Presented to:

By:

Date:

Suppression, Detection and Signaling Research and Applications Symposium (SUPDET)

Matthew E. Karp

September 2019

Methods for Characterizing Artificial

Smoke Generators for Standardizing

Inflight Smoke Detection Certification

Overview

– Section 1 Background

– Section 2 Methods and test equipment

– Section 3 Potential parameters

– Section 4 Conclusions

2

Test Apparatus

Section 1

Background

• Federal Aviation Administration (FAA) regulations require that a commercial

aircraft cargo compartment smoke detection system must provide visual

indication to the flight crew within one minute after the start of a fire 1.

• Further FAA guidance states that the smoke detection certification test is

designed to demonstrate that the smoke detection system will detect a

smoldering fire that produces a small amount of smoke 2.

• In an attempt to eliminate the frequency of false alarms, the FAA issued a

Technical Standard Order to adopt the Minimum Performance Standards of

smoke detector equipment, which includes criteria for resisting alarms from

nuisance sources such as water vapor, insecticide aerosols, dust and light 3.

3

1 Title 14 Code of Federal Regulations (CFR) Part 25.858, 2/10/1998

2 Federal Aviation Administration Advisory Circular 25-9A

3 TSO C1e, 8/19/2014

Section 1

Background Continued

• Due to health and safety concerns – artificial smoke generators are used for

inflight certification testing

• Aerosols created by artificial smoke generators must have similar size

distributions and light scattering characteristics in order for the false alarm

resistant smoke detectors to alarm

• standardizing the artificial smoke generators for the total quantity of aerosol

production, rate of aerosol production and repeatability of aerosol production

is necessary to ensure the reliability and integrity of the inflight smoke

detection certification test

4

Section 2

Methods and Test Equipment

• Testing is conducted in an environmental chamber

that is capable of simulating ambient conditions that

are typical to flight

• Blue and infrared, light scattering measurement

(LSM)

• Scanning mobility particle sizer (SMPS)

• 6 vertically aligned light obscuration meters

• Four aircraft manufacturer’s smoke generators and

settings are tested and compared

– Annotated as: MFR 1a, MFR 1b, MFR 1c, MFR 2a,

MFR 2b, MFR 3, and MFR 4

• Vane anemometers characterize smoke transport

5

Test Apparatus

Method

• A smoke generator is turned on to produce an

aerosol for 60 seconds to best represent the test

procedures outlined in 14 CFR 25.858

• Light obscuration and light scattering

measurements are taken through the entire test

• SMPS measurements are taken at 210 seconds

• Additionally, size distributions and light scattering

characteristics from smoke created by smoldering

foam, smoldering wood, and lithium-ion battery

thermal runaway vent-gas are measured and

compared to aerosols produced by the smoke

generators

6

Altitude Chamber

Smoke Transport

• A cone is connected to the

smoke generator’s chimney

• Attached to the cone is a vane

anemometer to measure the

volumetric flow rate

• The volumetric flow rate is

directly correlated with

chimney heat output

7

Siemens Cerberus with Volumetric Flow

Rate Cone

Vane

anemometer

Cone

Chimney

Smoke Generators

• Theatrical smoke generators use an inert gas to

propel mineral oil into a heat exchanger, where

the solution is vaporized to create smoke.

• The theatrical smoke exits through a chimney

incorporated with heaters to create a thermally-

buoyant plume

• Important variables of smoke generators

– Gas propellant

– Gas propellant pressure

– Chimney heater temperature

– Mineral oil characteristics

• Viscosity and refractive index

8

Simplified Smoke Generator

Schematic

Inert Gas Regulator

Mineral Chimney Stack

Oil and Heaters

Heat Exchanger

Nozzle

Mie Scattering Theory

• Mie Scattering Theory governs light

scattering by sub-micron particles

• A simplified approximation of Mie Scattering

Theory is given by van de Hulst [4]

𝑸 = 𝟐 −𝟒

𝒑𝐬𝐢𝐧 𝒑 +

𝟒

𝒑𝟐(𝟏 − 𝐜𝐨𝐬 𝒑 )

• Where Q is the efficiency factor of scattering

𝒑 =𝟒𝝅𝒂 𝒏 − 𝟏

λ

Blue Scattering Intensity

IR Scattering Intensity

0 0.2 0.4 0.6 0.8 1

0

0.5

1

1.5

2

2.5

3

3.5

4

0

2

4

0 0.25 0.5 0.75 1

Effi

cie

ncy

Fac

tor

of

Scat

teri

ng

Diameter, µm

This shows that the scattering intensity is

a function of

n, Refractive index of the particle

a, radius of the particle

λ, Wavelength of the incident lightMaxwell’s Equation plotted using 470nm and 850nm

wavelengths, refractive index 1.5

[4] van de Hulst, H. C. (1957). Light scattering by small particles.

New York: John Wiley and Sons. ISBN 9780486139753.

Electromagnetic Radiation

Scattering Measurement (LSM)

• Two LEDs

– 470nm blue LED forward

scattering angle of 135°

– 850nm IR LED forward

scattering angle of 135°

• Measurements used to calculate:

– Percentage of blue to blue and IR

light scattering intensity

– Utilize van de Hulst

approximation to characterize

very small particles

LEDs

Silicon

Detectors

45°

Smoke

Electromagnetic Radiation Scattering

Measurement (LSM)

Scanning Mobility Particle Sizer (SMPS)• How does the SMPS work?

– An impactor removes large particles and

measures flow.

– A neutralizer creates a well-characterized

charge distribution on the particles.

– Inside a Differential Mobility Analyzer

(DMA), the charged particles experience

an electrical field that separates particles

based on their electrical mobility and

outputs a monodisperse aerosol.

• Electrical mobility is inversely related to

particle size

– The condensation particle counter (CPC)

counts the monodispersed particles as

they exit the DMA.

11

Co

nce

ntr

atio

n

DiameterSMPS Output

SMPS Simplification

• Particle characterization

– LSM

– SMPS

• Light obscuration

– Transient obscuration

– Steady state obscuration

– Repeatability

• Ambient environment

• Smoke transport

LSM and Vane Anemometer Cone

Section 3

Potential Parameters

MFR 1

MFR 2

MFR 3

MFR 4

Smoldering Foam

Lithium Batteries

Smoldering Wood

0

100

200

300

400

0 0.25 0.5 0.75 1G

eo

met

rica

l Me

an D

iam

ete

r, n

m% Blue Signal

Particle Size - LSM and SMPS

• Smoke generators used by airline

manufacturers are capable of

producing an aerosol that is

comparable to particles from certain

smoke sources

• There is a strong negative correlation

between the percentage of blue to

blue and IR light scattering intensity

and geometric mean diameter for

individually sourced particles

• Strong correlation diminished when

comparing particles from multiple

sources with varying refractive

indexes

LSM %blue signal vs SMPS diameter

Time vs light obscuration

Various manufacturer smoke generator settings

Light ObscurationTransient

Obscuration

Steady State

Obscuration

• Transient Obscuration

– Initial 60 seconds

– Characterizes rate of aerosol

production

• Steady State Obscuration

– After 120 seconds when the

aerosol fully mixes

– Characterizes total aerosol

production

• Repeatability

– Relative deviation between tests

over 10 second increments

0

10

20

30

40

50

60

70

80

90

100

0 20 40 60 80 100 120 140 160 180

Ob

scu

rati

on

, %/f

t

Time, s

MFR 1a MFR 1b MFR 1c MFR 2a MFR 2b

MFR 3 MFR 4 foam wood

Transient Obscuration

Time vs light obscuration

Used to determine transient light obscuration

• Transient obscuration characterizes the rate of aerosol production

• The data points represent the time to surpass the marked light obscuration threshold

• The steeper the curve, the more rapid the aerosol production rate

• The highest point on the curve represents the maximum light obscuration reached

• There is a range of aerosol production rates used for smoke detection certification testing

MFR 1b

MFR 2b

MFR 3

MFR 1a

MFR 1c

MFR 4

MFR 2a

0 10 20 30 40 50 60

0

20

40

60

80

100

120

0

25

50

75

100

0 20 40 60%

ob

s/ft

Time, s

Steady State Obscuration

Time vs light obscuration

Used to determine steady state light obscuration

• Steady state obscuration characterizes the total aerosol production

• Tests conducted in a 7.8m3 vessel

• The average steady state obscuration is 32 %/ft with a standard deviation of 17 %/ft

• There is a range of total aerosol production used for smoke detection certification testing

– Two perspectives• This is ok, because of the varying

size of aircraft cargo compartments

• This is not ok, because a fires beginning does not discriminate against aircraft size

6

23

25

30

32

49

64

0 20 40 60 80

MFR 1a

MFR 2a

MFR 3

MFR 1b

MFR 2b

MFR 1c

MFR 4

Steady State Obscuration, %/ft

Repeatability

• The percent deviation of light obscuration per foot between a minimum of three tests over 10 second increments are calculated to determine test repeatability

• The initial 10 seconds are the least repeatable and arguably the most significant portion of certification testing

• The first 10 seconds have an average of 24% deviation between tests from an individual smoke generator

Time vs percent deviation

Used to determine repeatability

5, 24%

15, 15%

25, 9%35, 7%

45, 7%

55, 7%

0

10

20

30

40

50

0 20 40 60

Perc

ent

Dev

iati

on

Time, s

MFR 1a MFR 1b MFR 2a MFR 2b

MFR 3 MFR4 Average

Ambient Environment and

Steady State Obscuration

Steady State Light Obscuration by Ambient

Temperature

• Positive correlation between

ambient temperature and the

total aerosol production

• The significance of the

ambient temperature on a

smoke generator’s aerosol

production varies by smoke

generator and setting 32.128.4

22.621.7

48.828.4

30.216.6

27.225.1

10.05.5

0 20 40 60

MFR 2b 30C 101kPaMFR 2b 10C 101kPa

MFR 2a 30C 101kPaMFR 2a 10C 101kPa

MFR 1c 30C 101kPaMFR 1c 10C 101kPa

MFR 1b 30C 101kPaMFR 1b 10C 101kPa

MFR 1b 30C 69.7kPaMFR 1b 10C 69.7kPa

MFR 1a 30C 69.7kPaMFR 1a 10C 69.7kPa

Steady State Light Obscuration, %/ft

Ambient Environment and

Steady State Obscuration

Steady State Light Obscuration by Ambient

Pressure

• Weak negative correlation

between ambient temperature

and the total aerosol

production

• Increasing the ambient

pressure can, but does not

always, decrease the total

aerosol production 48.8

49.6

30.2

27.2

16.6

23.9

5.6

10.0

0 20 40 60

MFR 1c 30C 101kPa

MFR 1c 30C 69.7kPa

MFR 1b 30C 101kPa

MFR 1b 30C 69.7kPa

MFR 1b 10C 101kPa

MFR 1b 10C 69.7kPa

MFR 1a 30C 101kPa

MFR 1a 30C 69.7kPa

Steady State Light Obscuration, %/ft

Smoke Transport

20

• Volumetric flow rate characterizes

the smoke transport

• The volumetric flow rate is directly

correlated with chimney heat output

• There is a range of volumetric flow

rates used for smoke detection

certification testing

– Previous large scale testing has

shown that detection time is

significantly reduced by

increasing volumetric flow rate

Volumetric flow rate

Comparison volumetric flow rate of various smoke

generator settings with Siemens Cerberus and Aviator 440

13

13

13

13.2

16.8

16.8

19.4

MFR 2a

MFR 2b

MFR 3

MFR 1a

MFR 1b

MFR 1c

MFR 4

Volumetric Flow Rate, ft3/min

• Smoke generators are capable of producing an

aerosol with similar particle size distributions and

light scattering characteristics to real smoke sources

• There is a measurable difference between the total

aerosol production, rate of aerosol production and

smoke plum velocity between smoke generators

used by major airline manufacturers for certification

testing

• A smoke generator’s aerosol production can vary

with the ambient temperature range associated with

a typical aircraft flight profile

Section 4

Conclusions

Contact Information

- Matthew Karp

- Research Engineer

- FAA ANG-E211 Systems Fire Protection

- Phone: 609-485-4538

- Email: Matthew.Karp@FAA.gov

22