I found that error as well, when trying to play with a half core pseudopotential only few days ago. It depends on the way the element name is written in the UPF file:
element=" C" rather than element="C " in the PP_HEADER section. Then the specific routine misses the core element. It seems that we are working towards similar goals... maybe we should get in touch. Let me write you privately as well. Best, Guido Fratesi On 05/04/2012 03:16 PM, Giuseppe Mattioli wrote: > >> It is working fine, both with pw.x and xspectra.x > > Not exactly... pw.x works fine, xspectra works fine till the lanczos chain is > finished, but it is not > able to calculate the spectrum. It says something very strange... > > Estimated error at iter 2350 is 2.773650816161815E-003 > Estimated error at iter 2400 is 7.339432110204557E-003 > Estimated error at iter 2450 is 3.915220430921537E-003 > Estimated error at iter 2500 is 1.759769462142735E-003 > Estimated error at iter 2550 is 1.609732607968768E-003 > Estimated error at iter 2600 is 2.963442071041433E-003 > Estimated error at iter 2650 is 3.196232925421465E-003 > Estimated error at iter 2700 is 2.704414998145615E-003 > Estimated error at iter 2750 is 2.510644989742356E-003 > Estimated error at iter 2800 is 1.927644198666715E-003 > Estimated error at iter 2850 is 2.702595324285325E-003 > Estimated error at iter 2900 is 1.766250501508676E-003 > CONVERGED at iter 2950 with error= 7.723753097511076E-004 > Could not find the element C in the table of K edge energies! > > I've never seen that when using carbon NC or US gipaw PPs. > > Yours > > Giuseppe > -- Guido Fratesi Dipartimento di Scienza dei Materiali Universita` degli Studi di Milano-Bicocca via Cozzi 53, 20125 Milano, Italy Phone: +39 02 6448 5183 email: fratesi at mater.unimib.it
