Set the mirror and tester set up as shown . If the polishing has gone well ,the mirror should be almost spherical. The images shown in the diagram are for a perfectly spherical mirror. It is highly unlikely that your mirror will will be perfectly spherical but it is worth trying to get the mirror as close as possible, this will give a good starting point to start parabolisation .
Fig1 shows with the knife edge at position C the mirror will darken evenly all over as the knife edge is made to cut the light cone using the adjusting knob. with the knife edge inside focus image A will be seen, and out side image B. The diffraction lines seen on the two images can be used to ensure that the slit and the knife edge are parallel. They also give an indication of how good the figure is.
Using the normal parabolization stroke the surface of the mirror will change to give something like the above images. The amount the knife edge has to move from a to c to give a fully corrected mirror is given by hxh/r. where h is the position of the light hitting the mirror surface and r is the radius of curvature.
to ensure that the light is hitting the correct position on the mirror a screen is used placed over the mirror. with a least 4 cut outs on each side (testing setup). The average position for each set of two zones (h) measuring from the centre of the mirror is known . The knife edge is is positioned so that the required zone on each side of the mirror is made to go uniformly dark at the same time and a reading is taken from the micrometer. Starting from the centre zone the micrometer screw is used to move the knife edge along to take the next zone reading, the reading for each zone is then compared with the theoretical after subtracting a constant. when the readings compare reasonably well with the theoretical value the mirror is complete.
Before aluminizing the mirror , place the mirror in the telescope with an aluminized diagonal and carry out a star test the mirror will be reflective enough to perform a test. see star testing