CFL Error Using Multi-Scale Kain-Fritsch (MSKF) convection scheme WRF version 4.2

[jahokson]

Hi! I've been getting CFL error with my settings in WRF v4.2.

I adjusted some values (except epssm), as advised by this: https://forum.mmm.ucar.edu/phpBB3/viewtopic.php?t=9429, but the error still persists.
(When I edit epssm to 0.50, the run just stops without any error message other than showing the processes that got stopped. So, I remove the edit for that).

The problem only occurs in MSKF, and not other cumulus schemes like Kain-Fritsch Scheme (no.1).

Attached are the files for reference.

The error message is:
WRF NUMBER OF TILES = 1
d02 2015-10-15_00:00:48+**/** 148 points exceeded cfl=2 in domain d02 at time 2015-10-15_00:00:48+**/** hours
d02 2015-10-15_00:00:48+**/** MAX AT i,j,k: 75 56 13 vert_cfl,w,d(eta)= 21131.97 1.6148515E+07 4.5530438E-02

Thank you.

 

[Ming Chen]

Your namelist.input looks fine. The topography in D02 doesn't show any large gradient, which means that EPSSM is not an issue. This error indicates that something might be wrong in the model physics.
Can you do the following tests and see how the model works:
(1) Turn off adaptive time step, then run the same case
(2) change mp_physics = 40, 40, 40, run with and without the adaptive time step
(3) set diff_opt = 2, 2, 2, then rerun the same case (cu_physics = 11, mp_physics=8, adaptive time step off)
Please let me know the results. Thanks.

 

[jahokson]

All of the tests resulted to (almost same CFL) errors (see attached).
Thanks.

 

[Ming Chen]

I am not sure yet what is wrong in this case. Please avoid using the MSKF option at present.
You may also try to figure out poses;e reasons by running in debug mode. To do so, please follow the steps below:
(1) ./clean -a and ./configure -D
(2) recompile WRF
(3) run the failed case.
The code and line at which the error first appear will be provided in RSL files, from where you can try to figure out what is wrong.