To my knowledge, resonance happens where more than one lewis stucture can represent the compound with no shift in sigma bonds
so, suppose in a simple conjugated system even like 1-3 pentadiene , two forms are well known, i want to know if a third form where each atom has completely filled or completely empty p orbitals should also be taken into consideration, as technically, there is nothing wrong with it, I accept that it's "contribution" will be negligible, but it should be counted in all POSSIBLE resonating structures
https://commons.wikimedia.org/wiki/File:Naphthalene_resonance_structure.svg
similar argument here, a lot more can be created if we allow charge separation of pi bond
EDIT: The question is whether there is a convention of some sort which tells us which ones to count which ones not (if we cant tell the magnitude of difference in stability of the structure)