can be used with several different Ronchi rulings or a knife edge, they are interchangeable, but the Ronchi rulings are very fragile
The focuser uses a Ronchi grating of 180 lines/inch to achieve critical focus
The principle of operation is the same for knife edge or Ronchi. The Ronchi ruling is a periodic bar pattern that is positioned near the primary focus of the telescope. The periodic bar pattern interupts the diffraction pattern of a single star and thus acts as a spatial filter when it is exactly at focus (Ronchi pattern in the focal plane of the telescope). One can imagine the ruling bar blocking the star's central Airy disk at focus. The Stiletto optical system includes a re-imaging lens and a Ploessl eyepiece which image the pupil plane of the telescope when in focus. The telescope central obscuration is present in the imaged pupil.
When first set up, you see a bar pattern that changes spatial frequency as you attempt to focus. Increasing spatial frequency indicates movement away from focus. Simply adjust the focus to
reduce the bar pattern to a single “bar” that covers the entire pupil. At this point, you are in focus and the bar is either lighter or darker depending very critically on focus position. Focus is achieved quickly. The Stiletto is then removed and the camera is attached. The assumption is that the distance to camera focal plane is the same as the distance to the Ronchi ruling.
focus sensitivity will be a function of the Ronchi ruling spatial frequency, that is, the 300 line/inch ruling will produce a tighter focus than the 180 line/inch ruling. This would especially be important for faster optics with a smaller depth of field and smaller Airy disk. Compare the size of the bar to the Airy disk. Image brightness and contrast may work in the opposite direction, with the 180 line/inch image being brighter than the 300 line/inch image.
The knife edge is the ultimate focus aid, since it would be more sensitive to focus error than any Ronchi ruling. It blocks one half of the star's diffraction pattern. The claim is that it is more difficult to use, since it is easy to move right through focus without noticing it. Use of the Ronchi is more forgiving.
They are both spatial filters since they are at the Fourier plane of the telescope. The size of the Airy disk is linearly proportional to the focal ratio.
A 180 line/inch Ronchi ruling has a bar size of 141 micrometers, about twelve times larger than the Airy disk. So a single Ronchi bar may block the central order of the diffraction pattern (Airy disk) as well as quite a bit more of the circularly symmetric star diffraction pattern when it is in the focal plane. This is why, when looking into the Stiletto, the pupil image is either dark or light at focus -
dark if a bar is centered or near centered on the Airy disk, light if the space is centered or near centered on the Airy disk. Only a “single bar” is left when at focus.
For a telescope focused at infinity, i.e., a star, the Fourier plane and the image plane are the same. The magnitude squared of the Fourier transform of the amplitude and phase distribution across the
circular aperture of a telescope is the diffraction pattern seen in the focal plane. The image of a star is the resulting diffraction pattern. Again, the image of a star is the magnitude squared of the
Fourier transform of the aperture function. If the telescope has phase distortions or a central obscuration, the diffraction pattern represents the magnitude squared of Fourier transform of those phase and amplitude distortions.
Focus error, coma, astigmatism, etc. can all be viewed as phase errors. A special set (normal set) of mathematical functions called Zernike polynomials can be used to describe these and higher order phase distortions.
Using the knife edge is a good way to view higher order phase aberrations in your telescope or in the atmosphere.
As for references, try Fourier Optics by Joe Goodman to start and probably a Google search on Schlieren optics, Fourier optics, Foucault test, phase contrast microscope, spatial filtering, etc. may
be useful since all these terms all are related. Even analog communications theory with a vector representation of narrowband phase modulation is useful in gaining insight into this problem.