In this research two acoustic analogies, the linearized Euler equations and Lighthill’s equation are evaluated. These equations describe the propagation of acoustic waves. Both the linearized Euler equations and Lighthill’s equation use a right hand side source term that is determined by a fluid solver. When coupling the fluid solver with an acoustic solver for a low Mach number aeroacoustic problem, a difference in spatial and temporal scale can exist for the fluid and acoustic calculations. This research is based on aeroacoustic solvers in which the spatial grid for the acoustics and the fluid is exactly the same. Due to this condition it is shown that the ratio between the fluid and acoustic time step is governed by the inverse of the Mach number to ensure that both the Courant and acoustic Courant numbers are smaller than one. In the case of low Mach numbers (M < 0.1) this results in a time step difference between the acoustic and fluid simulation of at least a factor 10. Due to the time step difference, the source term in the fluid solver is not updated for every acoustic time step and hence the accuracy of the acoustic result is reduced. To remedy this decreased accuracy one could solve for the source term at every acoustic time step. However this leads to lengthy computations. The objective of this research is thus to improve the efficiency of the fluid acoustic coupling by increasing the acoustic accuracy for non-matching time steps in the fluid solver and in the acoustic solver. This research objective can be fulfilled by improving the source term transfer from the acoustic solver to the fluid solver by higher order temporal interpolation between different source levels. Furthermore the source term can be reconstructed at the nodes by implementation of a higher order time integration method in the fluid solver. In this research is investigated whether the improved temporal transfer and reconstruction of the source term can lead to an increased time step difference between the fluid and acoustic calculations for which an acceptable acoustic accuracy can be obtained. The approach taken in this research is based on firstly quantifying the effect of keeping an analytic monopole source term constant for multiple acoustic time steps. To adequately describe how the source term is updated, a source term Courant number is derived. This number is defined as the time interval between source term updates multiplied by the summation of the convective velocity and the speed of sound and divided by the spatial acoustic step. Then the source term levels are interpolated with Lagrange polynomials, to assess the improvement of the acoustic results. Furthermore this known analytic monopole source term is reconstructed by solving an ordinary differential equation that has the monopole as its exact solution. Both the spectral deferred correction method and the backward Euler method are used to solve this equation. The spectral deferred correction method is a higher order method that uses iterative low order substepping over Gauss Lobatto quadrature nodes. Using Gauss Lobatto quadrature allows for exact integration of the integral in the Picard formulation. In the last part of this research the spectral deferred correction method is implemented in an incompressible, transient fluid solver to construct higher order source terms. Lagrange interpolation is used to interpolate the source terms at the Gauss Lobatto node locations, to improve the source term transfer to the acoustic solver. The results from this research show that a large distortion of the acoustic results take place when the source term Courant number exceeds one. For the source term transfer to the acoustic solver, interpolation with Lagrange basis functions over Gauss Lobatto nodes delivered the most accurate acoustic results. In terms of higher order time integration, the reconstruction of the source term by solving an ordinary differential equation with the spectral deferred correction method showed the highest accuracy. The root-mean-square error of the acoustic result was decreased compared to a backward Euler method. The relative effect on the improvement of the acoustic accuracy proved to be highest by Lagrange interpolation over the Gauss Lobatto nodes compared to accuracy improvements at the nodes by the spectral deferred correction method. Finally based on the result from the implementation of the spectral deferred correction method in the incompressible, transient fluid solver, it can be noted that the spectral deferred correction method did not show the same error convergence as was found in the solution of an ordinary differential equation for the reconstruction of the analytic monopole. The Lagrange interpolation over the Gauss Lobatto node locations improved the overall acoustic result in the active source region of the flow. In the far field the acoustic results were similar for both zeroth order interpolation (which resembles the original piecewise constant source term) and Lagrange interpolation. ... Read more