Engine Nacelle Halon Replacement, FAA, WJ Hughes Technical Center Point of Contact :Doug Ingerson Department of Transportation Federal Aviation Administration.

Engine Nacelle Halon Replacement,FAA, WJ Hughes Technical Center

Point of Contact : Doug Ingerson

Department of Transportation

Federal Aviation Administration

WJ Hughes Technical Center

Fire Safety Section, AAR-422

Bldg 205

Atlantic City Int'l Airport, NJ 08405 USA

tel: 609-485-4945

fax: 609-485-7074

email: [email protected]

web page: http://www.fire.tc.faa.gov/

International Aircraft Systems Fire Protection Working GroupLondon, England 13-14 June 2002

Federal Aviation AdministrationWJ Hughes Technical Center, Fire Safety Section, AAR-440


Major Topics for Review :

Project Progress

Difficulties With Hot Surface Ignition

Response To Hot Surface Ignition Difficulties

Near Term Plans




PROJECT PROGRESS April 1ST, 2002 - Nacelle Fire Testing

Five tests run against established Halon distribution

Five tests resulted in complete extinguishment

Ran fuel/air mixture through test article looking for hot surface IGNITION after the successful five test sequence

Hot surface did not IGNITE mixture

Forward motion halted and an investigation began

April 1ST, 2002 - Test Conditions

Air flow : 2.2 lbm/s @ 100°F

Fuel : JP8, 0.2 gpm @ 157-160°F

Agent : Halon 1301 @ 5.2#, 100°F, 41 lbf/ft^3

Hot Surface Temperatures : Average of 4 thermocouples => 1065 - 1084°F

Single point maximums = 1245 - 1283°F

Preburn : 20 secondsInternational Aircraft Systems Fire Protection Working GroupLondon, England 13-14 June 2002



DIFFICULTIES WITH HOT SURFACE IGNITION

Massive Hot Plate Geometry Located at 12:00 in the core beneath fuel spray

350 lb steel assembly

Electrically heatedAPPROXIMATE

HOT PLATETHERMOCOUPLE

POSITIONS

AIRFLOWFASTENINGBOLTS




DIFFICULTIES WITH HOT SURFACE IGNITION

Reviewing Fire Test Requirements of the Minimum Performance Standard for the Engine and APU Compartments (MPSE)

Fire test requirements are a subset of the larger procedure

Run a minimum of five tests against the established Halon 1301 distribution

Demonstrate an extinguishment performance of 70 - 90%

Meaning that of 5 tests run, 4 result in complete extinguishment

The “failed” test requires that a hot surface IGNITE the fuel spray after the conditions of the initial flame extinguishment subside

Attain prescribed extinguishment performance by adjusting the fire intensity with the following parameters (in priority) :

Fuel flow rate

Hot surface temperature

Successful extinguishment is defined as an 8 second duration between initial flame extinguishment and the REIGNITION of the fire




RESPONSE TO HOT SURFACE IGNITION DIFFICULTIES

Evaluating Non-existent Hot Surface Ignition

Reasoned two options as allowing inadequate hot surface IGNITION threat

Insufficient fuel flow rate

Insufficient hot surface temperature

Fuel flow Rate : Test fixture being run at highest desirable fuel flow rate

At 0.2 gpm, liquid fuel observed to roll off core surfaces

Resultant test fire observed having a flame length of roughly 10 feet

Prior fires fueled at higher flow rates resulted in damage to the test section

Increasing fuel flow is deemed an undesirable option

Hot Surface Temperature : Only remaining remedy for existing geometry

Ran 15 tests with the existing massive hot plates, April 5-15, 2002

Tests configured to favor hot surface IGNITION





Massive Hot Plate Testing, April 5-15, 2002

Tests run to maximize hot surface ignition

no agent released

removed agent cooling - although perceived as negligible against a massive hot plate

removed agent interactions potentially retarding/complicating hot surface IGNITION behavior

fuel flow turned off for 8 seconds

duration began when the agent was supposed to be released

simulated the initial extinguishment by stopping fuel flow

removed the cooling effects of the fuel on the massive hot plate - again perceived as negligible against a massive hot plate

fuel flow resumed at the expiration of the 8 seconds without electrical ignition while looking for hot surface IGNITION





Time Line for Massive Hot Plate Tests Investigating Hot Surface Ignition

FIRE IGNITEDELECTRICALLY

FUEL FLOWTURNED OFF

FUEL FLOWRESUMED WITHOUT

ELECTRICAL IGNITION

8

SECONDSPREBURN COOL DOWN

30

SECONDS

5

SECONDS

FUEL FLOWTURNED OFF

BEGINTEST,t = 0





Massive Hot Plate Testing, April 5-15, 2002 (continued)

Test Parameters


Fuel : JP8, 0.2 gpm @ 146 - 166°F

Hot Surface Temperatures : Average of 4 thermocouples => 1095 - 1185°F

Single point maximums => 1180 - 1347°F

Preburn : 5 - 30 seconds

Secondary Fuel Flow Duration : 5 seconds

Test Results

One hot surface IGNITION out of 15 tests

Pertinent conditions for the hot surface IGNITION; average = 1180°F, maximum single point = 1328°F, preburn = 10 seconds

Two tests run with higher maximum single point temperatures AND equivalent or longer preburn durations without hot surface IGNITION





Massive Hot Plate Testing, April 5-15, 2002 (continued)

Anecdotal Experience

Massive hot plates formerly provided sufficient IGNITION energy to REIGNITE the fuel/air mixture

Improper fuel control operations during April 1 - 15 test sequences resulted in sporadic hot surface IGNITION

Fastening bolts shown in previous photograph observed glowing red during testing

Hot surface behavior during experience did not produce 100% IGNITION

Conclusions

Current massive hot plates are incapable of reliable hot surface IGNITION threat

Hot surface IGNITION itself is an enigmatic phenomena

Begin looking deeper for a solution to this difficulty





Considerations Regarding the Hot Surface and the MPSE

The concept of the MPSE relies on the hot surface IGNITION threat to push the established Halon 1301 distribution to near failure

This verifies that the fire is not too “weak” for a given agent distribution

“weak” fires can not withstand indirect extinguishment effects such as flame strain or other potentially unrecognized effects given new suppression technologies

A “weak” fire can be reasoned as a mechanism that would allow inadequate quantities of a replacement candidate/system to perform equivalently to Halon 1301

The current version of the MPSE will not allow an insufficient quantity of a replacement candidate/system to perform equivalently to Halon 1301, given the hot surface IGNITION threat

The hot surface IGNITION threat is a valid consideration for the engine application

The hot surface IGNITION threat is being used as a tool to verify the test fire is of sufficient intensity; not as a direct extinguishment goal





Review Experience/Literature Addressing Hot Surface Ignition

Review experience of others by contacting :

FAA, WJ Hughes Technical Center, C. Sarkos, R. Hill, D. Blake, and H. Webster

USAF, Wright-Patterson Air Force Base (WPAFB), J. M. Bennett and J. Tucker

USN, USN Lakehurst, W. Leach and M. Tedeschi

NIST, A. Hamins





Review Experience/Literature Addressing Hot Surface Ignition (continued)

Review Literature

AFRL-VA-WP-TR-1999-3068, “Aircraft Engine/APU Fire Extinguishing System Design Model (HFC-125),” J.M. Bennett, M.V. Bennett

WL-TR-95-3077, “Halon Replacement Program for Aviation / Aircraft Engine Nacelle Application Phase I, Operational Parameters Study,” M.L. Kolleck, J.M. Bennett, J.A.Wheeler, G.M. Caggianelli

NIST SP 890, Volume II, “Suppression of Engine Nacelle Fires,” A. Hamins, et al

AFWAL-TR-85-2060, Volume II, pt. I, “Vulnerability Methodology and Protective Measures for Aircraft Fire and Explosion Hazards / Fire Detection, Fire Extinguishment, and Surface Ignition Studies,” A.M. Johnson & A.F. Grenich

AFAPL-TR-79-2095, “Dynamic, Hot Surface Ignition of Aircraft Fuels and Hydraulics Fluids,” D.J. Myronuk





Results of Review

“weak” fire behavior observed in early work at FAA Technical Center

supports concept of challenging a given agent distribution; potentially to failure

MPSE addresses concept through hot surface IGNITION threat

Autoignition data produced via standardized test method DOES NOT relate to hot surface IGNITION phenomena

At a minimum, hot surface IGNITION alone is dependent upon multiple factors of :

aerodynamic characteristics of the compartment ventilation

the geometry of the fire zone

fuel volatility

characteristics of the material bathed in the fire - thermal conductivity, emissivity, etc.





Results of Review (continued)

Hot surface IGNITION has a threshold tendency which is dependent upon the installation

Hot surface IGNITION was addressed during the development of the HFC-125 design model produced from the Survivability group at Wright-Patterson Air Force Base

Hot surface IGNITION was used by the USN during system development for the F-18 E/F to validate Halon 1301 replacement






USAF/WPAFB experience

hot surface is 1” plate of 30” length and 100° arc on a 36” diameter

halon replacement effort was based on fire behavior (not gaseous concentration)

the “successful” replacement agent mass was a result of a minimum of 4 iterative cycles against the same test conditions

the test conditions were altered to represent a wide range of operational conditions

during phase one work, the significant fire dynamic parameters in an engine nacelle fire were uncovered by using the quantity of agent needed to extinguish any given fire as the control variable

during this work, hot surface IGNITION revealed itself as a problem that could alter the determination of the “successful” agent mass for a given test series

the final results :

hot surface IGNITION was found to be a MAJOR factor in determining how much agent was required to suppress a nacelle fire

if hydraulic fluid was used for testing, the spray was stopped at agent release

if JP8 was used for testing, the spray operated 5 seconds beyond agent releaseInternational Aircraft Systems Fire Protection Working Group

London, England 13-14 June 2002





USN/USN Lakehurst experience

Hot surface was a particular geometry of 4 tubes subject to flame impingement

Fire intensity was observed to vary with preburn duration

Quantity of replacement candidate was based on a fire threat sized to the existing Halon 1301 fire extinguishing system installed on the airframe

The tube array had a finite lifetime with respect to reproducible test results





Developed Plan to Alleviate Hot Surface IGNITION Difficulties

Desired to maintain a method to challenge the established agent distribution in the same spirit as the massive hot surface IGNITION threat

Conceived 3 options to potentially alleviate difficulties where each provides :

an IGNITION source to potentially REIGNITE the fuel/air mixture after initial extinguishment

better ability to characterize the performance of the established Halon 1301 gas distribution

Concept of “reignition delay” must now be defined as it is used as a metric for comparison

time between fuel/air mixture hitting a hot surface and its IGNITION

OR

time between no-flame-observed (“fire extinguishment”) and the subsequent REIGNITION of the fuel/air mixture





Developed Plan to Alleviate Hot Surface IGNITION Difficulties (continued)

Three options for evaluation; looking specifically for reliable secondary ignition

Tube array

heated strictly by preburn duration; tubes started at ambient temperature

expected increasing surface temperature (via increasing preburn) would decrease REIGNITION delay

ran through four versions before finding “reliable” configuration

Hot block

small block located in fuel/air flow field that was electrically heated

expected increasing surface temperature would decrease REIGNITION delay

ran through four versions before discontinuing effort

Continuously operating electrodes

used the electrical ignition source that initially starts the test fire

ran electrodes continuously throughout the length of the entire test

expected consistent reignition delay; agent mass and ventilation condition were reasoned as the sole impacts on any variation found in this concept





Tube Array

Array located in an intense region of the spray flame; determined visually

Tube array positioned downstream from flame stabilizing rib between 8” - 10”

Tubes bent on an approximate radius of 14-1/2”; supported approximately 2-1/2” above core surface

FUEL NOZZLES

2” TALL FLAME

STABILIZATION

RIB

TUBE ARRAY

AIRFLOW

FWD

UP

CORE





Tube Array (continued)

All tubes were 1/2”OD, grade 304 stainless steel

Supported in place with structure not subject to flame impingement

Tubes positioned by the lowest, most forward tube in the array

AIRFLOW

FWD

UPVERSION #1

Singular tube

VERSION #2

Tube spaced horizontally on 1” centers





Tube Array - Horizontal Assembly (continued)

AIRFLOW

FWD

FUEL NOZZLES

TUBE ARRAY





Tube Array (continued)

VERSION #3

Tubes aligned on an axis and rotated

approximately 45° from horizontal

about the lower, forward tube

AIRFLOW

FWD

UP

VERSION #4

Tubes stacked so centers

correspond to the corners of a rhombus

SURFACE TEMPERATURE

MEASUREMENT POINT





Tube Array - Diagonal Stack (continued)

FWD

UP

FUEL NOZZLES

TUBE ARRAY

AIRFLOW





Tube Array - Rhombus Stack (continued)

AIRFLOW

FWD

UP

FUEL NOZZLES

2” TALL FLAME

STABILIZATION

RIB

TUBE ARRAYCORE





Tube Array - Test Timeline

( identified in later discussion of the results as “NFN” fuel control )

FIRE IGNITED ELECTRICALLY;ELECTRODES TURNED OFF AFTER

OPERATING FOR 5 SECONDS

FUEL FLOWTURNED OFF

FUEL FLOWRESUMED WITHOUT

ELECTRICAL IGNITION

8

SECONDSPREBURN COOL DOWN

30

SECONDS

5 - 20

SECONDS

FUEL FLOWTURNED OFF

BEGINTEST,t = 0


Tube Array - Test Timeline (continued)

( identified in later discussion of the results as “Agent Released” )

FIRE IGNITED ELECTRICALLY;ELECTRODES TURNED OFF AFTER

OPERATING FOR 5 SECONDS

AGENTRELEASED

PREBURN COOL DOWN

30

SECONDS

5 - 20

SECONDS

FUEL FLOWTURNED OFF

BEGINTEST,t = 0





Tube Array - Observations

Reignition delay decreased with increased preburn

Threshold behavior observed but not consistent day-to-day

atmospheric conditions ( ?? )

tube material changing with increasing flame exposure ( !! )

Reignition delay values were scattered; occasionally, the same preburn time resulted in contradicting information

Reignition delay increased with increasing agent mass discharged

Agent storage behavior was erratic and may have affected test section air stream which then affected fire behavior preceding/during extinguishment

As tube array complexity increased, frequency of hot surface IGNITION increased





Tube Array - Observations (continued)


Tube Array - Observations (continued)

TUBE ARRAY VERSIONnumber of

testsnumber of ignitions

comments

Single tube, NFN 3 0

Four tubes, horizontal, NFN 5 1ignition not from the intended

surfaceFour tubes, diagonal stack, NFN 8 5Four tubes, rhombus stack, NFN 7 3Four tubes, rhombus stack, Agent

released20 16

totals 43 25





Hot Block Assembly

Versions #1 & #2 supported on core surface loosely by angle iron (concept proof)

All versions located in the spray zone of the fuel nozzles in the spray fire scenario

Positioned the base of block between 1” - 2” above core surface

Forward face positioned downstream from flame stabilizing rib between 8” - 15”

FUEL NOZZLES

2” TALL FLAME

STABILIZATION RIB

HOT BLOCK

AIRFLOW

FWD

UP

CORE





Hot Block Assembly (continued)

VERSION #2

ADDED 1” x 1/2” x 1/8”

PIECE OF STEEL ANGLE IRON

VERSION #1Basic features are :

• 1” x 2” x 6” mild steel block

• Two 3/4”OD x 6” long cartridge heaters

• Electrically heated

• 3 kW capability

SURFACE TEMPERATURE

MEASUREMENT POINT






VERSION #3VERSION #4

CHANGE STEEL ANGLE

IRON TO 2” x 1” x 1/8”

ADD STEEL

MOUNTING BRACKET,

2” TALL x 2-1/2” WIDE x 1/8”






FWD

UP

FUEL NOZZLES

HOT BLOCK WITH

2” x 1” x 1/8”

ANGLE IRON

CORE

AIRFLOW

ELECTRICAL

CONNECTIONS FOR

THE HOT BLOCK




2” TALL FLAME

STABILIZATION

RIB


Hot Block Assembly - Test Timeline

FUEL FLOW STARTED;IGNITION ELECTRODESWERE NOT ENERGIZED

COOL DOWN

30

SECONDS

9 - 30

SECONDS

FUEL FLOWTURNED OFF

BEGIN TEST, t = 0; HEATER BLOCK ATDESIRED SURFACE TEMPERATURE





Hot Block Assembly - Observations

Reignition delay decreased with increased surface temperature (time between initial fuel impingement and IGNITION)

Threshold behavior observed but not consistent day-to-day

atmospheric conditions ( ?? )

block material changing with increasing flame exposure ( !! )

aerodynamic flow around block very complex

Reignition delay values were scattered





Hot Block Assembly - Observations (continued)

0

2

4

6

8

10

12

14

16

18

20

22

1000 1100 1200 1300 1400 1500 1600 1700

Surface Temperature at Ignition (°F)

Tim

e D

elay

(s)

24Apr02

13May02

15May02

ALL OTHER DATES

NO HOT SURFACEIGNITION

BLOCK SITTING

ON 2” TALL

MOUNTING BRACKET

BLOCK SITTING

APPROXIMATELY 1”

ABOVE CORE SURFACE


Hot Block Assembly - Observations (continued)

IGNITION VERSIONnumber of

testsnumber of ignitions

comments

Simple block 2 0Block w/1" x 1/2" x 1/8"

Angle5 3

ignition occurring on forward face of block

Block w/1" x 1/2" x 1/8" Angle and mtg brkt

11 3ignition occurring on aft or bottom

faces of blockBlock w/2" x 1" x 1/8"

Angle and mtg brkt5 1

ignition occurring on aft or bottom faces of block

totals 23 7





Continually Operating Electrodes

Operated electrodes throughout the entire fire test

Core surface beneath spray fire was uncluttered

Electrodes run off transformer; 120VAC primary and 10,000 VAC secondary

Geometry of nozzles/electrodes had to be rigidly maintained





Continually Operating Electrodes (continued)

FWD

UP

AIRFLOW

FUEL NOZZLES

ELECTRODE PAIR





Continually Operating Electrodes - Test Timeline

FIRE IGNITED ELECTRICALLY;ELECTRODES CONTINUED

OPERATING

AGENTRELEASED

PREBURN COOL DOWN

30

SECONDS

10 - 20

SECONDS

FUEL FLOW &ELECTRODESTURNED OFF

BEGINTEST,t = 0





Continually Operating Electrodes - Observations

Reignition delay apparently not dependent upon preburn duration

Reignition delay not consistent day-to-day; atmospheric conditions suspected

Reignition delay increased with increasing quantity of agent released

Inconclusive behavior while keeping the discharged agent mass constant with varied fuel flow

REIGNITION occurred for every test run except one; the one that was not initially extinguished

Reignition delay times demonstrated scatter





Continually Operating Electrodes - Observations (continued)

0

1

2

3

4

5

6

7

8

9

10

11

12

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00

Reignition Delay (sec)

Wei

ght

of A

gen

t R

elea

sed

(lb

f)

10.4# [email protected] gpm, 14May 5.2# [email protected] gpm, elec pos #1, 17May

5.2# [email protected] gpm, 15May 5.2# [email protected] gpm, 16May

5.2# [email protected] gpm, elec pos #2, 17May 5.2# [email protected] gpm, 28May02

5.2# [email protected] gpm, 29May02



0

1

2

3

4

5

6

7

8

9

10

11

12

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50


Wei

ght

of H

FC

-125

Rel

ease

d (l

bf)



0.00

0.05

0.10

0.15

0.20

0.25

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00


JP8

Flo

w R

ate

(gp

m)


Comparison of Tube Array and Continually Operating Electrodes

Due to larger scale scatter with the hot block, it was dropped from further consideration

Sequence of 10 tests spread over 3 days; May 28 - 30, 2002

Five tests run against each concept

Test Parameters


Fuel : JP8, 0.2 gpm @ 155 - 161°F

Agent : Halon 1301 @ 5.2#, 100°F, 41 lbf/ft^3

Evaluating performance of each

between different days

statistical spread (consistency)





Comparison of Tube Array & Continually Operating Electrodes (continued)

2.82

2.32

3.37

4.06

2.8

2.29

2.84

4.47

2.67

4.25

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

continuouselectrode -28May02

continuouselectrode -29May02

rhombus stackedtube array -

29May02

rhombus stackedtube array -

30May02

Rei

gnit

ion

Del

ay (

sec)

Rhombus Tube Stack : Average = 3.80 sec, standard deviation = 0.676sContinuous Electrodes : Average = 2.58 sec, standard deviation = 0.258s


Conclusions.

Given the following :

testing occurred between April 1ST and May 30TH evaluating hot surface ignition based on difficulties experienced regarding this issue

ran 111+ tests in this period of time

discharged 220 lbf of HFC-125 and 80 lbf of Halon 1301

evaluated three potential ignition sources to produce a fire that could be used to characterize or challenge the established Halon distribution in agreement with the spirit of the MPSE, and doing so reliably





Conclusions (continued).

A recommendation is put forward to address the enigmatic problem of hot surface ignition with respect to the MPSE; this recommendation being to use continuously operating electrodes to relieve the difficulties associated with the hot surface IGNITION phenomena which would then permit forward motion in producing equivalency information for the relevant replacement candidates

the massive hot plates in the FAA simulator have exceeded their respective life spans

massive hot plates, based on the experience of personnel at WPAFB, are unreliable when considering hot surface ignition phenomena

hot surfaces are subject to degradation during continual fire exposure

the established Halon distribution profile called for in the MPSE must be challenged or characterized in some manner to acknowledge the peak ability of that very same Halon; circumstances otherwise would open the door to inadequate quantities of a replacement candidate/system as being labeled “equivalent” to the performance of Halon which would then lead to an imbalance in the “equivalent level of safety” conceptInternational Aircraft Systems Fire Protection Working Group

London, England 13-14 June 2002




Conclusions (continued).

The basis for recommending continually operating electrodes over the tube array and hot block concepts

regardless of how specified in the MPSE, the challenge to the Halon distribution in the engine application is a turbulent diffusion combustion phenomena; upon initial extinguishment, this phenomena inherently becomes a forcibly ventilated maelstrom requiring a reliable ignition source to obtain full threat ability

hot surface IGNITION, based on previously provided information, simply and utterly complicates the IGNITION component that leads to the secondary combustion process that is specified in the MPSE

issues complicating hot surface IGNITION are not restricted to combustion dynamics alone, but include the materials and geometry actually encompassed in the combustion volume as well

the simple comparison of reignition delay times and their respective statistical behavior for the Halon tests run on May 28 - 30 clearly suggests the continually operating electrodes are likely a better option to define the abilities of any associated Halon distribution




NEAR TERM PLANS

Continue working with the continually operating electrodes while moving in the direction to provide equivalence data

continue evaluating the performance of this concept

begin rationalizing modifications of the MPSE to incorporate this recommendation

Submit data and other pertinent information to the engine task group for review

Wait for task group comments or plan activities as needed to accomplish the feedback portion of the cycle




Engine Nacelle Halon Replacement, FAA, WJ Hughes Technical Center Point of Contact :Doug Ingerson Department of Transportation Federal Aviation Administration.

Documents

hot surface ignition

hot surface temperatures

hot surface ignition

massive hot plate ifuel

existing massive hot

electrical ignition

safety section

f fuel