[modeller_usage] How to properly set the fields 5 of the second line in the .aln file?

I have a two-chain protein to model, Chain A and Chain B.

In the PDB file, the Chain B residues are numbered from 3-33, so only 31 residues appear in the PDB. When "1 :A:+33 :B" is used for the fields 3-6 in the second line of the .aln file, I got error: "Number of residues in the alignment and  pdb files are different".

When it is changed to "1 :A:+31 :B", the same error occurred.

However, when only Chain B is modelled, "3 :B:+33 :B" can be run successfully.

So, it seems that, when Field 5 = n, it assumes that the last chain has in total n residues, not the last residue is numberred as n.

So how should I set the Field 5 number? Do I have to renumber the Chain B residues in the PDB from 1 to 31?

Thank you!