Adding water molecules

GROMACS version: 5.1.4
GROMACS modification: Yes/No
Dear All,
I want to simulated three different systems (each system contain a typical solute). How do I solve them in water so that is scientifically correct? Is it true that I use flags “-d 1.0” and “-bt box type” for solving? In this case I will have three systems with different numbers of water molecules. Is it possible to compare the structural and thermodynamic properties of these three different solutes while they solving in different number of water molecules? Should the number of water molecules necessarily be the same in all systems and use “-box x y z” flag in “gmx editconf” and then “-maxsol” flag in “gmx solvate”? These three solutes are nanotube and its derivatives.
Best regrads,

In the standard MD simulations (using periodic boundary condition) a solute in water solution has to be considered like a solute in a very dilute solution.
In general it is good to set systems in the same study in a similar way. For example generate the box with similar criteria using -d xxx and -bt xxx and then use gmx solvate
If the system have slightly different dimension, the use of -box x y z may be dangerous.
For example it may result in a box of too small dimension for the large system, if it is calibrate on the values used for the smaller system.

Best regards

Dear Alessandra
Many thanks.