A serious disadvantage of using a lattice of vortex rings to model a surface of vorticity, is that points close to the lattice surface may see large ‘holes’ in what from a larger distance appears as a smooth vorticity distribution. This failing is particularly important when two surfaces approach each other, such as occurs at the trailing edge of a wing, or when a free vortex passes near to a solid boundary. In these cases the error in induced velocity can adversely affect the overall flow computation. A relaxation process to find the force-free position of a free vortex close to a surface would obviously fail, for example. The obvious way of overcoming this problem is to improve the accuracy of the vorticity model whenever a point draws critically near to the vortex lattice.

Maskew suggested a way of overcoming this problem in two dimensions and his ideas have been extended in this note to three dimensions.