Results from Gaseous Methane/Oxygen Mixture Testing

Reema RevelesMike Bangham, James Allen, Jairus Schwartz, Ross

Lambert, and Brendan ManganBangham Engineering Inc.

2018 International Explosives Safety Symposium & Exposition

• LOX-LNG fueled launch vehicles

– BE-4 by Blue Origin

• Advantages of LOX/LNG

– Denser than LH2 requiring smaller

tanks

– Simpler turbopump design

– Unlike RP-1 methane can be

used to self-pressurize

– Cost

• DoD Safety Standards

Introduction

Blue Origin’s BE-3 throttles during acceptance testing. Credit: Blue Origin

Energetic liquid TNT Equivalence

LO2/LH2Larger of 8 W2/3, or,

14%W, where W is total weight of LO2/LH2

LO2/RP-1 10%

2

• Establish Quantity-Distance for LOX-LNG explosions

• Expect gaseous explosion envelopes cryogenic

• Methane – oxygen gas test setup

Methane/Oxygen Unconfined

Explosion/Combustion Test Series

3

Balloon sizes

6 ft.(7 tests)

12 ft.(18 tests)

14.5 ft.(5 tests)

16 ft. (15 tests)

Methane Mixture ratios

Stoichiometric

+-5% by volume

+-10% by volume

+-15% by volume

Test Site

Probe setup top view

4

MUCTA test site

• Pressure probes

– Piezoelectronic Quartz ICP blast

pressure pencil probe

Instrumentation

5

Metal Track Assembly

Angular

Adjustment

Cable

Protection

Fig: Blast pressure pencil

probe mounting assembly

12.0 ft. diameter balloon, 37%

CH4, 63% GOX by volume

• Photron AX200 monochrome camera– Max. pixel resolution: 1024 x

1024

– Max. frame speed: 6,500 fps at 1 megapixel

– Captures the combustion flame front

Cameras

6

• Phantom v2512 colored camera – Max. pixel resolution:1280 x

800

– Max. frame rate: 25,700 fps

– Wide-angle view capturing the shock wave created from the blast

Fig: High speed

camera tower with

accompanying support

and camera

enclosures

• Long distance acoustic sensors– PCB 377C10 piezoelectric microphone

• Triaxial Accelerometers mounted inside witness object– +-500g pk PCB Piezoelectric 356A02

Instrumentation

7

CAD of witness object

mounted on the pole with

pressure probe

Triaxial accelerometer

mounted to welded stud

plate inside witness

object

SRS of 16-ft diameter

balloon CH4-GO2

blast event at 50 ft.

from COE

• Sensor arm with pressure probes mounted inside the

balloon for the last test

– PCB Piezoelectronics model 113B26 ICP pressure sensors

Instrumentation

8

Fig: CAD diagram

of Sensor arm

dimensions and

specifications

Fig: Sensor

arm after the

test

• As 𝑟 → 𝑟0, i.e. at the edge of the gas cloud 𝑂𝑃 →180𝑝𝑠𝑖

• 𝐎𝐏𝐦𝐚𝐱~ 𝟐𝟐𝟎𝐩𝐬𝐢 to account for increase with eq. ratio and 10% margin

Overpressure curve fit

9

Impulse

10

• As Impulse decays with scaled distance, it increases with

increasing mass.

• The autoignition limit on impulse needs to be verified in the

cryogenic test

Impulse = 38.05 𝑚𝐶𝐻40.37 𝑟/𝑟0

−0.86 psi − ms

Increasing mass

16 ft Methane/Oxygen Video

Flame speeds

• Flame speed is based on

speed of burning gas

outwards from the ignitor till

it reaches the balloon ~560-

600ft/s

• Subsequent detonation

around 5000-15000 ft/s

100

1000

10000

0 5 10 15

Velocity(ft/s)

Time(ms)

12-ft.slowcombustion

12-ft.detonation

100

1000

10000

0 5 10 15

Velocity(ft/s)

Time(ms)

16-ft.slowcombustion

16-ft.detonation

12

• Equivalent weight of TNT

TNT Equivalency

13

• Inaccurate

representation of

vapor cloud

explosions

Overpressure comparison

14

• In general, TNT based predictions are higher than MUCTA in

the near field, and matches the far field data.

• The discrepancy increases with increasing fuel mass

• The comparison with cryogenic explosions might reveal a

larger discrepancy

1

10

1 10

Pea

k o

ver

pre

ssu

re (

psi

)

Scaled distance, Z (ft/lb1/3)

12-ft diameter balloon

Kinney prediction

Brode prediction

Kingery-Bulmash prediction

MUCTA experiment 1

10

100

1 10

Pea

k o

ver

pre

ssure

(psi

)

Scaled distance, Z (ft/lb1/3)

16-ft diameter balloon

Kinney prediction

Brode prediction

Kingery-Bulmash prediction

MUCTA experiment

• TNT predicted impulse is only dependent on the scaled distance

• MUCTA models impulse decay with scaled distance and a simultaneous rise with increasing mass of fuel involved in the explosion

• Limiting of impulse with autoignition needs to be modeled.

Impulse comparison

15

0

10

20

30

40

50

60

0 5 10 15 20 25 30

Po

siti

ve

ph

ase

Imp

uls

e (p

si-m

s)

Scaled distance, Z (ft./lb1/3)

MUCTA exp. 12-ft.

MUCTA exp. 16-ft.

TNT based prediction

Long Distance Acoustics

• Blast shock waves decay to acoustic pressure waves at far field distances from the COE

• Overpressure and impulse models are not applicable

• Community noise and safety around test ranges are affected by blast testing.

0.5

1

1.5

2

2.5

0 5 10 15 20

Sh

ock

Mac

h N

um

ber

Distance/massCH41/3, Z (m/kg1/3)

6-ft diameter12-ft diameter16-ft diameterCurve fit 2.4Z^-0.35

• ISO 9613-1:1993 standard is used to model the acoustic decay

• High frequency content is rapidly absorbed by air, low frequency content is transmitted through longer distances and through any structures in the way.

• The waves inside the room can get amplified and cause room modes to occur- result in building damage through a resonance phenomenon similar to the amplification of sound in a drum

Low frequency propagation

17

Long Distance PCB

Microphone 1 - 861 ft

Long Distance PCB

Microphone 2 - 966 ft

SPL = -11.18ln(x) + 233.25dB

120

130

140

150

160

170

180

190

200

0 500 1000 1500 2000

So

un

d P

ress

ure

Lev

el [

dB

]

Distance from COE [ft]80

90

100

110

120

130

140

150

160

170

180

10 100 1000 10000

Sou

nd P

ress

ure

Lev

el [

dB

SP

L]

Sound Frequency [Hz]

r/r0 = 6.5 r/r0 = 10 r/r0 = 25r/r0 = 75 r/r0 = 150 r/r0 = 250

Faster decaySlower decay

OP/p0 = 6.25(r/r0)-1.29

0

0.1

0.2

0.3

0.4

0.5

0.6

0 100 200

OP

/p0

Normalized Distance from COE (r/r0)

Slower decay than blast wave

• The images from the combustion zone shows that the methane-oxygen mixture transitions from a deflagration to detonation (DDT) in most of the tests.

• This bounds the pressure to an upper limit at the edge of the gas cloud irrespective of the volume

• Overpressures decay as a power law of r/r0

• Impulse decays with r/r0 and increases with mass.

• Acoustic decay is modeled to represent community noise and safety issues.

Conclusions

18

Future Work

19

• Phase 1 – gas explosion tests to set probable envelope -completed

• Phase 2 – cryogenic explosion tests to establish final QD and compare with TNT based equivalent weights -ongoing

• Phase 3 – ground impact to establish public safety risks -TBD

• ULA Contract FA8811-16-9-0004

• Douglas Gogel, ULA

• Redstone Test Center (RTC) Huntsville, AL

Acknowledgements

20

THANK YOU

21