UNCLASSIFIED, PUBLIC RELEASE Temperature Eigenfunctions for Accelerated Time-domain Transverse Mode Instability Simulation Generally, TMI does not affect low-power systems, meaning it is rarely observed in systems operating below several hundred watts. This makes it difficult to build and test a wide range of fiber lasers using different TMI mitigation designs. It would be helpful to have accurate and efficient simulations methods for rapidly testing fiber design parameters and TMI mitigation techniques. TMI simulation models are generally quite computationally intensive, so we have developed a new time-domain algorithm that calculates temperature as a superposition of the fiber’s thermal eigenmodes [1]. It accelerates computation of heat diffusion in the fiber and thermo-optic mode coupling. Over a range of simulation parameters, this model can run on average ~13.9 times faster than an older model using the standard finite-difference time-domain methods for heat diffusion. These modelling techniques have been used to simulate several TMI mitigation strategies. The most effective mitigation technique we’ve found uses intermodal phase modulation between the FM and the HOM. Using a specially-designed fiber with different strain-optic coefficients in the core and cladding, lateral compressive strain can cause a greater phase delay in the HOM, which extends out into the cladding, than in the FM, which is more tightly confined to the core. This modulates the intermodal phase, shifting the mode interference pattern (MIP). Power coupling in TMI is mediated by a thermo-optic long-period fiber grating (LPFG), which is generated from the beating of the MIP. Periodically modulating the intermodal phase along the full fiber length breaks the phase matching between MIP and LPFG, preventing power transfer from the FM to the HOM. Simulations show that this technique can increase the TMI threshold by a factor of ~4.2. References [1] J. Hunt and J. Talghader, "Temperature eigenfunction basis for accelerated transverse mode instability simulation," Opt. Express 32, 11979-11991 (2024).
UNCLASSIFIED, PUBLIC RELEASE
|