In the mid 20th century, R.A. Burton developed an alternative form of schlieren photography which is now usually called focusing schlieren or lens and grid schlieren,[1] based on a suggestion by Hubert Schardin.[2] Focusing schlieren systems generally retain the characteristic knife edge to produce contrast, but instead of using collimated light and a single knife edge, they use an illumination pattern of repeated edges with a focusing imaging system.
Basic diagram of a focusing schlieren system
The basic idea is that the illumination pattern is imaged onto a geometrically congruent cutoff pattern (essentially a multiplicity of knife edges) with focusing optics, while density gradients lying between the illumination pattern and the cutoff pattern are imaged, typically by a camera system. As in classical schlieren, the distortions produce regions of brightening or darkening corresponding to the position and direction of the distortion, because they redirect rays either away from or onto the opaque part of the cutoff pattern. While in classical schlieren, distortions over the whole beam path are visualized equally, in focusing schlieren, only distortions in the object field of the camera are clearly imaged. Distortions away from the object field become blurred, so this technique allows some degree of depth selection. It also has the advantage that a wide variety of illuminated backgrounds can be used, since collimation is not required. This allows construction of projection-based focusing schlieren systems which are much easier to build and align than classical schlieren systems. The requirement of collimated light in classical schlieren is often a substantial practical barrier for constructing large systems due to the need for the collimating optic to be the same size as the field of view. Focusing schlieren systems can use compact optics with a large background illumination pattern, which is particularly easy to produce with a projection system. For systems with large demagnification, the illumination pattern needs to be around twice the size of the field of view to allow for defocusing of the background pattern.[3][4]