[gmx-users] choosing box vectors

[gromacs query]

Dear All,

I have two different membrane systems (membrane with waters only) already
with me. So I dont need to add solvent molecules.

So to start with simulation I need box size (last line in gro file). I used
following command (for both cases):

editconf -f del.pdb -o del.gro -c -bt cubic


First case details:

    system size : 10.707 11.077  7.493 (nm)
    diameter    : 14.737               (nm)
    center      :  6.928  6.932  6.936 (nm)
    box vectors : 10.707 11.077  7.493 (nm)
    box angles  :  90.00  90.00  90.00 (degrees)
    box volume  : 888.68               (nm^3)
    shift       : -1.574 -1.394 -3.190 (nm)
new center      :  5.353  5.538  3.747 (nm)



Second case details:

    system size : 10.065 11.826  9.911 (nm)
    diameter    : 16.459               (nm)
    center      :  9.229  9.229  9.229 (nm)
    box vectors : 18.459 18.459 18.459 (nm)
    box angles  :  90.00  90.00  90.00 (degrees)
    box volume  :6289.31               (nm^3)
    shift       :  0.000  0.000  0.001 (nm)
new center      :  9.229  9.229  9.229 (nm)



As I have solvents already so my box size as a rule should be (Maximum X -
Minimum X), (Maximum Y - Minimum Y), (Maximum Z - Minimum Z). This indeed I
found to be correct for my first case in which box dimensions in gro file =
10.707 11.077  7.493

But why in my second case box dimension in gro file = 18.459 18.459 18.459.
Surprisingly its is equal to Maximum X + Minimum X of my system and all are
same being 18.459. Rather it should be (Maximum X - Minimum X), (Maximum Y
- Minimum Y), (Maximum Z - Minimum Z) as in first case.

Also in VMD I saw gro files. In first case pbc box completely fits the
membrane water system but in second case pbc box is un-necessarily large as
it is using 18.459 18.459 18.459


I am confused why in second case it chose Maximum X + Minimum X unlike in
first case?

please suggest,

thanks,
Jiom