can i use odd Monkhorst Pack Kpoint for small supercells

[magnones]

hi, i am in a trouble well i was using a odd Kpoint for calculation in a small cell containig 3 atoms it´s a Kpnt file like this
K-Points
0
Monkhorst Pack
11 11 11
0 0 0

but now i need to use a supercell based in the replication of 3x3x2 times the basic unit cell (containig 3 atoms) i was reading the manual and some handsons files from the site and they told me that if i would have used even Monkhorst pack Kpoint i would only need to decrease the Kpoint number for the same proportion as the basic unit cell is increase in this case x =3 and x= 2 in the x and y direction respectively

but in the case i use odd Kpoint which Kpoint can i use for me supercell? can i use odd Kpoint for me supercell ? jow many Kpoint should i use?

i would be so kindnees to answer my question it is important for me.

[admin]

yes of course. I would use a 3x3x5 mesh (if you want to keep tho the odd mesh).

[magnones]

thanks for your help i understand the way to choose the kpoint but i have another problem i did what you explain me to use i i had a message like this in the prompt What is that?



VERY BAD NEWS! internal error in subroutine IBZKPT:
Reciprocal lattice and k-lattice belong to different class of lattices. 48
POSCAR, INCAR and KPOINTS ok, starting setup


can you help me to solve the problem please?

thank you in advance

[admin]

the symmetrization routines sometimes fail for systems with low symmetry (unfortunately including hexagonal) if you use an even k-mesh (please have a look at the k-points.pdf file of the VASP-workshop tutorial). Please
--) use an odd k-mesh. If this does not solve the problem
--) turn off all symmetrisation (ISYM=0)