Nonclinear calculation with symmetry cannot converge

[Shen-Zhen-Xiong]

Describe the bug

nonclinear.tar.gz
Please see the example, I calculate a CrI3 system with nonclinear, only when symmetry=-1, it can converge, but for symmetry=0 or 1, it is not converged.

Expected behavior

No response

To Reproduce

No response

Environment

No response

Additional Context

No response

Task list for Issue attackers (only for developers)

• Verify the issue is not a duplicate.
• Describe the bug.
• Steps to reproduce.
• Expected behavior.
• Error message.
• Environment details.
• Additional context.
• Assign a priority level (low, medium, high, urgent).
• Assign the issue to a team member.
• Label the issue with relevant tags.
• Identify possible related issues.
• Create a unit test or automated test to reproduce the bug (if applicable).
• Fix the bug.
• Test the fix.
• Update documentation (if necessary).
• Close the issue and inform the reporter (if applicable).

[dyzheng]

To be honest, I think the convergence of symmetry=-1 is a lucky case.

I recommand you to try smaller mixing_beta and try to set mixing_angle .

[Shen-Zhen-Xiong]

I try mixing_beta = 0.1 and 0.02 , after 500 step in symmetry =1, it is not converged.
scf.log

[dyzheng]

@Shen-Zhen-Xiong have you tried to set 'mixing_angle', which refer to function specifically designed to optimize the convergence of noncollinear magnetic moments.

[dyzheng]

@WHUweiqingzhou any suggestion for this Issue ?

[dyzheng]

@Shen-Zhen-Xiong Another suggestion is symmetry should not be used for non-collinear spin case, it will lead to error result now.
Only symmetry = -1 is recommended.

[Shen-Zhen-Xiong]

I think symmetry = -1 should be recommended for non-collinear spin case, symmetry is very complex for non-collinear calculation. Maybe it is a good choice for default symmetry = -1 in non-collinear spin case?