RU2720263C1 - Method of internal feedback loop arrangement for phase synchronization of fibre lasers grid in systems of coherent addition of beams and device for implementation thereof - Google Patents

Abstract

FIELD: physics.SUBSTANCE: invention relates to laser equipment. Method of internal feedback loop for phase synchronization of fibre lasers in systems of coherent addition of beams is realized by device containing narrow-band laser, generating coherent, linearly polarized Gaussian beam, fibre splitter dividing radiation into N channels associated with N optical phase-shifting elements which control the phase of the optical wave, depending on the value of the applied control voltage, N fibre amplifiers having fibre output into free space. N fibre outputs form equidistant array of divergent beams with Gaussian intensity distribution, each of which consists of central maximum, containing ~ 92 % power and peripheral part, having N collimating lenses output, forming synthesized aperture consisting of clusters, each of which comprises central collimating lens and 6 peripheral collimating lenses assembled into hexagonal grid, which form in each channel flat wave front and direct radiation of all channels in parallel to each other. Small fractions of peripheral parts of Gaussian beams are intercepted after collimation by parabolic axial mirrors located behind plane of input pupil of each three of adjacent collimating lenses located in tops of triangle, one of which is located on optical axis of central collimating lens. Reflected from each parabolic mirror fraction of peripheral parts of radiation of neighbouring Gaussian beams, passing through the hole in the centre of three collimating lenses, are focused in plane of small diaphragm of photodetector, where their interference takes place. Signal from the photodetector is transmitted to the control controller, which generates iterative stresses in accordance with the stochastic parallel gradient descent method to control the corresponding phase-shifting cells, while maintaining the maximum attainable signal amplitude from the photodetector, which corresponds to phase synchronization of the laser beams.EFFECT: technical result consists in provision of possibility to obtain high power density in systems of coherent addition of beams.9 cl, 2 dwg

Description

FIELD OF THE INVENTION

The invention relates to laser technology and can be used in multichannel systems of coherent addition of laser beams in order to organize an internal feedback loop for phase synchronization of laser beams. The invention can find application in various fields of technology where the use of laser radiation with a high radiation power density is required, for the tasks of wireless optical communication, in medicine, in the military-industrial field, with remote sensing and processing of materials.

State of the art

To create high-power lasers, spectral (Dennis Lowenthal and Andrew Brown, Aculight Corporation, NASA Tech Briefs, Jan 2006) and coherent methods of addition of laser beams (Pyrkov Yu.N., Trikshev AI, Tsvetkov VB "Phase of several amplification channels in the coherent addition of laser beams ", Quantum, electron., 2012), which allow to increase the radiation power of the laser system by N times, and the radiation intensity in the far zone by N ² times, where N is the number of laser channels. Both methods demonstrate comparable performance indicators when adding relatively low-power laser beams. When using powerful laser beams of diffraction divergence, the most promising is the method of parallel coherent addition of N laser beams formed by a multichannel system with a synthesized aperture. However, in such systems, the deterioration in the quality of the resulting radiation can be associated with both depolarization of the channel radiation, incomplete filling of the resulting aperture with the radiation of the added channels, and wavefront distortions due to an error in the phase adjustment of different channels, which is the most critical parameter requiring priority stabilization. In such systems, it is necessary to arrange a phase shift control feedback loop.

Known methods for coherently adding laser beams, for example, the method (Yanxing Ma, Pu Zhou, Xiaolin Wang et al., "Coherent beam combination with single frequency dithering technique", Optics letters 35, 9 (2010), 1308-1310), based on the definition and alternately compensating for the relative phase difference in each channel.

An analogue of the present invention is a method (Pyrkov Yu.N., Tsvetkov VB, Kurkov A.S., Trikshev A.I. "Method for coherent addition of laser beams with synchronous detection and a device for coherent addition of laser beams with synchronous detection" patent RF invention No. 2488862), based on the determination of the phase difference of the radiation in each working channel relative to the radiation of the reference channel. The disadvantage of these schemes is the need to install a synchronous phase detector in each working channel, which reduces the noise immunity of the system, greatly complicates the design, increases the cost and mass-dimensional parameters of the system.

There are schemes (Volkov M.V., Garanin S.G., Dolgopolov Yu.V. et al. "Method for the coherent addition of laser radiation in multichannel cw lasers", RF patent for invention No. 2582300; Ling Liu, Mikhail A. Vorontsov, "Phase-Locking of Tiled Fiber Array using SPGD controller", Proc. Of SPIE 58650P, 2005), in which the radiation is extracted from the optical fiber using collimators assembled in a hexagonal order into one module, and part of the radiation at the system output is allocated using a beam splitter plate and the lens focuses on the photodiode through a diaphragm whose size is less than the bandwidth of erferentsionnoy pattern formed in the plane of the receiving photodiode area. The phase control in the channels is carried out by iteratively supplying control voltages to the phase modulators according to the stochastic gradient descent method. The disadvantages of these circuit solutions are the need to obtain a limited control beam using a signal reflected from an external remote target or to introduce a beam splitter plate into the light beams after the collimators, which significantly increases the dimensions of the system and introduces distortions into the wave fronts of the light beams passing through it, as well as the loss of a part of the energy when reflected from a beam splitter plate.

Closest to the claimed method is the scheme described in (M.A. Vorontsov, SL Lachinova LA Beresnev, T. Weyrauch "Obscuration-free pupil-plane phase locking of a coherent array of fiber collimators" JOSA Vol. 27, No. 11, A106-A121. 2010), in which a small portion of the laser radiation generated by the peripheral part of the Gaussian beam, which contains about 10% of the total beam energy, is used to operate the phase shift feedback loop. The fiber laser radiation divided into channels is directed to phase modulators, after which all the channels are exposed parallel to each other, and in each of them a flat wavefront is formed using a round-shaped collimating lens. Collimating lenses form a synthesized aperture consisting of clusters, each of which contains a central collimating lens and 6 peripheral collimating lenses assembled in a hexagonal lattice. Phase synchronization of the beams is ensured by an internal feedback loop using the peripheral parts of diverging Gaussian beams intercepted in front of the collimating lens location plane. Segments of off-axis parabolic mirrors or a mirror diffraction element intercepting a small fraction of the peripheral part of the Gaussian beam and focusing it in the plane of the diaphragm with a photodetector are used as a device for organizing the internal feedback loop. In this case, an interference pattern is formed in the plane of the diaphragm from three neighboring channels located at the vertices of the triangle, one of which is located on the optical axis of the central collimating lens. The maximum intensity intensity of the interference pattern depends on the relative phase difference Δϕ of the interfering light beams and reaches its extremum at Δϕ = n⋅2π, where n = 1, 2, .... The photodetector detects the intensity of the interference pattern and generates a signal at the input of a multi-channel controller, which supplies iterative voltages to the phase-shifting elements according to the stochastic parallel gradient method until the input signal reaches its maximum amplitude, forming a feedback loop. The peak intensity of the interference maximum is adjusted by moving the diaphragm with the photodetector. Organized in this way, the internal feedback loop, using the interference of peripheral Gaussian beams, allows you to get a limited reference beam without using a signal reflected from an external remote target or a bulky beam splitter that introduces distortions into the generated wave fronts.

A significant disadvantage of this method is the need for focusing in the receiver plane of the peripheral parts of diverging Gaussian beams, which causes aberration distortions of the wavefront and, as a result, blurring the interference pattern and reducing its contrast. This leads to an increase in the error in determining the magnitude of the phase mismatch and a decrease in the efficiency of the feedback loop, which in turn leads to a decrease in the achievable intensity of the central maximum of the far-field intensity distribution.

In turn, the implementation of this method requires the placement of a reflecting optical element in diverging beams, consisting of three segments of an off-axis parabolic mirror, which are very difficult to align, while inevitably there are aberration distortions of the wavefront due to the fact that the parabolic mirror does not work in a parallel beam, but diverging, and this leads to a decrease in the contrast of the interference pattern.

где Δх - ширина итерференционной полосы в опыте Юнга, Я - длина волны излучения, L - расстояние от источника оптического пучка до плоскости наблюдения интерференционной картины, d - расстояние между центрами двух оптических пучков, d_ТД - диаметр диафрагмы, что является необходимым условием формирования сигнала обратной связи для управления согласно методу стохастического градиентного спуска. В случае интерференции трех пучков, интерференционная картина представляет собой структуру световых пятен с размером пятна, равным Δх. В практически реализованных системах когерентного сложения пучков волоконных лазеров, фокусное расстояние коллимирующих линз

лишь в несколько раз превышает дистанцию между центрами пучков d (в прототипе

что ограничивает при разумных с точки зрения конструктивных параметров размерах системы, ширину интерференционной полосы условием Δх≤10λ, и накладывает серьезные ограничения на размеры диафрагмы и, соответственно, ограничивает мощность регистрируемого фотодиодом сигнала. Чувствительность современных фотоприемников, работающих в ближнем (800-1700 нм) ИК диапазоне составляет ~ 0,5 А/Вт, а темновой ток ~ 1 нА. Таким образом, падающая на фотоприемник мощность излучения должна составлять десятки - сотни нановатт. В связи с этим, возможность увеличения ширины интерференционной полосы, а, следовательно, и допустимого диаметра диафрагмы d_ТД, ограничивающей мощность падающего излучения, является важным фактором, обеспечивающим эффективность работы контура обратной связи.In this case, due to the location of the mirror element intercepting the peripheral parts of the Gaussian beams in front of the collimating lens location plane, another problem arises related to the need to form an interference pattern with a strip width in the receiver plane

where Δх is the width of the interference fringe in Jung's experiment, I is the radiation wavelength, L is the distance from the source of the optical beam to the observation plane of the interference pattern, d is the distance between the centers of two optical beams, d _TD is the diameter of the diaphragm, which is a necessary condition for signal formation feedback for control according to the stochastic gradient descent method. In the case of interference of three beams, the interference pattern is the structure of light spots with a spot size equal to Δx. In practically implemented systems of coherent addition of fiber laser beams, the focal length of collimating lenses

only several times greater than the distance between the centers of the beams d (in the prototype

which limits the size of the system, which is reasonable from the point of view of design parameters, the width of the interference band by the condition Δx≤10λ, and severely limits the aperture size and, accordingly, limits the power of the signal recorded by the photodiode. The sensitivity of modern photodetectors operating in the near (800-1700 nm) IR range is ~ 0.5 A / W, and the dark current is ~ 1 nA. Thus, the radiation power incident on the photodetector should be tens to hundreds of nanowatts. In this regard, the possibility of increasing the width of the interference strip, and, consequently, the allowable diameter of the diaphragm d _TD , limiting the power of the incident radiation, is an important factor that ensures the efficiency of the feedback loop.

Summary of the invention

The purpose of the invention is to create a method and apparatus for organizing an internal feedback loop for phase synchronization of an array of fiber lasers, which allows to increase the contrast of the interference pattern, to increase the achievable width of the interference band, to reduce the error value of the phase mismatch necessary to increase the efficiency of the feedback loop, and to obtain power density in systems of coherent beam addition.

This goal is achieved in that the proposed method and device for organizing the internal feedback loop, as well as the prototype, include laser radiation divided into channels directed to phase modulators. After passing through the phase modulators, all channels are amplified and aligned parallel to each other, while a flat wavefront is formed in each channel using collimating lenses. Collimating lenses form a synthesized aperture consisting of clusters, each of which contains a central collimating lens and 6 peripheral collimating lenses assembled in a hexagonal lattice. The phase synchronization of the beams is ensured by an internal feedback loop using the peripheral parts of the Gaussian beams. In contrast to the prototype, the peripheral parts of the Gaussian beams are intercepted after the laser radiation is collimated by a parabolic axial mirror located behind the plane of the entrance pupil of the collimator, and the axial point of which is located at the intersection of the medians of an equilateral triangle, the vertices of which lie on the axes of three neighboring collimating lenses, one of which is the central collimating lens for this cluster. The diameter of the parabolic mirror is chosen in such a way as to prevent the vignetting of the part of the Gaussian beam containing 92% of the power. The part of the multichannel radiation reflected from the parabolic mirror is focused through the hole in the geometric center of each triple of collimating lenses onto the diaphragm with a photodetector to record the maximum intensity of the interference pattern arising from the addition of wave fronts from three neighboring channels. In contrast to the prototype, hexagonal lenses are used as collimating, intercepting ~ 96% of the power of a Gaussian laser beam and glued together in a hexagonal order into a common module.

The advantages of the proposed method are that the interception and focusing in the plane of the diaphragm with the photodetector of the peripheral sections of the Gaussian beams is carried out after their collimation, which allows:

1) minimize the aberration distortions of wave fronts, preventing blurring of the interference pattern and increasing its contrast;

2) reduce the error value of the phase mismatch and increase the efficiency of the feedback loop, which leads to an increase in the achievable intensity of the central maximum of the far-field intensity distribution. The advantages of the proposed device are that the use of an axial parabolic mirror when organizing the internal feedback loop for phase synchronization in systems of coherent addition of fiber laser beams allows you to:

1) to simplify the adjustment;

2) increase the achievable value of the width of the interference pattern ~ (three) D _floor / D _z times compared to the prototype while maintaining the same structural (geometric) parameters;

3) set the bandwidth of the interference pattern in accordance with the selected size of the point aperture by selecting the focal length of the parabolic mirror;

4) to simplify the adjustment of interference fringes in the plane of the photodiode without moving the diaphragm with the photodiode, and by displacing the parabolic mirror along its optical axis and controlling its tilts.

The hexagonal shape of collimating lenses allows denser filling of the synthesized aperture with the radiation of folding channels and thereby increase the achievable power density in the system of coherent beam addition.

New for the method is: focusing in the plane of the photodetector of the peripheral parts of the Gaussian beams after their collimation, which eliminates the wavefront aberrations, eliminates blurring and increases the contrast of the interference pattern and increases the width of the interference strip.

New for the device is:

- the use of collimating hexagonal lenses glued together in a hexagonal order, which leads to a more complete filling of the synthesized aperture with laser radiation and, as a result, an increase in the achievable power density in the system of coherent beam addition;

- to intercept and focus the peripheral parts of the Gaussian beams in the plane of the photodetector, an axial parabolic mirror is used, located behind the plane of the entrance pupil of the collimating lenses, and the axial point of which is located at the intersection of the medians of an equilateral triangle, the vertices of which lie on the axes of the three neighboring collimating lenses, one of which is a central collimating lens for a given cluster;

- the diameter of the parabolic mirror is chosen in such a way as to prevent the vignetting of the part of the Gaussian beam containing 92% of the power;

- a hole is located in the geometric center of each triple of collimating lenses to prevent vignetting and refocusing of radiation reflected from the parabolic mirror.

Brief Description of the Drawings

In FIG. 1 shows an optical diagram of a device for organizing an internal feedback loop for controlling a phase shift in systems of coherent addition of fiber laser beams (for simplicity, only three adjacent channels are shown), comprising a fiber narrow-band laser (1), a fiber splitter (2); phase shifting cells (3); fiber amplifiers (4); 3-aperture collimator (5) with a hole in the center; axial parabolic mirror (6); aperture (7), photodiode (8) and controller (9).

The implementation of the invention

The device operates as follows: a laser beam from a radiation source — a fiber narrow-band laser (1) passes through a fiber splitter (2), where radiation is divided into N channels. To adjust the phase, the radiation is directed to phase-shifting cells (3) and then amplified using fiber amplifiers (4) and removed from the optical fiber using densely packed N-collimators assembled into a single module (5). Each collimator consists of a single hexagonal aspherical lens. The collimators are installed in parallel and direct the collimated beams to the target. Part of the radiation (the peripheral part of the collimated Gaussian beam, containing about 4% of the power) is reflected from a pre-configured axial parabolic mirror (6) located behind the plane of the entrance pupil and whose axial point is located at the intersection of the medians of an equilateral triangle, the vertices of which lie on the axes of three neighboring collimating lenses, one of which is the central collimating lens for a given cluster. The peripheral beams reflected from the parabolic mirror pass through an aperture located at the intersection of the medians of an equilateral triangle, the vertices of which lie on the axes of three neighboring collimating lenses, one of which is the central collimating lens, interfere in the plane of the diaphragm (7) and the power transmitted through the diaphragm is detected by a photodiode ( 8). The signals from the photodiode are fed to the controller (9), which controls the state of the phase-shifting elements according to the method of stochastic gradient descent. Preliminary adjustment of interference fringes in the plane of the photodetector is carried out by shifting the parabolic mirror along its optical axis and controlling its tilts. The diameter of the parabolic mirror (Fig. 2) is calculated according to the formula [1] in order to exclude the vignetting of the main beam directed to the target having a light diameter equal to the light diameter D of _0.89 Gaussian beam.

where D _floor is the total light diameter of collimated radiation, D _floor = 2NA⋅ƒ ' _count. where NA is the numerical aperture of the optical fiber,

ƒ ' _count - focal length of a collimating lens; D _0.89 - the diameter of the light beam taking into account the coefficient of interception of the beam by a collimator lens γ = 1 / 0.89;

где ω₀ - половина диаметра моды оптоволокна;

where ω ₀ is half the diameter of the mode of the optical fiber;

The radius of curvature of a parabolic mirror is calculated from two conditions. First, from the condition of the reflected light from the parabolic mirror without vignetting of the beam through the aperture diameter d _{of holes} disposed between the three faces of adjacent hexagonal lenses (FIG. 2).

Based on this condition, the focus of the mirror should be no less than:

Secondly, in order to match the width of the interference strip with the size of the diaphragm according to the formula [3]:

where, Δx is the width of the interference strip, _Dser is the diameter of the mirror.

The above formulas for calculating the design parameters of a parabolic mirror (radius of curvature, diameter) make it possible to design a device circuit consisting of commercially available components with a certain detectable interference bandwidth, taking into account the selected wavelength and specific characteristics of the optical fiber used.

Claims (14)

1. A method of organizing an internal feedback loop for phase synchronization of a fiber laser array in coherent beam addition systems, in which the radiation of a narrow-band laser generating a coherent, linearly polarized Gaussian beam is divided into N channels containing an optical phase-shifting element in each channel that controls the phase of the optical waves depending on the magnitude of the applied control voltage, and an amplifier having a fiber output in free space, so that N fiber outputs They indicate an equidistant array of diverging beams with a Gaussian intensity distribution, each of which consists of a central maximum containing ~ 92% of power and the peripheral part, while small fractions of the peripheral parts of neighboring beams are intercepted by the optical system and assembled in the plane of the small aperture of the photodetector, characterized in that diverging beams collimate, forming a flat wave front in each channel with a Gaussian intensity distribution, after which small fractions of the peripheral parts of neighboring beams are intercepted by the optical system so that their central maxima do not vignet, and are focused in the plane of the small aperture of the photodetector, where they interfere, while the signal from the photodetector is fed to the control controller, which generates iterative stresses according to the stochastic parallel gradient descent method to control the corresponding phase-shifting cells, maintaining the maximum achievable value of the amplitude of the signal from the photodetector, which corresponds to phase howling laser beams. 2. A device for organizing an internal feedback loop for phase synchronization of a fiber laser array in coherent beam adding systems, including a narrow-band laser generating a coherent linearly polarized Gaussian beam, a fiber splitter dividing the radiation into N channels connected to N optical phase-shifting elements that control the phase of the optical waves depending on the magnitude of the applied control voltage, N fiber amplifiers having a fiber output in free space, t so that N fiber outputs form an equidistant array of diverging beams with a Gaussian intensity distribution, each of which consists of a central maximum containing ~ 92% of power and a peripheral part having N collimating lenses at the output, forming a synthesized aperture consisting of clusters, each of which contains a central collimating lens and 6 peripheral collimating lenses assembled in a hexagonal lattice that form a flat wave front in each channel and direct the radiation of all channels parallel to each other, characterized in that small fractions of the peripheral parts of the Gaussian beams are intercepted after collimation by parabolic axial mirrors located behind the plane of the entrance pupil of each three neighboring collimating lenses located at the vertices of the triangle, one of which is located on the optical axis of the central collimating lens, and the fractions of the peripheral parts of the radiation of neighboring Gaussian beams reflected from each parabolic mirror, passing through a hole in the center of the triple collim lasing lenses are focused in the plane of the small aperture of the photodetector, where interference occurs, and the signal from the photodetector is fed to the control controller, which generates iterative voltages according to the stochastic parallel gradient descent method to control the corresponding phase-shifting cells, maintaining the maximum achievable signal amplitude from the photodetector, which corresponds to phase synchronization of laser beams. 3. The device according to p. 2, characterized in that the collimating lenses are hexagonal in shape and are tightly packed. 4. The device according to claim 2, characterized in that the dimensions of the collimating lens of a hexagonal shape, intercepting ~ 96% of the power of a Gaussian beam, correspond to the diameter of the inscribed circle, calculated based on the numerical aperture of the optical fiber (NA) and the focal length of the collimating lens (ƒ'col) according to the formula: D _sex = 2NA⋅ƒ'col. . 5. The apparatus according to claim 2, characterized in that to avoid vignetting of the reflected radiation from the parabolic mirror of the light beam when passing through the hole diameter d _{of holes} disposed between the three faces of adjacent hexagonal lenses, aperture size is chosen from the condition: d _holes ≤0,1547D _floor. 6. The device according to p. 2, characterized in that the diameter of the parabolic mirror is calculated based on the numerical aperture of the optical fiber (NA), the diameter of the mode field of the fiber

and working wavelength (λ) according to the formula:

7. The device according to p. 2, characterized in that the radius of curvature of the parabolic mirror is selected on the basis of the formula:

8. The device according to p. 2, characterized in that the radius of curvature of the parabolic mirror in order to match the width of the interference strip with the diaphragm size d _TD to fulfill the condition d _TD ≤Δx must satisfy the ratio:

9. The device according to p. 2, characterized in that the adjustment of the interference band to the maximum intensity in the plane of the photodiode is carried out by shifting the parabolic mirror along its optical axis and controlling its tilts.

Images