<div class="iw_mail" dir="auto" style="font-size: 13px;"><div>Dear QE users,</div><div>Hi.</div><div><br></div><div>I am trying to compute Hubbard parameters using HP code.</div><div>However, for some q-point of some atom I get divergence for some components of chi.</div><div>I have played with all alphamix, nmix, q_mesh, but did not succeed to manage convergence.</div><div>The output is:</div><div><br></div><div> =--------------------------------------------=<br id="isPasted"> START SOLVING THE LINEAR SYSTEM<br> =--------------------------------------------=<br><br> atom # 2 q point # 1 iter # 1<br> chi: 1 0.0155052037<br> chi: 2 -5.5123136053<br> chi: 3 0.0155052037<br> chi: 4 5.1960958877<br> chi: 5 0.0177200092<br> chi: 6 0.0177200092<br> chi: 7 0.0177200092<br> chi: 8 0.0177200092<br> chi: 9 0.0177200092<br> chi: 10 0.0177200092<br> chi: 11 0.0177200092<br> chi: 12 0.0177200092<br> Average number of iter. to solve lin. system: 36.2<br> Total CPU time : 31.0 s<br><br> atom # 2 q point # 1 iter # 2<br> chi: 1 -0.0255548541 residue: 0.0410600578<br> chi: 2 ************** residue: **************<br> chi: 3 -0.0255548539 residue: 0.0410600576<br> chi: 4 ************** residue: **************<br> chi: 5 0.0155525281 residue: 0.0021674812<br> chi: 6 0.0155525281 residue: 0.0021674812<br><br>----------------------------------------------------------------------------------------------</div><div>Could anybody please help and remind any other convergence switches which may be forgotten?</div><div>Any comments is highly appreciated.</div><div><br></div><div>Best regards,</div><div>Mahmoud<br></div><div><br></div><div><br></div><div class=""><div>-------------------------------</div><div>Mahmoud Payami</div><div>NSTRI, AEOI,Tehran, Iran</div><div>Email: mpayami@aeoi.org.ir</div><div>Phone: +98(0)2182066504</div><div>------------------------------------</div></div></div>