Some EURO bills will be used here to run the experiment.
Anyhow, at the time of writing the author used EURO-ersatz since the EURO bank-note shall be the sole cash currency in Europe but only in 2002 ;-)
Shown on the picture below.
Remember that the set up shown here is for information only. Bank-notes are not mandatory. They were used here for fun.
Any background with distinctive features or pattern shall fit.
Some EURO bills are arranged (each vertically on its side) in a circle around the front of the CP990 fisheye lens.
At mid distance, some sharpened nails are vertically pinned down on a blank sheet of paper. Check the vertical with a Square.
The Optical Axis of the lens is projected and drawn also on the paper as a straight line. Use of a Square for that purpose is recommanded.
On the test scene : After setting-up the experiment simply take a picture with the fisheye lens. You can set the camera in any mode which shall give you the maximum Horizontal Field Of View.
On your computer : transform this circular fisheye image into an equi-rectangular picture. You may use for this the Helmut Dersch's PTool. It's a plug-in for photoshop (or Graphic Converter on Mac). You may alternatively use PTPiker. You shall then get something like the following picture.
It is an extract from the larger equirectangular panorama. Point B is hidden by the bottom of the nail. Mark it with a pencil on the sheet of paper :
Observe carefully a nail on the screen of your computer. It hides some details of the background. Try to identify and remember the missing part.
Go back to the test scene and with a pencil (and possibly with the help of a Square) mark that point on the base of the vertical pattern (on the paper) which matches the identified missing part.
Draw the line which goes from this point through the "hole" made by the nail on the paper. This straight line shall cross the Optical Axis trace on the paper sheet at... the Nodal Point which correspond to the specific ß aspect angle of this specific nail.
Repeat the process for each nail. Notice that the smaller the ß angle (for the nails at the center of the test scene), the more difficult it is to accurately define the intersection.
The nodal point location depends actuallyon the aspect angle ( ß ) from the optical axis of the lens.
There is AN INFINITY of nodal points. They are located at a distance (D) that is about 17 or 20 mm (*) from the lens front to 4.5 mm from the same surface.
(*) In fact this experiment gives very satisfactory results for wide ß aspect angles. BUT when ß is small it is naturally difficult to accurately find the intersection point of the optical angle with the corresponding "Light ray".
As the most important range is for very wide angles and especially for ß=90 deg, then the process is satisfactorily helpful.
For a more accurate experiment test and result please go to this.