Pump architecture for double-passed fiber laser

Abstract

To reduce nonlinear impairments without sacrificing pumping efficiency, increasing fiber temperatures, or suffering the performance and integration issues due to a PSC at the output.SOLUTION: A fiber laser amplifier system includes a first dual-clad delivery fiber receiving a signal beam and a pump beam, a doped amplifying fiber coupled to the first delivery fiber and receiving the signal beam and the pump beam and amplifying the signal beam using the pump beam, and a second dual-clad delivery fiber coupled to the amplifying fiber and receiving the amplified signal beam and the pump beam. The system also includes an endcap having an input facet and an output facet. The input facet is coupled to the second delivery fiber and receives the amplified signal beam and the pump beam. The output facet passes the amplified signal beam and reflects the pump beam back onto the second delivery fiber to be directed back to the doped amplifying fiber.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001]

[0001] This disclosure generally relates to fiber laser excitation architectures that reduce the effects of nonlinearity and thermal damage, and more particularly to fiber laser excitation architectures that include a dichroic endcap that double-passes the pump light to reduce the effects of nonlinearity and thermal damage. Prior Art

[0002]

[0002] High-power laser amplifiers have many applications including industrial, commercial, military, etc. Laser amplifier designers are constantly researching ways to increase the power of laser amplifiers for these and other applications. One known type of laser amplifier is the fiber laser amplifier. The fiber laser amplifier employs a doped fiber, such as a fiber doped with ytterbium (Yb) or a fiber doped with thulium (Tm), receives a seed beam and a pump beam that amplifies the seed beam, and generates a high-power laser beam. The fiber has an effective core diameter of about 10 to 20 μm or more.

[0003]

[0003] Directed energy (DE) systems, which emit high-energy light beams directly towards a target, are rapidly becoming a reality in real-world operational environments. Beam delivery of individual multi-kW class lasers and high-energy and peak-power pulsed irradiators to distant beam directors and combiners is a major driving force for DE systems because it is reliable, robust, and efficient. Fiber laser amplifiers have proven desirable as an energy source for DE systems due to their high efficiency, high power scalability, and outstanding beam quality. Fiber laser systems employ multiple fiber laser amplifiers, which combine amplified beams in some way to generate even higher power. The design challenge of this type of fiber laser amplification system is to combine beams from multiple fiber amplifiers so that the beam provides a single beam output with uniform phase across the entire beam diameter, and to focus the beam to a small focal spot, thereby defining the beam quality (far field) by focusing the combined beam to a small, far-field spot.

[0004]

[0004] In one known design of a multifiber laser amplifier called coherent beam combining (CBC), a master oscillator (MO) generates a seed beam, which is split into multiple split seed beams, each having a common wavelength, and each seed beam is amplified. The amplified seed beam is then guided to a diffractive optical element (DOE) or other optical system, which combines the multiple coherent amplified beams into a single output beam. The DOE has a periodic structure formed within the element, so that when the individual amplified beams, each having a slightly different angular direction, are redirected by the periodic structure, all the beams diffract from the DOE in the same direction.

[0005]

[0005] In another known multiple fiber laser amplifier design called spectral beam combining (SBC), multiple master oscillators (MOs) generate multiple seed beams at multiple different wavelengths and amplify each seed beam. The amplified seed beams are then emitted toward a diffraction grating or other wavelength-selective element, and the diffraction grating combines the fiber beams of different wavelengths into a single combined output beam. Because the diffraction grating has a periodic structure formed in a grid, when the individual amplified beams, each having a slightly different wavelength and angular direction, are redirected by the periodic structure, all of the beams diffract from the diffraction grating in the same direction.

[0006]

[0006] In a typical known fiber amplifier stage, the fiber gain medium is formed as a double-clad fiber, with typical cross-sectional diameters of approximately 20 μm for a Yb-doped single core, approximately 400 μm for the excitation cladding, and approximately 550 μm for the outer acrylate coating. The core numerical aperture (NA) is selected to ensure high beam quality for the signal light in a single transverse mode operation, and is typically NA ~ 0.06. To enable coupling of low-luminosity diode excitation light, the cladding NA is given as large as possible, but is typically NA ~ 0.46. The desired low-power seed beam light is injected into the fiber core, and at the same end of the fiber, the diode excitation light is simultaneously injected into the cladding. As the excitation light propagates downstream through the fiber, it is absorbed as it traverses the core, and the seed beam is amplified within the core to multi-kW levels at output. Typically, the fiber length is selected to be long enough to absorb more than 95% of the diode excitation light.

[0007]

[0007] In packaged narrow-linewidth, multi-kW fiber laser amplifiers, the excitation and seed beams are typically not coupled while passing through free space, but rather injected in the amplifier's multi-stage chain using specialized all-fiber components. In one specific design, the seed beam is often amplified to the 10W class in a multi-stage pre-amplifier before being injected into the core of a double-clad gain fiber. The diode excitation light is injected into the cladding of the gain fiber using a tapered fiber bundle pump-signal combiner (PSC). After amplification, the output end of the gain fiber is spliced ​​into a passive distribution fiber. The residual excitation light is stripped from the cladding using a cladding light stripper (CLS). A cladding light stripper removes the acrylate coating in a window-like manner, and the exposed cladding glass surface is roughened or brought into contact with an adhesive of matching refractive index to outcouple the residual excitation light. A wedge-shaped anti-reflection (AR) coated end cap is spliced ​​to the output end of the distribution fiber to extend the fiber modes, avoiding damage to the output facets and suppressing back reflection of the signal light. Typically, the amplifier operates at a saturation gain of over 20 dB, and a return loss of -60 dB from the end cap is required to avoid instability or undesirable power extraction in the return direction. [Overview of the Initiative] [Problems that the invention aims to solve]

[0008]

[0008] Power scaling of narrow-linewidth Yb-doped fiber amplifiers is currently limited by two separate nonlinear optical interferences: stimulated Brillouin scattering (SBS) and self-phase modulation (SPM), with SBS being the primary nonlinear interference. SBS is a nonlinear effect in which the laser field generates a phase grating due to electrostriction within the fiber core, reflecting some fraction of the beam propagating in the forward direction. If the effective reflectivity of the grating becomes too high, the output power from the fiber amplifier decreases, and the lost power is reflected in the reverse direction towards the lower-power components upstream, ultimately causing catastrophic damage. Since the threshold for SBS is inversely proportional to the spectral brightness (~power / linewidth), SBS limits the usable power from a single-frequency fiber laser to several hundred watts. To increase the threshold power of SBS, a common method is to broaden the fiber laser input seed spectrum to a multi-GHz domain using phase modulation, or equivalently, frequency modulation (FM). This shortens the optical coherence length and therefore reduces the SBS gain. As the power of the Yb-doped fiber amplifier increases to multi-kW levels, or as the distribution fiber length increases, a wider FM linewidth is required to suppress SBS. Typically, for Yb-doped fiber amplifiers, the linewidth increases almost linearly with power up to around ~10-20 GHz / kW. Reducing the SBS suppression linewidth, i.e., increasing the SBS suppression fiber spectral brightness, makes it possible to scale the beam-coupled fiber laser system to even higher power levels.

[0009]

[0009] Optical interference SPM is parameterized by B-integrals, i.e., nonlinear phase shifts, and can degrade beam coherence by converting low levels of uncontrolled amplitude modulation (AM) into phase noise. This nonlinear effect can suppress the efficiency of the CBC, i.e., the beam quality of the SBC, and therefore degrade the performance of the fiber laser system. Specifically, this can result in loss of spectral brightness or loss of optical coherence. To avoid or reduce these effects, it is generally desirable to limit the amount of AM, also known as relative intensity noise (RIN), that propagates within the seed beam seeding the fiber amplifier. Techniques can be implemented in the fiber amplifier to broaden the spectrum of the seed beam to provide frequency modulation without amplitude modulation, and if the seed beam is only frequency modulated, the Kerr nonlinearity that drives SPM does not cause problems. That is, it does not produce time-dependent nonlinear phase shifts in the seed beam. However, if AM is superimposed onto the seed beam by FM / AM conversion, whether intentionally or unintentionally, SPM can cause nonlinear spectral expansion of the beam emitted from the fiber amplifier, potentially degrading beam quality during SBC.

[0010]

[0010] These optical interferences typically limit the spectral brightness of the output beam, i.e., the power per unit optical linewidth, or kW / GHz. These optical interferences become more severe as fiber power increases and fiber length increases. As fiber power increases to multi-kW levels or as distribution fiber length increases, increasingly wider FM linewidths are required to suppress SBS. In co-pumped fiber amplifiers, the magnitude of the interference typically increases as the integral of the signal power over the entire fiber length (colloquially known as the product of effective power and length). Therefore, one path to tuning a fiber laser to increase power while maintaining a narrow spectral linewidth is to reduce nonlinear optical interferences by decreasing the effective length of the fiber.

[0011]

[0011] It is well known that shorter fiber lengths are advantageous for reducing nonlinear optical interference and enabling adjustment of spectral brightness, but other design considerations also impose a limit on the minimum practical fiber length in amplifiers. For example, one design constraint is that most of the excitation light must be absorbed in order to ensure high optical-to-optical conversion efficiency. Yb-doped fibers suitable for high-power amplifiers typically have cladding excitation absorption efficiencies of ~1 to 1.5 dB / m. This means that a gain fiber length of about 10 m is required to absorb >95% of the total excitation light, or 13 dB. Another design constraint is that the generation of waste heat per unit fiber length must be kept to a minimum. If the length of a fiber amplifier emitting a fixed power level while maintaining the same total excitation absorption is cut in half (for example, by changing the excitation wavelength to improve its effective absorption characteristics), the waste heat per unit length doubles, raising the fiber temperature. If the fiber overheats, its acrylate coating may burn, leading to a catastrophic failure. Waste heat is particularly limiting for Tm-doped fibers emitting in the 2 μm band. This is because their excitation absorption efficiency (~6 dB / m) is typically 4 to 5 times higher than that of Yb-doped fibers emitting in 1 μm with a similar shape, and their quantum defects (waste heat fraction, ~35%) are 3 times higher. Therefore, the length of the co-excitation fiber amplifier is determined by a compromise that balances the competitive design requirement of suppressing SBS / SPM (shorter fibers are better) with maintaining high absorption efficiency and low temperatures (longer fibers are better). Further expansion of this tradeoff space is needed for fiber amplifier excitation architectures to adjust to even higher spectral brightness.

[0012]

[0012] One well-known fiber amplifier architecture that enables higher performance by reducing SBS and SPM switches the direction of the excitation light from copropagation with the seed beam to backpropagation with the seed beam. The excitation signal combiner is often located at the high-power output end of the amplifier, and the excitation cladding light is emitted backward in the opposite direction to the seed beam. This architecture has the advantage of redistributing the laser gain in the amplifier toward the output end, thereby reducing the effective power-length product, i.e., the integral of power over the entire fiber length. However, the back-excitation architecture also has a variety of well-known drawbacks, which have hindered its widespread adoption in fiber laser amplifiers commonly used for beam coupling.

[0013]

[0013] As a result of reverse excitation, a very high peak thermal load is generated near the output end of the fiber, causing both the excitation and seed beam light to reach their maximum intensity, and thus the laser extraction to become significantly saturated, potentially causing thermal damage to the fiber. A PSC at output imposes extra losses on the high-power seed beam compared to co-excitation. This is because additional splicing is required and due to the insertion loss of the tapered fiber bundle (TFB) combiner itself. Splicing between the gain fiber and the PSC is difficult because its performance must be optimized simultaneously for both low excitation loss (usually requiring "hot" splicing to completely melt the outer glass cladding for a smooth transition) and low signal loss (usually requiring "cold" splicing to prevent material diffusion outside the core). As a result, losses are generally greater than when the excitation and seed beam splicing can be optimized separately. PSCs must be able to handle and properly sink uncontrolled signal power loss from spliced ​​and internal TFB structures. This is particularly difficult with Tm-doped fibers, where scattered 2μm seed beam light can be absorbed by most of the fiber acrylate coating, potentially burning them out. PSCs impose a considerable (typically 0.5–1m) extra fiber length at the amplifier output, which can partially offset the nonlinear length reduction inherently brought about by inverse excitation.

[0014]

[0014] Both co-excitation and reverse excitation can be performed simultaneously by bidirectional excitation or bi-pumping. This bi-pumping method does not yield as much reduction in the effective power-length product as reverse excitation alone, but it is still an improvement over co-excitation. By splitting the excitation power between the two ends of the fiber, the thermal load can be divided more uniformly than with co-excitation or reverse excitation alone. However, even bi-pumping presents challenges in integration and power handling when the amplifier has a PSC at its high-power output end. Therefore, there is a need for an improved fiber excitation architecture that does not sacrifice excitation efficiency, does not increase fiber temperature, does not suffer from performance and integration problems due to PSC at the output, and reduces nonlinear interference. [Brief explanation of the drawing]

[0015] [Figure 1] This is a schematic block diagram of a fiber laser amplification system including a double-pass excitation architecture with dichroic end caps. [Figure 2] Figure 1 shows the gain fiber in the fiber laser amplification system. [Figure 3] This is an equispaced diagram of a dichroic endcap showing the excitation beam light. [Figure 4] This is a constant-width diagram of a dichroic end cap showing the signal beam light. [Figure 5] This is a schematic block diagram of an SBC fiber laser amplification system including the back path excitation architecture. [Figure 6] This is a schematic block diagram of a CBC fiber laser amplification system including the back path excitation architecture. [Modes for carrying out the invention]

[0016]

[0021] The following discussion of embodiments of the present disclosure relates to fiber laser excitation architectures that include dichroic end caps that double-pass the excitation light to reduce the effects of nonlinear and thermal disturbances, but this is essentially illustrative and is not intended to limit the present disclosure or its applications or uses.

[0017]

[0022] Figure 1 is a simplified block diagram of a fiber laser amplification system 10. The fiber laser amplification system 10 includes a single amplification channel 12 having a seed or signal beam source 14, the beam source 14 generating a low-power signal beam with a center wavelength λ on a fiber 16. The beam source 14 may include a main oscillator (MO), such as a single-longitudinal mode distributed feedback (DFB) diode laser oscillator, and a frequency modulator, such as an electric photon modulator (EOM). The EOM can receive an applied voltage supplied by an amplified radio frequency (RF) electrically driven signal, such as white noise or a pseudo-random bit sequence (PRBS), from an RF source (not shown) for frequency modulation broadening, so that the modulated signal beam has a substantially broadened linewidth, thereby suppressing stimulated Brillouin scattering in a downstream high-power fiber amplifier. A low-power pre-amplifier 22 receives the broadened signal beam. The preamplifier 22 can be a single fiber amplifier or a series chain of fiber amplifiers, and boosts the beam power to a level suitable for seeding a high-power fiber amplifier (typically about 10W). Optical isolators 20 and 24 on each side of the preamplifier 22 allow the signal beam to pass through but prevent reflected amplified light from returning to the beam source 14. Failure to do so could result in damage.

[0018]

[0023] The signal beam, along with multiple excitation beams from the excitation diode 28, is coupled in an excitation-signal combiner 30, such as a suitable tapered fiber bundle, and sent to a dual-clad distribution fiber 32, so that the excitation light propagates through the fiber cladding and the signal light propagates through the fiber core. The dual-clad distribution fiber 32 is spliced ​​into a doped gain fiber 34, which uses the excitation beam to amplify the signal beam in a co-excitation manner. Figure 2 shows a diagram of the gain fiber 34, which has a 20 μm core 52, a 400 μm internal cladding layer 54, and a polymer coating 56. The distribution fiber 32 has the same structural elements as the gain fiber 34, except that the core of the dual-clad distribution fiber 32 is not doped with laser gain material.

[0019]

[0024] The gain fiber 34 is spliced ​​into a dual-clad distribution fiber 36, which has the same structural elements as the gain fiber 34, except that the core of the distribution fiber 36 is not doped with laser gain material. The distribution fiber 36 is coupled to a dichroic end cap 40 that addresses the expansion of the amplified signal beam, thereby reducing the optical power density when the signal beam reaches the air interface; otherwise, the distribution fiber 36 may be damaged. As will be discussed in detail below, the output end of the end cap 40 is covered with a dichroic and anti-reflective coating or layer 42, which reflects the excitation beam wavelength and allows the signal beam wavelength to pass through. The dichroic layer 42 is similar to a known AR coating. An AR coating is typically a laminate of thin dielectric layers with optical properties desired to prevent as much of the signal beam as possible from being reflected back toward the end cap 40, but the material and thickness of layer 42 are designed to also reflect the excitation beam wavelength. The reflected excitation beam is emitted in the reverse direction toward the cladding of the dual-clad distribution fiber 36, so that the reflected excitation beam provides additional signal beam amplification in the gain fiber 34 in a backpropagation manner. This effectively doubles the fiber absorption length of the gain fiber 34, enabling a reduction in peak fiber thermal load equivalent to the benefit of bidirectional excitation, without adding complexity, impacting performance, or incurring the development cost of a reverse excitation coupler. In addition, by shifting the laser gain and signal power toward the output end of the gain fiber 34, the effective nonlinear interaction length of the dual path is reduced to only 70% of that of a co-excitation fiber of equivalent length. Combined with the halving of the thermal load, the final effect of implementing dual-path excitation to reduce nonlinear interference is three times greater compared to co-excitation.

[0020]

[0025] Figure 3 is an equiwidth diagram showing the end cap 70 through which the propagating excitation light 72 passes, and Figure 4 is an equiwidth diagram showing the end cap 70 through which the propagating signal light 74 passes. The excitation light 72 is a lower quality beam than the signal light 74 and has a larger beam-parameter product, so the excitation light 72 diverges more than the signal light. The end cap 70 is shown as a non-limiting example of an end cap that can be used as the end cap 40 in system 10, and other configurations of the end cap 40 can be prepared according to the discussion herein. The end cap 70 includes a glass body 76 having a tapered section 78, an entrance facet 80, a straight section 82, and a semicircular curved exit facet 84. As will be discussed in detail below, a short coreless fiber 86, for example a few millimeters long, is optionally connected to one end of the entrance facet 80, and a distribution fiber 36 is spliced ​​to the other end of the coreless fiber 86. The coreless fiber 86 has the same structural elements as the gain fiber 34, except that it does not have a core. The tapered portion 78 reduces the amount of glass that must be heated when the coreless fiber 86 is optically welded to the incident facet 80, so that the fiber 86 and the incident facet 80 melt simultaneously. The dichroic coating 88 is deposited on the extrusion facet 84 so that a large proportion of the excitation light 72 is reflected in the opposite direction toward the fiber 86 and enters the distribution fiber 36, and then enters the gain fiber 34, and a large proportion of the signal light 74 passes through the extrusion facet 84. In other words, the coating 88 acts as an AR coating with respect to the signal light 74.

[0021]

[0026] When using the coreless fiber 86, its length is not important. However, its maximum length is limited by the divergence of the signal light 74 at the incident facet 80 if the length is too long. Generally, to avoid significant clipping, the length of the coreless fiber 86 should be less than nD / 4*NA. Here, n is the refractive index of the fiber glass, D is the diameter of the coreless fiber 86, and NA is the divergence half-angle of the signal fiber mode. For a fiber core diameter of 20μm and a wavelength of 1μm, a typical NA is 0.035. In the assembly process, first, a long coreless fiber 86 is spliced to the cored distribution fiber 36, and then it is cleaved to the desired length from the splice joint. Then, the assembled body with the terminated coreless fiber is welded to the end cap 70. Since the cladding layer 54 of the dual-clad gain fiber 34 and the pump light 72 traveling in the distribution fiber 36 have a typical NA of ~0.46, the pump light 72 begins to diverge at the welded joint between the distribution fiber 36 and the coreless fiber 86. Therefore, for an efficient optical path recovery scheme, the pump light 72 must be re-imaged onto the fiber cladding layer 54. This is achieved by configuring a spherical injection facet 84 on the dichroic end cap 70. Since the center of curvature of the facet 84 is located at the welded joint between the coreless fiber and the end cap, the remaining cladding pump light 72 is retro-reflected and passes through both the distribution fiber 36 and the gain fiber 34 twice.

[0022]

[0027] In a perfectly matched system with an excitation NA of 0.46, a fiber cladding diameter of 400 μm, and an end cap length of 8 mm, and a refractive index n of 1.45 (fused silica), approximately 99.3% of the reflected excitation light 72 is reproduced within the cladding layer 54. When coupled with a coating with a reflectivity of 99%, a final net return coupling efficiency of 98% is achievable. The critical tolerance for high excitation coupling efficiency includes ensuring minimal deformation of the welded joint surface of the coreless fiber, and further ensuring that the tip of the coreless fiber is welded as close as possible to the center of curvature of the injection facet 88 in the lateral direction. For the end cap 70, ray tracing shows that the lateral tolerance required to keep the reflection loss of excitation light 72 within 1% of the optimal value is ±5 μm in the specific example discussed herein. Also, the welded joint relative to the thickness of the injection facet on the axis must be exactly one radius of curvature of the injection facet 84. Ray tracing demonstrates that the thickness tolerance is approximately ±20 μm for a 1% bond loss. This tolerance can be achieved by manufacturing the end caps and shortening the stem length by polishing to match the radius of curvature.

[0023]

[0028] Also, the end cap 70 is configured to control unwanted back reflection of the signal light 74. For a high-gain (>20 dB) amplifier, suppressing the reflection loss of the signal light 74 to about -60 dB is important for preventing instability. When the distribution fiber 36 is directly and perfectly aligned and spliced to the end cap 70, almost 100% of the reflected signal light 74 is reproduced within the core 52. With only the coating 88, only about -30 dB of suppression can be obtained. Therefore, such a configuration would be useless. However, splicing within the coreless fiber 86 can prevent the reproduction of the signal light 74 into the core 52. After the splice joint between the coreless fiber 86 and the distribution fiber 36, the signal light 74 is no longer induced into the core 52 and freely expands along the length of the coreless fiber 86. Thus, as shown by the signal light 74, the reflection image from the facet 84 of the core 52 is at a point 90 within the body 76 of the end cap 70, several millimeters away from the splice joint between the distribution fiber 36 and the coreless fiber 86, and thus reduces the amount of reflected signal light 74 incident on the core 52. A simple calculation shows that this results in a -30 dB form of geometric coupling loss to the core 52, which combined with the -30 dB loss from the coating results in a final reflection loss of -60 dB, meeting the desired requirement.

[0024]

[0029] Alternatively, a similar morphological reflection loss coefficient of -30 dB can be achieved by directly overlapping the distribution fiber 36 to the end cap 70 without using a coreless fiber, resulting in a lateral shift of ~10 to 20 μm. Therefore, the reflected signal light 74 is reproduced on the cladding layer 54 at a point offset from the core 52 with a diameter of ~20 μm. In this case, the reflected image of the core 52 is formed directly at the splice junction located at the incident facet 80, but because it is laterally shifted by twice the eccentricity of the core, it couples to the cladding layer 54 rather than the core 52. In this case, the excitation coupling efficiency drops by several percent because it is also reproduced with a lateral shift. Therefore, although this configuration is feasible, it may be less advantageous than using a coreless fiber 86.

[0025]

[0030] Typically, applying a fiber amplifier requires low cladding light (usually <1% of the total power). In a hypothetical double-path excited fiber amplifier with a total excitation absorption (13dB) of 95%, the excitation power emitted at the seed end is attenuated by only 6.5dB at the end cap 70. The dichroic coating 88 provides an additional attenuation of -20dB, so the pump power fraction is only -26.5dB lower than the signal for the boundary case of a 100% quantum efficiency laser. This is about 0.2% of the output signal light. If this power level becomes a problem (which is usually not the case), the power level can be further reduced by trading the coating performance between AR and HR, and further reduced by trading back a certain length from the extended gain fiber.

[0026]

[0031] The dimensions of the end cap 70 correspond to (support) the ~4 mm spacing or pitch between fibers in the array. A smaller pitch is considered more valuable for the SBC. In this case, the 4 mm lateral dimension of the exit facet 84 needs to be shortened. To do this, the thickness of the end cap needs to be proportionally reduced to prevent clipping of the excitation light 72, but this increases the peak intensity of the signal light 74 on the exit facet 84 to a level that could potentially cause damage. However, by extending the length of the coreless fiber 86, the signal intensity can be reduced and additional beam spreading of the signal light 74 on the end cap 70 can be provided without affecting the pump spreading. As previously explained, the length of the coreless fiber 86 is limited by signal power clipping on the outer diameter (OD) of the coreless fiber 86 at the welded joint of the end cap.

[0027]

[0032] As a specific example, we will consider the structure (design) of an end cap based on end cap 70, but with half the pitch (2 mm instead of 4 mm). In a typical LMA Yb-doped fiber with a core diameter of 20 μm, the signal NA is 0.035 in air, corresponding to a half-angle divergence of 25 mrad in SiO2. If system 10 is designed to clip on a 400 μm OD coreless fiber at a beam radius of 4, the length of the coreless fiber 86 can be 400 μm / (4 * 25 mrad) = 4 mm without causing significant clipping. Also, the thickness of the end cap 70 will be 4 mm, compared to a total signal beam expansion length of 8 mm.

[0028]

[0033] Figure 5 is a simplified block diagram of the SBC fiber laser amplification system 96. The system 96 includes multiple amplification channels 98, each amplification channel 98 having an MO 100 that generates a signal beam, and the MO 100 in different channels 98 generates signal beams at different wavelengths. The signal beam is sent to an EOM 102, which receives an applied voltage supplied by an RF driver 104 to correspond to frequency modulation broadening. The signal beam is then amplified by a fiber amplifier 106, and the amplified signal beam is sent to a beam launcher 108 having end caps 110 of the type discussed earlier. The end caps 110 are configured to accommodate fibers in all of the channels 98, and many of the end caps 70 may be configured to be joined together. Alternatively, the flat input surface of a monolithic microlens array may be joined to the flat output surface of a multi-fiber end cap array to form a multi-faceted end cap array suitable for welding multiple fibers. The amplified beam is then sent through free space to the SBC-coupled optical element 112. Since the SBC optical element 112 includes a grating (not shown) with a periodic structure formed on the grating, when the individual amplified beams, each having slightly different wavelengths and angular directions, are redirected by the periodic structure, all of the beams are diffracted from different gratings in the same direction to form a combined output beam.

[0029]

[0034] Figure 6 is a simplified block diagram of the CBC fiber laser amplification system 116, where elements similar to those in system 96 are identified by the same reference numbers. System 116 includes a single MO 100 that generates a signal beam. The signal beam is split into multiple signal beams by a beam splitter 118, and these multiple signal beams are amplified by an amplifier 106. The amplified beam from the beam launcher 108 is sent to a CBC optical element 122, which combines all of the amplified beams into a composite output beam.

[0030]

[0035] The foregoing discussion merely discloses and describes exemplary embodiments of the present disclosure. From this discussion, and from the accompanying drawings and claims, it will be readily apparent to those skilled in the art that various changes, modifications, and variations are possible without departing from the spirit and scope of the present disclosure as set forth in the following claims.

Claims

1. 1. A fiber laser amplification system comprising: at least one signal beam source for generating a signal beam; at least one excitation beam source that generates an excitation beam; a beam combiner for combining the signal beam and the excitation beam; a first dual-clad delivery fiber coupled to the beam combiner to receive the combined pump and signal beams; a doped amplifying fiber coupled to the first delivery fiber to receive the combined pump and signal beams, the doped amplifying fiber using the pump beam to amplify the signal beam; a second dual-clad delivery fiber coupled to the doped amplifier fiber for receiving the amplified signal beam and the pump beam; an end cap including an input facet and an output facet, the input facet being coupled to the second delivery fiber and configured to receive the amplified signal beam and the pump beam, and the output facet being configured to pass the amplified signal beam, reflect the pump beam towards the second delivery fiber, and launch the pump beam back towards the doped amplifying fiber; A fiber laser amplification system comprising:

2. 10. The system of claim 1, wherein the output facet includes a dichroic coating that is anti-reflective at the wavelength of the signal beam and highly reflective at the wavelength of the pump beam.

3. 10. The system of claim 1, wherein the output facet is spherically curved.

4. The system of claim 1 , wherein the second delivery fiber is directly welded to the input facet.

5. 10. The system of claim 1, further comprising a coreless fiber coupled between the second delivery fiber and the input facet.

6. 10. The system of claim 1, wherein the end cap is configured such that the pump beam reflected by the output facet is replicated on a cladding layer of the second delivery fiber.

7. 10. The system of claim 1, wherein the end cap is configured such that any signal light reflected by the output facet is reproduced on a cladding layer offset from the core of the second delivery fiber.

8. 2. The system of claim 1, wherein the end cap includes a tapered section adjacent the input facet and a straight section adjacent the output facet.

9. 10. The system of claim 1, wherein the end cap is approximately 8 mm long and 4 mm wide.

10. 10. The system of claim 1, wherein the beam combiner is a tapered fiber bundle.

11. The system of claim 1 , wherein the at least one excitation source is a plurality of excitation sources.

12. 10. The system of claim 1, wherein the fiber laser amplification system is a coherent beam combining (CBC) fiber laser amplification system, and the at least one pump beam source, the beam combiner, the first delivery fiber, the doped amplifier fiber, and the second delivery fiber are part of one fiber channel among a plurality of fiber channels.

13. 10. The system of claim 1, wherein the fiber laser amplification system is a spectral beam combining (SBC) fiber laser amplification system, and the at least one pump beam source, the beam combiner, the first delivery fiber, the doped amplifier fiber, and the second delivery fiber are part of one fiber channel among a plurality of fiber channels.

14. 1. A fiber laser amplification system comprising: at least one signal beam source for generating a signal beam; at least one excitation beam source that generates an excitation beam; a beam combiner for combining the signal beam and the excitation beam; a first dual-clad delivery fiber coupled to the beam combiner to receive the combined pump and signal beams; a doped amplifying fiber coupled to the first delivery fiber to receive the combined pump and signal beams, the doped amplifying fiber using the pump beam to amplify the signal beam; a second dual-clad delivery fiber coupled to the doped amplifier fiber for receiving the amplified signal beam and the pump beam; a coreless fiber coupled to the second delivery fiber for receiving the amplified signal beam and the pump beam; an end cap including an input facet and an output facet, the input facet being coupled to the coreless fiber and receiving the amplified signal beam and the pump beam, the output facet being spherically curved and including a dichroic coating that passes the amplified signal beam, reflects the pump beam toward the coreless fiber, and ejects the pump beam back toward the second delivery fiber and the doped amplifying fiber, the reflected pump beam being replicated on a cladding layer of the coreless fiber, and any signal light reflected from the output facet being replicated at the end cap; A fiber laser amplification system comprising:

15. 15. The system of claim 14, wherein the end cap includes a tapered section proximate the input facet and a straight section proximate the output facet.

16. 15. The system of claim 14, wherein the end cap is approximately 8 mm long and 4 mm wide.

17. 15. The system of claim 14, wherein the fiber laser amplification system is a coherent beam combining (CBC) fiber laser amplification system, and the at least one pump beam source, the beam combiner, the first delivery fiber, the doped amplifier fiber, and the second delivery fiber are part of one fiber channel of a plurality of fiber channels.

18. 15. The system of claim 14, wherein the fiber laser amplification system is a spectral beam combining (SBC) fiber laser amplification system, and the at least one pump beam source, the beam combiner, the first delivery fiber, the doped amplifier fiber, and the second delivery fiber are part of one fiber channel of a plurality of fiber channels.

19. 1. An optical end cap comprising an input facet and an output facet, the input facet being coupled to a dual-clad delivery fiber and configured to receive an amplified signal beam and a pump beam, and the output facet being configured to pass the amplified signal beam, reflect the pump beam towards the delivery fiber, and launch the pump beam back towards a doped amplifying fiber.

20. 20. The end cap of claim 19, wherein the output facet is spherically curved and includes a dichroic coating that is anti-reflective at the wavelength of the signal beam and highly reflective at the wavelength of the pump beam.

Images