heat flux study and flame propagation evaluation of ......heat flux summaryheat flux summary bunsen...

Heat Flux Study and Flame Propagation Evaluation of Composite MaterialsEvaluation of Composite Materials

Daniel SlatonTechnical Fellow

Boeing Commercial Airplanesg C pFlammability & Airworthiness

Ricardo Andrade AguilarRicardo Andrade AguilarBoeing Commercial Airplanes

Flammability & Airworthiness

FAA Fire Test Working GroupToulouse, FranceJ 20 21 2012

1

June 20-21, 2012

Copyright © 2012 Boeing. All rights reserved.

Test Method Development OverviewTest Method Development OverviewThe FAA is developing new proposed requirements for non accessible areas FAATC task groups arenon-accessible areas. FAATC task groups are developing new test methods for evaluating flame propagation.

This presentation describes research evaluating the behavior of common materials used in the inaccessible areas under three different test methods:

F Bl k• Foam Block• Radiant Panel• Meeker Burner

The goal of this evaluation was compare test methods and determine if there is correlation.

Summary of test results and recommendations are

2

Summary of test results and recommendations are presented.


AgendaAgenda

Heat Flux Study Heat Flux Study Material Test Matrix Foam Block Testing

1. Foam Block Rig 2. Test Results3. Photos4 Conclusions/Observations4. Conclusions/Observations

Radiant Panel Testing 1. Insulation Test Method (25.856(a))2. Ducting Test Method (low heat flux)2. Ducting Test Method (low heat flux)3. 30 degree orientation4. Conclusions/Observations

Meeker Burner Study

3

Final Remarks


Heat Flux from Foam BlockHeat Flux from Foam Block Calorimeter 1 Reference: FAA Presentation:

Development of a Flame Propagation Test Method forPropagation Test Method for Structural Composite Materials in Inaccessible Areas, 10/19/11.

Calorimeter 1 (60sec)Calorimeter 1 (60sec)

Peak Heat Flux= 55 Kw/m^2

4

Peak Heat Flux= 55 Kw/m 21 Minute Total Heat Flux= 36 Kw-min/min^2


Heat Flux from Meeker and Bunsen BurnerHeat Flux from Meeker and Bunsen Burner

• Calorimeter range: 0-12 w/cm^2

• Conversion Factor: .605 Mv/w/cm^2 + 6%

• Calorimeter location: 1” from base

Temp 1" above burner= 2265F + 50

Meeker:

burner 2265F + 50Heat Flux= 113 kw/m^2

Temp 1" above burner= 1800F

Bunsen Burner:

5

Heat Flux= 74 kw/m^2


Time to Reach Foam Block Heat FluxTime to Reach Foam Block Heat Flux

6Copyright © 2012 Boeing. All rights reserved.

Heat Flux Comparison for 60 Seconds:Heat Flux Comparison for 60 Seconds:Total Heat Flux for 60 Seconds

112

100

120

7480

min

/m^2

)

3640

60

tal H

eat

Flux

(Kw

-m

+ ????

1720

Tot

radiant

Flame15 sec

7

0Foam Block Bunsen Burner Meker Radiant Panel

radiant heat

Meeker


Heat Flux SummaryHeat Flux Summary

Bunsen and Meeker burner can reach the same heat flux as the foam block in less time. Bunsen and Meeker burner tests can generate higher heat flux at a

localized area compared to FB. Using a Meeker burner shows potential as a more stringent test

method and better represents the intermediate scale foam block orientation.

Simpler test for airplane certification- Simpler test for airplane certification- Enhances safety due to higher heat flux


Test Method ComparisonTest Method Comparison

ARAC Goals Current Foam New New Horiz.ARAC Goals CurrentBunsenBurner

Foam Block

New Bunsen Burner

New Meeker Burner

Horiz. RadiantPanel

Enhance Safety: * Greater Application* Larger Ignition/Fuel Source

Simple Test Method 3 1 3 3 2

Fire Threat Correlation* Method & requirements define

correlation potential? ? ?

1 Large samples, configuration specific, many part configurations, variation in foam2 Variation from calibration, complex heat flux/pilot flame contribution, non‐representative test samples, 3 Easy to setup and repeat, accommodates unique sample constructions

highly

p

9

highly

somewhat

no


Material Test MatrixMaterial Test MatrixCategory Description

063 Aluminum Sheet.063 Aluminum Sheet.036 Aluminum Sheet.5" Marinite Board.036"-.072" Structural Bondend Al5lb Nomex core, Low density12lb Nomex core High density

Controls

12lb Nomex core, High density7lb Kevlar core, High density5lb Kevlar core, Low density9lb Kevlar core, High density6lb Alum core, Low density

Floor Panels

8.5lb Alum core, High density5lb Nomex core, Low density.013 Rigid, FR Plastic Sheeting.045 Rigid, FR Plastic Sheeting.070 Rigid, FR Plastic Sheetingg , g.070 Rigid, FR Plastic Sheeting .013 Rigid Woven FG.050 Rigid Woven FG.070 Rigid Woven FG027 Rigid Woven FG

Cargo Liner

10

.027 Rigid Woven FG14.0 ± 1.0 oz.yd^2, Flexible32.0 ± 3.0 oz/yd^2, Flexible


Material Test Matrix (continued)Material Test Matrix (continued)250F Polyester fiberglass fabrica. 4 ply (.036")b 8 ply ( 072")b. 8 ply (.072 )c. 12 ply (.120")250F cure epoxy fiberglassa. 4 ply (.044")b. 8 ply (.095")c, 12 ply (.135")250F cure epoxy fiberglass, FRa. 4 ply (.030")b. 14 ply (.105")c, 24 ply (.180")250F cure epoxy carbon fabric

4 l ( 036")

Composite Laminates

a. 4 ply (.036")b. 10 ply (.090")c, 16 ply (.145")350F cure epoxy fiberglass a. 4 ply (.018")b. 14 ply (.065")b. 14 ply (.065 )c, 24 ply (.106)a. Fiberglass/Crushed Coreb. Carbon Fiber/Crushed CorePhenolic HC/foil dec 0.35 Phenolic FR HC 0.35

Sidewall - Crushed Core

11

Phenolic HC 0.35Phenolic HC 0.375Phenolic HC 0.47 Phenolic HC/foil dec 0.5

Stowbins/closets


Foam Block TestingFoam Block Testing• Built according to FAA Specs Test Data Recorded:

B l th• 30 degrees from horizontal.

• Polyurethane Foam with 10cc of heptane

8 Th l diff l i

• Burn length • Maximum Temperature• “Smoke Time”

• 8 Thermocouples at different locations.

Front edge of shroud

12FAA Rig

Boeing Rig


Foam Block Results: Burn LengthFoam Block Results: Burn LengthThese results represent the burn

434445464748

length results of one sample per configuration.

323334353637383940414243

Edge of Shroud

212223242526272829303132

urn

Leng

ht (i

n)

Polyester Fiberglass laminate

350F Cure Epoxy FG laminate

250F Cure Epoxy Carbon Fabric Laminate

250F Cure Epoxy FG FR Laminate

250 F Cure Epoxy FG Laminate

Cargo liner

Sidewalls/Stowage/Closets Floor Panels

101112131415161718192021Bu

0123456789

10

Base Line

13Could not read burn length Copyright © 2012 Boeing. All rights reserved.

Foam Block Results: Smoke TimeFoam Block Results: Smoke Time480 250 F Cure

Epoxy FG

Smoke Time: 713sec

360

420

250FP l t

350F Cure Epoxy FG

250F Cure Epoxy Cabon Fabric

250F Cure Epoxy FG FR

Baseline

Floor Panels

me

(sec

)

240

300

250F Polyester FG

FR

Cargoliner

Sidewall/Stowage/Closets

Smok

e Ti

m

120

180

0

60

14

Note: Smoke time is def ined as the time when all eight thermocouples reached Copyright © 2012 Boeing. All rights reserved.

Foam Block Results:Max TemperatureFoam Block Results:Max Temperature

1500 0

1600.0350F Cure Epoxy FG

250F Cure Epoxy

250 F Cure Epoxy FG

Floor Panels

1100 0

1200.0

1300.0

1400.0

1500.0

250F Polyester FG

Epoxy Cabon Fabric 250F Cure

Epoxy FG FR

Epoxy FG

Cargo Liner

Baseline Sidewalls/Stowage/Closets

800.0

900.0

1000.0

1100.0

Deg

rees

F

400.0

500.0

600.0

700.0

0.0

100.0

200.0

300.0


Foam Block Temperature ProfilesFoam Block Temperature Profiles Foam Block Results Polyester Fiberglass Laminate

140012 Ply8 Ply4ply

1400.063 AluminumMarinite BoardAluminum Skin

Temperature Profiles for Foam Block Rig Testing "Controls"

600

800

1000

1200

Tem

pera

ture

(F)

4ply

600

800

1000

1200Aluminum Skin

Tem

pera

ture

(F)

0

200

400

0 30 60 90 120 150

Time (sec)

T

0

200

400

0 30 60 90 120 150

Time (seconds) Time (sec)( )

350 Cure Epoxy Fiberglass Laminate

1000

1100

1200

1300

1400

24 Ply14 ply4 ply

1000

1200

1400

250F Cure Epoxy Fiber Glass Laminate

12 Ply

8 ply

4 Ply

400

500

600

700

800

900

1000

Tem

pera

ture

(F)

400

600

800

1000

Tem

pera

ture

(F)

16

0

100

200

300

0 30 60 90 120 150

Time (sec)

0

200

0 30 60 90 120 150

Time (sec)


Foam Block Test Results: Observations

General Testing- Smoke time and max temperature results had significant variation –unable to draw any

conclusions

Laminates:- Burn length beyond flame impingement varied from 0”- 9”g y p g- Thickness pattern : Thickest laminates generate the lowest burn length and thinnest laminates

record the highest burn length. - Polyester Fiberglass laminates record the highest burn lengths.

Cargo Liners:Cargo Liners:- Burn length beyond flame impingement varied from 0” to 7”- Maximum Temperatures ranged from 550F to 1100F

Sidewalls/Storage/Closets:- Burn length beyond flame impingement varied from 4” to 13”- All panels recorded similar maximum temperatures from 1080F to 1290F- Honeycomb core did not present any signs of char.

Floor Panels:

17

Floor Panels:- Burn length beyond flame impingement varied from 3” to 13.5”- Maximum temperatures recorded ranged from 950F to 1450F


Foam Block Post test PhotosFoam Block Post-test Photos

18

250F cure epoxy carbon fabric 4 ply 350 Epoxy Fiber Glass 4 ply250F Cure Epoxy Fiberglass 14 ply



cut off area

19

Floor Panel 12lb nomex core, 4plies of uniglass tape.



20

0.25” Honeycomb/ phenolic prepreg

0.10” Side Wall Crushed Core - Test Face

0.10” Side Wall Crushed Core - back

0.25” Honeycomb/ phenolic prepreg



Rigid Cargo Liner, t = 0.013”- Test side

Rigid Cargo Liner (front side). Specimen still in test fixture. Tedlar

l d ff d i th t t

Rigid Cargo Liner, t = 0.013”- Test side

21

pealed off during the test.


R di t P l T ti S Insulation RP Method - 25 856 (a)

Radiant Panel Testing Summary Insulation RP Method - 25.856 (a) Ducting Test Method 30 degree orientation 30 degree orientation Conclusions/Observations


Radiant Panel Test SpecsRadiant Panel Test SpecsRequirements

Radiant Panel Test Specimen Orientation

Radiant Heat Source (W/cm^2)

Heat Soaked Time (sec)

Pilot Flame Time (sec)

Burn Length(in)

After Flame Time (sec)

Test Per 25.856 (a) Horizontal 1.7 0 15 2 3Ducting Method Horizontal 1.3 60 15 2 4530 Degree Method 30degrees 1.7 60 15 N/A N/A


Radiant Panel Test Results Floor Panels Burn LengthFloor Panels - Burn Length

5Standard Test

4

4.5 Low Heat Test30 Degree Test

Full length 18"

3

3.5

gth

(in)

Full length 18"

1 5

2

2.5

Bur

n Le

ng

0 5

1

1.5

24

0

0.5

8.5lb Alum core, HD 6lb Alum core, 2 LD 9lb Kevlar core, HD 5lb Kevlar core, LD 7lb Kevlar core, HD 5lb Nomex core, LD 12lb Nomex core, HD


Radiant Panel Test Results Fl P l E ti i hi TiFloor Panels – Extinguishing Time

60Standard Test

60 S d

45

50

55 Low Heat Test30 Degree Test

60+ Seconds

35

40

(sec

onds

)

20

25

30

Extin

guis

hing

Tim

e (

5

10

15

E

25

08.5lb Alum core, HD 6lb Alum core, 2 LD 9lb Kevlar core, HD 5lb Kevlar core, LD 7lb Kevlar core, HD 5lb Nomex core, LD 12lb Nomex core, HD

00 0


Floor Panel Test Result Observations:Floor Panel Test Result Observations:30 Degree Orientation Test Method:- Nomex core panels recorded a full burn length for the 30 degree test. All other

panels recorded burn length less than 2”. Unclear how skin layup is involved with the results.

- Nomex core panels had extinguishing times above 45secs for the 30 degree- Nomex core panels had extinguishing times above 45secs for the 30 degree tests. Unclear how skin layup is involved with the results.

- During the 30 degree test we witnessed skin delamination during preheat. This test configuration is very stringent in terms of heat flux evenly across the entire g y g ypanel surface

Ducting Test Method:- After flames for ducting method were generally the shortestg g y25.856(a) Test Method:- 25.856 (a) test obtained longer after flame time for aluminum core panels.


Radiant Panel Test ResultsC Li B L thCargo Liners - Burn Length

7Standard Test

L H t T t

6

Low Heat Test

30 Degree Test

11”

4

5

th (i

nche

s)

2

3

Bur

n Le

ngt

1

27

0.070", Rigid FRPlastic Sheeting

.045" Rigid FR Plastic Sheeting

.013", Rigid FRPlastic Sheeting

.050" Rigid Woven FG

.013", Rigid Woven FG

Flexible 14.0 ±1.0oz/yd2

Flexible 32.0 ± 3.0 oz/yd2


Radiant Panel Test Results C Li E ti i hi TiCargo Liners – Extinguishing Time

28

30Standard Test

Low Heat Test

22

24

26

2830 Degree Test

117 sec

56 sec

16

18

20

Tim

e (s

ec)

10

12

14

Extin

guis

ing

4

6

8

28

0

2




.050" Rigid Woven FG

.013", Rigid Woven FG

Flexible 14.0 ±1.0oz/yd2

Flexible 32.0 ± 3.0 oz/yd2


Cargo Liners Test Result ObservationsCargo Liners Test Result Observations- Generally good flame propagation resistance for cargo liner materials. - Burn length for rigid fine weave recorded the highest burn lengths and after

flame time (this was observed at the thinnest thickness of .013” but not thicker configuration at .050”).

- Due to unique products (resin and reinforcement differences) it is difficult to- Due to unique products (resin and reinforcement differences) it is difficult to draw firm conclusions comparing the three different test methods.


Radiant Panel Test ResultsH b P l B L thHoneycomb Panels – Burn Length

3Standard Test

Low Heat Test

2.5

30 Degree Testing

1.5

2

gth

(inch

es)

1

Burn

Len

g

0.5

30

0Phenolic HC/foil dec

0.35" Phenolic FR HC

0.35" Phenolic HC

0.35" Phenolic HC

0.375"Phenolic HC

0.47" PhenolicHC/foil dec

0.5"


Radiant Panel Test Results H b P l E ti i hi TiHoneycomb Panels – Extinguishing Time

4Standard Test

3

3.5

Low Heat Test

30 Degree Testing

2.5

(sec

onds

)

1.5

2

ngui

sing

Tim

e (

0.5

1Exti

31

0Phenolic HC/foil

dec 0.35" Phenolic FR HC

0.35" Phenolic HC

0.35" Phenolic HC

0.375"Phenolic HC

0.47" PhenolicHC/foil dec

0.5"Copyright © 2012 Boeing. All rights reserved.

Honeycomb Panel Test Result ObservationsObservations

- Burn length results were all below 2’’ for all panels and test methods- After flame time was below 4 seconds for all panels and test methods

Note: Only tested one ply skins with different core thicknesses. All phenolic prepregs.


Radiant Panel Test ResultsC it L i t B L thComposite Laminates - Burn Length

4

Standard Test

3

Low Heat Test30 Degree Testing

th

2

Bur

n Le

ng

1

0

Note: bars off the chart represent a burn lengths of above 10" to full burn Copyright © 2012 Boeing. All rights reserved.

Radiant Panel Test ResultsC it L i t E ti i hi TiComposite Laminates – Extinguishing Time

464850

343638404244

Standard Test

Low Heat Testec)

22242628303234 Low Heat Test

30 Degree Testing

hing

Tim

e (s

e

10121416182022

Ext

ingu

ish

All values off02468

10

34

All values off the chart extinguished after 60 +sec


Composite Laminate Test Result ObservationsObservations- Burn length using 25.856(a) tests was very similar ranging from .5” to 2”.- Burn length using the ducting method gave burn lengths below 2” for all but one

4 ply sample. - 30 degree test method gave the highest burn lengths and after flame times- 4 out 5 “4ply” laminates tested by the 30 degree method recorded values from

11” to full length.- Polyester FR laminates had very low extinguishing times for all three test

methods and all three thicknessesmethods and all three thicknesses.- In general, the thinner the laminate the higher the burn length and extinguishing

time.Specific resin system will influence how thickness influences test results- Specific resin system will influence how thickness influences test results.


Radiant Panel Test Photos:Radiant Panel Test Photos:

25.856 (a)

4ply 250F Epoxy Fib l

4ply Polyester FR Fiberglass

4ply Epoxy Carbon

Ducting

Fiberglass Fiberglass

Ducting Method

36

12 ply 250F Epoxy Fiberglass

Rigid Cargo Liner Phenolic Honeycomb, t = 0.35”


Radiant Panel PhotosRadiant Panel Photos

30 degree Test Method

4ply 250F Epoxy Fiberglass

8ply 250F Epoxy Fiberglass

8ply Polyester FR Fiberglass

4ply 250F Epoxy Carbon

250F


Foam Block & Radiant Panel TestG l C l i / Ob tiGeneral Conclusions/ Observations- Different flame dynamics

Foam Block: Unstable flame and heat loss (no radiant heat source)Radiant Panel: Constant flame & radiant heat source.

- Different flame orientationsFoam block: Applies heat directly underneath.Radiant Panel Tests applies the flame to the top of the test co pon at an angleRadiant Panel: Tests applies the flame to the top of the test coupon at an angle.

- Visibility/Witnessing Foam Block: Test configuration does not allow to visually witness the test. Therefore, there is no way to identify after flame time. Temperature profiles provides some qualitative indication of continued combustion of the test sample.Radiant Panel: Can witness the entire testing without any visual constrains.

- Test results:Foam block: Difficult to read burn lengths for floor panels and stowage bins/closet panels.g p g pRadiant Panel: Easy to observe flame propagation.Both tests can only be compared by using burn length results not extinguishing times.

- Conclusion:No clear correlation of testing results between RP tests and Foam Block test

38

No clear correlation of testing results between RP tests and Foam Block test.

For laminates, a correlation does not exist. Due to the low number of samples tested, repeatability in the test methods have not been determined.


Meeker Burner Test StudyMeeker Burner Test Study

Meeker Burner has a wider opening and hotterMeeker Burner has a wider opening and hotter flame than the Bunsen burner. It can generate a heat output of 3.5kWh and flame

temperatures of 2250°F + 50S l l t d t 30d f h i t l Sample located at 30degrees from horizontal. Flame applied to test specimen for 30 seconds.


Initial Meeker Test StudyInitial Meeker Test Study

.045" Rigid Cargo Liner A

.070" Rigid Cargo Liner B

.065” 14 ply 350F cure Epoxy Fiberglass

.018” 4 ply 350F cure Epoxy Fiberglassp y p y g

.030” 4 ply Epoxy FR Fiberglass

.036” 4 ply Polyester FR Fiberglass

.044” 4ply 250F cure Epoxy Fiberglass

.50” Fiberglass Epoxy Floor Panel, 12lb Nomex.50 Fiberglass Epoxy Floor Panel, 12lb Nomex


Meeker Test Results: Burn LengthMeeker Test Results: Burn Length

10

8

10

6

Leng

th (i

n)

4

Bur

n L

2

41

0.045" Rigid .070" Rigid 350F cure e/fg

14ply350F cure e/fg

4plyEpoxy FR 4 ply Polyester 4 ply 250F cure e/fg

4ply12lb Nomex,

HD,


Meeker Test Results: After Flame TimeMeeker Test Results: After Flame Time

55

60

45

50

)

Extinguished after 90 sec

30

35

40

uish

ing

Tim

e (s

ec)

15

20

25

Extin

gu

5

10

42

0.045" Rigid

BMS 8-2.070" Rigid

BMS 8-2350F cure e/fg

14plyBMS 8-139

350F cure e/fg4ply

BMS 8-139

Epoxy FR 4 plyBMS 8-151

Polyester 4 plyBMS 8-80

250F cure e/fg4ply8-79

12lb Nomex,HD,

BMS 4-17


Observations/PhotosObservations/Photos

43

350 F Epoxy Fiber Glass 14 ply 250 Epoxy FR Fiber Glass 4 ply .070 Rigid , FR Plastic Sheeting


Meeker Temperature ProfilesMeeker Temperature ProfilesAluminum Base Line

2000Chan 1Chan 2Chan 3Chan 4

1800

2000


Chan 1

Chan 2

Chan 3

800

1000

1200

1400

1600

1800

Tem

pera

ture

(F)

800

1000

1200

1400

1600

Tem

pera

ture

(F)

Chan 4

0

200

400

600

0 10 20 30 40 50 60

Time (sec)

0

200

400

600

0 10 20 30 40 50 60 70 80 90 100

Time (sec)( )

1600

1800

2000

350F Epoxi Fiber Glass 14 ply

Chan 1

Chan 2

Chan 3

Chan 42000

2500

350 Epoxi Fiber Glass 4 Ply

Chan 1

Chan 2

Chan 3

Chan 4

600

800

1000

1200

1400

Tem

pera

ture

(F)

1000

1500

Tem

pera

ture

(F)

44

0

200

400

0 20 40 60 80 100 120

Time(sec)

0

500

0 20 40 60 80 100 120

Time (sec)


Meeker Burner ObservationsMeeker Burner Observations Additional Testing needs to be done to draw any conclusions about similarity

to foam block or radiant panel resultsto foam block or radiant panel results. Limited testing indicates closer correlation to Foam Block test method than

radiant panel. Holding fixtures and test enclosures need more development Need to develop Holding fixtures and test enclosures need more development. Need to develop

a way to control air/gas mixture. Meeker burner allows for easy observation of ignition, propagation and after

flame time. Recommended Next steps:

1. Further Meeker burner evaluations to standardize burner test setup.2. Evaluate additional laminates (different thicknesses and quantities) to understand

repeatability. 3. Evaluate and compare samples in other flame propagation test methods (e.g. Vertical RP)


Final RemarksFinal Remarks Using the radiant panel at a 30 degree orientation is not representative of the

intermediate scale foam block.intermediate scale foam block. 25.856(a) & ducting test methods need further work to understand the viability

and correlation to intermediate scale foam block. The combination of heat flux from the radiant panel and pilot flame needs to be better understood in any RP test configurationtest configuration. Challenges of Radiant Panel Test Methods

- Calibration- Test complexityTest complexity- Test configuration limits ability to test complicated part designs (more adaptability to

parts using meeker).- Multiple configurations and test setups/requirements (Not material universal)

Using a Meeker burner shows potential as a more stringent test method and better represents the intermediate scale foam block orientation.

- Simpler Test for airplane certification- Enhances safety due to higher heat flux (comparative to FB)

46

Enhances safety due to higher heat flux (comparative to FB)- Supports ARAC goals


H t Fl d Fl P ti E l tiHeat Flux and Flame Propagation Evaluation

Thank You!

heat flux study and flame propagation evaluation of ......heat flux summaryheat flux summary bunsen...

Documents