Hello, 1-) You have to select the parameter you will change, i.e. study the effect in convergence. Here on called P1. Select the other parameter in which you are studying the effect of P1 (maybe the total energy or a property like the gap). Example of parameters used as P1: meshcutoff and number of k points 2-) Now you have to run several calculation, for the same input, changing only P1 (at least 10-15 times) My advise is to do a shell script (or Python, etc.) or use something like ASE to make the whole process automatic. 3-) Do a graph of P2 (Y) vs P1 (X)
[]'s, Camps On Thu, Dec 17, 2015 at 11:17 AM, Muhammad Ilyas Physics Department < [email protected]> wrote: > Dear > can you tel me the complete procedure for convergence test in siesta > ? i will be very thankful to you > regard ilyas > PhD Student > > On Mon, Dec 14, 2015 at 9:46 PM, Matías Soto C. <[email protected]> > wrote: > >> Dear Rajan, >> >> How large are the variations of the atoms that deviate from the 2-D plane >> when relaxed below 0.02 eV/Ang? It could be that these variations are >> within others have found when relaxing graphene sheets or nanoribbon. >> >> Also, check your Mesh Cutoff value. Do you see a change in the total >> energy calculation if you increase this value? And make sure other values >> are converged, such as kgrid and lattice vectors. >> >> Best regards, >> >> -- Matias >> >> >> Matias Soto C. >> Ph.D. Program >> MSNE Department >> Rice University >> >> On Mon, Dec 14, 2015 at 3:22 AM, RAJAN SINGH <[email protected]> >> wrote: >> >>> Dear Siesta Users >>> >>> I am trying to relax graphene nanoribbon using siesta, but each time >>> when it converges it change 2-D structure to 3-D. >>> >>> I am attaching my fdf files for reference. >>> >>> Please help me with this. >>> >>> Thank You >>> >>> Regards >>> Rajan Singh >>> >> >> >
