Note: Descriptions are shown in the official language in which they were submitted.
CA 02238971 1998-0~-28
GR 95 P 3844 IN ~ILE~ THIS A~L~DL~
TRANSLATI~N
Description
Method and arrangement for measuring a measured variable,
in particular an electric current, with a high measuring
resolution
The invention relates to a method and a device
for measuring a measured variable.
Optical measuring arrangements are known for
measuring an electric current in a conductor which are
based on the magnetooptic Faraday effect, and t~ f~fs~rs)
are also designated as magnetooptic current transformers~
In a magnetooptic current transformer, linearly polarized
measuring light is transmitted through a Faraday element
which is arranged in the vicinity of the conductor and
consists of an optically transparent material exhibiting
the Faraday effect. The magnetic field generated by the
current effects a rotation of the polarization plane of
the measuring light by a rotationalrangle p which is
proportional to the path integral ~ the magnetic field
along the path covered by the measuring light. The
proportionality constant is called the Verdet constant V.
The Verdet constant V depends generally on the material
and the temperature of the Faraday element and on the
wavelength of the measuring light used. In general, the
Faraday element surrounds the conductor, with the result
that the measuring light circulates the conductor at
least once in a virtually closed path. The rotational
angle p is essentially directly proportional in this case
to the amplitude I of the current to be measured, in
accordance with the relationship
p = N ~ V ~ I (1),
N being the number of the circulations of the measuring
light around the conductor. The Faraday rotational
angle p is determined polarimetrically Dy polarization
analysis of the measuring light which has passed through
the Faraday element, in order to obtain a measuring
signal for the electric current. Single-channel polariz-
ation evaluation and dual-channel polarization evaluation
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 2 -
are known for the purpose of polarization analysis.
In the case of single-channel polarization
evaluation, after traversing the Faraday element the
measuring light is fed to a polarizer as analyser, and
transmltted
the measuring light p~3cd by the polarizer is converted
by a photoelectric transducer into an electric signal as
measuring signal S. This measuring signal S corresponds
to the light intensity of the light component, projected
onto the polarization axis (transmission axis) of the
polarizer, of the measuring light and, neglecting dis-
turbing influences such as temperature changes and
vibrations, has the general form
S = So/2 ~ (1 + sin(2p + ~
= So/2 ~ (1 + sin(2-N-V-I + ~)) (2).
Here, S0 is the constant ~;mllm amplitude of the measur-
ing signal S, which corresponds to the case in which the
polarization plane of the measuring light is parallel to
the polarization axis of the polarizer (m~;mllm trans-
mitted light intensity). ~ is a constant offset angle for
a current zero (I = 0 A) and depends on the polarizer
angle between the polarization plane of the measuring
light ~or ~ unc~lng into the Faraday element and the
polarization axis of the analyser. If this polarizer
angle is equal to 45~, ~ = O (IEEE Transactions on Power
Delivery, Vol. 7, No. 2, April 1992, pages 848 to 852).
In the case of a dual-channel polarization evalu-
ation, after traversing the Faraday element the measuring
light is separated by an analyser into two linearly
polarized light components L1 and L2 having polarization
planes directed perpendicular to one another. Polarizing
beam splitters such as, for example, a Wollaston prism or
else a simple beam splitter with two downstream
polarizers whose polarization axes are rotated by ~/2 or
90~ relative to one another are known as analysers. The
two light components L1 and L2 are converted in each cas~
by an assigned photoelectric transducer into an electric
intensity signal T1 or T2, respectively, which is propor-
tional to the light intensity of the respective light
component L1 or L2, respectively. A measuring signal
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 3 -
T = (T1 - T2)/(Tl + T2) (3)
which corresponds to the quotient of a difference and the
sum of the two intensity signals T1 and T2 (WO 95/10046)
is formed from these two electric signals. Neglecting
disturbing influences, this measuring signal T is equal
to
T = sin(2p + ~) = sin(2-N-V-I + ~) (4),
~ being an offset angle for I = 0 A which depends on the
angle between the ~olarization plane of the measuring
at incidence lnto
light fo l~ hi ~ in~ the Faraday element and a prime
optical eigenaxis of the analyser.
Both in the case of single-channel and in the
case of dual-channel polarization analysis, the measuring
signal S in accordance with equation (1) or T in accord-
ance with equation (4) is thus a periodic, sinusoidalfunction of twice the rotational angle 2p with the
period ~. It therefore holds that
S(p + n ~) = S(p) and T(p + n ~) = T(p)
with the whole number n. The periodicity of the measuring
signals S and T follows from the fact that polarization
planes, rotated by an integral multiple of ~ or 180~
relative to one another, of the measuring light cannot be
distinguished from one another polarimetrically.
That is why, although in accordance with equation
(1) the Faraday measuring angle p is a linear and thus
unique function of the current I, the measuring signals
S and T of a polarimetric magnetooptic current
transformer are, by contrast, unique functions of the
measuring angle p only over a ~;mum angular range of
~/2 (or 90~) for the measuring angle p. It is therefore
possible to use the known polarimetric magnetooptic
current transformers to measure uniquely only those
electric currents which lie in a current measuring range
(current measuring interval) MR of interval length IMRI
and corresponding to the said m~; mllm angular range of
~/2 (or 90~) for the measuring angle p. In accordance
with equation (1), the current measuring range MR has a
m;~r; mllm Of
IMRI = ~/(2-N-V) (5).
CA 0223897l l998-0~-28
GR 95 P 3844 IN - 4 -
It may be seen from equation (5) that the magnitude IMRI
of the current measuring range MR of a magnetooptic
current transformer can be set by the choice of materials
with different Verdet constants V for the Faraday element
and/or by the number N of circulations of the measuring
light around the conductor. A larger current measuring
range is obtained if the product N V in the denominator
is set to be smaller. However, such a choice of a larger
current measuring range MR inescapably entails a reduced
measuring resolution MA of the current transformer for a
prescribed display resolution. The measuring resolu-
tion MA is defined here and below as the absolute value
¦MS ¦ of the measuring sensitivity MS of the current
transformer. The measuring sensitivity MS corresponds to
the slope of the characteristic of the magnetooptic
current transformer at a working point, and in the case
of single-channel evaluation in accordance with equa-
tion (2) is equal to
MS = dS/dI = S0 ~ N ~ V ~ cos(2 N V I + ~) (6)
and in the case of dual-channel evaluation in accordance
with equation (4) is equal to
MS = dT/dI = 2 ~ N ~ V ~ cos(2 ~N-V-I + ~) (7).
It is evident at once from equations (6) and (7) that a
reduction in the product N-V leads in the case of both
evaluation methods to a reduction in the measuring
resolution MA = IMSI.
EP-B - 0 088 419 discloses a magnetooptic current
transformer in which there are arranged parallel to one
another about a common conductor two Faraday glass rings
which consist of Faraday materials with different Verdet
constants, and thus have current measuring ranges which
are inherently different for each. Each Faraday glass
ring is respectively assigned a transmitting unit for
transmitting linearly polarized measuring light into the
glass _ing, and a dual-channel evaluation unit for
calculating a respective measuring signal for the respec-
tive Faraday rotational angle. The two measuring signals
of the two evaluation units are fed to an OR gate which
determines a maximum signal from the two measuring
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 5 -
signals. This m~; mllm signal is used to switch over
between the measuring ranges of the two glass rings.
Different measuring ranges of the two glass rings can
also be achieved in the case of identical glass material
for both glass rings by using measuring light of differ-
ent wavelengths. In this case, the wavelength dependency
of the Faraday rotation is exploited.
"International Conference of Large Hig~ Voltage
Electric Systems", CIGRE, Paris, 28.8-3.9.1988,
0 Conference Proceedings, T, Pref. Subj. 1, Vol. 34,
Issue 15, pages 1 to 10 discloses a fibre-optic measuring
arrangement with a first magnetooptic current transformer
for measuring nominal currents and with a second magneto-
optic current transformer for measuring overcurrents. The
first current transformer for measuring nominal currents
contains an optical monomode fibre which surrounds the
conductor in the form of a measuring w;n~;ng with N
turns. Linearly polarized light traverses the measuring
winding, is retroreflected from a mirror into the fibre
and traverses the measuring winding in the opposite
direction a second time (reflection type). In this case,
the Faraday rotational angle is doubled, while the
undesirable temperature-dependent effects of circular
birefringence of the fibre material cancel each other out
precisely. After traversing the measuring winding twice,
the light is subjected to dual-channel polarization
evaluation. The second magnetooptic current transformer,
which is provided for protective purposes, likewise
comprises a monomode fibre which surrounds the conductor
in the form of a measuring winding with one measuring
turn. By contrast with the first current transformer,
which is provided for measuring purposes, the second
current transformer is of the transmission type, that is
to say the linearly polarized measuring light is sub-
jected to polarizatior. analysis after traversing themeasuring winding only once.
"SENSOR 93 Kongre~band IV Bll.l, pages 137 to
144" discloses a magnetooptic current transformer for
protective purposes for measuring alternating currents,
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 6 -
in which after traversing a Faraday optical fibre
linearly polarized light is split into two component
light signals and each of these component light signals
is fed to an analyser. The eigenaxes (polarization axes)
of the two analysers are directed towards one another at
an angle of 45~ or 58~. The light intensities passed by
the analysers are firstly normalized by dividing by their
dc components, which are obtained by peak value
rectification. Subsequently, a product of the normalized
signals is formed and this product is then
differentiated. The Faraday rotational angle is obtained
directly by integration. A signal is thereby obtained
which is proportional to the current and is not therefore
subject to any limitations on the measuring range.
However, this method is comparatively expensive.
EP-B-0 208 593 discloses a magnetooptic current
transformer in which, after traversing a Faraday optical
fibre surrounding a conductor, linearly polarized measur-
ing light is split by a beam splitter into two component
light signals and each of these component light signals
is fed to an analyser. The eigenaxes of the two analysers
are directed at an angle of 0~ or 45~, respectively, to
the launching polarization of the measuring light. The
result is a first, sinusoidal signal at the output of one
analyser and a second, cosinusoidal signal at the output
of the other analyser. These two signals are each ambigu-
ous, oscillating functions of the current in the conduc-
tor which are phase-shifted relative to one another by an
angle of 90~. A unique measuring signal is now composed
of these two ambiguous signals by comparing the sign and
the absolute values of the measured values of the first,
sinusoidal signal and of the second, cosinusoidal signal.
As soon as the absolute values of sine and cosine are
equal, that is to say in the case of an integral multiple
of 45~, a switchover is made, as a func~ion of the sign
of sine and cosine, from a unique branch of the first,
sinusoidal signal into a unique branch of the ~econd,
cosinusoidal signal, or vice versa. This method is an
incremental method, and so in the event of a failure of
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 7 -
the electronic system of the current transformer it is
necessary for the working point at zero current to be
reset again.
It is the object of the invention to specify a
method and an arrangement for measuring a measured
variable from a prescribed measuring range and, in
particular, for measuring an electric current in a
conductor from a prescribed current measuring range in
which a high measuring resolution is achieved.
This object is achieved in accordance with the
invention by means of the features of Claim 1 and
Claim 8, respectively.
The method for measuring a measured variable from
a prescribed measuring range comprises the following
method steps:
a) derivation for the measured variable of a first
measuring signal which is a unique function of the
measured variable in the prescribed measuring range,
b) derivation for the measured variable of a second
measuring signal, which is an essentially periodic
function of the measured variable with a period
which is shorter than twice the interval length of
the measuring range, and
c) derivation for the measured variable, from the two
said measuring signals of a third measuring signal
which, in the prescribed measuring range, is a
unique function of the measured variable and has at
least the same measuring resolution as the second
measuring signal.
The arrangement for measuring a measured variable
from a prescribed measuring range contains
a) a first measuring device for generating a first
measuring signal which, in the prescribed measuring
range, is a unique function of the measured
variable,
b) a second measuring device for generating a second
measuring signal which is an essentially periodic
function of the measured variable with a period
which is shorter than twice the interval length of
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 8 -
the measuring range, and
c) a signal processing-unit which is connected to the
first measuring device and the second measuring
device and derives from the first measuring signal
and the second measuring signal a third measuring
signal which is a unique function of the measured
variable in the prescribed measuring range and has
at least the same measuring resolution as the second
measuring signal.
The third measuring signal combines in itself the
advantage of the larger measuring range of the first
measuring signal and the advantage of the higher measur-
ing resolution of the second measuring signal, and
therefore renders it possible to measure the measured
variable uniquely in the prescribed measuring range with
a high measuring resolution. Conversely, the third
measuring signal has a large measuring range for a
prescribed measuring resolution. The method and the
arrangement are not incremental. In the event of a
malfunction or a failure of the current transformer, the
latter can be taken back immediately into operation
without prior calibration. The functional reliability is
thus always ensured.
Particular refinements and developments of the
method and of the arrangement in accordance with the
invention follow from the respectively dependent claims.
As a consequence, in one embodiment the first
measuring signal is also a periodic function of the
measured variable.
In a preferred embodiment, the measuring range is
subdivided with the aid of the first measuring signal
into uniqueness ranges over which the second measuring
signal is a unique function of the measured variable, and
the third measuring signal is composed of the branches of
the second measuring signal over these uniqueness ranges.
The third measuring signal can be derived from
the first measuring signal and the second measuring
signal with the aid of a previously determined table of
values, or else by computation.
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 9 -
The measuring method and the measuring arrange-
ment are preferably used-to measure an electric current
in a conductor from a prescribed current measuring range.
For this purpose, at least two Faraday elements surround-
ing the conductor are provided. A first linearly polar-
ized light signal is sent through a first of the two
Faraday elements at least once, and a first evaluation
unit derives the first measuring signal for the current
from a rotation of the polarization plane of this first
light signal after traversal of the first Faraday
element. A second linearly polarized light signal is sent
at least through a second of the two Faraday elements at
least once, and a second evaluation unit derives the
second measuring signal for the current from a rotation
of the polarization plane of this second light signal
after traversal of at least the second Faraday element.
In a special embodiment of the measuring method,
the second linearly polarized light signal traverses both
the first Faraday element and the second Faraday element
at least once in each case. The second measuring device
of the measuring arrangement then also contains the first
Faraday element of the first measuring device, and the
two Faraday elements are connected optically in series
via optical connecting means. In this embodiment, one
light source suffices for transmitting linearly polarized
measuring light, since the measuring light is split by
the optical connecting means into two light components of
which one light component is provided as first light
signal and the other light component is provided as
second light signal. In addition, a compact design of the
arrangement is possible, since the two Faraday elements
can be arranged next to one another in a space-saving
fashion.
Reference is made, for the purpose of explaining
the inv ntion, to the drawing, in which
Figure 1 shows a design principle of an arrangement for
measuring a measured variable from a prescribed
measuring range,
Figure 2 shows a diagram of three different measuring
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 10 -
signals for the measured variable,
Figure 3 shows an arrangement for measuring an electric
current from a prescribed current measuring
range, having two separate Faraday measuring
devices,
Figure 4 shows an arrangement for measuring an electric
current from a prescribed current measuring
range, having two Faraday elements which are
connected optically in series and operated in
transmission mode in each case,
Figure 5 shows an arithmetic unit for determ; n; ng the
second measuring signal,
Figure 6 shows an arrangement for measuring an electric
current from a prescribed current measuring
range, having a first Faraday element operated
in transmission mode, and a series circuit
composed of this first Faraday element and a
second Faraday element operated in reflection
mode, and
Figure 7 shows an arrangement for measuring an electric
current from a prescribed current measuring
range, having a first Faraday element operated
in reflection mode, and a series circuit,
operated in transmission mode, composed of this
first Faraday element and a second Faraday
element,
all the figures being diagrammatic illustrations. Mutual-
ly corresponding parts are provided with the same refer-
ence symbols.
Figure 1 shows a measuring arrangement for
measuring a measured variable I in a prescribed measuring
range MR. The measuring arrangement contains a $irst
measuring device 5 and a second measuring device 6, as
well as a signal processing unit 12. The measured vari-
able I is present in ~ch case at an input 5A of the
first measuring device 5 and at an input 6B of the second
measuring device 6.
The first measuring device 5 converts the
measured variable I into a first measuring signal M1
CA 02238971 1998-05-28
GR 95 P 3844 IN - 11 -
which is a unique function of the measured variable I at
least over the prescribed measuring range MR. It there-
fore holds that
M1 (Ia) ~ M1 (Ib) (8)
for two arbitrary, mutually differing values Ia and I~ of
the measured variable from the prescribed measuring range
MR. The first measuring device 5 thus has a unique
characteristic in the measuring range MR. The first
measuring signal M1 is fed from an output 5B of the first
measuring device 5 to a first input 12A of the signal
processing unit 12.
The second measuring device 6 generates from the
measured variable I a second measuring signal M2 which,
at least over the prescribed measuring range MR, is a
periodic function of the measured variable I with a
period P2. It therefore holds that:
M2 (I) = M2 (I + n ~ P2)
where n is a whole number, and the characteristic of the
second measuring device 6 is periodic, at least over the
measuring range MR. The period P2 of the second measuring
signal M2 is shorter than twice the interval length
2 IMR¦ of the measuring range MR, that is to say
P2 ~ 2 ¦MRI (10).
Because of the condition (lO), the second measuring
signal M2 is, by contrast with the first measuring sig-
nal M1, not a unique function of the measured variable I
over the measuring range MR.
In the measuring range MR, possibly with the
exception of individual subranges, the measuring
resolution
MA(M2) = IdM2/dI I (11)
of the second measuring signal M2 and of the second
measuring device 6 is higher than the measuring
resolution
MA(M1) = ldM1/dII (12)
of the first measuring signal Ml and of the associated
first measuring device 5. The second measuring signal M2
is transmitted from an output 6B of the second measuring
device 6 to a second input 12B of the signal processing
-
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 12 -
unit 12.
The signal processing unit 12 now forms from the
two measuring signals M1 and M2 present at its inputs 12A
and 12B a third measuring signal M3 which, over the
entire measuring range MR, is a unique function, on the
one hand, and has at least the measuring resolution
MA(M2) of the second measuring signal M2, on the other
hand. This third measuring signal M3 can be tapped at an
output 12C of the signal processing unit 12.
Figure 2 shows in a diagram an exemplary embodi-
ment of the three measuring signals M1, M2 and M3 derived
in a measuring arrangement in accordance with Figure 1.
A prescribed measuring range for the measured variable I
plotted on the abscissa is denoted by MR and corresponds
to a preferably closed interval [IA' IB] ~f values of the
measured variable I between a first interval limit IA and
a second interval limit IB. This measuring range MR thus
has the length IMRI = I IB ~ IA I -
The first measuring signal M1 is a periodic and
preferably sinusoidal function of the measured variable Iand oscillates between a maximum value Max~Ml) and a
m;n;mllm value Min(M1) with the period P1. The first
measuring signal M1 traverses the central value Cen(M1) =
0.5 ~ (Max(M1) + Min(M1)) between the two extreme values
Max(M1) and Min(M1). The first measuring signal M1 is a
unique function of the measured variable I over the
measuring range MR, that is to say it satisfies the
condition (8). In the exemplary embodiment represented,
the measuring range MR lies inside a half period Pl/2 of
the first measuring signal M1 between the m;n;mllm value
Min(M1) and maximum value Max(M1) of said first measuring
signal M1, in which range a sinusoidal function is known
to be unique, and the central value Cen(M1) of the first
measuring signal M1 corresponds to the value of the first
measuring signal M1 at the midpoint
IM = ~ . 5 ~ (IB + IA) ~f the measuring range MR, that iR to
say Cen(M1) = M1(IM) . The first measuring signal Ml
increases in the measuring range MR with increaRing
measured variable I with a positive slope in a strictly
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 13 -
monotonic fashion, but can also be a strictly
monotonically decreasing-function with a negative slope.
At least over the measuring range MR, the second
measuring signal M2 is a periodic and, preferably,
sinusoidal function of the measured variable I, and
oscillates between a m~X;mum value Max(M2) and a m;n;mum
value Min(M2) with the period P2. Three values of the
measured variable I for which the second measuring sig-
nal M2 assumes its m;n;mum value Min(M2) are denoted by
Io~ I2 and I4, whereas three values for which the second
measuring signal M2 assumes its m~X;m~m value Max(M2) are
denoted by I1, I3 and I5. It holds that Io ' I1 ~ I2 ~ I3
c I4 ~ I5. The second measuring signal M2 traverses a
central value Cen(M2) = 0.5 ~ (Max(M2) + Min(M2)) between
two extreme values Max(M2) and Min(M2). In the exemplary
embodiment represented, the measuring range MR comprises
two periods 2-P2 ~f the second measuring signal M2, the
second measuring signal M2 assuming its mean value
Cen(M2) in each case at the end points IA and IB as well
as at the midpoint IM of the measuring range MR, that is
to say it holds that Cen(M~ = M~(IA) = M~ IB) = Mz~IM).
The second measuring signal M2 represented thus fulfils
the previously mentioned conditions (9) and (10).
For the purpose of greater clarity, the first
measuring signal M1 is represented on a larger scale than
the second measuring signal M2, and thus has a flatter
course in general in comparison with the second measuring
signal M2, and is, moreover, illustrated offset with
respect to the second measuring signal M2. The value
ranges [Min(M1),Max(M1)] of the first measuring signal M1
and [Min(M2),Max(M2)] of the second measuring signal M2
can also overlap. In particular, the two central values
Cen(M1) and Cen(M2) can be equal, preferably Cen(M1) =
Cen(M2) = 0 in appropriate measurement units. The measur-
ing resolution MA(M1) of the first measuring signal Ml isplainly lower over the largest part of the measuring
range MR than the measuring resolution MA(M2) of the
second measuring signal M2. An exception is formed only
by small ranges around the values of In of the measured
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 14 -
variable I with n = O, 1, 2, 3, 4 or 5, in which the
second measuring signal-M2 assumes its extreme values
Max(M2) and Min(M2), respectively, and in which the
measuring resolution of the second measuring signal M2 is
therefore zero, that is to say MA(M2(In)) = O. The
m~x;ml-m measuring resolution MA(M1(IM)) of the first
measuring signal M1 at the midpoint IM of the measuring
range MR iS plainly lower than the maximum measuring
resolution MA(M2)(IM) ) of the second measuring signal M2,
which in the exemplary embodiment shown is assumed five
times over the measuring range MR, and in particular at
the midpoint IM of the measuring range MR.
The third measuring signal M3 is now derived in
the following way from the first measuring signal M1 and
the second measuring signal M2. The measuring range MR
can be subdivided into individual uniqueness ranges over
which the second measuring signal M2 is a unique function
of the measured variable I. These uniqueness ranges
(quadrants) are separated from one another by lines in
the diagram represented and correspond to the intervals
[IA~I1]~ [I1,I2], [I2~I3], [I3,I4] and [I4,IB] between two
neighbouring extreme values of the second measuring
signal M2. The first uniqueness range [IA' I1] and the last
uniqueness range [I4, IB] are only P2/4 long in this case,
while the inner uniqueness ranges [I1,I2], [I2,I3] and
[I3,I4], lying therebetween, have the m~;mllm length P2/2
in each case. It is, however, impossible to determine
simply from the measured value M2(I) alone, which is
delivered by the second measuring signal M2, in which of
the uniqueness ranges the current measured value I lies.
This ambiguity of the second measuring signal M2 is now
resolved with the aid of the first measuring signal M1.
The individual branches of the second measuring signal M2
in the uniqueness ranges are combined to form the new
measur_ng signal M3 by transformations (operations) such
as translations and, as the case may be, reflections.
These operations differ in general for each uniqueness
range. The first step is now to use the current value
M1(I) of the first measuring signal M1 to determine the
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 15 -
uniqueness range in which the current value of the
measured variable I lies, after which the operation
assigned to this uniqueness range is used to determine
the measured value M3(I) of the third measuring signal M3
for the current value of the measuring variable I. For
the exemplary embodiment represented, the following
operations, in particular, are suitable for deriving the
third measuring signal M3 from the second measuring
signal M2 with the aid of the first measuring signal Ml;
10M3(I): = M2(I) for M1(IA) S M1(I) ~ M1(I1) (13a)
M3(I): = 2-A2 ~ M2(I) for M1(I1) ' M1(I) ~ M1(I2) (13b)M3(I): = M2(I) + 4 A2 for M1(I2) ' M1(I) ~ M1(I3) (13c)
M3(I): = 6-A2 ~ M2(I) for M1(I3) ' M1(I) ~ M1(I4) (13d)
M3(I): = M2(I) + 8 A2 for M1(I4) ' M1(I) ' M1(IB) (13e),
A2 being the prescribed maximum amplitude of the second
measuring signal M2, with
A2 = 0-5 ~ (Max(M2) - Min(M2)) (14).
Because of the uniqueness of the first measuring
signal M1, the conditions, on which the operations (13a)
to (13e) are based, placed on the value M1(I) of the
first measuring signal M1 correspond precisely to the
uniqueness ranges [IA,I1] to [I4,IB] of the second mea~ur-
ing signal M2.
A further embodiment (not represented) for
deriving the third measuring signal M3 is given by the
following operations in which, by contrast with the
exemplary embodiment represented, the second measuring
signal M2 is not reflected in the uniqueness ranges:
M3(I): =M2(I) for M1(IA) ' M1(I) ~ M1(I1) (15a)
3 ) 2(I) + 2 A2 for M1(I1) c M1(I) ~ M1(I2) (15b)
3 ) 2(I) + 4 A2 for M1(I2) c M1(I) ~ M1(I3) (15c)
M (I): =M2(I) + 6-A2 for M1(I3) ' M1(I) ~ M1(I4)
M3(I): =M2(I) + 8 A2 for M1(I4) ' M1(I) ' M1(IB) (15e)-
In this embodiment in accordance with the oper-
ations (15a) to (15e), by contrast with the exemplary
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 16 -
embodiment shown in accordance with the operations (13a)
to (13e), the individual-branches of the resultant third
measuring signal M3 are not continuously juxtaposed.
The operations (13a) to (13e) or (15a) to (15e)
can be carried out, in particular, with the aid of a
table of values (look-up table) determined experimentally
in advance or by calculation. The table of values assigns
the value pair of the measured values Ml(I) and M2(I) of
the first measuring signal M1 and the second measuring
signal M2, respectively, a value M3(I) of the third
measuring signal M3 as a unique measure of the measured
variable I. The signal processing unit 12 then preferably
contains a memory for storing the values of the table of
values which have been determined in advance, an analog-
to-digital converter for converting the current values of
the first measuring signal Ml and of the second measuring
signal M2 into a digital value in each case, and a digi-
tal signal processor or microprocessor for comparing
these measured digital values with the values stored in
the table of values, and for assigning the value M3(I),
which is likewise digital.
The said operations (13a) to (13e) or (15a) to
(15e) can, however, also be carried out computationally.
The measured value M1(I) of the first measuring signal M
is then compared to the values, determined in advance, of
the first measuring signal M1 at the interval limits In
of the uniqueness ranges, and the second measuring signal
M2 is subjected to the associated arithmetic operation as
a function of this comparison. In this embodiment, the
signal processing unit 12 contains, for example, a
corresponding number of comparator circuits for comparing
the current value M1(I) of the first measuring signal M
with the values, determined in advance, of the first
measuring signal M1 at the interval limits In of the
uniqueness ranges, and an analog ari_hmetic unit with
analog components such as subtracters, adders and
inverters, or an analog-to-digital converter and a
downstream digital arithmetic unit for carrying out the
arithmetic operations. As an alternative to this, the
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 17 -
signal processing unit 12 can also contain an analog-to-
digital converter for converting the current values of
the first measuring signal M1 and of the second measuring
signal M2 into in each case one digital value, and a
downstream digital signal processor or microprocessor
which compares the digital value of the first measuring
signal M1 with the stored digital values Ml(In) at the
interval limits In and then carries out the associated
digital arithmetic operations.
The current value of the measured variable I can
now be determined uniquely from the third measuring
signal M3 obtained, by applying the inverse function M3~
of the third measuring signal M3 to the value M3(I~
determined for the third measuring signal M3, because it
holds that M3~1 (M3(I)) = I.
In the case of the represented sinusoidal second
measuring signal M2 with the amplitude A2 and the
period P2, the measured value I for the measured variable
is yielded, for example, as:
I = Ios(Ml) + I(M2) (16),
the measured value I being composed of an offset measured
value IoS(Ml) which depends on the first measuring signal
M1 and is constant in each case for a uniqueness range,
and a component I(M2) which is a continuous function of
the second measuring signal M2.
The offset measured value IoS(Ml), respectively
constant in the uniqueness ranges of the second measuring
signal M2, of the measured value I is given in the exemp-
lary embodiment represented and in accordance with the
operations (13a) to (13e), and is also given in the
exemplary embodiment in accordance with the operations
(15a) to (15e), by the equations
IOS (M1) = IA for M1(IA) C M1 (I) ~ M1(I1) (17a)
IOS (M1) = IA + ~ ~ 5 P2 for M1(I1) c M1(I) c M1(I2) (17b)
IOS (M1) = IA + P2 for M1(I2) c M1(I) c Ml(I3) (17c)
IoS(Ml) = IA + 1 ~ 5 P2 for M1(I3) c M1(I) ~ M1(I4) (17d)
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 18 -
IOS (M1) = IA + 2 P2 for M1(I4) c M1(I) 5 M1 (IB) (17e).
The continuously variable component I(M2) in the
uniqueness ranges of the second measuring signal M2 is
I(M2) = (P2/2~) arcsintM2/A2) (18a)
for M1(IA) C M1 (I) ~ M1(I1) or
M1(I2) c M1(I) ~ M1(I3) or
M1 (I4) 5 M1 (I) 5 M1 (IB)
and
I(M2) = (P2/2~) arcsin(_ M2/A2) (18b)
for M1(I1) 5 M1(I) c M1(I2) or
M1 (I3) 5 M1 (I) ~ M1 (I4) -
In an advantageous embodiment, the current value
of the measured variable I can thus also be determined
directly from the first measuring signal M1 and the
second measuring signal M2 by the signal processing
unit 12, for example again with the aid of a table of
values, or else by computation with the aid of a pro-
cessor. In this case, the third measuring signal M3 is
equal to the identical function over the measuring
20 range MR, that is to say
M3(I) = I- (19)
Since the methods described are not incremental methods,
the functional reliability is always given.
For the purpose of a higher measuring resolution,
in a particular embodiment the first measuring signal Ml
can also be used as the third measuring signal M3 in
ranges around the points I~ at which the measuring resol-
ution MA(M2) of the second measuring signal M2 vanishes.
For this purpose, an additional comparison of the measur-
ing resolutions MA(M1) of the first measuring signal M1and MA(M2) of the second measuring signal M2 can be
carried out, and the first measuring signal M1 can be
used as the third measuring signal M3 as long as it holds
that MA(M2) ~ MA(M1). It is preferable to determine in
advance the ranges in which this condition holds. When
the condition is no longer fulfilled, the third measuring
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 19 -
- signal M3 is derived again as previously described.
The second measuring signal M2 can also be
another periodic function of the measured variable I, for
example a linear sweep function (saw-tooth function). The
first measuring signal M1 can also be a linear function
o~ the measured variable I or can, in the exemplary
embodiment of Figure 2, be approximated by a linear
function at the midpoint IM ( linear interpolation).
The previously described embodiments of the
measuring arrangement and of the measuring method in
accordance with the invention are preferably used to
measure an electric current I as measured variable with
the aid of the Faraday effect. As already described, in
the case of a polarimetric magnetooptic current trans-
former it is generally the case that sinusoidal orcosinusoidal measuring signals are generated in accord-
ance with one of the equations (2) or (4), which cor-
respond to the exemplary embodiment shown in Figure 2.
An exemplary embodiment of such an arrangement
for measuring an electric current I in a conductor 2 is
represented in Figure 3. TWO Faraday measuring devices 5
and 6 are assigned to the conductor 2. Each Faraday
measuring device 5 and 6 respectively has a Faraday
element 3 and 4 exhibiting the magnetooptic Faraday
effect, a light source 9 and 12, respectively, for
transmitting a linearly polarized light signal L1 and L2,
respectively, into the Faraday element 3 and 4, respect-
ively, and an evaluation unit 7 and 8, respectively, for
evaluating a polarization rotation (Faraday rotation) of
the linearly polarized light signal L1 and L2~
respectively, after at least one traverse of the Faraday
element 3 and 4, respectively, as a consequence of the
magnetic field generated in the conductor 2 by an
electric current I. The two Faraday elements 3 and 4
prefer_bly surround the conductor 2, with the result that
light running through the Faraday element 3 or 4
circulates the conductor 2 along a virtually closed light
path at least once.
In one embodiment, the first Faraday element 3
CA 0223897l 1998-0~-28
GR 95 P 3844 IN - 20 -
- and/or the second Faraday element 4 can be formed using
solid bodies exhibiting -the Faraday effect, preferably
made from glass. It is also possible to provide one or
else several bodies for each Faraday element 3 or 4. A
single coherent body is then designed as a polygonal
annular body in general. Such solid bodies, in particular
bodles
annular ~A~, are mechanically robust and exhibit
virtually no circular birefringence. In another embodi-
ment, the first Faraday element 3 and/or the second
Faraday element 4 are formed by one or in each case one
monomode optical fibre which preferably surrounds the
conductor 2 in the form of a measuring winding with at
least one turn.
The polarization of the first light signal L1
after traversing the first Faraday element 3 is evaluated
by the first evaluation unit 7 of the first measuring
device 5 for a first measuring signal M1 for the electric
current I, while the polarization of the second light
signal L2 is evaluated by the second evaluation unit 8 of
the second measuring device 6 for a second measuring
signal L2 for the electric current I. In this case,
single-channel or dual-channel polarization evaluation
can respectively be used for each of the two evaluation
units 7 and 8, thus resulting, in particular, in
sinusoidal measuring signals M1 and M2 in accordance with
the relationships (2) or (4). Again, in order to correct
falsifications of the measuring signals it is possible in
principle to provide all the additional measures which
are known for suppressing or compensating linear and/or
circular birefringence, for example as a consequence of
temperature changes or vibrations, in the Faraday
elements 3 and 4 and the optical transmission links.
As in the embodiment in accordance with Figure 1,
the two measuring signals M1 and M2 are fed to the signal
processing unit 12 ail further processed for a third
measuring signal M3 for the electric current I in the
conductor 2. The measuring sensitivity of the first
Faraday element 3 is lower than the measuring sensitivity
of the second Faraday element 4 and is selected to be so
CA 0223897l l998-0~-28
GR 95 P 3844 IN - 21 -
high that the first measuring signal M1 in a prescribed
current measuring range MR is a unique measure of the
current I. The measuring sensitivity of the second
Faraday element 4 is, by contrast, selected such that the
period P2 of the second measuring signal M2 is shorter
than twice the length of the current measuring range MR.
The second measuring signal M2 is thus not a unique
function of the current I in the current measuring
range MR. In order to set different measuring sensitiv-
ities of the two Faraday elements 3 and 4 it is possible,for example, to use materials with different Verdet
constants for the two Faraday elements 3 and 4, respect-
ively, or else to use Faraday elements 3 and 4 with
different light path lengths along the magnetic field
generated by the current I.
The prescribed current measuring range MR prefer-
ably reaches from an overcurrent measuring range, which
typically lies in the range between 10 kA and at least
100 kA, as far as a nominal current measuring range which
usually lies below the overcurrent measuring range by a
factor of about 10 to 30. The nominal current measuring
range is preferably covered in this case by a uniqueness
range of the second measuring signal M2 which is half a
period P2 long. The third measuring signal M3, which is
derived using the method and the arrangement in accord-
ance with the invention, permits measurement both of
nominal currents for measurement or meter applications
and of overcurrents for protective purposes with the same
measuring resolution.
Apart from the embodiment, described with the aid
of Figure 3, with two separate measuring devices 5 and 6,
embodiments are also possible in which the two Faraday
elements 3 and 4 are connected optically in series via
optical connecting means, and a first linearly polarized
light signal L1 traverses only ti~ first Faraday
element 3 and a second linearly polarized light signal L2
traverses the first Faraday element 3 and the second
Faraday element 4. Such embodiments are disclosed in the
older German Patent Applications P 44 29 909.5 and
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 22 -
P 44 31 615.1, which were not published prior to the date
of filing of the present application and whose content is
also incorporated into the present application. Figures
4 to 6 show exemplary embodiments with Faraday elements
3 and 4 connected in this way in series.
In the embodiment in accordance with Figure 4,
the first Faraday element 3 and the second Faraday
element 4 are connected optically in series via a three-
port coupler 18 as optical connecting means. An optical
fibre which surrounds the conductor 2 in the form of a
measuring winding is respectively provided both for the
first Faraday element 3 and for the second Faraday
element 4. It is preferable to provide annealed optical
fibres which are distinguished by low linear birefring-
ence and virtually negligible circular birefringence. Inaddition to the selection of different materials for the
two Faraday elements 3 and 4, it is also possible, in
this embodiment, to vary the number of the turns of the
measuring windings in order to match the current measur-
ing ranges. The measuring winding of the second Faradayelement 4 in this case preferably has more turns than the
winding of the first Faraday element 3. A transmitting
unit 10 generates linearly polarized measuring light L
which is preferably launched into the first Faraday
element 3 via an optical fibre 50 which maintains polar-
ization. A laser diode, or else a light source (for
example LED) with a downstream polarizer can, for
example, be provided as transmitting unit 10. A low-
birefringence optical fibre (LoBi fibre) can be provided
as the optical fibre 50 which maintains polarization. A
splice 53 preferably joins the optical fibre 50 to the
measuring winding of the first Faraday element 3. The
three-port coupler 18 is designed in such a way that a
first portion of the measuring light L, which is launched
into the first Faraday element 3 by the transmitting u~ t
10 and runs through the first Faraday element 3, is
launched as first light signal Ll into the first
evaluation unit 7, and a second portion of the measuring
light L running through the first Faraday element 3 is
CA 0223897l l998-0~-28
GR 95 P 3844 IN - 23 -
launched as second light signal L2 into the second
Faraday element 4. A beam-splitting, semi-transparent
mirror which is arranged at an angle of generally 45~ to
the beam direction of the incident measuring light L, for
example a beam-splitting splice, or else a Y-coupler, in
particular a fibre coupler, can be provided as three-port
coupler 18. A first port 18A of the three-port coupler 18
is optically connected to the first Faraday element 3, a
second port 18B is optically connected to the first
evaluation unit 7, and a third port 18C is optically
connected to the second Faraday element 4. An optical
fibre 70 which maintains polarization is preferably
provided for transmitting the first light signal L1 from
the three-port coupler 18 to the first evaluation unit 7.
The second light signal L2 traverses the second Faraday
element 4 only once and is then directly launched into
the second evaluation unit 8, preferably likewise via an
optical fibre 60 which maintains polarization, and can be
connected via a splice 64 to the measuring winding of the
second Faraday element 4 and/or can be designed as a LoBi
fibre. The second Faraday element 4 is thus operated in
transmission mode with respect to the second portion L2
of the measuring light L.
The first evaluation unit 7 evaluates the Faraday
rotation of the polarization plane of the first light
signal L1 for a first measuring signal M1 which can be
tapped at an output of the first evaluation unit 7. The
first Faraday element 3 is thus operated in transmission
mode. The second evaluation unit 8 evaluates the Faraday
rotation of the polarization plane of the second light
signal L2 for a second measuring signal M2, which can be
tapped at an output of the second evaluation unit 8. The
entire Faraday rotation of the polarization plane of the
second light signal L2 is composed in this case of a
first Faraday rotational angle component, produced in the
first Faraday element 3, and a second Faraday rotational
angle component, effected in the second Faraday
element 4. If the Faraday rotation was performed in the
same direction in the two Faraday elements 3 and 4, in
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 24 -
accordance with an identical sensor circulation of the
transmitted light in the first Faraday element 3 and
second Faraday element 4 relative to the direction of
flow in the conductor 2, then the entire rotational angle
is equal to the sum of the two individual rotational
angles. In the case of opposite directions of rotation,
the resultant rotational angle corresponds, by contrast,
to the difference between the two individual rotational
angles.
The advantageous embodiments, shown in Figure 4,
of the two evaluation units 7 and 8 in each case carry
out dual-channel polarization analysis of the associated
light signal L1 and L2, respectively. Means 72 which àre
optically coupled to the port 18B of the three-port
coupler 18 via the optical fibre 70 which maintains
polarization, for example a polarizing beam splitter,
preferably a Wollaston prism, or else a beam splitter and
two optically downstream analysers which cross at a
prescribed angle, are provided in the first evaluation
unit 7 for the purpose of splitting the first light
signal L1 into two linearly polarized component light
signals L11 and L12 with different polarization planes. It
is possible, in particular, to provide as the optical
fibre 70 which maintains polarization an optical fibre
with a high linear birefringence (HiBi fibre), whose
eigenaxes of the linear birefringence are adjusted to the
eigenaxes of the Wollaston prism 72. Furthermore, the
evaluation unit 7 contains photoelectric transducers 74
and 75 for converting the component light signals L11 and
L12 into an electric signal R11 and R12, respectively, in
each case as a measure of the intensity of the respective
component light signal L11 and L12, respectively, and
electronic means 76 for deriving the first measuring
signal M1 from these two electric signals R11 and Rl2. As
first measuring signal M1, the electronic means 76
preferably determine a quotient signal
M1 = (Rll-Rl2)/Rll+Rl2) (20)
from the difference R11-R12 and the sum R11+R12 of the two
electric signals R11 and R12. This quotient signal is
CA 0223897l l998-0~-28
GR 95 P 3844 IN - 25 -
largely freed from intensity fluctuations in the trans-
mitting unit 10 or in the transmission links for the
measuring light L and the first light signal L1, and it
holds in general that:
Ml = sin(2 Ns Vs a + ~) (21)
with the Faraday rotational angle a of the first light
signal L1, the number NS of the turns of the first
Faraday element 3, and the Verdet constant Vs ~f the
material of the first Faraday element 3, as well as the
constant offset angle ~. NS is typically between 1 and 3.
The component light signals L11 and L12 can be transmitted
to the transducers 74 and 75 in a free-beam arrangement,
or else via optical fibres. The outputs of the photo-
electric transducers 74 and 75 are respectively connected
electrically to an input of the electronic means 76.
The second evaluation unit 8 is constructed in a
similar way to the first evaluation unit 7. Means 82, for
example a Wollaston prism, which are optically connected
to the second Faraday element 4 via an optical fibre 60
which maintains polarization are provided for the purpose
of splitting into two linearly polarized component light
signals L2l and L22 with different polarization planes the
linearly polarized light component L2 transmitted through
the second Faraday element 4. The optical fibre 60 is
connected to the fibre of the second Faraday element 4
via a splice 64. As the optical fibre 60 which maintains
polarization, it is possible, in particular, to provide
an optical fibre with a high linear birefringence (HiBi
fibre) whose eigenaxes of the linear birefringence are
adjusted to the eigenaxes of the Wollaston prism 82.
Furthermore, the second evaluation unit contains photo-
electric transducers 84 and 85 for converting these
component light signals L2l and L22 into in each case an
electric signal T21 and T22, respectively, as a measure of
the intensity of he respective component light
signal L21 and L22, respectively, and electronic means 86
for deriving the second measuring signal M2 from the two
electric signals T21 and T22. As second measuring signal
M2, a quotient signal
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 26 -
M2 = (T21-T22)/(T21+T22) (22)
of normalized intensity-is preferably derived by the
electronic means 86 from the difference T2l-T22 and the
sum T21+T22 of the two electric signals T21 and T22. This
second measuring signal M2 depends on the total Faraday
rotational angle ~ of the second light signal L2 in the
following way:
M2 = sin(2-(Ns-vs+NM-vM)-~B + ~) (23)
with the number of turns NM~ the Verdet constant VM of
the second Faraday element 4, and a constant offset angle
~. The number of turns NM of the second Faraday element
4 is generally selected to be between 10 and 50.
In a preferred embodiment, the electronic
means 76 and/or 86 contain an analog-to-digital converter
for digitizing the two signals R11 and R12 of the trans-
ducers 74 and 75, or T21 and T22 of the transducers 84
and 85 and a downstream digital arithmetic unit for
calculating the first measuring signal M1 or the second
measuring signal M2.
In another embodiment, represented in Figure 5,
the electronic means 86 contain analog arithmetic com-
ponents. Quicker signal processing can be achieved
thereby. Provision is made of a subtracter 31 and an
adder 32 at whose inputs the two electric signals T
and T22 of the transducers 84 and 85 (not represented)
are present in each case. The outputs of the
subtracter 31 and of the adder 32 are connected in each
case to an input of a divider 3 3. As second measuring
signal M2, the divider 33 forms the quotient signal (T21-
T22)/(T21+T22) from the difference signal T21-T22 at the
output of the subtracter 31 and the sum signal T21+T22 at
the output of the adder 3 2. A corresponding circuit can
also be provided for the electronic means 76 of the first
evaluation unit 7.
Figure 6 shows a further ~mbodiment of the
measuring arrangement. The first Faraday element 3 and
the second Faraday element 4 are connected optically in
series via an optical four-port coupler 19 with four
ports 19A to l9D as optical connecting means. The second
CA 02238971 1998-05-28
GR 95 P 3844 IN - 27 -
evaluation unit 8 is optically connected to the port l9D
of the four-port coupler 19. Assigned to the second
Faraday element 4 are optically reflecting means 40, for
example a mirror, which retroreflect the second light
signal L2 into the second Faraday element 4 after a
single traverse of the second Faraday element 4. The
retroreflected second light signal L2' traverses the
second Faraday element 4 in the reverse direction a
second time and is then fed to the second evaluation
unit 8 via the four-port coupler 19. The port l9B con-
nected to the second Faraday element 4, and the port l9D
connected to the second evaluation unit 8, of the four-
port coupler 19 are optically coupled to one another for
the purpose. Likewise optically coupled to one another
are the port l9A, connected to the first Faraday
element 3, and the port l9C, connected to the first
evaluation unit 7, of the four-port coupler 19. A beam
splitter, for example, with a semitransparent mirror
arranged obliquely relative to the light propagation
direction, or a fibre-optic coupler can be provided as
the four-port coupler 19. In this embodiment in
accordance with Figure 6, the second Faraday element 4 is
operated in reflection mode. The second light signal L2'
evaluated by the second evaluation unit 8 has a Faraday
rotation of its polarization plane which is composed of
the Faraday rotation in the case of a traverse of the
first Faraday element 3 and twice the Faraday rotation in
the case of a traverse of the second Faraday element 4.
In the embodiment in accordance with Figure 7,
the first Faraday element 3 is operated in a reflection
arrangement. An optical coupler 13 connects the first
Faraday element 3 both to the transmitting unit 10 for
transmitting linearly polarized measuring light L and to
the first evaluation unit 7. This coupler 13 can be a Y-
fibre coupler, or else a beam splitter formed by means 3fa semitransparent mirror. The optical connecting means 15
for optically coupling the two Faraday elements 3 and 4
are now designed in such a way that a portion of the
measuring light L running through the first Faraday
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 28 -
element 3 is retroreflected as first light signal Ll into
the first Faraday element 3, and another portion is
passed as second light signal L2 and launched into the
second Faraday element 4. For this purpose, the optical
connecting means 15 preferably contain a semitransparent
mirror 35 which is arranged essentially perpendicular to
the beam direction of the incident measuring light L.
However, it is also possible to provide a beam splitter
with a semitransparent mirror directed at an angle, for
example 45~, to the beam direction of the incident
measuring light L, and a further mirror, arranged in the
beam path of the component measuring light reflected at
this semitransparent mirror, for the purpose of retro-
reflecting this component measuring light to the semi-
transparent mirror. The first light signal Lretroreflected by the optical connecting means 15 tra-
verses the first Faraday element 3 in the rever~e direc-
tion a second time and is fed via the optical coupler 13
to the first evaluation unit 7. Because of the non-
reciprocity of the Faraday effect, the first lightsignal L1 experiences a Faraday rotation which is twice
as large as in the case of only a single traverse of the
Faraday element 3. By contrast, effects of a possible
circular birefringence in the first Faraday element 3
cancel one another out because of their reciprocal prop-
erty. The evaluation unit 7 derives the first measuring
signal M1 from the first light signal L1. By contrast,
after the first Faraday element 3, the second light
signal L2 passed by the optical connecting means 15 also
traverses the second Faraday element 4, and is fed to the
second evaluation unit 8 after traversing the second
Faraday element 4. The second light signal L2 arriving at
the evaluation unit 8 has a polarization plane which has
been rotated both in the first Faraday element 3 about a
first Faraday rotational angle and in the second Faraday
element 4 about a second Faraday rotational angle. The
resultant total rotational angle of the second light
signal L2 is evaluated in the evaluation unit 8 for a
second measuring signal M2.
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 29 -
In the embodiment represented, a fibre spool
surrounding the conductor 2 and made from an optical
monomode fibre is respectively provided both for the
first Faraday element 3 and for the second Faraday
element 4. The optical connecting means 15 are then
formed with the aid of a semitransparent coating which
serves as a mirror 35 and is applied by sputtering or
chemical deposition to the fibre end, preferably of the
first Faraday element 3 and with the aid of mechanical
connecting means for connecting the fibre end provided
with the mirror to the neighbouring fibre end of the
other Faraday element. The mechanical connecting means
can be a detachable plug-and-socket connection, or else
an undetachable splice, for example a capillary tube from
the Nippon Electric Glass company.
The second evaluation unit 8 is designed in a
fashion similar to Figure 4. By contrast, in the embodi-
ment represented, the first evaluation unit 7 contains
polarizing means lOB which are connected optically
between the first Faraday element 3 and the coupler 13,
and a photodetector 79 which is optically coupled to the
coupler 13. A light source lOA which is optically coupled
to the coupler 13 is provided for transmitting light L.
The coupler 13 can be connected to the polarizing
means lOB via an optical fibre 50'. In this embodiment,
this fibre 50' can be a simple teleco~mnnication fibre
without properties of maintaining polarization, since the
polarizing means lOB linearly polarize the light L only
directly before its entry into the series circuit of the
two Faraday elements 3 and 4 at the input of the first
Faraday element 3. Consequently, it is also possible to
provide a non-polarizing, simple light source as the
light source lOA. The polarizing means lOB and the light
source lOA together form the transmitting unit 10 for
laun~hing linearly polarized measuring light L into the
first Faraday element 3. The polarizing means lOB are
provided at the same time as an analyser for the first
light signal Ll which is reflected by the semireflecting
connecting means 15 and has its polarization plane
CA 02238971 1998-05-28
GR 95 P 3844 IN - 30 -
rotated. The analyser passes only the component Ll' of
the first light signal L1-which is projected onto its set
polarization axis. The light component L1', passing
through the analyser, of the first light signal L1 is fed
via the coupler 13 to the photodetector 79 and converted
there into an electric signal as first measuring sig-
nal M1. This first measuring signal M1 is proportional to
the light intensity of the light component L1'. It there-
fore holds that
M1 = K cos2(0.5 Ns Vs ~) (24)
with the Faraday rotational angle a of the first light
signal L1, and a proportionality factor K.
It is also possible to carry out a method for
compensating temperature and/or vibration because of the
high measuring resolution required for norn;~l currents.
In principle, all analog and digital evaluation
methods for detecting the polarization state of linearly
polarized light can be used to evaluate the Faraday
rotational angle in the first evaluation unit 7 and in
the second evaluation unit 8. In particular, the embodi-
ments of the evaluation units 7 and 8 in accordance with
Figures 4, 5 and 7 can be combined with one another at
will. Preferably, the two light signals L1 and L2 are
evaluated by means of dual-channel polarization analysis.
However, it is also possible in each case to provide
single-channel evaluation for each light signal Ll
and L2. The measuring signals Ml and M2 obtained are then
of the sinusoidal configuration as in the relationship
(23) or (24) and thus fulfil, in conjunction with
appropriately selected measuring sensitivities NS Vs of
the first Faraday element 3 and NM-VM ~f the second
Faraday element 4, the preconditions for deriving the
third measuring signal M3 in accordance with one of the
embodiments previously described.
The optical coupling of the various optical
components of the measuring arrangement is preferably
supported by collimator lenses (Grin lenses) for focusing
the light.
In a particular embodiment (not represented), it
CA 02238971 1998-0~-28
GR 95 P 3844 IN - 31 -
is also possible to use a plurality of linearly polarized
measuring light signals of wavelengths which differ and
are generally close to one another, in conjunction with
wavelength-sensitive optical connecting means 15, 18
or 19. Crosstalk between protective and measurin~
channels can thereby be avoided.
