[phenixbb] Water molecules whose existence is contingent on the alt conf of a side chain are deleted in refinement

Hi, I have a model with a Tyr side chain with two conformations. When it is in the A conformation three water molecules occupy the site of the B conformation. I have defined these water molecules to have the alt loc A (with no B versions) and set their occupancy equal to that of the A version of the Tyr. During refinement, using the GUI, these water molecules disappear during the "add/remove water" stage. My guess is they are running afoul of dist_sol_mol_min. I can't find where I can set this parameter to some, very small, value that will allow these waters to live and prosper. If this is my problem, how can I disable this check? Dale Tronrud