http://www.k-wave.org/forum/topic/convergence-in-non-homegenous-media Support for the k-Wave MATLAB toolbox en-US Tue, 12 May 2026 22:34:47 +0000 http://bbpress.org/?v=1.0.2 q http://www.k-wave.org/forum/search.php http://www.k-wave.org/forum/topic/convergence-in-non-homegenous-media#post-8372 Thu, 16 Dec 2021 10:57:53 +0000 Bradley Treeby 8372@http://www.k-wave.org/forum/ <p>Hi Frank,</p> <p>Have you tried increasing the simulation resolution further? For such an extreme impedance contrast, 16 PPW may not be enough. If this is too computationally demanding, you could make a smaller simulation domain (or set up something in 1D) to see how many PPW you will need to reach convergence.</p> <p>Brad. </p> http://www.k-wave.org/forum/topic/convergence-in-non-homegenous-media#post-8332 Fri, 22 Oct 2021 15:24:59 +0000 s148275 8332@http://www.k-wave.org/forum/ <p>Hello,</p> <p>I am trying to simulate the wave propagation of a focused (continuous) wave through a layered medium. The media are: water (35 mm), oil (25 mm), polymer (2 mm) and steel (5 mm), respectively. A linear array (fc=2.72MHz) is placed at the beginning of the water and focused on the polymer by using delays. Attenuation is not included.</p> <p>The system works quite well if I do not include the steel after the polymer layer. By reducing cfl or increasing ppw, the maximum pressure amplitude in the powder is converging. However, when using the steel layer (c0=5000 and rho0=4000) the simulation is not converging for increasing values of ppw (see <a href="https://ibb.co/Zdh7gvS" rel="nofollow">https://ibb.co/Zdh7gvS</a> for image). It seems that the convergence get worse if the grid size is decreased, which doesn't make sense?</p> <p>Since the impedance mismatch is quite large between the polymer and steel, it is almost a perfectly reflecting surface. Therefore, I tried replacing the steel with a dirichlet BC with particle velocity is zero. Unfortunately, again no convergence with ppw was found (see <a href="https://ibb.co/TqnR6HX" rel="nofollow">https://ibb.co/TqnR6HX</a> for image). Increasing the PML size did also not solve this problem. However, it does have an effect on the maximum pressure in the powder as shown in this image: <a href="https://ibb.co/T4LTfJs" rel="nofollow">https://ibb.co/T4LTfJs</a>.</p> <p>Any ideas why all of this is happening and how to solve this?</p> <p>Best,<br /> Frank </p>