Generally, I know that SOC + magnetic DFT calculations can be challenging to 
achieve convergence.
So, I suggest you should decrease your conv_thr criteria. (Of course, you have 
to check that enough to see the physical properties you want.)
Plus, you have to set a good initial guess(atomic position, direction of spin, 
etc.)

GiMyung Park
Yonsei University
-----Original Message-----
From: "Barsha Pal"<[email protected]>
To: <[email protected]>;
Cc:
Sent: 2025-11-22 (토) 16:42:04 (GMT+09:00)
Subject: [QE-users] SOC SCF not converging

Hello,
I am running with this system card, a scf run, but the scf is not converging 
till 140 steps, what can I do to finish the job ? Also for soc calculations to 
converge is there any important parameter that I am missing ?


    degauss     =  0.01
    ecutwfc     =  170
    ecutrho     =  680
    ibrav       =  6
    nat         =  8
    ntyp        =  3
    occupations = "smearing"
    smearing    = "mv"
    noncolin = .true.
    lspinorb = .true.
    starting_magnetization(1) =   0.5
    starting_magnetization(2) =  -0.5
/

&ELECTRONS
   conv_thr         =  1.00000e-08
   electron_maxstep =  600
/

ATOMIC_SPECIES
Ga1  69.723000d0   Ga_ONCV_PBE_FR-1.0.upf
Ga2  69.723000d0   Ga_ONCV_PBE_FR-1.0.upf
P    30.973800d0   P_ONCV_PBE_FR-1.1.upf





Ashley Cooper
Phd
Assam University Silchar
_______________________________________________________________________________
The Quantum ESPRESSO Foundation stands in solidarity with all civilians 
worldwide who are victims of terrorism, military aggression, and indiscriminate 
warfare.
--------------------------------------------------------------------------------
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list [email protected]
https://lists.quantum-espresso.org/mailman/listinfo/users

Reply via email to