Note: Descriptions are shown in the official language in which they were submitted.
3L2~ 3~
N OPTICAI FREQUENCY CONVERT~R DEVICE AND A RATE GYRO
CO~YTAINING SUCH A DEVICE
BACKGROUND OF THE INVENTION
The invention covers an integrated optical frequency converter device.
Classical op~ical frequency conver~ers are well known. The frequency
converter most usually used is certainly the one based on acousto-optical
interactlon. In this method, an acoustic network which propagates in a
5 medium produces periodic refractive index variations in the form of a
travelling wave. This moving network diffracts the light. If the in~eraction
lenght is sufficient, a single order may predominate. In the diffracted order
(~" D~ the optical wave frequency ~ has been modified by a quantity eyual
to the acoustic wave frequency Q .
Hense, ~D = ~+ Q
Fundamental frequency rejection may be excellent because the con-
verted wave and the direct wave ~undiffracted) are then separated in space.
It is then possible to study what this gives in integrated optics. Under
this name are designed thin layer monilithic s~ructures intended for lumi-
nous signal processin~ which are obtained by the techniques of deposit,
diffusion and engraYing by masking analogous to those used in the production
of electronic integrated circuits. With these techniques, in particular, i~ is
known how to produce linear structures characterized by a refractive index
higher than that of the surrounding medium and forming wave-guides along
20 which the li~ht propagates by a series of total reflections or progressive
refractions. 1~ is known how to combine two such wave-guides by arranging
them in para~lel one with respect to the other along part of the path to
produce directional couplers. Thanks to the vanishing wave phenomenon, the
energy carried in the first wave-guide is passed progressively in to the
25 second wave-guide and it is found that the maxirnum energy is transferred
at the end of a certain lenght, called ~he coupling lenght, which depends on
the geometrical and optical parame~ers of the structure and9 in particular,
on the value of the refractive indices of the materials formin~ the two
waves-guides and the me~ium which separates ~hem. Then $he energy passes
. ~
progr2s~sively from the second guide into the first and so on. When an
electro-optical material is used for one of the materials forming the wave-
guides or the medium which separates them, it is also known how ~o cause
the index to vary under the effect of an electric field, which makes it
5 possible, by acting on the coupling lenght, to control electrically the part of
the energy transfered from one wave-guide to the other. It can be seen that
it is also possible to make a ligh~ modulator by arranging, in parallel with
the wave-guide wh;ch carries the luminous wave, a section of wave-guide in
which a more or less large part of this energy will be transferred.
Also, there are frequency converters intended to produce, from a
guided electromagnetic radiation of frequency ~, a guided electromagnetic
radiation wh~se frequency is a multiple of the frequency ~ . These conver-
ters are used, in particular, in the integrated optical field, thus named by
analogy with electronic in~egrated circuits, which are monolithic structures
15 using thin layers.
Converters of the type already described have been produced in
integrated optics but they require the use of a flat wave-guide and this is
not applicable to microguides. Techniques usable with microguides have
already been suggested in which electro-optical modulation can be used.
20 lhis may then be a serrodyne or balanced rnodulator system. Such an optical
frequency conver~er contains a wave-guide used as phase mo~ulator, which
is controlled by a signal in the form of a saw-tooth. Such a si~nal has the
same effect as a voltage gradient which allows a variation of the index as a
function of time. It may also be an acoustic modulation in which a TE type
25 wave is converted into a TM type with a change in frequency. In this case,
the application of a transverse electric field enables the pass-band of a TE-
TM rnode acousto-optical converter to be modified by the colinear interac-
tion of acoustic surface waves and a guided optical wave.
These two techniques have several disadvantages .
- the two waves propagate in the same wave-guide (the frequency
translated wave and the fundamental wave) ans this can cause problems in
separating them;
- in certain cases, the effectiveness of the conversion is very closely
related to the wave shape (case of the serrodyne translator);
- 3 - 1~55~
- i.n the case of a TE-TM conversion, one of the
problems that may be met with is the extreme sensi-tivity
of the device to wave-length (variation of ~/KTE - ~/KT~)
which may, however, allow this type of device to be used
as a filter.
The device of the invention enables these dis-
advantages to be attenuated. In this devi.ce, the converted
and unconverted waves are separated in space because the
diaphony is related to simple geometrical parameters and
can be reduced arbitrarily. Also, the two waves keep
the same polarizing. The device can be extended and used
in the case in which it is required to produce frequency
filters.
SUMMARY OF THE INV13NTION
In accordance with the present invention, there
is provided a rate gyroscope including an interferometer
for measuring a non-reciprocal phase-shift undergone by
first and second radiations circulating in opposite direc-
tions in a ring wave-guide, comprisingO
said ring wave-guide with ends;
a monochromatic luminous source;
means Eor photo-detecting an interference of
said first and second radiations;
an optical separator and mixer connec-ting the
ends of said ring wave-guide to said luminous source and
said photo-detecting means;
an electrically controlled optical phase shifter;
and
- at least one optical frequency converter device
in an optical path of said ring wave-guide, said optical
frequency converter device including:
a flat substrate;
~SS13~
at least two wave-guides with different
characteristics integxated into said substra-te, one of which
is intended to receive an incident wave, said two wave-
guides being parallel with one another over a predeter-
mined length and separated from one another by a distance
such that the incident wave radiation is transferable
from one wave-guide to another; and
means for generating an acoustic wave
collinear with an incident wave propagating direction
in one of the two wave-guides, the generating means being
arranged between the two wave-guides to produce frequency
conversion when radiation is transferred from one wave-
guide to the other.
In aeeordanee with the invention, there is also
provided a rate gyroseope ineluding an interferometer for
measuring a non-reeiproeal phase-shift undergone by first
and seeond radiations cireulating in opposite directions
in an optical fiber ring wave-guide, comprising:
said optical fiber ring wave-guide with ends;
a monochromatic luminous source;
means for photo-detecting an interference of
said first and seeond radiations;
an optieal separator and mixer conneeting the
ends of said ring wave-guide to said luminous souree
and said photo-deteeting means, said luminous source,
photo-detecting means and separator and mixer being inte-
grated on a single substrate;
first and second wave-guides integrated onto said
substrate, said first wave-guide having a first end cou-
pled to said monoehromatie luminous source, and said
second wave-guide having a .first end coupled to said
photo-detecting means, said first and seeond wave-guides
having respective seeond ends coupled to different ends
of said optical fiber ring;
::
: ~ :
- 3b -
an electro-optical phase modulator comprising
a pair of electrodes positioned one each on opposite
sides of one of said first and second wave-guides; and
at least one optical frequency converter device
in an optical path of said ring wave-guide, said optical
frequency converter device including:
at least two wave-guides with different
characteristics integrated into said substrate, one of
which is intended to receive an incident wave, the two
wave-guides being parallel with one another over a pre-
determined length and separated from one another by a
distance such that the incident wave radiation is trans-
ferable from one wave-guide to another; and
means for generating an acoustic wave colli-
near with an incident wave propagating direction in one
of the said two wave-guides, the generating means being
arranged between the two wave-guides to produce frequency
conversion when radiation is transferred from one wave-
guide to the other.
According to the invention, there is further pro-
vided an integrated ring interferometer, comprising:
a substrate;
a source branch port;
a de-tection branch port;
a first Y-shaped wave-guide integrated into
said substrate, said first wave-guide having first and
second branches and a trunk portion, said first branch
being coupled to said source branch port and said second
branch being coupled to said detection branch port;
a monomode filter formed by a metallization over
said trunk portion;
a second Y-shaped wave-guide integrated into
said substrate, said second wave-guide having first and
second branches and a -trunk portion, said trunk portion
- 3c - ~ ~5S~3~
of the second wave-guide being coupled to said trunk
portion of said first Y-shaped wave-guide;
a first phase modulator coupled to said first
branch of said second Y-shaped wave-guide, said first
phase modulator being constituted by a pair of electrodes
on opposite sides of said first branch of the second
wave-guide;
a second phase modulator coupled to said second
branch of said second Y-shaped wave-guide, said second
phase modulator being constituted by a pair of electrodes
on opposite sides of said second branch of the second
wave-guide;
a first optical frequency converter integrated
into said substrate and optically coupled to said ~irst
branch of said second Y-shaped wave-guide, said first
optical frequency converter including:
third and fourth wave-guides integrated
into said substrate and having different optical charac-
teristics from one another, said third wave-guide being
optically coupled to said first branch of said second
Y-shaped wave-guide, and said fourth wave-guide having
a portion that is substantially parallel with a portion
of said third wave-guide over a predetermined length
thereof;
first means for generating an acoustic
wave collinear with an incident wave propagating direc-
tion of said third wave-guide, said generating means
being positioned so that an acoustic wave generated
thereby produces a frequency shift in an optical wave
coupled between said third and fourth wave-guides;
a second optical frequency converter integrated
into said substrat:e and optically coupled to said second
branch of said second Y-shaped wave-guide, sai d second
optical ~requency converter including:
..~
- 3d -
fifth and si~th wave-guides inteyrated into
said substrate and having different optical characteristics
from one another, said fifth wave-guide being optically
coupled to said second branch of said second Y-shaped
wave-guide, and said sixth wave-~uide having a portion
that is substantially parallel with a portion of said
fifth wave-guide over a predetermined length thereof;
second means for generating an acoustic
wave collinear with an incident wave propagating direction
of said fifth wave-guide, said generatiny means being
positioned so that an acoustic wave generated thereby
produces a frequency shift in an optical wave coupled
between said fifth and sixth wave-guides; and
an optical fiber ring having a first end optically
coupled to said fourth wave-guide and a second end opti-
cally coupled to said sixth wave-guide.
BRIEF DESCRIPTION OF THE URAWINGS
Other characteristics and advantages of the
present invention will appear in the following non-
restrictive description made with reference to the
accornpanying drawings in which:
Figures 1 to 3 which are labelled as " PRIOR ART"
show a device known in the art;
Figure 4 shows the device in accordance with the
invention;
Figure 5 which is labelled as " PRIOR A~T"
shows a device known in the art;
Figures 6 and 7 show a system containing the
device in the invention; and
- Figure 8 which is disposed on the same sheet
of formal drawings as Figure 5 shows a variant of the
system.
~ 3e ~
DETAILED DESCRIPTION OF THE PREFERRED EM _DIMENTS
Figures 1 and 2 show a sectional view and a
view from above respectively of a switch produced in
': . : '
' "' ''
~uides 1 and 2 are inserted in the sub~trate 3. The materlal through~ which
csuplin~ is made is the one forming susbstrate 3. To insert wave-guides 1
and 2, it Is possible9 for example, to cause titanium to diffuse in a substrate
formed of a monocrystalline shee~ of lithium niobate (LiNbO3~ e tita-
5 niurn, in the diffusion zone, replaces the niobium in part to ~ive a mixedcompound with the formula LiTiXNbl ~03 with a refractive index higher
than that of the pure nioba~e. These diffused ~ones, with an index higher
than that of the substrate, form wave-guides I and 2. If the diffusion
temperature is higher than the Curie point for the ma~erial, the cooling
I0 phase which follows is used to subject the plate to a uniform electric field in
order to polarize the plate uniformly and produce a "sin~le domaine"
structure.
When a voltage is applied between the electrodes 10 and 20, a
distribution of field lines is produced which is shown as reference 4 on figure
15 1. The field component in the directlorl C perpendicular the substrate
surface 23 has the sarne absolute value but opposite direction in one guide
and the other, which gives refractive index var;ations of the same absolute
value and opposite sign. Nonetheless, the existence, in a direction perpendi-
cular to the direction of the substrate axis C, the susbtrate having its
20 extraordinary index, of a non-zero field component, and the fact that the
electric field applied also causes the index value to vary in the part of the
substrate 22 contained between the two wave-guides cause a certain
dissymetry of the phenomenon. The coupling obtained varies in accordance
with the polarity of the voltage applied between the electrodes 20 and 21.
2s The polarity of the voltage supplying the maximum coupling can be deduced
from the crystallographic orientation of the material forming the substrate~
If this orientation is unknown, it is very easy to determine experimentally
the optimum polarity by the measurement of the luminous intensity trans-
mitted by ore of two wave-guides fortwo polari~ies of opposite signs.
If the metallic electrodes are arranged directly on the wave-guide
surfaces, the existence of a vanishin~ wave moving in the relatively
abs~rbent metallic medium can cause energy losses in the coupler. To avoid
them it ls possible to fit in between as shown ln figure 1, a transparent
dielectric layer 11 and 21 between wave-guides 1 and 2 and electrodes 10
and 20. This isolating layer is made of a material wi~h good tr~nsmission for
the iuminous wave-len~ht carried by the wave-guide and a refractive Index
lower than that of the wave-guide. Silicon dioxide (SiO2) is a ma~erial
perfec~ly adap~ed to the case previously described in which the substrate is
5 formed by lithium niobate.
The two wave-guides9 as shown in figure 2, are parallel one with ~he
other on a rectilinear section of lenght L, which is a flmction of the
parameter, called the coupling lenght, which will be defined later. 1 he
distance between the rectilinear parallel parts is of a value d which must
10 not exceed a few wave-lengths (calculated in the medium separating the two
wave ~uides) of the light carried by the wave-~uides. The two waves-guides
are formed by the ~ame electro-optical ma~erial which, when subjected to
an electric field, has a refractive index that varies as a function of the
applied field value. The refractive index of this material is so chosen that,
15 even in the presence of the applied electric field, it remains higher than the
index of the material forming substrate 3.
Because of the electro-optical character of the material forming
wave-guides 1 and 2, the distribu~ion of the ~ield lines in the wave-guides
produces within them refractive index varia~ions roughly equal in absolute
20 value but of opposite signs.
When a wave is carried by wave-guide, part of the energy propagates
outside the wave-guide, in the medium which surrounds it in the form of a
vanishing wave. The amplitude of this wave decreases exponentially on
leaving the wave-guides walis. If a second wave-guide is arranged parallel to
25 the first one, it picks up progressively, throu~h this vanishing wave, the
energy carried in the first wave-guide and, the closer are the wave-guides,
the more quickly it does it. AEter a given distance, called the coupling
lenght, which depends on the geometrical and optical parameters of the two
wave-guides and of the medium separating them (and of the refractive
30 indices in particular), the maximum of energy has been transferred from the
first wave-guide to the second. Beyond this lenght, the reverse phenomenon
occurs. The energy transfers pro~ressively from the second wave-guide to
the ~irst to leave ~he minimum value in the second wave-guide. Any
:' :
modification of the index of one of the media present acts in one direction
or the other along the coupling lenght.
In the device shown in figures 1 and 2, the lenght L can be cho~en
equal to th~ couplin~ len~h~ in the absence of this applied electric field.
5 Because of the perfect symetry of the two waves-guides in the coupling
zone, the energy transfer is complete from the first wave-~uide to the
second (or from the second to the first). The application of a voltage
between electrodes 20 and 21 reduces the coupling lenght and part of the
energy is retransferred from the second wave-guide to the first (or from the
10 first to the second). The final result is then that, as the voltage is increased,
the energy transferred from the first wave-guide to the second (or from the
second to the first), measured at the end of the coupling zone, is reduced to
reach a zero value. The coupling between the two wave-guides thus
decreases from 100 % to 0 % when the voltage applied to the electrodes
15 increases. The result would be ~he same if the lenght L was made equal to
and odd multiple of the coupling lenght with a zero field.
It is also possible to give the lenght L a value equal to an even
multiple of the coupling lenght with a zero field. The energy transferred at
the output, from one wave-guide to the other, increases from zero when the
20 voltage applied between the electrodes increases from zero.
A device has then made which, when controlled by an electric signal,
enables part or all of the energy carried by one wave-guide to be switehed
to the other associated with it in the coupling zone.
It goes without say;ng that, if one of the two wave-guides is limited to
25 a section whose minimum lenght is the lengh~ L of ~he coupling zone, this
device enables the energy carried in the other wave-guide to be modulated
no%.
In the case in which the two wave-guides are different, a periodic
structure made behveen them can make it possible to increase the exchan-
30 ges between them. When the wave carried in one wave-guide has the same
propagation speed as one of the orders diffracted in the other wave-guide,
there is then an energy exchange.
To produce this exchange several means can be used, in particular the
production of an electric field between two electrodes, for example, of
-
' - .,
.
periodic structures 18 and 29 fitted on one side and on the other of ~he two
wave-guides 5 and 6 as shown in figure 3. A luminous wave 24 propagating in
the first wave-guide produces, by coupling due to the presence o~ a
polarization Vo9 a coupled wave 25 which propagates in the second wave-
5 guide 6. It can also be the production of a network engraved in the substratebetween the two wave-guides. In the device of the in~ention, acoustic waves
12 are produced by the électrodes 13 and 14 in the form of interdigi~al
combs, at whose terminals a genera~or V is connected, which propagate
between the two waves-guides as shown in figure 4. However, the electrodes
10 can be deposited on a thin layer 26 of a piezoelectric material, a ~inx oxide(ZnO) for example, the thin layer being deposited on substrate 3 consisting o~
another material, silicon dioxide for example. Thin layer 26 can be made of
the same material as the substrate, crystalline quartz, gallium arsenide or
lithium niobate for example.
The device in accordance with ~he invention has be advantage of
allowing an adjustement of the coupling between wave-guides 5 and 6 which
is a function of the acoustic wave frequency. This acousto-optical deflector
allows a frequency translation. The luminous waves carried by one of the
wave-guides 5 and diffracted by these acousto-optical waves are then
20 converted in frequency and transmitted in the second wave-guide 6. These
two wave-guides are not necessarily of the same width.
If a medium 30 is considered in which a beam of elastic waves 31 of
frequency f is propagating, as shown in figure 5, if an incident luminous
beam 32 is passed in this medium, a ~roup of diffracted beams 33 is obtained
25 with frequencles F + + kf, in which k is a positive or negative whole number.The sinusoidal variation oE the index, produced by the elastic wave,
has an effect on the luminous wave analogous to that of a phase network.
The luminous beam 32 penetrating the crystal 30 in parallel with the elastic
wave planes is separated into several beanns symmetriAcally inclined with
30 respect ~o ~he incident beam by angles ~ N: Sin ~N = k ~ in whlch ~ is the
wave plane step and ~ the incident beam wave-lenght. However, the elas~ic
beam thiclsness e must be less than a critical value ec. The side waves are
produced all long ~he carrier wave path inside the ultrasonic beam and not
only at the output, on the frontier. If, in throught, the elastic beam is
divided into thin slices parallel to the propagation direction, for each of
these slices parallel to the propagation direction, for each of these slices9
~he preceding spectral analysis is valid. The frequencies Q ~ k~ and ~he
direction of propagation ~ N oE ~he side wavcs are the same for the absclssa
slices x and x ~ Q . iE9 for a given order, the contribu~ions of these two sl~esQ apart are added, there is phase opposition for a dis~ance Q N = A
N The interference of the waves emitted by the slices QN apart may
then be destructive. If the width o~ the beam is greater than Q N~ the effect
of a slice is cancelled by ~he silice Q N away. Under the best conditions, the
elastic beam 2thickness e must not exceed a first order critical value:
ec= Rl= A .
For a Bragg angle incident of the luminous beam 32 with respect to
~he elastic wave planes, the interaction is the biggest because it enables the
interferences to be made constructive for the first order of angular
frequency Q + ~. Hence, it only supplies a single deviated beam.
The device as in the invention uses a directional coupler whose two
wave-guides are not identical. In this case, if B /K 1 and B/K~ are the
propagation constants of the modes in these two wave-guides of the coupler~
the relative energy in one of the wave-guides when the other is energized
will be written: E = 12 2 Sin2V 1 + ~B 7i~cL in which L is
l ~ a~ l~c
the interaction lenght, c is the coupling constant and ~B = 2N (~ IK1 -
C2) in which ~ is the wave-lenght in a vacuum. The relative energy present
in this wave-guide at the coupler output depends on three parameters, L, c
and ~ . If ~B is large with respect to c, it can be sen that, in any case, no
25 matter what the value of L, the maximum energy exchanged can be smallO
~or example, if c _ 1.510 4 /um9 2 ~ = 0-0001, EMAX = 0.0017 and if
c - 1.5 10-4 /um, 2Tr ~= O-Ol~ EMAx = 0.000017
These values are very small and can be reduced further arbitrarily by
changing the lenght L.
1~ is known ~hat, if the propagation constants in the two ~ave-guides
are made to vary periodically and the corresponding period is carefully
chosen, the exchange between the two wave-guides can be increased by
comprising the ~ ~ with the network vector K.
.
~25~
The interaction is then vrit~en, because of the conservation of the
nts ~ 29 l e- ~ K - ~ 2/K) = ~ in which A is the
network period.
Hence, if the network is formed, as produced in the device of the
5 invention shown in figure 4, by an acoustic wave propagating colinearly ~o
the optical wave, there will be a frequency translation of the coupled wave.
The effectiveness of the interaction depends on the value of the index
variation induced by ~he acoustic wave and hence nf the power injected. A
directional coupler produced in lithium niobate (LiNbO3) by titanium diffu-
10 sion can be taken as an example. The variation in index corresponding totitanium is usually of the order of: ~ n ~ 510 3.
1~ can be seen then that it is possible to produce the two wave-guides
with ~ B/K = 210 3. This can be obtained by changlng the width and/or the
thickness of the titanium for these two wave-guides in the coupler. For an
15 interaction lenght of lO mm9 the maximum energy exchanged will be: for
)~ = 0.83/um, EMAX = 4 10 4. The acoustic wave-length required for com-
pensation will be: 415tum, i.e. in the case of lithium niobate (LlNbO3) a
frequency transla~ion of 7.2 MHz. The wave picked up at the output of the
second wave-guide (which was not initially energized) will then be obtained
20 with a frequency translation of 72 MHz and the maximum quantity of
fundamental in this wave-guide will be - 33 d~ with respect to the total
optical energy.
The device as in the invention can also be produced by making one of
the wave-guides by proton exchange and the other by titanium dif Eusion ~or
25 the two by proton exchange but different characteristics3. In this case,
~ ~/K ~ 0.1 can be obtained with an interaction lenght of lO mm, the
maximum energy exchanged: - 6~ dB OI the total energy is ob~ained with an
acous~ic wave-lenght of 8.3 /um, i.e. an acoustic frequency of the order of
361 MHz.
Hence, in the device of the invention shown in figure 4, a wave 23
passed in the firs~ wave-~uide causes by coupling the existence of a wave 25
in ~he second wave-guide, this wave then heing translated in fre~uency.
Sever~l wave-guide configurations are possible, wlth a substrate 3 of
lithium niobate, for exampleO lhe two wave-guides are obtained by diffusion
of titanium in the substrate. The waves L~uided in ~he two wave-guides are
either twc TE waves or two TM waves. ~ of the order of a few l0 3 i5
then obtained.
Ig is also po~sible to have a crossed interaction, i.e. a TE wave in the
5 first wave_guide and a TM wave in the second or YiC2 versa. K of the order
c>f 0.l is then obtained
One Qf the two wave-guides can be obtained by titaniuM diffusion and
the second one by proton exchange. If an axis C perpendicular to the
substrate surface is considered, there is a TM wave in each of the two wave-
10 guides. There could also be twn TE waves. ~ of the order of 0.l is then
obtained. The two waves-guides can be obtained by proton exchange but
their characteristics must then be different. ~ ~ û.l is then obtained.
By changing the acoustic frequency, which can vary from l0 to 3D0
MHz, a tunable filter can be obtained. The birefringence of the material
varies as function of the frequency.
The pass-band of the device in the invention is a function of the
optical wave to acoustic wave interaction lenght. The greater the number of
wave planes in the acoustic wave seen during this coupling9 the narrower the
pass-band.
The device described here can be used then as a filter by using, for
example, ~he variation of the birefringence of a rna~erial with the wave-
lenght. It can be considered then that it is a TE (TM) wave in the first wave-
guide and TM (TE) wave in the second which are coupled by means oE the
acoustic wave. In this case, for lithium niobate, ~ ~ B /KTM - ~B KTE) ^ 0.l
and again an acoustic wave frequency of about 361 MHz. This filter i5
adjustable because it is sufficient to change the acoustic wave frequency.
In the device of the invention, electrodes can be deposited, on one side
and on the other of the two wave-guides, for example, or on these wave-
guides. An insulating buffer layer can also be deposited between the
electrodes and the substrate. The electric field produced between these two
electrodes enables the device of the invention ~o be adjusted in its initial or
its final state.
The device of the invention has an application in the field of the
optical fibre ra~e gyro.
~55~3~
11
Figure 6 shows schematically a ring interferometer of known design. A
laser source S transmits a beam of parallel rays 41 to a separa~or deYice
formed by a semi-transparent sheet M.
A certain number of mirrors, 15~ M2, M3, define an optical path
5 forming the interferometer ring. This ring can be made, for example~ with a
monomode op~ical fibreO The sensitiYi~y of the measurement is increased by
the use of a long optical path. This ring is looped back on the separator
device M which also acts as a mixer device and determines an outpu~ branch
43. Two waves, then, propagating in opposite directions, pass trough the
10 ring, one clockwise (direction 2) and the other anticlockwise (direction 1).
These two waves recombine on the separatin~ sheet M. The result of this
recombinaison can be seen in the output branch 43 with detector D. Part of
the beams is picked up a~ain in the input arm by the separating sheet M' and
passes through filter device F again. At the output the two waves recombine
15 on separating sheet M~. The result of this recombina~ion can be seen in the
output branch 44. The fact that the filter device F has been fitted in the
interferometer input arm makes the arm strictly reciprocal. Hence, a wave
contained in a single optical mode passes through it. This filter device
consists of a mode filter followed by ~ polarizer. The incident beam 41
20 passes throu~h this filter and the fraction which comes out is in a single
mode. It is possible then to consider either the emerging beam 43 correspon-
ding to the Interference of the two beams which have not passed through the
mode filter device or the part of the beams which is picked up again in the
input arm by semi-transparent sheet M. This part of the beams passes
25 through filter device F again. At its output, the two beams passed through
arm 44 by means of semi-transparent sheet M' are contained in the same
mode and this makes the interferometer insensitive to 'reciprocal' disturban
ces.
If ~ ~ is the difference in phase between the two waves which
30 propagate in opposite directions in the ring and PS is the optical power
which can be measured in output branch 44, then, when there is 'non
reciprocal' disturbances; ~ 4 is zero.
If a rate gyro using this ring interferometer is considered, a 'non
reciprocal' disturbance will be produced when the rate gyro is spun. The
~s~
12
phase difference ~ ~ is not zero and ~ Q in which n Is the speed of
rotation and a = k ~C in which k is a constant dependin~ ~n the rate gyro
geometry7 L the lenght of the optical path9 A the length o~ ~he ligh~ wave
~mitted by the laser source S and C the speed of the li~ht in rin~ 42. When
the speed of rota~ion n increases, ~he pha.se di~ference ~ ~ increases in ~he
same proportion because coefficient a remains constant. The optical power
PS changes according ~o a cosine law.
~, .
Ps = PlS + P25 + 2 Y PlsP2~ Cos ~ ~ ~ ) in which PlS corresponds to
direction 1 and P2S to direction 2. The measurement sensitivity for a ~iven
dPS
value ~ ~ is given by the derivative of Ps~ d(~ )
2V~Sin ~ ~ ~ 3
The interferometer sensitivity is very low if the phase difference
15 is little different from zero. This is the case in a rate gyro if it is required
to measure small rotational speeds Q . The variation in optical power in the
output branch is shown in the dia~ram in figure 7.
It may be considered that the terms PlS and P2S are equalO It follows
that, for a phase difference ~ = 7r, the power detected is a minimum. It
20 passes through a maximum PSmax when Q ~ = 0; 21r and so on.
To increase the interferometer sensitivity, a constant 'non reciprocal'
bias can be introduced in the phase of the two waves circulatin~ in opposite
directions in order to move the interferometer operating point.
In the case of a func~ion varying In accordance with a cosine law9 the
25 greatest sensitivity point is obtained for angles ~2k + 1) ~r/2 in which k is a
whole number. A bias can then be chosen which introduces a phase variation
for each wave with an absolute value of rr /4 but with opposite signs. In the
absence of 'non reciprocal' disturbance, ~he phase diff erence then be-
comes: ~ ~' = Q~ + ~ ~ o in which ~ ~ o = 7r/2. This is point A in figure 7.
As shown in figure 6, it is possible to add in the wave path in ring 42 a
phase modulator 45 which uses a rexiprocal effect to obtain better sensitivi-
ty with the device. This modula~or is so energized as to produce a phase
variation in $he wave passin~ through it. This variation is periodic, the
period being 2 T in which T iS the time for a wave to pass in the ring.
5~3~
The difference ~hen becomes ~ + ~ (t-T ) in which each o~ the
waves circulatin~ in the opposite direction undergoes this phase shift when
it passes through the modulator with ~ ~t~ = ~ (t + 2T )-
The operating point then describes the curve P5 = ~( a~ ) in figure 7
5 symmetrically between a pair of end points.
The device (reciprocal phase modulator) which enables the disturbance~ (t~ to be introduced can~ with advantage, be divided into two devices, 4S
and 46, one at each end of the path as shown in figure 6, one giving the
phase shift ~ l(t) and the other the phase shift ~ 2(t). These phase modula~or
10 devices, symmetrically placed at the two ends of the optical path may be in
opposition. This arrangement adds additional symetry to the phenomena and
reduces the second order errors due to possible non-linearity in the
modulators.
The ideal is to work at points A and B on the curve shown in figure 7
15 to start with. To work at A, ~ l(t) = 4 and ~ 2(t~ = - 74 and then to work
at point B, ~ 1(t) = - 4 and ~2(t~
This result can be obtained by using two square-shaped signals with
two levels,- 4 and 1l .
If the phase modulation signals are frequency F, if the gyroscope is not
20 rotating, a rectified signal at frequency 2F is obtained after detection.
However, if the gyroscope is rotating, frequencies F and 2F are obtained.
This device has the disadvantage of not lncluding a zero technique. Also9 the
measurement is not linear.
If a zero method is required, a non reciprocal effec~ compensating the
25 rotational effect must be considered. A component at frequency F in the
signa3 detected which is zero must then be obtained. The modified parame-
ter enabling the rotational speed to be found is then measured.
The Iield applied to the modulator electrode terminals can be altered
if it is electro-optical. The difference of frequency in the modes which are
30 propagating can be altered and this résults in a phase shift at the detector
output.
The device in the invention finds its application in this optical fibre
rate gyro field ln which two frequency converters o~ the invention can be
arranged in the two arms working at frequencies such that the non
~5~3~
reciprocity, introduced by the fact that the two waves in the interferometer
are not at the same frequency9 compensates for that due to the Sagnac
eff ect.
Two converters, 62 and 63, arranged beside the modulators, 45 and 46,
5 as shown in figure 6 can be considered.
l he deYice in the invention then makes possible digital adjustement. If
two frequertcy converters are placed beside ~he two modulators9 i~ is
possible to compensate for the component of frequency F which is due to
the Sagnac effect, when there is rotation. There are then two frequencies,
10 Fl and F2, in the two converters.
In the standby condition, ~l = F2 should be obtained. When ~he
gyroscope rotates at constant speed, frequencies Fl and F2 beat and the
number of beats can be counted.
The progress made in the production of low loss op~ical fibres makes it
15 possible to uce optical fibres to produce these ring interferometers as it has
already been said. An example of the production of a ring interferometer
complying with the invention is shown in figure 8. The fibre 52 wound round
itself forms ring 42 of the interferometer. The various branches of the
interferometer are made of in~egrated optics. The wave-guides are made by
20 integration in a substrate. The substrate can be chosen~ for example, from
among the following materials: lithium niobate or tantalum niobate in
which, to make the wave-guides, titanium or nlobium respec~ively have been
diffused.
The frequency conver~er is broken down into two converters, 54 and
25 S5, placed at the ends of the fibre. These conver~ers, are the devices
already described in the invention which make it possible to compensate for
the Sagnac effect, when the two frequencies of the two acoustic waves ~58,
59), generated by the elec~rodes ~S6, 57) are altered. The phase modulators,
60 and 61J shown by the electrodes placed on one side and the other of ~ach
30 o~ ~he wave-guides are in the loop to make it possible to find out the
instants at which the gyroscope rotates. In this case, a component of the
signal at frequency F is detected as it has already been explained.
The optical radia~ion separators consist of monomode wave-guides
connected between themselves to form Ys, these Ys being connec~ed
::L25~
between themselves by one of their branches acting the role previously
played by the semi-transperent sheets in figure 6. The l,vave-guide ~8 actrs
~he role of the monomode filter in figure 1, a polarizer being made, for
example, by metallizing 49 on the substrate surface over wave-guide 48.
The devlce in the invention also finds applications in optical 2elecom-
munica~ions to multiplex/demultiplex optical waves in wave-length.
, .
. ' ~ ' .,
' ~ '' -
