging from CD + 03 (no premixing) to@ = 2.464. The Reynolds numbersrange from 3 1 (0 + a) to 167 (<D =2.464), which are all within the laminarregime.

Gas temperatures are measured usinguncoated thermocouples. Speciesconcentrations are determined byextracting gas samples and analyzingthem using online mass spectrometry. Inparticular, species concentrations ofmethane, formaldehyde, acetylene,oxygen, water, carbon dioxide, as wellas C2H20 and other higher hydrocarbonsare measured. The experimentaluncertainty is described in detail;temperature measurements have anabsolute uncertainty of 50 K and arelative uncertainty of 10 K, and aspatial resolution of 0.3 mm. Theconcentrations of methane, acetylene,oxygen, water, and carbon dioxide haveabsolute uncertainties of 3 0 % .Measurements of formaldehyde andother higher hydrocarbons have relative

uncertainties of 30% and absoluteuncertainties up to three times higher.All flow rates are accurate to within 5%.Quantities derived from the flow ratesare accurate to within 10%.

Experimental data include maximumflame temperature tabulated for eachequivalence ratio, and graphs oftemperature as a function of axialposition. The mole fractions of CI&, 02,H20, COz, CO, HZ , OH, C2H2, andCzH20 are also plotted as a function ofaxial position. This paper also presentsthe results of a computationalsimulation, which are quitecomprehensive in nature. In general, thenumerical results w e r e i n goodagreement with the experimental data.

This paper will prove particularly usefulfor validating the chemistry module ofthe fire code, because of the largenumber of species considered, and thedetailed error analysis. It may also beuseful to compare the results obtainedwith the fire code to the numericalsimulation in this work. Radiant heatflux and the velocity field are notm e a s u r e d i n this experiment.


4 8

Delichatsios, M. A., and Orloff, L., 1988, “Effects of Turbulence onFlame Radiation from Diffusion Flames,” Proc. 22”d Symposium(International) on Combustion, The Combustion Institute, pp. 1271-1279.

This paper discusses the effect ofturbulence on soot formation andradiation heat transfer. Theexperimental apparatus consists of anozzle that discharges fuel verticallyupwards into the flame. A small coflow a.-

isoof oxygen is used to attach the flame tothe nozzle exit. Burner diameters of 1mm,1.5mm,2mm,3mm,and4mrnwere used. The entire apparatus is

s


surrounded by a water-cooled enclosure.Flames of CzH2, CJH~, C& w e r estudied.

Total flame radiation is measured usinga wide-angle radiometer. Radiation perunit flame height was measured usingtwo slit-radiometers. No experimentaluncertainty is provided with any of themeasurements.

Data include plots of turbulent radiantfractions (the ratio of total radiant powerto the theoretical heat release rate) as afunction of total heat release rate,dimensionless radiant power per unitheight as a function of dimensionlessheight, and radiant fraction as a functionof several different dimensionless scales.

Although radiation data is presented, thispaper is of limited use to the presentstudy because no error analysis is given,radial variation of properties are notprovided, the non-dimensionalpresentation of the data is inconvenientfor validation, and no other data (i.e. thetemperature field, velocity field, andturbulence statistics) are provided.

4 9

Faeth, G. M., Gore, J. P., Chuech, S. G., and Jeng, S. M., 1989,“Radiation from Turbulent Diffusion Flames,” from Annual Review ofNumerical Fluid Mechanics and Heat Transfer Vol. 2, Tien, C. L. andChawla, T. C. Eds., Hemisphere Publishing Corp., New York, U. S. A.,pp. l-37.

This work presents the radiat ionproperties of luminous and non-luminous turbulent diffusion flames. Itcontains no original experimental data,but presents a collection of results fromsix published sources. The flamesstudied included H2 (Re = 3000 and5722), CO (Re = 11400, 7475, and13140), C&, (Re = 2920, 5850, 11700,and 4.6-5.0 x 106), and C2& (Re = 6370and 12740.) In all cases, the difmsionflames were formed by injecting the fuelupward through a vertical burner tube.With one exception, all flames were instill air, at normal temperature andpressure.

A considerable amount of data ispresented in this work. Data includeplots of the mass fraction of combustionproducts as functions of equivalenceratios for Hz/air and C&/air diffusionflames, soot volume fraction as afunction of equivalence ratio for aC$!l.,Jair diffusion flame, mole fractions,temperature, and velocity as a functionof axial distance for a CO/air diffusionflame, and mass fraction andtemperature as a function of axialdistance for a Cl&/air diffusion flame.Radiation measurements include spectralradiation intensity plots for all types offlames, and total radiation heat flux as a

function of axial distance for a Cl&/airdiffusion flame.

This paper contains few details ofexperimental apparatus and noinformation about error and uncertainty,but these details can be found in thecited references.

This paper should prove to be useful forcode validation because of thecomprehensive quality of the data, andbecause many different fuel types areconsidered. This paper also contains anexcellent summary of the availableexperimental data for turbulent diffusionflames up to that time.


5 0

Faeth, G. M., Gore, J. P., and Sivathanu, Y. R., 1988, “Radiation fromSoot-Containing Flames,” from Proceedings of AGARD Conference No.422 Combustion and Fuels in Gas Turbine Engines, North AtlanticTreaty Organization, pp. 17-l-17-11.

The purpose of this paper is to presentthe radiation properties of turbulent,soot-containing, luminous diffusionflames. Four flames were studied; twowere fueled with ethylene with Reynoldsnumbers of 6,3 17 and 12,740, and twowere fueled with acetylene withReynolds numbers of 5,300 and 9,200.The fuel flows upwards through a burnerpassage with an exit diameter of 5 mm.The fuel is then ignited, and the flame isattached with a small coflow ofhydrogen. The entire apparatus iscontained within a screened enclosure tominimize the effects of ambient drafts.

Data include plots of the mole fractionsand mass fractions of the combustionproducts as a function of equivalenceratio and soot volume fraction as afunction of mixture fraction. Themixture fraction, mole fractions andvelocity fluctuations are plotted as afunction of axial location. Radialvariations of soot volume fractions andmonochromatic transmittncies (at 632.8and 5 14.5 run) are plotted at differentaxial locations, and radiant heat flux isplotted as a function of axial position.Also, spectral radiation intensitiesmeasured at the axis of the flame areplotted at three different axial heights.

Velocity measurements were made witha Doppler anemometer, and haduncertainties less than 5% with 95%confidence. Gas species concentrationswere measured using a gaschromatograph. Uncertainties in species

concentrations were less than 15% in theflaming region (95% confidence) butwere higher in the over-fire region.These measurements were in turn usedto calculate mixture fractions, which haduncertainties of -20, -50, and -100% atf= 0.04, 0.004, and 0.0005, respectively.Extinction measurements were madeusing lasers at 514.5 and 632.8 nm, andwere used to calculate the sootabsorptivitiy and volume fraction. Thesemeasurements had uncertainties of lessthan 10% and 20%, respectively.Spectral radiation intensities weremeasured using a monochromator, witha view angle of 1.2”, and radiation heatfluxes were measured using a wide-angle (150”) water-cooled radiometer.Uncertainties in these measurementswere less than 15% and lo%,respectively.

This paper may prove to be very usefulfor validating the radiation andchemistry modules o f theSlEWSYRINX c o d e , s i n c e t h ecombustion products composition andradiation intensity are described in detailand an extensive error analysis isprovided. .

SIERRA Code

5 1

Gore, J. P., and Faeth, G. M., 1986, “Structure and Spectral RadiationProperties of Turbulent Ethylene/Air Diffusion Flames,” Proc. 21thSymposium (International) on Combustion, The Combustion Institute,pp. 15214531.

This paper describes an experimentaland theoretical study of the structure andproperties of two round, turbulent,luminous ethylene/air diffusion flames,with Reynolds numbers of 6,370 and12,740 at the burner exit. The apparatusis contained within a screened enclosure,to minimize the effects of ambientdrafts.

Data include mean and fluctuatingvelocities, concentrations of thecombustion products, and soot volumefractions. Extensive thermal radiationmeasurements are also presented,including the absorption at 632.8 nmmeasured through the diameter of theflame, the spectral intensity distributionmeasured at different heights above theflame, and the radiant heat flux atdifferent axial and radial distances fromthe flame axis.

Velocity measurements are made using alaser Doppler anemometer.Uncertainties in both mean andfluctuating velocity components are lessthan 5%. Uncertainties in speciesconcentrations are less than 15%. Sootabsorption and volume fraction isdetermined using a laser. Uncertaintiesin extinction coefficient and soot volumefraction measurements are less than 10%a n d 20%, respectively. Spectralintensities and radiation fluxes haveuncertainties of less than 15% and lo%,respectively.

This paper will prove quite usefulbecause of the breadth and quality ofdata provided, and the detailed erroranalysis. The paper will be particularlyuseful for validating the radiationmodule of the code, since extensiveradiation data is provided.


Gore, J. P., Faeth, G. M., Evans, D., and Pfenning, D. B., 1986,“Structure and Radiation Properties of Large-scale Natural Gas/AirDiffusion Flames,” Fire and Materials, Vol. 10, pp. 161-169.

This paper describes the flame structureand properties of a large-scale turbulentnatural gas/air diffusion flame. Theexperimental apparatus consists of aburner injecting natural gas (about 95%methane by volume) into still air. Twodifferent burners were used, with innerdiameters of 76 and 102 mm. Sevendifferent flames, with chemical energyrelease rates ranging from 135 to 210MW were studied, all having Reynoldsnumbers near 3 x 106. The tests wereconducted outdoors, with ambient windspeeds ranging from 0.2 to 2.1 m/s.

Temperature measurements were madeusing an array of 20 thermocouples, andradiant heat flux was measured usingfive water-cooled heat flux transducers,located at various positions throughoutthe experiment. The locations andorientations of these radiometers areprovided in detail. One radiometer had aview angle of 7”, while the other hadview angles of 150”. Total radiationfluxes are found by integrating spectralradiation intensities over all wavelengthsand paths through the flame. Theconcentrations of N2, 02, CO, HzO, andH2 were measured using a gaschromatograph.

Data include plots of the mass fractionof each species as a function of fuelequivalence ratio, mean temperaturemeasurements along the axis of theflame, and the spectral radiation as afunction of wavelength measured at oneaxial station. The total radiation heatflux evaluated at different detectors for

two different flames is presented intabular form. No experimentaluncertainty analysis is provided,although thermocouple errors wereestimated to have caused the meantemperatures to be over-measured by 10K in the flame region. Richardsonextrapolation suggests that discretizationerrors in total heat flux calculations areless than 10%.

This paper should prove quite useful forvalidating the radiation module of theSIEWUEGO or SIERRA/SYRINXcode, because of the comprehensivenature of the radiation measurements.Although the experimental uncertainty isnot provided, good agreement betweenthe data and predicted results as well asthe results from other published sourcessuggest that the data is generally of agood quality. Unfortunately, thevelocity and temperature distributions inthe fire are not provided.


Gore, J. P., Jeng, S. -M., and Faeth, G. M., 1987, “Spectral and TotalRadiation Properties in Hydrogen/Air Diffusion Flames,” Journal ofHeat Transfer, Vol. 109, pp. 165171.

This paper describes the structure andradiation properties of hydrogendiffusion flames. The apparatus consistsof a 5 mm diameter water-cooled burnerthat injects hydrogen vertically upward,where it is ignited. The burner andflame are contained within a 115 mmdiameter quartz tube, which in turn wascontained within a screened enclosure tominimize the effects of drafts. A coflowof air is used to attach the flame to theburner exit. Two flames were studied,with Reynolds numbers of 3000 and5720 at the burner exit.

Presented data include plots of the massfraction of the combustion products as afunction of the equivalence ratio andalso the mole fractions of combustionproducts, mean temperature, velocity,and velocity fluctuations along the fireaxis as a function of height. Radiationdata include spectral radiation intensitiesfor radial paths through the axismeasured at three different heights andtotal radiative heat fluxes, measured atthe centerline and plotted as a functionof height, and measured at the burnerexit plane, and plotted as a function ofradial location.

Velocity measurements were made usinga laser Doppler anemometer.Uncertainties in mean and fluctuatingvelocity measurements were less than5%, with 9 5 % confidence.Temperatures were measured using athermocouple; no uncertainty isspecified, but the data are not adjusted

for radiation losses from the bead so themeasurements are thought to be 100-200K too low in the hottest parts of theflame. Species concentrations weremeasured using a gas chromatograph;these measurements have uncertaintiesof less than 15%. Spectral radiationintensities were measured using anarrow-angle (1.2”) monochromatorwith a pyroelectic detector, and totalheat flux measurements were madeusing a wide-angle (150’) sensor.Uncertainties in these measurements areless than 20% and lo%, respectively.

This paper will be extremely useful forvalidating the SIERRA/FUEGO orSIERILWYRINX code. The radiationdata is very comprehensive, and thevelocity and temperature fields are alsowell specified. Since these types offlames are non-luminous, this data willbe particularly useful for validating theperformance basic radiation codewithout soot effects.


5 4

Giilder, 6, 1992, “Soot Formation in Laminar Diffusion Flames atElevated Temperatures,” Combustion and Flame, Vol. 88, pp. 74-82.

The purpose of this paper is to determinethe influence on flame temperature onsoot formation for different fuels. Theexperimental apparatus consists of a 3mm diameter stainless steel burner tube,which is concentrically located within a100 mm diameter converging air nozzle.Both the fuel and air streams arepreheated by electric resistance heaters.The fire was contained in a flameenclosure with glass windows, whichprovides optical access while eliminatingthe effects of ambient drafts. Datafrom twenty different laminar flames arepresented, each with a differentcombination of fuel preheat temperatureand mass flow rate. The fuels wereEthylene, Propylene, Isooctane, andPropane, although the Propane flamedata was taken fi-om previouslypublished source.

Line-of-sight soot volume fractions arecalculated based on the transmission of amultilane laser beam, at wavelengths of830 nm, 632.8 nm, and 515 nm. Theoptical path length (flame diameter) wasdetermined optically, using a readingtelescope. These measurements had anuncertainty ranging between 0.125 and0.25 mm, or less than 1%. sootbrightness temperature was determinedusing a dual-wavelength disappearingfilament pyrometer. The temperatureuncertainty at calibration was fl5 K.

Presented data include a tabledocumenting the reactant temperature,fuel mass flow rate, smoke point height,line-of-sight maximum soot volumefraction, adiabatic equilibrium

5 5

temperature, and the maximum line-of-sight soot surface temperature for eachflame. Plots of soot volume fraction andsoot temperature as a function of heightabove the burner at different preheattemperatures are also included.

Although momentum-driven diffusionflames are not expected to beencountered in the accident scenario,simulating these experiments shall proveto be especially useful for validating sootproperties since the temperature andsoot-volume fractions in these flames iswell documented. Unfortunately, theradiant intensity and velocity fields arenot measured.


.

Giilder, 0. L., Thompson, K. A., and Snelling, D. R., 2000, “Influence ofthe Fuel Nozzle Material on Soot Formation and Temperature Field inCoflow Laminar Diffusion Flames,” Proceedings, Corn bustion InstituteCanadian Section, Spring Technical Meetings, Ottawa ON.

This paper describes the effect of fuelnozzle material on the properties of amomentum-driven diffusion flame. Theexperimental apparatus consists of a 10.9mm diameter fuel tube, centered in a 100mm diameter air nozzle. Fuel tubesmade of aluminum, steel, and Pyrexglass were studied. The fuel and air aredischarged into a screened enclosure andignited. Two different fuels, ethyleneand propylene, are studied. The ethyleneand propylene have flow rates of 194ml/min and 40 ml/min respectively, andthe air has a flow rate of 284 l/min.

The fuel tube temperature is measuredusing two thermocouples. Radiantintensity is measured with aspectrometer. N o experimentaluncertainties are included for thesemeasurements. Soot temperature andvolume fraction is calculated from thespectral intensity measurements.

Data include plots of radial soot volumefraction profiles and soot surfacetemperature profiles, evaluated at eightaxial stations for the propylene flameand six axial stations for the ethyleneflame.

This paper may be used to validate thesoot chemistry m o d u l e o f theSIIBRALFUEGO or SIERRA/SYRINXcode, since the soot temperature andvolume fraction distributions are wellcharacterized. Unfortunately, no

experimental uncertainty is reported, andalthough spectral intensity was measuredto calculate the soot properties, theintensity values are not reported in thepaper.


5 6

Jeng, S. M., and Faeth, G. M., 1984, “Species Concentrations andTurbulence Properties in Buoyant Methane Diffusion Flames,‘” Journalof Heat Transfer, Vol. 106, pp. 721-72’7.

This paper presents the speciesconcentrations and turbulent statistics fora turbulent, buoyant, methane diffusionflame. The turbulence data is then usedt o evaluate different turbulencemodeling procedures. The experimentalapparatus consists of natural gas (-95%methane by volume) being injectedupward through a liquid-cooled burnerhaving an exit diameter of 5 mm. Theflames are attached to the burner by asmall coflow of hydrogen. The entireapparatus is contained within a screenedenclosure to minimize drafts. Threedifferent flames with exit Reynoldsnumbers of 2920,5850, and 11700, werestudied.

Data include plots of species massfraction as a function of fuel equivalenceratio, temperature and speciesdistributions along the flame axis, andradial variation of temperature andspecies concentration, measured at threeaxial locations. Turbulence statisticsinclude radial distributions of Reynoldsstress and velocity fluctuations,evaluated at four axial locations.

Mean and fluctuating velocities weremeasured u s i n g a laser Doppleranemometer; these results have anexperimental uncertainty of less than10%. Species concentrations aremeasured using a gas chromatograph.U n c e r t a i n t i e s i n compositionmeasurement were estimated to be lessthan 15%. Temperature measurementsare made using small thermocouples, but

details on the measurements anduncertainty analysis are not provided.This paper will be moderately useful forthe present study. Although turbulencemodeling is not the main emphasis of thestudy, this paper will be useful forvalidating the turbulence module of thefire code. The temperature profilewithin the flame is also reported indetail, especially for the Re = 2920 case.Unfortunately, this paper does notcontain any radiation measurements, andalthough the mean velocity profile wasmeasured experimentally, it is notexplicitly reported in this study.


5 7

Jeng, S. M., and Faeth, G. M., 1984, “Radiative Heat Fluxes nearTurbulent Buoyant Methane Diffusion Flames,” Journal of HeatTransfer, Vol. 106, pp. 886-888.

The purpose of this technical note is toevaluate methods for predicting totalradiant heat fluxes from buoyant,turbulent difmsion flames, by comparingthe model predictions with experimentaldata. The experimental apparatusconsisted of a 5mm burner injectingvertically upward into still air. Flameswere attached to the burner using a smallcoflow of hydrogen. Three flames, withReynolds numbers of 2920, 5850, and11700, were studied.

Total radiation measurements were madeusing a wide-angle (150’) gas-purged,water-cooled sensor. The experimentaluncertainty of heat flux measurementswas less than 10%.

Data include the total radiative heat fluxplotted as a function of axial heightmeasured parallel to the flame axis at aradial distance of 575 mm, and the radialdistribution of total radiative heat fluxmeasured at the flame base.

This paper should prove useful forvalidating the radiation module of theSlERRARJEGO or SIERRASYRTNXcode, since total radiant heat transfer isthe quantity of interest in the presentstudy. Moreover, the experimentalapparatus and location of the sensors iswell defined, and should be relativelyeasy to model. Unfortunately, thevelocity and temperature distributions inthe flame region are not provided.

!%RRA Cdde

5 8

Kennedy, I. M., Yam, C., Rapp, D. C., and Santoro, R. J., 1996,“Modeling and Measurements of Soot and Species in a LaminarDiffusion Flame,” Combustion and Flame, Vol. 107, pp. 368-382.

This paper proposes a new analyticalmodel for laminar, soot-laden flames,which is evaluated by a comparison withexperimental data. The experimentalapparatus consists of a coaxial laminarethylene-air burner, consisting of an 11.1mm diameter brass fuel tube, surroundedby a 102 mm diameter air channel. Theflame was shielded by a brass cylindricalchimney to minimize the effects ofambient drafts. The volume flow ratesof the ethylene and air were 3.85 cm3/sand 713 cm3/s, respectively.

The composition of the combustiongases was determined using a massspectrometer, and were independentlyverified by gas chromatography analysis.Both of these techniques yieldedagreements within +20%. Laser-inducedflorescence was used to measure the OHconcentration field. Thesemeasurements were obtained from aprevious study. Temperature, velocity,and soot volume fraction data is includedfrom other published sources.

Reported data include the radial velocity,temperature, C~HJ and OH mole fractionprofiles at two axial positions, the sootvolume fraction along the axis plotted asa function of height, the radialdistribution of soot volume fraction atfour axial locations.

This paper may prove somewhat usefulin the present study, especially forverifying the chemistry module of theSIERRA/FUEGO or SIERRA/SYRINXcode. Also, the temperature and velocity

6 1

distributions are reasonably welldocumented. The applicability of thisstudy is limited, however, sincemomentum diffusion flames are unlikelyto be encountered in the codeapplication.

Applicabil i ty to 7SIERRA Code

Liu, F., Guo. H., Smallwood, G.J. and Gilder, &,2001, “Effects of Gasand Soot Radiation on Soot Formation in a CoFlow Laminar EthyleneDiffusion Flame,” in Radiation III: Third International Symposium onRadiativeTransfer, M. Pinar Mengiiq and N. Selcuk, eds., Begell House,New York, 2001.

This paper is chiefly concerned withdeveloping a complete numerical modelof radiation from sooting flames inwhich gas radiation is also present, andcomparing the relative effects of sootradiation and gas radiat ion ontemperature and soot prof i les . Indeveloping the numerical model,

module of the SlJ3WVFUEGO orSIEFGL4t3YRINX code.

zt,

however, comparison is made betweenpredictions of the model under variousdegrees of approximation o f t h eradiative processes and experimentalmeasurements of a carefully controlled

8


laminar diffusion flame. Theexperimental results are from anothersource (Guilder et al., 2000), which isalso included in this summary.

Presented data include temperature andsoot volume fractions presented as acontour plot over the entire domain ofthe flame. These results are comparedwith the results of various simulations.No experimental uncertainty is specifiedfor the data.

The results from this paper aresomewhat useful for the presentapplication, although it contains noradiation or velocity measurements.Also, the data is presented in a graphicform that makes direct quantitativecomparison to other resul ts verydifficult. Nevertheless, the soot data isquite comprehensive, and this paper maybe used to validate the soot-chemistry

6 2

Markstein, G. H., 1977, “Scaling of Radiative Characteristics ofTurbulent Diffusion Flames,” Proc. 1 Gh Symposium (International) onCombustion, The Combustion Institute, pp. 1407-1419.

This paper presents scaling laws for theradiation properties of propane diffusionflames. The experimental apparatusconsists of a 19.06 mm diameter verticaltube, terminated by a 12.7 mm nozzle.Propane (95% purity) was passedthrough the burner and ignited. Twelvedifferent flames were studied, with fuelflow rates ranging from 44 to 412 cm3/s.

Two radiometers were used to measureradiation from the fire. The totalradiative flux is measured using a wide-angle radiometer. The total fluxreceived from a thin horizontal slice ofthe fire is also measured by placing a slitaperture over the lens of the wide-angleradiometer. The intensity is measuredusing a horizontal beam radiometer.Both radiometers were aimed so theirhorizontal optical axes intercepted thevertical axis of the fire. The radiometerswere traversed in the z-direction in orderto measure the variation of heat flux andintensity with height. No experimentaluncertainty is specified for the data.

Data include plots of the total radiativepower as a function of fuel flow rate andradiative power per unit height andintensity as a function of height, plottedfor three different fuel flow rates. Thelast two properties are also plotted indimensionless form for all twelve-flowrates.

This paper should prove somewhatuseful for validating the radiationmodule of the SIERRMFUEGO orSIERRA/SYRINX code, since the

radiation properties of the fire in theaxial direction are well characterized.Unfortunately, no measurements ofproperty variation in the radial directionor temperature and velocity data arereported. Also, the fires arecharacterized entirely by the fuel flowrate; no other data, such as the exitReynolds number, is given. This maymake computer modeling of the firedifficult.

yiiziy+SIERRA Code

6 3

Mbiock, A., Teerling, J., Roekaerts, D., and Merci, B., “Application ofthe BEM and Analysis of the Role of Radiation Effects in LabscaleTurbulent Diffusiuon Flames,” in Radiation III: Third InternationalSymposium on RadiativeTransfer, M. Pinar Mengiiq and N. Selcuk, eds,Begell House, New York, 2001.

The purpose of this report is todemonstrate the influence of radiation inlab scale turbulent diffusion flames, andto demonstrate the superiority of usingthe Boundary Element Method overperforming an optically thin radiationanalysis, for predicting the properties ofthese flames. The experimentalapparatus consisted of a 7.2 mmdiameter nozzle with a 18.2 mmdiameter pilot. The nozzle is surroundedby an annular coflow tube. A mixture of25% CH4 and 75% air is passed throughthe nozzle, while a lean premix of CzH2,HZ, air, and CO2 is passed though thepilot tube. Air flows through the coflowannulus with a velocity of 0.9 m/s.

The mixture fraction, temperature, andspecies concentrations were obtainedusing a Roman/Raleigh/LIT technique at16 locations along the jet axis. A wide-angle radiometer measured the totalincident radiation at one location in anattempt to characterize the radiantfraction. No error analysis is provided.

Data include plots of velocity, mixturefraction, and temperature as a fimctionof axial location.

This reference may prove useful forvalidating the SIERRAFUEGO orSIERRA/SYRINX code, although theexperimental apparatus is not welldefined in the paper, which may makemodeling difficult. (A more detaileddescription may be available in one of

the references.) Also, the, flameproperties are not measured off the fireaxis, and no uncertainty analysis isprovided.


6 4

Sandia-Livermore website with Datahttp://www.ca.sandia.gov/tdf/Workshop/DatDwnld.html

Archive:

A Workshop was organized to toidentify experimental data sets onturbulent nonpremixed flames that areconsidered to be reasonably complete,accurate, and appropriate for the purposeof validating state-of-the-art combustionmodels, with particular attention tofundamental issues of turbulence-chemistry interaction. The data setslisted are considered to be usefulbecause they include both detailed scalarmeasurements (temperature, majorspecies, and some minor species) andvelocity measurements in the flameswith relatively simple geometries,simple fuels, and well defined boundaryconditions.In making these data available to thecombustion community, the organizersand contributors have made an effort toprovide complete documentation. Thisdocumentation should includeinformation on experimentaluncertainties. Data sets include:

?? Nonreacting Jet of Propane intoAir (Sandia)

?? H2 and H2/He Jet Flames(SandiaETH Zurich)

? H2/N2 Jet Flames(TU Darmstadt)

? H2/N2 Jet Flames(DLR Stuttgart)

. co/H2/N2 Jet Flames(SandiaETH Zurich)

?? CH4/H2/N2 Jet Flame (DLRStuttgart/TU DarmstadtSandia)

? NEW Scalar data (28 Apr 2000)

6 5

? Piloted Jet Flames:? Piloted CH4/air Flames

(Sandia/TU Darmstadt)? TNF3, TNF4 & TNF5 Targets? Piloted Natural Gas Flames

(TU Delft/Sandia)?? TNF3 Target? Piloted Flames of Various Fuels

(USydneySandia)? Bluff Body Flames:?? CH4/H2 Flames

(USydney /Sandia)? TNF3, TNF4 & TNF5 Target?? Bluff Body Flames of Various

Fuels (USydneySandia)?? Swirl Flames:?? Tecflam Swirl Burner?? TNFS Target

There are some data sets listed (but notlinked) that have not been used as targetsin these first rounds of workshop activitybut may be of interest.

Sivathanu, Y. R., Gore, J. P., 1993, “Total Radiative Heat Loss in JetFlames from Single Point Radiative Flux Measurements,” Combustionand Flame, Vol. 94, pp. 265270

This paper focuses on developing amethod for estimating total radiantoutput of turbulent jet flames based onthe measurement of radiative heat flux ata single location. The measurements ofheat flux are carried out along the baseof the cylinder at the exit plane of theburner and parallel to the axis of theflame until the radiative flux decreasesto the low measurement limit of thegauge.

The measurements mentioned in thisstudy are complementary to the ones inreferenced dissertations in the studies byGore (1986), Sivathanu (1990), Dolinar(1992) and Skinner (1991). The newmeasurements of radiative heat flux at aradial distance to flame height ratio of0.5 are performed for six differentflames with fuels CH4, C2H2 and C2H4.

Measurements and predict ions ofradiative heat flux at the radiometerlocation to the amount of radiativeenergy lost to the environment ispresented as a function of height ofdetector above injector exit parallel tothe axis of jet to visible flame lengthratio for a fixed value of detectordistance radially outward along the exitplane of the burner to visible flamelength ratio. Some additionalinformation from the dissertationsmentioned above is also tabulated.

purposes. The data presented here is notcomplete as it is, as the authors do notpresent confidence limits; together withthe dissertations it might be valuable toprovide some addition information.


This study briefly adds on someadditional data to the existing datapresented in the dissertations, whichcould be really helpful for providingdetailed experimental data for validation

6 6

Sivathanu, Y. R., Kounalakis, M. E., and Faeth, G. M., 1990, “Soot andContinuum Radiation Statistics of Luminous Turbulent DiffusionFlames,” Proc. 23rd Symposium (International) on Combustion, TheCombustion Institute, pp. 1543-1550.

This paper describes the soot andspectral radiation properties of a highlybuoyant propylene flame burning in stillair. The apparatus consists of a 50 mmdiameter burner injecting verticallyupward within a screened enclosure.Flames with Reynolds numbers of 750and 1370 were studied.

An optical probe was used to measurethe probability density function of thesoot, and the under- and over-fireabsorption and under-fire intermittency.The laser absorption at 632.8 nm wasmeasured for 10 mm optical paths withinthe flame. Transient spectral radiationintensities were measured using amonochromator. The experimentaluncertainties are estimated to be lessthan 40 % for the soot volume fractionsand less than 15% for the under-fireintermittency, at 95 % confidence. Theexperimental uncertainties of the spectralradiation intensity measurements areestimated to be less than 20 %.

Data include plots of under-fireintermittency and mean and fluctuatingsoot absorption and probability densityfunctions of soot absorption in the over-and under-fire regions plotted as afunction of radial location. Theprobability density function of spectralradiation intensities at 1000 nm and2300 nm are also graphed. Mean andfluctuating soot statistics, intensities, andthe transmittance are presented in tabularform.

6 7

This data is of limited usefulnessbecause most data is presented as plotsof probability density functions, andmeasurements are only taken at one axiallocation. Nevertheless, the experimentaluncertainties are detailed and the dataappears to be of good quality. Thisparticular paper may be useful forvalidating the soot chemistry module ofthe SEWUEGO orSIERRA/SYRINX code.

4

ApplicabilitySIERRA Code

Snelling, D. R., Thomson, K., Smallwood, G. J., We&man, E., Fraser,R., and Giilder, 6. L., 1999, “Soot Surface Temperature Measurementsin a Laminar Diffusion Flame,” Proceedings, Corn bustion InstituteCanadian Section, Spring Technical Meetings, Edmonton AB.

6 8

This paper presents a new technique formeasuring the soot surface temperatureand volume fraction by spectralmeasurement of radiant emission. Theexperimental apparatus consists of a 10.9mm diameter fuel tube centered within a100 mm air nozzle. The fuel-flow rate is194 ml/min and the airflow rate is 284l/min. The fuel, ethylene, and air aredischarged and ignited within a screenedenclosure to reduce the effects ofambient drafts on the flame.

Radiation emission is measured using aspectrometer. Each measurementconsists of a spectrum averaged fromfive one-second exposures . Nouncertainty estimates are provided forthese measurements. The flametemperature and soot volume fraction arenot measured directly, but are inferredfrom the measured spectral intensities.Temperature data obtained usingcoherent anti-Stokes RamaIlspectroscopy (CARS) is also presented.These measurements have error bars thatsuggest an uncertainty of less than 3%,but the origin of this data is not clear.

Data consists of plots of intensity anddifferential intensity (with respect toradial position) at eight differentwavelengths, flame temperature, andsoot volume fraction as a function ofradial position, measured at a height of30 mm from the fuel tube outlet.

This paper is of limited use for codevalidation. The spectral intensitydistribution is only measured at one axiallocation and has no reportedexperimental uncertainty. Also, thevelocity distribution is not provided.


Souil, J. M., Joulain, P. and Gengembre, E., 1984, “Experimental andTheoretical Study of Thermal Radiation from Turbulent DiffusionFlames to Vertical Target Surfaces,” Combustion Science andTechnology, Vol. 41, pp. 69-81

This study concentrates on radiative heattransfer to vertical target surfaces. Theflames considered here are propanediffusion flames issuing from a circularburner of 0.3 m diameter. The detailedexplanation of the experimental setup isreferenced to Gengembre et al. (1983),which is in French. The experimentalvalues available in that reference areused in this study as flame physicalparameters for the radiative fluxcalculations presented.

The measurements of radiative flux areby water-cooled, wide-angle radiometersviewing the flame at a distance Ro=3 mfrom the burner axis, and the fluxmeasurements are collected as a functionof the height of the sensor above thebase of the burner.

This study focuses on the forwardradiative transfer from the flames tovertical target surfaces for the study offire progress in a room fire situation. Anarray of 4 wide-angle radiometers isused with sensitive surfaces located in avertical plane at the same height fromburner. The first sensor is locatedperpendicular to the burner axis and theothers are located 0.25 m apart from thefirst one and each other on a horizontalline on the same side of the burner axis.Radiative fluxes are presented for fourpoints at different heights for 4 differentsets of total fire heat outputs anddistance to fire configurations. Theseare 15.8 kW and 0.3 m, 15.8 kW and 0.4m, 22.9 kW and 0.3 m, 37.9 kW and 0.3m. The temperature distributions in the

6 9

flame axes at different heights are alsopresented for a total heat output of 15.8,22.9 and 37.9 kW.

Two approaches are used forcalculations of the heat flux to comparethe results; one that uses detailedexperimental information of temperatureand absorption coefficient distributionsand one based on isothermal andhomogeneous absorption coefficient.The flame shape and the absorptioncoefficient distribution for the firstapproach is referenced to Gengembre etal (1983), while the isothermal valuesare specified here.

This paper overall can be very useful forSandia’s validation purposes. Althoughthe flame is not a pool fire flame, it is aturbulent difmsion flame with a definedshape, temperature and absorptioncoefficient distribution. The maindrawback of the paper is that i treferences this information to papers inFrench. If the referenced paper isavailable and translated, this paper is canb e v e r y useful to validate theSIERRA/SYRINX code that calculatesthe radiation transfer.


You, H-Z., and Faeth, G. M., 1982, “Buoyant Axisymmetric TurbulentDiffusion Flames in Still Air,SS Combustion and Flame, Vol. 44, pp. 261-275.

This paper presents the structure ofturbulent buoyant diffusion flames,which are then used to evaluatetheoretical models. The experimentalapparatus consists of a vertical burnertube, which discharged natural gas into ascreened room in order to minimize theeffects of ambient drafts. The naturalgas was ignited at the burner exit, andwas attached to the lip of the burner.Flames with heat outputs of 1.67 and8.51 kW were studied, whichcorresponded to Reynolds numbers of 70and 356 at the burner exit, respectively.

Data include plots of the axial variationof mean temperature, mean velocity,velocity fluctuations, Reynolds stress,and the mass fractions of Nz, 02, CO2,H20, CH& CO, and HZ, as well as radialplots of these quantities at three axialstations for both flames. The axialvariation of radiant heat flux is alsoplotted for both flames.

Mean velocity, velocity fluctuations, andReynolds stresses were measured with alaser Doppler anemometer. The totalexperimental uncertainties are notincluded for these methods, but gradientbroadening errors (which are typicallythe dominant source of error) areestimated to be less than 10%. Thetemperature field is determined using athermocouple array. Error bounds arepresented graphically with thetemperature data. Gas composition wasdetermined with a gas chromatograph.Radiant heat flux was measured using awater-cooled, gas-purged stationary

sensor. No experimental uncertaintiesare included with these measurements.

This paper should prove to be veryuseful for the purposes of the presentstudy, since the velocity and temperaturefields are well defined, and axialvariation of radiant heat flux ispresented. Also, the experimentalapparatus is thoroughly described, whichshould facilitate numerical simulation.Unfortunately, the experimentaluncertainty for most data is notprovided.


7 0

ENCLOSURE FIRES

7 1

Brehob, E. G., and Kulkarni, A. K., 1998, “Experimental Measurementsof Upward Flame Spread on a Vertical Wall with External Radiation,”Fire Safety Journal, Vol. 31, pp. 181-200.

The purpose of this paper is to study theeffect of incident radiation on the flamespeed along a vertical surface, whichoccurs in enclosure fire conditions.Infrared heaters provided up to 15kW/m2 of incident radiation on samplesof practical building materials, and theflame (or pyrolysis) height is recorded asa function of time. This data is to beused to develop and validate a numericalflame spread model.

The experimental apparatus consists of a30 cm x 120 cm panel , mountedvertically so that the longer side isperpendicular to the ground. A propaneline burner was used to ignite the loweredge of the sample. Two electric-powered infrared heater panels wereangled towards the sample to simulatethe radiant heat transfer to the samplethat would occur if it were placed in aburning room.

Materials tested were clear solidpolymethyhrethacrylate CpmA),Douglas fir particleboard, fire retardantplywood, poplar, cotton textile,hardboard, and gray laminatedcardboard. Data include graphs of theflame height as a function of time fordifferent levels of incident radiant heatflux, sample preheat, and igniterstrength. The total heat feedback wasalso measured at five heights along theparticleboard sample face, usingGardon-type radiant flux gages. Noerror analysis is included.

This paper is moderately useful forvalidating the fire module of theSlERRA/F~GO or SIlXRA/SYRl-NXcode. The experimental data is usefulfor validating flame propagation inenclosure fires. The radiation feedbackdata is of particular interest since itdescribes the effect a fire has on anobject located outside of the fire, whichis a scenario that is relevant to thepresent study. Unfortunately, nodifferential (velocity and temperature)measurements were made in the firefield, and no experimental uncertainty isprovided. Moreover, the materialsconsidered are common buildingmaterials, and not likely to beencountered in the present application.

AmGcabilitv to$IkRRA &de

I 7 2

Butler, B.W. and Webb, B.W., 1993 “Measurement of Radiant HeatFlux and Local Particle and Gas Temperatures in a Pulverized Coal-Fired Utility-Scale Boiler,” Energy and Fuels, Vol. 7, pp. 835-841.

The paper presents experimentalmeasurements of local particle and gastemperatures in an industrial scale boilerwhere the bulk of the energy transferoccurs by radiation, and radiation isdominated by flame emission rather thanwall effects. The boiler is an 80 MWtangentially fired unit operated by NewYork State Electric and Gas. Theanalysis is carried out using twodifferent coals. The proximate andultimate coal analyses, furnace geometryand operating parameters are presentedin detail.

Spatially resolved gas and particletemperature data are presented. Thesewere acquired concurrently withmeasurements of gas species, gasvelocities, particle size distribution,number density, particle velocity andchar burnout data reported elsewhere.The uncertainty in measuredtemperatures is presented as 540 K. Theradiant heat flux on the walls ismeasured at different heights and axiallocations. The authors present a roughoverall energy balance to verify theirheat flux measurements. This paper,accompanied with several related M.S.theses and Ph.D. dissertations referencedin the paper, provides very detailedinformation to model an industrial scalefurnace.

may be extremely difficult if notimpossible to accurately model a coal-fired reactor with the SIERRABUEGOor SIEFCLVSYRINX code.

*Applicability toSIERRA Code

7 3

This paper bay be useful for validationgthe radiation m o d u l e o f theSIERFMFUEGO or SIERRA/SYRINXcode, given the comprehensive nature ofthe presented data. Unfortunately, it

Butler, B. W. and Webb, B. W., 1991, “Local Temperature and WallRadiant Heat Flux Measurements In an Industrial Scale Coal FiredBurner,” Fuel, Vol. 70, pp. 1457-1464.

This paper presents experimental datagathered at the New York State Electricand Gas Goudey plant, in boiler no. 13,which is an 80 MW pulverized coalcombustor fired tangentially. E0

s

The data include wall radiant heat flux,measured around the periphery at sixdifferent elevations. The comprehensivedata set presented includes the local gastemperature at four axial positions,boiler operating conditions, proximate,ultimate and particle size analyses ofcoal.


The uncertainty of the measuredtemperatures and the heat fluxes arepresented as f30 K and +5%,respectively. The overall energy balanceestimation that is presented roughlystates the confidence to the gastemperature and radiation heat fluxmeasurement.

This paper presents a study that isperformed to provide data to verifyradiation models developed for coalcombustion systems. Although SandiaNational Laboratories interests do notinvolve coal combustion in enclosures,use of this paper with several referencedtheses and dissertations could be helpfulto validate a radiation solver as theypresent a complete data set so that aradiation problem can be set in isolationfrom turbulence and chemical kinetics.

7 4

Dembsey, N. A., Pagni, P. J., and Williamson, R. B., 1995,“Compartment Fire Experiments: Comparison with Models,” FireSafety Journal, Vol. 25, pp. 187-227.

This paper presents experimental datafrom compartment fires for the purposesof evaluating the performance of twonumerical models. The experimentalapparatus consists of a room withdimensions 2.5 m x 3.7 m in plan, and aheight of 2.0 m. The compartment hadone doorway with dimensions of 0.76 mx 2.0 m, centered on one of the shortersides. The walls and ceiling consist ofceramic fiberboard backed by gypsum,and are covered with a stainless steelsheeting coated with an industrial heat-resistant paint. A 0.61 m x 1.22 mporous propane surface burner waslocated at a height of 0.61 m above thefloor, and at different horizontallocations for each test. Heat wassupplied at a steady rate between 330and 980 kW for the duration of eachexperiment. The experiment started atthe moment of ignition, and ended whenthe rate of wall temperature changedecreased below 3 Wmin, at which pointthe system is said to be in quasi-steadystate.

Temperature profiles are measured usingthermocouple arrays located on thecompartment walls, floor, and suspendedwithin the enclosure. Radiation errorswere corrected using five aspiratedthermocouple probes, located throughoutthe enclosure. Radiation heat fluxes arecalculated using the wall and floortemperature histories, and by assumingthat the walls and ceiling are semi-infinite solids. No experimentaluncertainty estimates are provided.

Data include a plot of the maximum gastemperature as a function of fire size(heat rate), and a plot of the gastemperature as a function of time.Radiation data include plots of the netincident heat flux on a wall as a functionof elevation, as well as the averagecombined convection/radiation heattransfer coefficient and average incidentheat flux as a function of fire size.Tables of the steady state gastemperatures 0.1 m below the ceiling,and the bulk centerline temperature,ceiling jet temperature, flameentrainment height, mean flame height,average floor radiant heat fluxes, andaverage net heat fluxes for eachexperiment are also included.

This paper shall be somewhat useful forthis project, given the comprehensivenature of the temperature measurements.Also, the detailed description of the testcompartment should facilitatecomputational modeling. Unfortunately,the nature of the heat flux calculationsrenders the accuracy of these dataquestionable.


7 5

Foley, M., and Drysdale, D. D., 1995, “Heat Transfer from FlamesBetween Vertical Parallel Walls,” Fire Safety Journal, Vol. 24, pp. 53-73.

The purpose of this paper is to study heattransfer from flames to vertical surfaceswith an emphasis on the modeling offlame spread in confined spaces withbulk storage, particularly warehouses.The experimental apparatus consists ofan instrumented vertical surface,exposed to flames from a propane lineburner. The walls were two 610 mm x813 mm x 25 mm Monolux(incombustible) boards, oriented so thatthe 813 mm edge was vertical. Anincombustible baseboard was placedbetween the boards that could beremoved to allow airflow vertically intothe space between the walls; otherwise,air was entrained from the horizontal gapbetween the walls. The separationbetween the walls ranged from 60 mm toinfinity (where only one wall wasincluded.) A 600 mm long propane lineheater was located between the twowalls and rested on the base, parallelwith the 610 mm side. Fuel flow rates of5 and 9 Vmin were used, correspondingto 7 and 12.7 kW respectively.

Four Gardon-type radiometers wereinstalled at four vertical stations on oneboard; these could be moved betweenfour horizontal stations at each elevation,resulting in a 16-point heat fluxdistribution over the wall. Although noerror analysis is included for thesemeasurements, one quarter of theexperiments were duplicated and theresults were found to have a repeatabilityof 9%.

Data include plots of heat flux as afunction of wall separation and heightabove burner. Contour plots of heat fluxtaken at 16 measurement locations withopen and closed base configurations arealso included.

This paper should be very useful forvalidating the radiation module of theSIERRAAJEGO or SIERRA/SYRINXcode. The experimental data describesthe radiation heat transfer between a fireand an object located outside of the fire,which is a scenario of particular interestto the present study. The experimentalconfiguration is described in detail, andshould be easy to simulate. Also, theheat-flux measurements are quitecomprehensive, and are presented in aconvenient manner. Unfortunately, adetailed uncertainty analysis is notprovided, and velocity and temperaturedata are not presented.

Applicabil i ty to .SIERRA Code

7 6

Gengembre, E., Cambray, P., Karmed, D., and Bellet, J. C., 1984,“Turbulent Diffusion Flames with Large Buoyancy Effects,”Combustion and Flame, Vol. 41, pp. 55-67.

This paper presents the chemicalcomposition, temperature, and velocityfields of a turbulent diffusion flame forthe purposes of modeling room fires.The experimental apparatus consists of a30 cm burner that contains about 2000juxtaposed alumina tubes (5 cm long, 4mm ID, 7 mm OD). The burner isoriented vertically, and propane is fedthrough the burner assembly and is thenignited. The burner assembly sits in thecenter of an uncovered 4 m x 4 m x 3.6m enclosure, which minimizes theeffects of ambient drafts on the flame.The enclosure walls have 1.2 m2apertures to permit combustion airflow.Flames with heat outputs of 16, 23, and38 kW were studied.

Combustion products were measuredusing gas chromatography or by specificinfmred analysis. No uncertaintyestimates are provided for thesemeasurements. The temperature field isdetermined using a thermocouple array,and an error of up to 60 K due toradiation effects is thought to occur inthe hottest area of the flame. Velocitymeasurements are made by laserDoppler anemometry. No error analysisis provided for these measurements.

7 7

Data include plots of the mole fractionsof combustion products (CO, CO2,C&HIII, 02, and Nz), mean temperature,temperature fluctuations, YES verticaland radial velocity components plottedas a function of axial location (scaled bythe flame heat output) and radial locationat the burner exit plane.This paper may be used to validate thefluid mechanics module of theSIERRA/FUEGO or SIEF&4BYRINXcode, since the velocity and temperatureprofiles of this fire are well documented.Unfortunately, no radiationmeasurements are presented, and adetailed uncertainty analysis is notprovided.


Hwang, Yuh-Long and Howell, John R., 2001, “Local Furnace Data andModeling Comparison for a 600MWe Coal-Fired Utility Boiler?accepted, Journal of Energy Resources Technology, March.

This paper presents measurements insidea coal-fired 606 MWe utility boiler.Complete specification of the geometry,fuel (including proximate analysis),burner locations and angle of firing,overfire air darnper settings, and excess02 are given. Experiments for manydifferent combinations were performed.

M e a s u r e m e n t s o f temperaturedistributions within the furnace weremade on four sides at insertion depths ofup to 2.41 m from the wall and atvarious furnace elevations. Thesemeasurements are given for eachcombination of input parameters, alongwith radiative and total heat fluxes at thewall and stack NO, and excess 02concentrations. Temperaturemeasurements are stated to be within*ST, radiative flux measurementswithin &50 kW/m’ (in ranges up to 500kW/m*) a n d total heat fluxmeasurements within *60 kW/m2 (inranges up to 600 kW/m2). Stack NOxmeasurements are given as *5 percentand stack excess 02 as f 4 percent.

Measurements are compared withpredictions from a 3-dimensionalcomprehensive code deve loped a tBrigham Young University with varyingdegrees of agreement. Temperature andflux predictions are in reasonableagreement, but the NO, predictions werequite poor. This is ascribed to the needfor tine grid resolution to resolve theNO, chemistry as well as a poor modelfor NO, in the version of the code usedfor prediction. This paper provides themost detai led data available forverification of a code for very large coal-fired systems.

For the purposes of the present study, thegeometry and the presence of complexcoal chemistry and the lack of soot andash concentration measurementsprobably make the work less useful thanpapers that describe less complexsystems.

*Applicability toSIERRA Code

7 8

Ingason, H., 1998, “Modeling of a Two-Dimensional Rack StorageFire,” Fire Safety Journal, Vol. 30, pp. 47-69.

This paper presents a two-dimensionaltheoretical model for simulating the firein a rack-storage facil i ty (e.g. awarehouse), which is validated by acomparison to experimental data. Theexperimental apparatus consists ofrectangular Navilite boxes (0.59 m x0.235 m x 0.22 m) held up by twonarrow steel columns. The rack storageis two boxes wide and four boxes high.A 590 mm ling propane burner wasplaced on the floor centered between thetwo racks, oriented parallel to the 0.59 medge of the boxes.

nrE0

Temperature and velocity weremeasured using an array of K-typethermocouples and bi-directional probes,located along the centerline of thevertical flue between the two racks.

Data include a plot of air mass flow ratesat different t iers, flame heights,centerline temperature, and velocity as afunction of burner heat output. Noexperimental uncertainty is reported forthis data.

The applicability of this paper to thepresent project is very limited. Theexperiment is unlike any accidentscenario of interest. Also, no radiationdata is included, and temperature andvelocity measurements are taken at veryfew stations. Furthermore, noexperimental uncertainty is includedwith the data.

7 9

Ingason, H. and de Ris, J., 1998, “Flame Heat Transfer in StorageGeometries,” Fire Safety Journal’, Vol. 31, no. 1, pp. 30-60.

This paper presents the heat fluxdistributions to the four square steeltowers, of size 0.3x0.3x1.8 m, exposedto turbulent buoyant flames. Threedifferent fuels, carbon monoxide,propane, and propylene are used, leadingto flames with different soot content.The fuel is supplied from a circular gasburner at the bottom. The geometry ofthe system is presented thoroughly forthe four steel towers that represent atypical storage arrangement in awarehouse. The entire apparatus isenclosed by four walls made of gypsumboard, to minimize the effects ofambient drafts.

The temperatures and heat fluxes aremeasured by the use of thermocouples at52 different locations and the locationsof the thermocouples are presented in thepaper. Total heat flux is measured bywelding a thermocouple between a sheetof stainless steel and an insulating layer.The total heat flux incident on the steelis calculated based on the rate oftemperature rise measured by thethermocouple. Total heat flux was alsomeasured at one location using athermopile detector. Intensitymeasurements were made byradiometers that were located at threeheights and aimed at the fire axis.

The temperature of gas before ignition,ambient temperature, chemical andconvective heat release rates, total mass-entrainment rate of air, the averagevelocity at the floor and the Reynolds

nurnber are provided for severaldifferent configurations. The total heatflux to the surfaces of the square towersare presented with the temperatures ofthe surfaces and gas. The flame radianceand centerline gas velocity are alsopresented. The paper also presentscorrelations developed based on theresults for the range of sootinessbetween 0 and 5.6 where theexperiments are carried out. Themaximum errors for the measured andcalculated values are estimated to bebelow 5 %.

Although this paper describes the gastemperature within the flame, theemphasis is on the interaction of a firewith its surroundings rather than theproperties of the fire itself. Accordingly,this paper would be most useful forvalidating the SIEICELWFUEGO orSIERRA/SYRINX code when it is usedto predict the heat transfer to an objectthat is located outside the pool fire.

I

Applicabil i ty to -SIERRA Code

8 0

Pavlovit, P., Jovi& L., JovanoviC, L., and Afgan, N. H., 1974, “Steadyand Unsteady Heat Flux Measurement on the Screen Tube of a PowerBoiler Furnace,” from Heat Transfer in Flames, Afgan, N. H. and BeerJ. M. Eds., Scripts Book Co., Washington, U.S.A..

The primary purpose of this paper is todescribe the development of a radiantheat flux meter, which is then used tomeasure the heat flux incident on boilerscreen tubes in a pulverized coal furnacefrom a 200 MW thermal power station,although an unspecified liquid fuel wasused in this experiment. The heatingvalue of the fuel is provided, but the fuelcomposition is not included.

The radiant flux is calculated based onthe thermal resistance between the topsurface of the flux meter, which isexposed to incident radiation, and thebottom surface, which is welded onto thescreen tube of the boiler. By comparingthe temperatures of the two surfaces(measured using thermocouples), theincident radiant heat flux is calculated.

The experimental data of interestconsists of total radiant heat flux andtemperature measurements made ateighteen different stations on the furnacewall (nine stations at three levels, on twoadjacent surfaces). Contour plots of theradiant heat flux, relative to themaximum heat flux, and the temperaturedistribution over the furnace wall arepresented for two different steady-stateoperating conditions. An error analysisfor this experiment is not included; otherexperiments, however, demonstrate thatthe static performance of the flux meteris accurate to within 5%.

This paper will likely prove not beuseful for validating the

SIEWUEGO or SIERRlVSYRINXcode. None of the data is tabulated, andabsolute magnitudes of the heat fluxmeasurements are not presented.Moreover, since the exact compositionof the fuel is not specified, numericalsimulation o f t h i s experiment isimpossible.

I *Applicability toSIERRA Code

8 1

Spearpoint, M. J., and Dillon, S. E., 1999, “Flame Spread ModelProgress: Enhancements and User Interface,” National Institute ofStandards and Technology Technical Report NIST GCR 99-782.

The purpose of this work is to enhancethe Quintere Flame Spread Model,.QFSM, which computes the wind-aidedand opposed flow flame spread front aswell as the associated burn-out fronts fora compartment lined with variousmaterials. These modifications arebased on the results of recent publishedworks, which are repeated in the report.In particular, the height of the ignitionsource flame, its extension along thehorizontal ceiling, and the incident heatflux to a target material are addressed.The performance of the code is assessedby comparing i ts predictions toexperimental results.

The ,experimental apparatus is an IS09705 room/comer test, which consists ofa 2.4 m x 3.6 m x 2.4 m rectangularenclosure with a gas burner in one of thebottom comers. The gas burner has a0.17 m square area, and burns propane.The data include the flame height andthe flame extension along the ceiling ofthe enclosure (which occurs when theflame tip reaches the enclosure ceiling)plotted as a function of the thermaloutput of the burner, and the heat fluxemitted from the burner flame as afunction of the flame height.

The data also includes the time requiredfor the fire to reach flashover, i.e. whenthe thermal output exceeds 1 MW.These experiments were done for a widevariety of wall materials, and the

experimental results generally compare

8 2

very poorly to the numerical simulations.The time rate of energy release withinthe enclosure is also documented foreach experiment.

Although this report contains plenty ofdata, the purpose of the test facility is toevaluate residential and industrial firescenarios, so the experimental conditionsdo not correspond to the presentapplication. Also, most of the data ismacroscopic in nature; there is nodescription of the flame structure,temperature or velocity fields, or theradiant heat-flux distribution. Thisgreatly limits the usefulness of the datafor the purposes of code validation.


Tree, D.R., Black, D.L., Rigby, J.R., McQuay, M.Q. and Webb, B.W.,1998, “Experimental Measurements in the BYU Controlled ProfileReactor,” Progress in Energy and Combustion Science, Vol. 24, pp. 355383

This paper presents a summary of theexperiments and some results that havebeen carried out in the small scaleBrigham Young University controlledprofile reactor. These results have beenused for model validation, investigationof new concepts in combustion, andtesting and development o f n e wmeasurement techniques.

the medium, surface or particles over thewhole spectrum but these propertiescould be obtained elsewhere.

The results presented in the paperinclude experiments carried out withnatural gas combustion in a swirlingflow and pulverized coal combustion.Detailed information about themeasurement techniques and predictionson measurement sensitivities is alsopresented. Although extensive data arepresented of distributions of flow,temperature and the concentration ofspecies and particles, the total andradiative heat fluxes on the walls ofcombustion chambers are presented onlyfor pulverized coal combustion cases.(It is mentioned that these values werealso measured for the natural gascombustion cases.) Extensiveinformation is referenced to severalPh.D. dissertations. The heat fluxdistributions are presented in sectionthree of the paper. The uncertainty ofthe measurements of temperatures isestimated as 2.5 percent in the flamezone is presented as 10 percent for theheat flux on the walls. Monochromatictransmission rate at a wavelength of 632m-n is also measured with a meanuncertainty of 3 percent. No results arepresented for the radiative properties of

The paper p rov ides a detaileddescription of the furnace geometry,which should facilitate numericalmodeling. Also, the comprehensiven a t u r e o f the data (radiation,temperature, and velocity measured atmany locations) and the detai leduncertainty analysis makes this paper apotentially valuable source of data.Unfortunately, most of the data wasobtained during operation withpulverized coal, which may be difficultif not impossible to model withSIERRA/FUEGO or SIERRASYFUNX.


I 8 3

REFERENCES

Bennett, B. A., McEnally, C. S., Pfefferle, L. D., and Smooke, M. D., 2000,“Computational and Experimental Study of Axisymmetric Coflow PartiallyPremixed Methane/Air Flames,” Combustion and Flame, Vol. 123, pp. 522-546.

Bouhafid, A., Vantelon, J. P., Joulain, P., and Fernandez-Pello, A. C., 1988, “On theFlame Structure at the Base of a Pool Fire,” Proc. 22”d Symposium (International) onCombustion, The Combustion Institute, pp. 1291-1298.

Brehob, E. G., and Kulkarni, A. K., 1998, CCExperimental Measurements of UpwardFlame Spread on a Vertical Wall with External Radiation,” Fire Safety Journal, Vol.31, pp. 181-200.

Brosmer, M.A. and Tien, C.L., 1987, “Radiative Energy Blockage in Large PoolFires,” Combustion Science and Technology, Vol. 51, pp. 21-27.

Burgess, D., and Hertzberg, M., 1974, “Radiation from Pool Flames,” from HeatTransfer in Flames, N. H. Afgan and J. M. Beer, Eds., Scripta Book co.,Washington D. C., Ch. 27, pp. 413-430.

Butler, B. W. and Webb, B. W., 1993, “Measurement of Radiant Heat Flux andLocal Particle and Gas Temperatures in a Pulverized Coal-Fired Utility-ScaleBoiler,” Energy and Fuels, Vol. 7, pp. 835-841.

Butler, B. W. and Webb, B. W., 1991, “Local Temperature and Wall Radiant HeatFlux Measurements In an Industrial Scale Coal Fired Burner,” Fuel, Vol. 70, pp.1457-1464.

Choi, M.Y., Hamins, A., Rushmeier, H., and Kashiwagi, T., 1994, “SimultaneousOptical Measurement of Soot Volume Fraction, Temperature, and CO2 in HeptanePool Fire,” Proc. 2fh Symposium (International) on Combustion, The CombustionInstitute, pp. 1471-1480.

Crauford, N. L., Liew, S. K., and Moss, J. B., 1985, “Experimental and NumericalSimulation of a Buoyant Fire,” Combustion and Flame, Vol. 61, pp. 63-77.

Delichatsios, M. A., and Orloff, L., 1988, “Effects of Turbulence on Flame Radiationfrom Diffusion Flames,” Proc. 2yd Symposium (International) on Combustion, TheCombustion Institute, pp. 1271-1279.

Dembsey, N. B., Pagni, P. J., and Williamson, R B., 1995, “Compartment FireExperiments: Comparison with Models,” Fire Safety Journal, Vol. 25, pp. 187-227.

84

Faeth, G. M., Gore, J. P., Chuech, S. G., and Jeng, S. M., 1989, “Radiation fromTurbulent Diffusion Flames,” from Annual Review of Numerical Fluid Mechanicsand Heat Transfer Vol. 2, Tien, C. L. and Chawla, T. C. Eds., HemispherePublishing Corp., New York, U. S. A., pp. l-37.

Faeth, G. M., Gore, J. P., and Sivathanu, Y. R., 1988, “Radiation from Soot-Containing Flames,” from Proceedings of AGABD Conference No. 422 Combustionand Fuels in Gas Turbine Engines, North Atlantic Treaty Organization, pp. 17-1-17-11.

Fire and Blast Engineering Project, Fire and Blast Information Group (FABZG), UKSteel Construction Institute, Final Report, 1997

Fischer, S., Hardouin-Duparc, B., and Grosshandler, W., 1987, “The Structure andRadiation of an Ethanol Pool Fire,” Combustion and Flame, Vol. 70, pp. 291-306.

Foley, M. and Drysdale, D.D., 1995, LLHeat Transfer from Flames Bewteen VerticalWalls,” Fire Safety J., Vol. 24, pp. 53-73.

Gengembre, E., Cambray, P., Karmed, D., and Bellet, J. C., 1984, “TurbulentDiffusion Flames with Large Buoyancy Effects,” Combustion and Flame, Vol. 41, pp.55-67.

Gore, J. P., and Faeth, G. M., 1986, “Structure and Spectral Radiation Properties ofTurbulent Ethylene/Air Diffusion Flames,” Proc. 21th Symposium (International) onCombustion, The Combustion Institute, pp. 1521-1531.

Gore, J. P., Jeng, S. -M., and Faeth, G. M., 1987, “Spectral and Total RadiationProperties in Hydrogen/Air Diffusion Flames,” Journal of Heat Transfer, Vol. 109,pp. 165-171.

Gore, J. P., Faeth, G. M., Evans, D., and Pfenning, D. B., 1986, “Structure andRadiation Properties of Large-scale Natural Gas/Air Diffusion Flames,” Fire andMaterials, Vol. 10, pp. 161-169.

Gregory, J.J., Keltner, N.R., and Mata, R, 1989, “Thermal Measurements in LargePool Fires,” Journal of Heat Transfer, Vol. 111, pp. 446-454.

Gritzo, L.A., Gill, W., and Nicolette, V.F., 1998, “Estimates of the Extent andCharacter of the Oxygen-Starved Interior in Large Pool Fires,” from Vev Large-Scale Fires, Keltner, N. R, Alvares, N. J., and Grayson, S. J. Eds., ASTM SpecialTechnical Publication 1336, pp. 84-98.

Gillder, ii., 1992, “Soot Formation in Laminar Diffusion Flames at ElevatedTemperatures,” Combustion and Flame, Vol. 88, pp. 74082.

85

Gilder, G.L., Thompson, K.A., and Snelling, D.R, 2000, “Influence of the FuelNozzle Material on Soot Formation and Temperature Field in Coflow LaminarDiffusion Flames,,, Proc. Combustion Inst. Canadian Section, Spring Tech. Mtng.,Ottawa OM.

Hamins, A., Klassen, M., Gore, J., and Kashiwagi, T., 1991, ‘<Estimate of FlameRadiance via a Single Location Measurement in Liquid Pool Fires,,, Combustion andFlame, Vol. 86, pp. 223-228.

Hamins, A., Klassen, M.E, Gore, J.P, Fischer, S.J. and Kashiwagi, T., 1994, “HeatFeedback to the Fuel Surface of Pool Fires,,’ Combustion Science and Technology,Vol. 97, pp. 37-62.

Hogben, C.D.A., Young, C.N., Weckman, E.J., and Strong, A.B., 1999, “RadiativeProperties of Acetone Pool Fires,,’ Proc, 1999 National Heat Transfer Conference,Albuquerque, NM, August.

Hwang, Y.-L. and Howell, J.R, 2001, “Local Furnace Data and ModelingComparison for a 600MWe Coal-Fired Utility Boiler,,, accepted, Journal of EnergyResources Technology, March.

Inamura, T., Saito, K. and Tagavi, K.A., 1992, “A Study of Boilover in Liquid PoolFires Supported on Water. Part II: Effects of In-Depth Radiation Absorption,,,Combustion Science and Technology, Vol. 86, pp. 105-119.

Ingason, H., 1998, “Modeling of a Two-Dimensional Rack Storage Fire,,, Fire SafegJournal, Vol. 30, pp. 47-69.

Ingason, H. and de Ris, J., 1998, “Flame Heat Transfer in Storage Geometries,,, FireSafety Journal, Vol. 31, no. 1, pp. 30-60.

Jeng, S. M., and Faeth, G. M., 1984a, “Radiative Heat Fluxes near TurbulentBuoyant Methane Diffusion Flames,,, Journal of Heat Transfer, Vol. 106, pp. 8%888.

Jeng, S. M., and Faeth, G. M., 1984b, “Species Concentrations and TurbulenceProperties in Buoyant Methane Diffusion Flames,” Journal of Heat Transfer, Vol.106, pp. 721-727.

Jeng, S. M., Chen, L. D., and Faeth, G. M., 1984, “Nonluminous Radiation inTurbulent Buoyant Axisymmetric Flames,,, Combustion Science and Technology,Vol. 40, pp. 41-53.

Jeng, S. M., Chen, L. D., and Faeth, G. M., 1982, “The Structure of BuoyantMethane and Propane Diffusion Flames,‘, Proc. lgh Symposium (International) onCombustion, The Combustion Institute, pp. 349-358.

86

Joulain, P., 1996, “Convective and Radiative Transport in Pool and Wall Fires: 20Years of Research in Poitiers,” Fire Safety JournaZ, Vol. 26, pp. 99-149.

Kennedy, I.M., Yam, C., Rapp, D.C., and Santoro, RJ., “1996, “Modeling andMeasurements of Soot and Species in a Laminar Diffusion Flame,” Combustion andFlame, Vol. 107, pp. 368-382.

Klassen, M., Gore, J. P., and Sivathanu, Y. R., 1992, “Radiative Heat Feedback in aToluene Pool Fire,” Proc. 24’h Symposium (International) on Combustion, TheCombustion Institute, pp. 1713-1719.

Koseki, H., 1999, “Large-Scale Pool Fires: Results of Recent Experiments,” FireSafety Science, Vol. 6, pp. 115-132.

Koseki, H., 1994, “Boilover and Crude Oil Fire,” Journal of Applied Fire Science,Vol. 3, no. 3, pp.243-271.

Koseki, H. and Yumoto, T., 1998a, “Burning Characteristics of Heptane in 2.7 mSquare Dike Fires,” Fire Safety Science- Proc. of Second Int. Symp., pp. 231-240.

Koseki, H. and Yumoto, T., 199813, “Air Entrainment and Thermal Radiation FromHeptane Pool Fires,” Fire Technology, Vol. 24, pp. 33-47.

Koseki, H., Iwata. Y., Natsume, Y., Takahashi, T. and Hirano, T., 2000,“Tomakomai Large Scale Crude Oil Fire Experiments,” Fire Technology, Vol. 36,no. 1, pp.24-38.

Kramer, M. A., Greiner, M., Koski, J. A., Lopez, C., and Sho-Antila, A., 2001,“Measurement of Heat Transfer to a Massive Cylindrical. Object Engulfed in aRegulatory Pool Fire,” Presented at the 2001 ASME National Heat TransferConference, Anaheim CA., June 10-12.

Liu, F., Guo. H., Smallwood, G.J. and Giilder, 6., 2001, ‘(Effects of FGas and SootRadiation on Soot Formation in a CoFlow Laminar Ethylene Diffusion Flame,” inRadiation III: Third International Symposium on RadiativeTransfer, M. PinarMengilc and N. Selcuk, eds, Begell House, New York, 2001.

Markstein, G.H., 1981, “Scanning-Radiometer Measurements of the RadianceDistribution in PMMA Pool Fires,” Proc. lfh Symposium (International) onCombustion, The Combustion Institute, pp. 537-547.

87

Markstein, G. H., 1977, “Scaling of Radiative Characteristics of Turbulent DiffusionFlames,” Proc. 16’h Symposium (International) on Combustion, The CombustionInstitute, pp. 1407-14.19.

Mbiock, A., Teerling, J., Roekaerts, D., and Merci, B., “Application of the BEM andAnalysis of the Role of Radiation Effects in Labscale Turbulent Diffusiuon Flames,”in Radiation III: Third International Symposium on RadiativeTransfer, M. PinarMengtic and N. Selcuk, eds, Begell House, New York, 2001.

Modak, A.T. and Croce, P.A., 1977, “Plastic Pool Fires,” Combustion and Flame,Vol. 30, pp. 251-265.

Nakos, J.T., Gill, W., and Keltnerm N.R., 1991, “An Analysis of FlameTemperature Measurements using Sheathed Thermocouples in JP-4 Pool Fires,”Proc. 1991 ASME/JSME Thermal Engineering Conference, March 1991.

Orloff, L., 1981, “Simplified Modeling of Pool Fires,” Proc. lgh Symposium(International) on Combustion, The Combustion Institute, pp. 549-561.

Pavlovic, P., JoviC, L., Jovanovic, L., and Afgan, N. H., 1974, “Steady and UnsteadyHeat Flux Measurement on the Screen Tube of a Power Boiler Furnace,” from HeatTransfer in Flames, Afgan, N. H. and Beer J. M. Eds., Scripts Book Co.,Washington, U.S.A..

Russell, L.H., and Canfield, J.A., 1973, “Experimental Measurement of HeatTransfer to a Cylinder Immersed in a Large Aviation-Fuel Fire,” Journal of HeatTransfer, Vol. 95 C, pp. 397-404.

Sandia-Livermore website with Data Archive:http://ww~v.ca.sandia.gov/tdf/Workshop/DatDwnld.html

Shinotake, A., Koda, S., and Akita, K., 1985, “An Experimental Study of RadiativeProperties of Pool Fires of an Intermediate Scale,” Combustion Science andTechnology, Vol. 43, pp. 85-97.

Sivathanu, Y.R, Gore, J.P., 1993, “Total Radiative Heat Loss in Jet Flames fromSingle Point Radiative Flux Measurements,” Combustion and Flame, Vol. 94, pp.265-270

Sivathanu, Y.R. and Gore, J.P., 1992, “Transient Structure and RadiationProperties of Strongly Radiating Buoyant Flames,” ASME Journal of Heat Transfer,Vol. 114, pp. 659-665

Sivathanu, Y. R., Kounalakis, M. E., and Faeth, G. M., 1990, “Soot and ContinuumRadiation Statistics of Luminous Turbulent Diffusion Flames,” Proc. 23rdSymposium (International) on Combustion, The Combustion Institute, pp. 1543-1550.

Snelling, D.R., Thomson, K., Smallwood, G.J., Weckman, E. Fraser, R., and Giilder,G.L., 1999, “Soot Temperature Measurements in a Laminar Diffusion Flame,” Proc.Combustion Inst. Canadian Section, Spring Technical Mtng., Edmonton, AB.

88

Souil, J. M., Joulain, P. and Gengembre, E., 1984, “Experimental and TheoreticalStudy of Thermal Radiation from Turbulent Diffusion Flames to Vertical TargetSurfaces,” Combustion Science and Technology, Vol. 41, pp. 69-81

Spearpoint, M.J., and Dillon, S.E., 1999, “Flame Spread Model Progress:Enhancements and User Interface,” National Institute of Standards and TechnologyTechnical Report NIST GCR 99-782.

Tree, D.R., Black, D.L., Rigby, J.R., McQuay, M.Q. and Webb, B.W., 1998,“Experimental Measurements in the BYU Controlled Profile Reactor,” Progress inEnergy and Combustion Science, Vol. 24, pp. 355-383

Weckman, E.J., and McEwen, C.S., 1991, “The Time Dependent Structure of aMedium-Scale Methanol Pool Fire,” Proceedings of the ASME/JSME JointConference, 1991, Vol. 5, pp. 270-275.

Weckman, E. J., and Strong, A. B, 1996, “Experimental Investigation of theTurbulence Structure of Medium-Scale Pool Fires,” Combustion and Flame, Vol.105, pp. 245-266.

You, H-Z., and Faeth, G. M., 1982, “Buoyant Axisymmetric Turbulent DiffusionFlames in Still Air,” Combustion and Flame, Vol. 44, pp. 261-275.

Zhang, X. L., Vantelon, J. P., and Joulain, P., 1993, “Thermal Radiation from aSmall-Scale Pool Fire: Influence of Externally Applied Radiation,” Combustion andFlame, Vol. 92, pp. 71-84.

Zhang, X. L., Vantelon, J. P., Joulain, P. and Fernandez-Pello, A. C., 1991,“Influence of an External Radiant Flux on a 15-cm-Diameter Kerosene Pool Fire,”Combustion and Flame, Vol. 86, pp. 237-248.

89

APPENDIX A: DATA CONTENT OF REFERENCES

90

Table 1: Summary of Lab-Scale Pool Fires References

Rankings are from 1 (complete data) to 3 (sparse data).

91

Table 2: Summary of Large-Scale Pool Fire References

Koseki et al., 2000 3 1 3 2Koseki and Yumoto, 1988a 1 2 1 2Kramer et al., 200 1 1 1

Nakos et al., 1991 1

Russell and Canfield, 1973 1 1


92

Table 3: Summary of Momentum-Driven Fire References


93

Table 4: Summary of Enclosure Fire References


94

APPENDIX B: List of ContactsN A M E AFFILIATION

Altenkirch, RobertAnnamalai, K.

Brewster, M.QuinnCharette, Andre

Delil, AdDupoirieux, Francis

:actory Mutual ResearchFaeth, Gerry

Farouk, BakhtierFelske, Jim

Fiveland, WoodyFlamand, Eric

Gore, JayGriffith, James

Grosshandler, BillGulder, Dr.Omer

Hamdan, FadiHamins, AnthonyHirano, Toshisuke

w s uTAMUUIUC

U. of Quebec-(Canada)NAL (Netherlands)

ONERA (France)

U of MichiganDrexel

SUNY BuffaloABB

Int. Flame Res Inst, Ijmuiden, NetherlandsPurdueSwRINIST

NRC (Canada)UK Steel Construction Inst.-- FABIG

NISTTokyo U. (retired)

Joulain, Pierre Laboratoire de Combustion et Detonique” (LCD), PoitieKashiwagi, Takashi

Kosecki, H.Krier, Herman

McCartney, CameronMcGrattan, Kevin

Menguc, PinarNat1 Fire Lab

Puzach, SergeyQuintiere, James

Ross, HowardSelcuk, NevinSirignano, Bill

Soutiani, DrAnouarFire Tech Dept

Tain, Jeanvan der Meer, Th. H.

NISTJapan Nat1 Res. Instit. Fire/Disaster

UIUCNRC- Canada

NISTU. of Kentucky

NRC CanadaThermal Phys. Dept., Acad. State Fire Serv., Russia

UMD College ParkNASA Glenn

METU (Turkey)UC Irvine

CNRS (France)SWRI

CNRS (France)University Twente, Enschede, Netherlands

van de Kamp, W. Flame Res. Org.,Corus (Hoogovens), Ijmuiden, Neth.Webb, Brent BYU

We&man, E.J.(Beth) U. Waterloo (Canada)Yang, Jiann NIST

95

e-mail

[email protected]@mengr.tamu.edu

[email protected]@uqac.uquebec.ca

[email protected]@onera.ti

[email protected]@coe.drexel.edu

[email protected]@usppl.mail.abb.com

[email protected]@swri.edu

[email protected]@nrc.ca

f.hamdan@,[email protected]@peach.ocn.ne.jp

[email protected]@nist.gov

[email protected]@uiuc.edu

[email protected]@nist.gov

[email protected]

[email protected]@ens.umd.edu

[email protected]@em2c.ecp.t?

taineBem2c.ecp.fr

[email protected]@sunwise.uwaterloo.ca

Distribution

1 J.Nakos 9132 0 5 5 51 0 S.Burns 923 1 08191 L.Gritzo 9132 08211 J.Moya 9130 08241 M.Pilch 9 1 3 3 0 8 2 81 A.Black 9 1 3 3 0 8 2 81 B.Blackwell 9 1 3 3 08281 S.Kearny 9112 08341 S. Kempka 9141 0 8 3 51 P.Desjardin 9132 08361 W. Gill 9132 08361 V.Nicolette 9132 08361 J.Suo-Anttila 9 1 3 2 08361 S.Tieszen 9132 08361 C. Moen 8 7 2 8 9042

1 0 Dr. John R. Howell

121

The University of Texas at AustinMechanical Engineering DepartmentMail Code C2200Austin, TX 78712Kyle Daun111 W. 38* Street Apt 201Austin, TX 78705-1554-5 1Hakan Erturk1652 W. 6ti Street Apt CAustin, TX 78703-5042-75Ad Deli1National Aerospace LaboratoryP.O. Box 153Emmeloord 8300ADThe NetherlandsProf. James QuintiereDepartment of Fire Protection Engineering0 15 1 Glenn L. Martin HallUniversity of Maryland-College ParkCollege Park, MD 20874

Central Technical FilesTechnical LibraryReview and ApprovalDesk for DOE/OSTI

8945-l 901809616 08999612 0612

9 6

Documents

~ibliography ofThermal tion Data for FireJohn R. Howell, K~Daun.and Hakan Erturk Prepared by Sandia National Laboratories Albuquerque, NewMexico 87185 and Uvennore, Califomia 94550