Copyright – P. J.
Smith
Despite the apparent
simplicity of a Pinhole set-up I strongly recommend, especially for beginners,
you regard it as the least desirable option for Ronchi testing and use in preference
a slit, which is more forgiving.
LED or Bulb with
Pinhole
One favourite is a LED with
the front lens sanded off. The frosted
remaining plastic helps with diffusion and a pinhole can be placed very close
to the small emitter in the LED. It is
cool, requires little energy, and is small enough to be placed near axis
without obstructing the eye. It is
simple. Despite this, perfectly even
illumination is difficult to achieve.
Small incandescent bulbs
also work and if a LED is unavailable will be quite adequate.
Using another lens to focus
the diverging rays from the light source towards the pinhole ensures more even
illumination. A microscope objective is
great for this and is often used in professional equipment.
Another point source is obtained
by shining a light beam onto a small spherical reflector such as a polished
ball bearing.
LEDs of 5 and 10 mm
diameter are readily available. 5 mm is
usually sufficient and is easier to mount close to the axis. It is probably easier to align the larger
one with the pinhole.
It is better to grind off
the collimating portion of the LED so the small chip source can be placed
closer to the pinhole. Leave this a
ground finish so a separate diffuser is not needed.
The Viewing Grating is of
course placed close alongside the source.
No alignment of Grating is required but the return beam is often harder
to locate compared to a slit or grating source..
No examples of test rigs
using pinholes are given but it would be easy to convert the Basic
Rig or Deluxe Rig. In a similar way, a lash up using modelling
clay to hold pinhole assembly and grating is easy.
Remote light source
It may be convenient to
locate the luminous device in a remote position and channel the light to a
pinhole, slit, or grating source.
Advantages are :-
1.
Keeps
heat remote from the optical path.
2.
Obstructs
light less.
This may be implemented by
:-
(a) A beam of light from, for example, a
slide projector. This can be placed to
one side and the beam aimed at a small mirror or prism or spherical reflector
such as a polished ball bearing. A more
modern variation is to use a laser beam.
(b) A glass rod or sheet that constrains
the light by total internal reflection, leading it to a small pinhole or
slit. A rod of glass may be truncated
to increase the throughput. An example of a device using a microscope slide to
channel light is given under The Brute
although it actually uses a slit.
(c) An optic fibre whose end may act as
a pinhole. The end should be well
finished for even illumination. A
variation is to form a small ball lens on its end by fusion. This gives an even, wide cone of light. It is
difficult to couple a light source to the optic fibre without much loss of
light. One makeshift way is to drill a tiny hole in the front of a LED, insert the fibre, and bond in place with
some transparent glue. By the time the light has traversed the length of the
fibre, any uneven illumination has usually been smoothed out. The end should be polished flat or a minute
bead formed on the end to provide an evenly illuminated cone of light. This is harder than it sounds but once done
the result is a fine point source. A
laser is not needed but is very convenient and probably easier and cheaper to
obtain than a good narrow collimated beam of white light these days.
The Laser Diode
The modern
Laser Diode without a collimating lens provides a possible ‘pinhole’
source. See Laser Diode.
Some Examples
In both of these examples a Laser beam shines
on a small steel ball. The ball should be
polished to ensure an even spread of light emanating from the virtual
pinhole. The laser beam can simply
shine through the centre of the glass in the primary since it need not be free
of aberrations until after it is reflected from the steel ball.
Here two different features are shown.
The laser beam can be off axis and the testing telescope in this case
may be less than full diameter as long as we assume a rotationally symmetric
test surface.
GOTO RONCHI INDEX