Same transitional flow model runs fine in COMSOL 6.1 but crashes in COMSOL 6.3

Please login with a confirmed email address before reporting spam

Hi everyone, I’m running into a version-related issue with COMSOL. A transitional flow model that simulates successfully in COMSOL 6.1 crashes when simulated in COMSOL 6.3. The geometry, mesh, physics settings, and solver configuration are essentially the same; the main difference is the software version. The computer configuration is as follows: CPU: Intel(R) Xeon(R) Gold 6246R CPU @3.4GHz GPU: NVIDIA RTX A6000


1 Reply Last Post 8 oct. 2026, 19:48 UTC−4
Robert Koslover RF/Antennas

Please login with a confirmed email address before reporting spam

Posted: 2 hours ago 8 oct. 2026, 19:48 UTC−4

There are often subtle differences between how problem solutions converge in different versions of Comsol Multiphysics. If your model in version 6.1 was on the edge of failing to converge or was close to a (for example) divide by ~zero error, it might fail in version 6.3. And likewise, one that failed in 6.1 might work in 6.3. I suggest that you post your .mph file to the forum. Among other things, take a look at your mesh quality and your solver settings. Be sure that these make good sense to you for your particular physics. If I recall correctly, poor mesh quality often leads to solver-related issues which may be unpredictable and vary between software versions. But there are many other possibilities too, such as subtle differences in boundary conditions, discretization, and just about any automatically-set/default values that differ between version 6.1 and 6.3. Most of the time, you are better off using the latest version of Comsol Multiphysics, but sometimes older models really do require a few tweaks to execute well in a new version.

-------------------
Robert A. Koslover
There are often subtle differences between how problem solutions converge in different versions of Comsol Multiphysics. If your model in version 6.1 was on the edge of failing to converge or was close to a (for example) divide by ~zero error, it might fail in version 6.3. And likewise, one that failed in 6.1 might work in 6.3. I suggest that you post your .mph file to the forum. Among other things, take a look at your mesh quality and your solver settings. Be sure that these make good sense to you for your particular physics. If I recall correctly, poor mesh quality often leads to solver-related issues which may be unpredictable and vary between software versions. But there are many other possibilities too, such as subtle differences in boundary conditions, discretization, and just about any automatically-set/default values that differ between version 6.1 and 6.3. Most of the time, you are better off using the latest version of Comsol Multiphysics, but sometimes older models really do require a few tweaks to execute well in a new version.

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.