Source: sci.optics & sci.techniques.microscopy
Reference: See original posts for authors and their original text
Critical illumination is the configuration of the illumination in a microscope in which the source of the illumination (lamp) is focused by the condenser on to the specimen (slide). This can result in un-even illumination of the specimen if there are spatial variations in the lamp, though the spatial variations can be mitigated using a frosted or opal glass plate (in what plane?). The iris is located at the entrance pupil to the condenser lens.
Koehler illumination is the configuration of the illumination in a microscope in which the source of the illumination (lamp) is focused by the condenser on to the entrance pupil (image of the aperture stop) of the microscope (generally lies somewhere inside the objective). This results in even spatial illumination of the specimen.
It appears that we can visualize Koehler illumination assuming the light source to be very small; sample (specimen) non-scattering, and condenser with no aberrations. With no loss of generality, we can consider a photo enlarger (microscope minus the eyepiece). In this case, the objective lens becomes unnecessary (we need it back if we re-introduce scattering in the specimen), the system works as a pinhole camera, although no light is lost. The depth of field is infinite since the sample is projected through the pinhole.
Both types of illumination have their specific roles in Microscopy
Both critical and Koehler enable maximum resolution to be obtained.
In terms of imaging fidelity, the two cases are identical. When using critical illumination, the primary feature in the object plane is typically the lamp filament instead of the sample of interest.
What is important is that one of the two illumination methods are used.
Critical probably may be better for fine detail. That is probably because having light come to the specimen from different angles smoothes out the diffraction artifacts. Indeed, if the light entering the specimen was a sheer parallel beam, from Koehler applied to a "point" source, then the image-forming rays would have to be diffracted/refracted etc, from the specimen to enter the objective as a proper cone centered on the specimen, thus introducing more fringes, etc. But if we use critical, the light is more directly "shadowed" through the specimen because many double cones of light are coming from the condenser, focusing to a point at the specimen, and proceeding directly to the objective (as indeed consistent with our being able to see any detail on the source, were there any.) In this sense, I don't think the two are as optically equivalent as some here think. BTW: you can get better results with realistic diffuse sources, by having them a bit out of focus at the specimen/image plane.
Actually the difference between critical illumination and Kohler illumination is were does it start? In Kohler illumination the system starts with the filament in the lamp. Actual Kohler (after Herr Dr. Prof. Kohler, Carl Zeiss, 1905) starts with a filament able to fill the field size. Koehler lamps have a very larger filament that is flat so it is able to fill the field. Kohler lamps are hard to find now.
Critical illumination uses a piece of ground glass as the first position, anything behind it is fine. It's easy and simple to build and works just great, most Zeiss instruments use it. The critical quality is the ability to center the condenser.
Fourier optics is easy to bring to cases when plane waves are converted into point images (secondary sources). The basic layout looks like "plane-f-lens-f-plane". In our system, the distances are longer (2f for 1:1 imaging of the source into the pupil). We could mentally split the condenser into 2 lenses and then try to think in terms of Fourier optics. Still, does not Fourier optics deal with diffraction and coherent light? Neither appears absolutely required to describe a microscope.
In terms of Fourier optics, Koehler and critical illumination are transforms of each other. Critical is called "incoherent" and Koehler is called "coherent" for historical reasons, but the coherence properties of the two illumination schemes are identical. That is, the aberrations present in the condenser lens do not degrade the imaging porperties of the objective lens when using critical/Koehler illumination. For more detail, Born and Wolf, section 10.6.2, has a discussion.
There may be a good reason for critical being called "incoherent" and Koehler "coherent." An extended source must be used for critical, and thus rays are coming into the specimen plane from many directions. In the case of Koehler, if we used a small filament, the rays would be coming into the specimen as a parallel beam extending across most of the diameter of the top lens of the condensor, assuming a fully opened iris opening. (The simple way to describe Koehler is that the rays are focused "at infinity" at the specimen plane, and thus are a parallel beam from a point source, whereas for Critical, the source is focused at the specimen plane, and would converge to a point at that plane.
light from a lamp is collected by a mirror which surrounds it.
light is collected by a system of lenses.
an image of the source is formed in the entrance pupil of the optical system< by the illumination system. (see US Patent 1143287, 1915)
an image of the source is formed on the object