Why is the assumption of spheres being point charges invalid for small separations? I have a hollow charged sphere and I am measuring the force between it and a grounded metal plate. How does the assumption that they are point charges (using method of image charges) become invalid for small separations? 
So how is the force affected at small distances due to the charge being distributed on the surface close to the plate? 
Also sidenote- how does the plate being finite affect the use of the method of image charges for this situation?
 A: With respect to its potential and electric field outside the sphere, a charged sphere with uniform surface charge density can be replaced by a point charge at the center of the sphere. This is also a good approximation in the case of a non-uniform charge on the surface when you look at the electric field or potential at a large distance compared to the diameter of the sphere. If you consider a conducting sphere, the charge will be uniformly distributed over the surface only when it is far away from any other (charged) bodies. When the charged sphere approaches the grounded metal plate, at distances comparable to the diameter the electric field of the induced charges on the grounded plate will destroy the uniform distribution of surface charges on the sphere. Thus the charges on the sphere cannot any longer be represented by a point charge in the center of the sphere and the simple method of an image charge to obtain the electric field cannot be used when the sphere is near the conducting plate. 
A: If the sphere is non-conductive and non-polarizable, it can be replaced by a point charge at any non-zero distance from the plate. Also, the image charge method is then valid.
