A Knife-edge Tester An overall view and diagram of details

This is based on a design suggested in "How To Make a Telescope" By Jean Texereau, published by Willman-Bell,Inc. - Prints out on 2 pages, Portrait orientation,legal size paper 8.5" x 14"

Page updated July 6th - 2001



Parts required:
  1. A 6v.or 12v. auto lamp bulb - frosted by rubbing with emery cloth.
  2. Transformer for the lamp. Use a 1 amp. fuse in the primary.
  3. A slit-plate - brass, see details for making it in the second drawing.
  4. Plywood base and knife edge carriage (can be made of metal or plastic if desired)
  5. A round steel rod about 1/4" diameter 10" long.
  6. Small thin glass plate on base for the carriage adjusting screw to run on -about 1" wide x 2" long.
  7. A 1/4 x 20 screw about 2 1/2" long, for the lead-screw to drive the carriage along the steel runner.
  8. 2 or three 1/4 x 20 brass nuts for the lead screw to run through. These are tightened enough to allow the lead screw to turn without back-lash. They are soldered to a brass support plate and provide very smooth running of the knife-edge carriage.
  9. Lead-screw knob about 1 1/4" diameter.
  10. Knob about 1" diameter, to drive the knife edge carriage sideways to move the knife in and out of the reflected beam from the mirror.
  11. Assorted hardware and screws to complete construction.


The next page shows the details of the way I made the tester that I use.


Important notes below

The hole in the slit-plate (behind the slit) should be about 1/4 inch diameter. Also the hole in the lamp housing which coincides with the one behind the slit when it is set in place can be made slightly larger (perhaps 3/8 inch to make it easier to find and position the reflected image (slit-plate removed) when setting up, prior to making measurements.

The knife-edge attached to the top of the carriage upright, is made in the same way as the two pieces on the slit-plate. It is a piece of brass fastened to the carriage by a single screw and nut, so that it can be tilted to align the knife-edge parallel to the reflected image of the slit. The edge is chamfered with a small file and then the edge is made flat and even by rubbing against a flat piece of ground-glass. This can produce an edge which is almost optically flat...and is much preferred to using a razor-blade or other piece of metal - the edge of which may not be quite as accurately flat.

Afew extra notes - The slit width used should be set as close as possible to about 3 to 5 thousandths of an inch, although the unit will work with a wider slit the narrower the slit the better! The two plates forming the slit should also be parallel as closely as you can make them. A couple of strips of typewriter paper can be used to help in this respect. When setting up for a test the slit-plate is taken off temporarily so that you can find the reflected image more easily, and then after aligning the reflected image between the knife-edge and the lamp housing, the slit-plate is replaced for making the tests.

Usually the scale on the tester is marked off in 10th's of an inch and the knob (if you are using a 1/4 x 20 thread screw) is marked in intervals to represent thousandths of an inch. A pointer of some kind should be made to read the scale on the knob's circumferernce. Generally you can divide the circumference of the knob into divisions each equalling 5/1000th's of an inch and then guestimate the distances in between to the nearest thou.(One full turn of a 1/4 x 20 screw = 50 thousandths of an inch) so the marks on the knob scale would go from '0' to '50'. Two full turns of the knob should move your longitudinal scale by 1/10th inch. So your pointer for the knob should be set so that when the other scale is at a 1/10th position, the scale reading on the knob is set at zero.

So you would (for example) start at a position where the longitudinal scale read 0.1 inch and then add (say) 12 thou read off the side of the knob...which would give 0.112 as your reading. The next window in the Couder screeen after moving the knife-edge further away from the mirror might be 0.1 plus 75 thou read off the knob scale...giving you 0.175. etc. It is best if the longitudinal scale is marked with the numbers ( .1 .2 .3 etc.) going away from the mirror...so that your total readings increase as you move the knife-edge away from the mirror. The absolute value of the numbers does not matter - what you are after is the relative spacing between each reading as the knife-edge is moved towards or away from the mirror for the darking position at each window of the Couder screen.

If you are using metric measurements, then make your scales accordingly. The 'coarse' scale on the carriage can be marked in millimeters. The lead screw for the in and out movement of the knife-edge carriage can be a metric screw with threads of 1mm. Then the scale on your drive-knob for the carriage can be marked with 10ths of a millimeter on its circumference, as one complete turn will give you one millimeter of movement. You could also 'estimate' the third decimal (100th's mm ) by estimating between the 10 marks on the edge of the drive-knob.

Return to FVAS Home Page
Return to Ronchi