Improved collimation test.

I t is frequently necessary to test and verify the degree of collimation, or the bundle divergence, of a nominally collimated (laser) bundle used, say, for interferometer illumination. The collimation achieved depends upon the method of observing it. Poor collimation may result from a lightsource of too large an extension, or from zonal defects in the collimator objective or from defocusing. The latter cause only is considered here. Two methods are commonly used for determining the degree of collimation of a collimator consisting of a pointsource Q, and a collimator objective 0 (Fig. 1): autocollimation and shearing interferometry. Autocollimation off a plane mirror, as good as it might be, verifies collimation only by implication (if the source, usually a tiny pinhole, and its image are of equal size, collimation is good). The bundle divergence detected by shearing interferometry depends on the bundle diameter, and the sensitivity of this method extends to extremely small angles only, if the field diameter is large. An improved interferometric method of testing and verifying collimation is shown here. The method bears some similarity to inverting shear interferometry, which typically employs a Michelson interferometer with a cube corner reflector in one arm and a plane reference mirror in the other arm, Fig. 2, (Peck and Bertolotti et al.). The difference between this Michelson inverting shear interferometer and the modified version of it to be described here lies only in the relative proportions of the components, but this difference leads to some appreciable improvements. In the Michelson inverting interferometer, the cube corner reflector and the plane reference mirror both cover about the same bundle diameter (Fig. 2). The inversion of the bundle takes place upon reflection by the cube corner reflector (CCR). The shear between the bundles reflected by the CCR and by the