[QE-users] effective charges sum far from zero

[Ronald Cohen]

Unfortunately the old run (which had dft+u also) crashes now in ph.x with
Calculating the dnsorth_cart matrix...
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source

ph.x 00000000011A1B4A Unknown Unknown Unknown
libpthread-2.26.s 0000151D8E9112D0 Unknown Unknown Unknown
ph.x 00000000006E57F0 dwfc_ 41 dwfc.f90
ph.x 0000000000689263 dnsq_orth_ 187 dnsq_orth.f90
ph.x 00000000005BD424 phq_init_ 309 phq_init.f90
ph.x 0000000000588164 initialize_ph_ 98 initialize_ph.f90
ph.x 000000000051C1C9 do_phonon_ 85 do_phonon.f90
ph.x 0000000000412ABD MAIN__ 78 phonon.f90
---

I am  getting effective charges that sum far from zero. I am using the latest develop code of ph.x but I found old runs from 2017 that show the same effect in a similar (but different) system. I can rerun the old job and make sure I get similar results, but this is a big system so maybe someone knows why this is happening? I am using DF2 vdW, pbe gga, and it is spin polarized, so everything has to work!


I attach old and new (in the tar file)  runs. Any help[ would be greatly appreciated. I also attached a file that lists the sums.

Sincereky,

Ron

---
Ronald Cohen
Extreme Materials Initiative
Carnegie Institution for Science
5251 Broad Branch Rd., N.W.
Washington, D.C. 20015
rcohen at carnegiescience.edu
office: 202-478-8937
skype: ronaldcohen
twitter: @recohen3
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210809/99624652/attachment.html>