Fiber Bragg grating temperature sensor-A review

Fiber Bragg grating temperature sensor

－

A

review

Abstract

:In recent years, considerable progress has been made in optical fiber sensors using a fiber Bragg

grating (FBG). This is because such FBG sensors have many advantages for practical applications: ease of

multiplexed operation, compactness, quasi-point sensing capability and simple structure.

Fiber Bragg

gratings (FBG)have generated much

interest for use as sensors for strain, temperature, and other physical

quantities because of their important properties such as immunity to electromagnetic noise, high sensitivity,

compactness, and simplicity of fabrication.

OCIS codes: (060.2370) Fiber optics sensors; (230.1480) Bragg reflectors.

1. Introduction

Fiber Bragg grating (FBG) has been intensively studied

and developed as an optical sensor

for various sensing applications, such as health monitoring of civil structure, non- destructive

testing of composite materials, smart structure, and traditional stain, pressure, and temperature

sensing [1]. To realize practical sensor systems with multiplexing capability, various kinds of

FBG

array

interrogation

techniques

have

been

suggested

[2].

Recently,

the

demand

for

warning the abnormal temperature increase at restricted spaces has been rapidly increasing. In

particular, due to different tolerable thresholds required at different locations, it is necessary

to

construct

the

low

cost

and

easy

interrogation

configuration.

In

this

letter,

a

novel

multi- point

temperature

warning

sensor

using

a

multi-channel

matched

FBG-based

multi-wavelength

pulsed

laser

is

proposed

and

demonstrated.

The

sensor

has

several

advantages,

including

flexible

setting

of

the

tolerable

temperature,

simple

structure,

high

signal-to-noise ratio, low system cost and quick response.

2. Background

（

1

）

Intracore

fiber

Bragg

gratings

for

strain

measurement

in

embedded

composite structures

The Bragg wavelength is given by

?

B

?

n

eff

?

(1)

where

n

eff

< br>

is

the

effective

refractive

index

of

the

fiber

(modal

index)

and

L

is

the

Bragg

grating

period.

We

obtain

the

Bragg

wavelength

shift

?

?

B

in

the

temperature

sensor

by

differentiating Eq. (1) with respect to temperature:

?

?

B< /p>

?

2

?

n

eff

?

?

2

n

eff

,

(2)

For a bare FBG, with temperature variation, the second term on the right-hand side of Eq. (2)

is negligible because the thermally induced fiber elongation effect (

??

) is much smaller than

the

effect of the

refractive index change. Attaching a metal strip to the

FBG can make the

- 1 -

fiber elongation [the second term on the right-hand side of Eq. (2) much larger than that of

bare FBG

.

（

2

）

Measuring Thermal and Mechanical Stresses on Optical Fiber in a DC

Module Using Fiber Bragg Gratings

FBGs

can

be

used

as

sensors

to

monitor

stress

and

temperature

during

fiber

processing,

handling, installation, and in service events. When an FBG is subjected to a combination of

mechanical and thermal loading, the return Bragg wavelength will shift proportionately to the

magnitude

of

the

load.

It

is

important

to

be

able

to

decouple

the

mechanical

and

thermal

response of the reflected Bragg wavelength, in order for this sensor to achieve its intended

usefulness of obtaining an accurate measure of mechanical stress when temperature varies.

The shift in the return Bragg wavelength as a function of temperature and stress is given by

[3]

λ

m =

?

B

k

?

?

?< /p>

?

?

B

kT

?

T

?

?

B

(3)

where:

?

m

measured Bragg wavelength;

?

B

Bragg

wavelength

at

a

reference

condition

(usually

at

room

temperature

and

stress

free);

?

?

change

in

stress,

where

it

is

assumed

that

stress

is

linearly

related

to

strain;

Δ

T

change in temperature;

k

?

stress coefficient;

k

T

temperature coefficient;

?

B

k

T

=

m

?

, grating stress sensitivity (calibration constant);

?

B

k

T

?

m

T

,grat ing temperature sensitivity or (calibrationconstant).

According to (3), the temperature coefficient

k

T

is

independent of wavelength. Therefore,

one only needs to measure the change in temperature to account for the thermal contribution

to

the

shifted

measured

wavelength

λm

from

the

initial

reference

wavelength

?

B

.

The

thermal

and

stress

parameters

k

T

and

k

?

,

respectively,

in

(1)

have

been

reported

in

a

variety of published studies and are summarized in Table I. In most of these studies,

k

T

and k

σ

are not given explicitly but can be calculated by dividing the published temperature

and

stress

sensitivities

(calibration

constants

for

the

grating)

by

the

unperturbed

Bragg

wavelength

λ

B of the grating. The thermal and stress coefficients were converted to SI units

when appropriate. Note that Table I is not an exhaustive review on the subject, but should

- 2 -

provide enough data for comparison.

The dependence of the index of refraction of glass on temperature is reported to cause an

equivalent nonlinear error of 2

?

C over a temperature range of

?

30 to 80

?

C [11]. However,

this

thermal

nonlinearity

effect

in

the

glass

fiber

is

not

sufficient

to

be

detected

by

the

wavelength measurement system used in most studies and will be assumed to be negligible in

this

study.

Several

researchers

[7],

[12]

who

have

measured

linear

temperature

ranges

for

Bragg- grating responses have also assumed the thermal nonlinearity effect in glass fiber to be

negligible. Whereas there is experiment-to-experiment variability in the reported temperature

response of uncoated FBGs in Table 1, their behavior is essentially linear with temperature

between

?

40 and 100

?

C.

One study [12] found the protective polymer coating to have a pronounced effect on the

measured thermal coefficient at low temperatures. The thermal behavior of the coated grating

becomes nonlinear for temperatures below 0

?

C.

From

the

statistical

portion

of

Table

I

,

The

variation

in

the

stress

coefficients

is

significantly greater than the variation in the thermal coefficients. This is a surprising result,

since the maximum wavelength shift for a temperature differential of 100

?

C is close to 1 nm,

while a change in stress of 0.2 GPa will produce a wavelength shift close to 4 nm. The data

from Echevarria et al.

[4] indicates that the large variation in the stress coefficient is real and

significant. Echevarria used the first- and second-order diffraction peaks from a single Bragg

grating to monitor wavelength changes and found that the two resulting thermal coefficients

were essentially identical and the stress coefficients varied significantly.

The

work

by

Shu

et

al

.

[9]

suggests

that

the

method

for

making

an

FBG

affects

the

measured

thermal

and

stress

coefficients. The

effect of

drawing

tension [10]

has

also

been

shown

to

affect

the

photosensitivity

of

codoped

core

fibers

and

the

formation

of

Bragg

gratings. These differences in the manufacturing process of the FBGs may help explain the

variations in the measurements of the thermal and mechanical coefficients within a particular

study and between studies [7].

For

FBGs

manufactured

using

similar

techniques

and

precisely

controlled

processes,

(3)

will generate

a

wavelength in dependent thermal coefficient [4], [11].

Echevarria

et al.

[4]

monitored the first- and second-order diffraction wavelengths from a single grating to obtain

thermal

coefficients

within

0.7%

for

a

wavelength

difference

of

768

nm

between

the

two

diffraction

orders.

In

the

study

by

Flockhart

et

al.

[11],

three

FBGs

written

at

separate

wavelengths

were

thermally

characterized.

The

fiber

gratings

were

written

into

hydrogen

loaded fiber by a two-beam holographic exposure using a frequency- doubled argon-ion laser.

The two-beam holographic writing process with the frequency- doubled

argon-ion laser is

a

more stable method than the more common process of using phase masks and an eximer UV

source [5]. The three gratings produced thermal coefficients within 0.2% over a wavelength

range of 53 nm. It should be noted that nominal values for the thermal coefficients found in [4]

and [11] differ by 1.4%. The orientation of the Bragg grating within the optical fiber does not

appear to have a significant impact on the thermal coefficient for a tilt angle of 5.5

?

[8].

Table

1,

PUBLISHED

THERMAL

AND

STRESS

COEFFICIENTS

FOR

FBGS

OF

VARIOUS WAVELENGTHS, RECOAT STATUS, AND GRATING

- 3 -