U.P.B. Sci. Bull., Series , Vol. , Iss. , 20 1 ISSN 1223 -7027 [628225]

U.P.B. Sci. Bull., Series …, Vol. …, Iss. …, 20 1 ISSN 1223 -7027
LABVIEW BASED DATA A CQUISITION SYSTEM FO R AN
OPTICAL FIBER CURREN T SENSOR
Author1, Author 2
The polarization effects in optical fibers have determined a background for a
new generation of powerful and sensing -oriented technique. The aim of t his paper is
to present a LabVIEW software analysis model used to acquire data from an optical
fiber current sensor based on Faraday Effect , using a NI -DAQ acquisition board .
The optical fiber current sensors give an advantage in carrying out the current
measurement with insulating materials. Tests were conducted to show experimental
results and their limitations concerning errors in using such sensors for
measu rements in electrical networks. Optical fibers have particular polarization
properties.
Keywords : LabVIEW, Optic al fiber, Cu rrent sensor , Faraday Effect .
1. Introduction
This paper aim to present a LabVIEW based data acquisition system for an
optical fiber current sensor. The current sensor work based on Faraday Effect . The
phenomenon of the Faraday Effect was first observed by Michael Faraday in
1845 . It is described as the result from the interaction between a magnetic field
and the light propagated through a medium . The Effect occurs when the rotation
of a linearly polarized wave passes through a th ickness of a tr ansparent
medium [1]. The rotation of the plane of polarization is direct proportional to the
component of the magnetic field that is parallel to the propagation direction. So is
possible to measure an unknown current using an optical fiber current sensor, a NI
DAQ acquisition board and a beam splitter.
Using LabVIEW software, a model to calculate the unknown current was
implemented, starting from the 2 voltages obtained at the output of the beam
splitter.
If the medium is a non -birefringent optical fiber, the Faraday rotation
angle ( θideal) is proportional to the Verdet constant of the medium (V), the applied
magnetic field (B) and the interaction length (L), acco rding to the following
equation [2]:
ideal VBL MVI
, (1)

1 From

2 From

Author 1, Author 2, Author 3

where:
μ is the magnetic permeability of the medium ( value for silica glass, μ ≈ μ 0 =
4π×10-7 N/A2),
M is the number of turns of wire , and
I is the current applied to the solenoid,
0 B nI
is the applied magnetic field and
/ n m L
is the turns density. It is assumed that the magnetic field is uniform and
parallel to the propagation axis [2].
For non -negligible linear birefringence of the fiber, which can be induced
due to bending or twisting intrinsic as a result of manufacturing imperfections, the
Faraday rotation can be exp ressed using the integral approach of the Ampere’s
law for magnetic field, as follows:

0sin( ) sinLideal
non idealL VBV B z dz LL      
(2)
Where
2 / ,xy n n n n        stands for the intrinsic linear
birefringence, λ is the wavelength of light.[3]
2. Low and high birefringence fibers measurement
An ideal isotropic fiber has no birefringence. It propagates any state of
polarization launched into the fiber unchanged. Real fibers possess some amount
of anisotropy owing to an accidental loss of circular symmetry. This loss is due to
either a noncircular geometry of the fiber or a nonsymmetrical stress field in the
fiber cross -section.
When birefringence is introduced into an isotropic fiber, the circula r
symmetry of the ideal fiber is broken thus producing the anisotropic refractive
index distribution into the core region. The asymmetry results from either intrinsic
birefringence including a geometrical deformation of the core and stresses
induced during the manufacturing process or material anisotropy due to induced
(extrinsic) elastic birefringence. [4]
Polarization -maintaining fibers can be divided into the following two
categories: (i) high -birefringence (Hi -Bi) fibers and (ii) single -polarization sing le-
mode (SPSM) fibers. In the case of highly birefringent fibers, the propagation
constants of the two orthogonally polarized modes are made quite different from
each other so that the coupling between the two modes is greatly reduced. [5]
To prevent these type of effects have been developed single -polarization
single -mode (SPSM) optical fibers. Three basic types of the SPSM fiber are
elliptical -core fiber, stress -induced bi -refringent fiber, and side -pit fiber. [6]
Low-birefringence optical fiber (known also as spun fiber) can transmit
polarized light over large distances with minimum error. Unlike regular PM fibers

LabVIEW based data acquisition system for an optical fiber current sensor

it relays both linear and circular polarization and does not require axial alignment
of the fiber when splicing or connectorizing. Spun fiber str ucture is designed to
spread non -uniformities in silica along all possible directions, effectively
cancelling out total fiber birefringence. [7] Major spun fiber parameters , such as
device sensitivity, number and size of fiber coils can be optimized depend ing on
application requirements .[8]
Table 1
Specifications
LB650
Operating wavelength 600-900 nm
Cut-off wavelength < 580 nm
Beat length
Spin period
Attenuation
Mode field dimensions 4 mm
3 mm
6 dB/km
6 um
Cladding diameter 125 um
Coating diamete r 250 um
Core -clad concentricity
Cladding offset
Coating material <0.5 um
<5 um
acrylate
Proof test
Bending radius 100 kpsi
>20 mm
3. The Verdet constant
The Verdet constant is the proportionality constant between the angle of
rotation θ of plane polar ized light and the product of the path length l through the
sample and the applied magnetic fi eld B. It is the proportionality constant in[1]:
V l B  
(3)
With a value about 2.65 x 10-4deg/A, the Verdet constant is a
magnetooptic coeff icient that appears in case of low -birefringence fibers. In the
current sensor applications, the fiber generally loops one or more times around the
conductor. For the linear birefringence this is induced in the fiber by the
bending. [9]
Using this approach one can obtain when the Faraday Effect is employed
in a sensor, glass (either bulk or fiber) is generally used, the angle θ expressed as:
LV H dl  
, (4)
where V = Verdet constant;
H = Amplitude of the magnetic field;
L = Length over whic h H acts.

Author 1, Author 2, Author 3

With the Wollaston prism (PBS) one can obtain two different voltages, u1(t) and
u2(t) with the expressions:
2
10( ) sin ( ( ))u t I t 
(5)
2
20( ) cos ( ( ))u t I t 
, (6)
where θ is the angle between the two components at the output of PBS.
If one divide these two voltages, u 1(t) and u 2(t), the result will be:
2 1
2()( ( ))()uttg tut
(7)
Simultaneously, it is assumed that:
( ) ( )of t V l H t  
, (8)
where V is the voltage, l of is the length of circuit and H(t) is the intensity of the
magnetic field.
The a mplitude of the magnetic field, H(t) it is expressed as:
( ) ( )H t k i t
(9)
And if we replace equation (9) in ( 8) the result will be:
( ) ( ) ' ( )of t V l H t k i t    
(10)
From (10) one can obtain the unknown current as follows :

( ) ( ) 'i t t k (11)
So the measurement chain provide a direct relation between the unknown
current and the rotational angle θ. Such a sensor has several advantages,
especially at high values of currents [6]. The practical implementa tion of this
solution using trans -impedance amplifiers eliminate complicated solutions needed
to stabilize the parameters of laser diode. Also the solution eliminates the non –
linearity of Polarized Beam Splitter (PBS) using the reverse formula implemented
in LabVIEW. Last but not least, the schematic is not affected by superior
harmonic components as it happens in conventional solutions.
-100 -50 0 50 10000.10.20.30.40.50.60.70.80.91
Teta [grad]U1, U2 [u.r.]

Fig. 1. The evolution of U 1(θ) and U 2(θ) function of splitting angle θ

LabVIEW based data acquisition system for an optical fiber current sensor

4. Experiment and Results
The work of developing this solution includes an electric assembly that
generates the unknown current from the autotransformer, then an optical fiber
around the current circuit that wil l polarized the light, also a laser diode as an
emitter source of light, and a photodiode as a receiver, a PBS prism to split the
light into two proportional fraction, and some LEM type sensors to supplementary
measure the unknown current.
Also to calculat e the current an NI acquisition board and LabVIEW
software are used . The experiment s describe some tests performed using a non-
invasive current sensors based on Faraday Effect, whe re a single polarized laser
signal is sent to a polarizing beam splitter (PB S) prism and so two angular
components are separated.
These two components are proportional with the unknown current that
created the Faraday Effect. Figure 2 presents the practical implementation that
includes the Polarized Beam Splitter (PBS), the NI DA Q – NI Data Acquisition
Module .

Fig. 2. Practical implementation showing the polarized light been sent to the PBS prism

The signals are processed using a NI -DAQ acquisition board and an
interface , a LabVIEW software tool. In order to obtain the sensiti vity threshold
and transfer characteristic , the sensor characteristics are tested .[10]

Author 1, Author 2, Author 3

Fig. 3. Waveform from LabVIEW application

Using the LabVIEW application it can be seen from (a) to (e) where:
(a) Is the input signal, the unknown current measured with LEMs (a
transducer employing the Hall Effect to measure DC and complex
waveform AC) ;
(b) The electrical voltage U 1(t) measured at the output of the Wollaston prism ;
(c) FD_LD, the control voltage of the photodiode;
(d) u2(t) measured as u 1(t);
(e) The unknown current mea sured by the system.

The purpose of the system is that the result acquired from (a) to be the same with
the result calculated at point (e). [10]

Furthermore, f igure 4 present the implementation using LabVIEW software of a
Butterworth filter working at 2kHz for measurement of u 1(t) and u 2(t).

LabVIEW based data acquisition system for an optical fiber current sensor

Fig. 4. Implementation of a Butterworth filter for measurement of u1(t) and u2(t) at 2kHz

Comparing the results between the current calculated in LabVIEW with
the current measured with LEM ’s sensors, not only the wa veform is identical, but
also the errors are minimal after a certain threshold is overpassed.
This threshold, around 0.5A is necessary to obtain good polarization for
the light in the optical fiber. Currents under 0.5A do not generate enough
electromagnet ic field that can polarize the light in a manner observable by our
measurement device.

Fig. 5 Relative error of measured current

The obtained results obtained show that for current under 0.5A, the
polarization is not strong enough so the errors are ex tending beyond an admitted
range of 1%. In the same time, greater the current is, greater the polarization and
the error decrease. Between 0.5A and 5.5A there is a good interval in which the

Author 1, Author 2, Author 3

sensor function very well, the average error being very small, a round 0.45%.
However, for current greater than 3.5A, an overheating reaction can be observed,
so a cooling system must be added in order to preserve the same error level.
6. Conclusions
Practical implementation of such non -invasive current sensors is not an
easy task, because the measurement method is an indirect one: first light
components should be obtained, then a proportional voltage are acquired and from
here LabVIEW calculate the current in function of polarization angle. But the
comparative results are good enough to overpass this drawback. When compar ing
the errors for the measured current with LEM sensors and the calculated current in
LabVIEW , one can observe that these have very small values – under 0.5%, for an
interval between 0.5A and 5.5A. Th e current under 0.5A doesn’t produce enough
polarization in order to obtain a useful light beam for PBS, and for currents higher
than 3.5A it appears a thermic deviation caused by the overheating of the circuit.
Otherwise , there is no theoretical maximum v alue of current that can be measured.
It needs to take into account that the PBS is a sensitive device that needs to be
used and handling carefully. To have some robustness of the sensor it is necessary
to have a mechanical hull.
R E F E R E N C E S
[1]. S. Suchat , P. Viriyavathana , P. Jaideaw , N. Haisirikul , W. Kerdsang , and S. Petcharavut ,
„Measurement of the Verdet Constant in Di®erent Mediums by Using Ellipsometry Technique ”,
in Progress In Electromagnetics Research Symposium Proce edings, Suzhou, China , Sept. 12 -16,
2011 , Pages 803 -806.
[2] Martha Segura, Natasha Vukovic, Nicholas White, Tim M ay-Smith, Wei H. Loh, Francesco
Poletti and Michalis N. Zervas , “Low birefringence measurement and temperature dependence in
metre -long optical fibers ”, in Journal of Lightwave Technology, DOI 10.1109/JLT.2014.2387291,
Issue Date: June 15, 2015, Pages: 2697 – 2702.
[3] J. L. Cruz, M.V. Andres, and M. A. Hernandez , “Faraday Effect in standard optical fibers:
dispersion of the effective Verdet constant ”. Appl. Opt. 35 (6), Pages: 922 -927, 1996.
[4] T.R. Wolinski , “Polarization in optical fibers”, in Proceedings of the IV International Workshop
NOA’98, Miedzyzdroje 1988, Vol. 95(1998), Acta Physica Polonica A, No. 5.
[5] https://spie.org/samples/TT90.pdf , Birefringence in Optical Fibers: Applications .
[6] Okoshi, T. ; “Single -polarization single -mode optical fibers ”, in Quantum Electronic s, IEEE
Journal (Volume: 17, Issue: 6), 06 January 2003 , Pages: 879 – 884.
[7] IVG Fiber , “Spun fiber”, http://www.ivgfiber.com/spun_fiber.htm , www.ivbfiber.com.
[8] IVG Fiber , Datasheet of “PM fibers for Faraday sensing”, www.ivbfiber.com .
[9] Zhanbing Ren, Yu Wang, and Philippe -Alain Robert, Senior Member, IEEE , “Faraday Rotation
and its Temperature Dependence Measurements in Low -Birefringence Fibers”, in Journ al of
Lightwave Technology, Vol. 7, No. 8, Pages 1275 -1278 , 1989 .
[10] Marcel Stanciu, Octavian Ghita, Adela Vintea, Sabina Potlog, “Design of a current sensor
based on optical fibers ”, in Proceedings of The 9th International Symposium on Advanced Topics
in Electrical Engineering, Pages: 961 -964, May 7 -9, 2015, Bucharest, Romania.