Hi, the first thing to do would be to decrease the TimeStep, you could also turn on CFL reporting to see what is the value.

Pozdrawiam,
Stanisław Gepner

13 mar 2019 13:19 Moto Light <redamansyhh@gmail.com> napisał(a):
Hi All,
i am runing case 3D using 32 modes and using new version of Nektar++4.4.1

and after 2 sec in 3D solution i had this error 
////////////////
CG iterations made = 5001 using tolerance of 1e-09 (error = nan, rhs_mag = inf)
//////////////////
and the simulation has been ended 

any one have idea what i can do to solve it

wish you great day

Reda