How can I properly define SPECIES_LCAO_STATES in CASTEP? 

In the user guide, 

By default, the number of states will be the appropriate number to complete the valence shell to the next noble gas. If shallow core states are excluded from a pseudopotential, the value of SPECIES_LCAO_STATES for that species should be included in the cell file to ensure a meaningful basis set is used. 

Then, if I want to see the unoccupied d state for the Si atom, should I change SPECIES_LCAO_STATES of Si from 2 (default) to some another larger number?

If SPECIES_LCAO_STATES indicates the number of l-momentum channels used to project the DOS, does SPECIES_LCAO_STATES of Si for the d-orbital projection is 4 or 5? What is correct? 

SPECIES_LCAO_STATES for Si = 2 denotes 3s - 3p
SPECIES_LCAO_STATES for Si = 4 denotes 3s - 3p - 4s - 3d
SPECIES_LCAO_STATES for Si = 5 denotes 3s - 3p - 4s - 4p - 3d

When if I want to calculate the unoccupied d state for the Al atom, what should I do?