ANALYSIS OF LARGE ELECTRON-OPTICAL SYSTEMS WITH MANY INTERACTING OPTICAL-ELEMENTS

The analysis of large electron optical systems depends on the accuracy of the field computation of the system. In turn, the accuracy of the computed fields is strongly dependent on the size and complexity of the system under investigation and the fineness used to convert the system to a discrete numerical model. This results in severe demands on computer memory and speed. In addition, the analysis of these complex electron optical systems having multiple optical elements requires a measure of flexibility in "adjusting" each element's contribution to the system independently. With this flexibility the entire system does not have to be solved again when a change is made to one element of the system. Two techniques have been developed to resolve these two limitations. One technique allows a large system to be broken into smaller sections by establishing the section boundary conditions using current loops. This permits the system problem to be solved using finer and finer discrete models of each section to increase the accuracy of the system solution. The second technique permits the field contribution of each optical element of the system to be computed individually, but as though all the elements of the system were operated at their nominal excitations.