Anonymous @ on
small decay length in the relaxation energy appears to cause non-convergence
Find information on common issues.
Ask questions and find answers from other users.
Suggest a new site feature or improvement.
Check on status of your tickets.
Status: Closed
2 Like 0 Dislike
Samarth Agarwal @ on
The problem is because of the adaptive energy integration which tries to add more nodes till the integral converges. The upper bound for the number of energy nodes that can be added will be properly set in the next release.
A small decay length translates into a very small value for the optical potential for a given relaxation energy below the conduction band edge if the relaxation model is chosen to be exponentially damped. To avoid convergence problems it might be worth trying a small value of relaxation energy(1e-6eV) with the relaxation model chosen to be independent of energy.
Report abuse
Anonymous @ on
Can this instability be fixed in the code?
Report abuse
Samarth Agarwal @ on
An upper bound for the number of nodes added by the adaptive grid has been set.
Another change which does not allow an indefinite addition of energy nodes between two starting nodes has also been incorporated.
Report abuse