Trouble with elph matrix calculations when I use the ibrav=0

[jgSi1113]

Dear all,

I am doing some superconductivity study with SSCHA.

The hamonic phonon and electron-phonon coupling matrix are calculated by the following workflow by revised QE_v5.1: scf==> ph==>elph.

When I use a ibrav /=1, the 3rd step (namely elph calculations) goes well. But if I use ibrav ==0, the 3rd step always send error message. The error message says:
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = RANK 0 PID 45510 RUNNING AT node108 = KILLED BY SIGNAL: 9 (Killed)
I tried resubmit the job to a fat node with 3T RAM, the error message is robust!

After many careful test, I found that the problem is not a real RAM problem, but from ibrav. When I use the ibrav=0, the dynamical matrix has 4 lines more than the ibrav /=0, namely the
"Basis vectors
4.843202867 0.000000000 0.000000000
0.000000000 4.843202867 0.000000000
0.000000000 0.000000000 4.843202867".
Such slight difference may induce some errors when the ph.x read the prefix.dyn*.

For the system likes SrB3C3, we can use the ibrav=1 to solve this problem, but for the system like CaH6, the ibrav could not be well defined with a non-zero number. So how to fix it?

Any suggestions and discussions will be greatly appreciated.

Best & Regards!
Jianguo Si

[jgSi1113]

It was solved with the help of Dr. Hao Gao.

Jianguo Si

[jgSi1113]

Just remove the four lines without any other changes!