Focus Tools
32 JANUARY 2020 • SKY & TELESCOPE
There is, though it produces a complicated image to
interpret. The answer is to simply replace the long slits of the
Bahtinov mask with a mesh grid, eliminating the need for
any precision cutting while still producing a usable diffrac-
tion effect.
First, you need to fi nd an inexpensive mesh that’s rigid and
suffi ciently large enough to allow the fi rst-order diffraction
spots to appear between 10 to 30 arcseconds from the central
peak. The grating relationship, which allows for calculating the
diffraction angle based on the grating period and the wave-
length of illumination, predicts this spacing for mesh periods
of roughly 11 to 3.5 mm. An extensive search identifi ed an
ideal material known as plastic canvas that you can purchase
online or at most hobby and craft stores. Plastic canvas is nor-
mally used to provide a backing for needlework through which
brightly colored yarn can be strung to make simple pixel-like
decorative patterns. We found it to be nearly perfect for our
use — it’s rigid, accurate, inexpensive, and easy to cut.
The three spots at the end of the fi rst-order diffraction
image are angled in the classic 40°, 12.7-mm period Bahti-
nov pattern. The outer spots of the triplet, which arise from
the two angled areas of the mask, are separated by about 6
arcseconds. You should select a mesh period that will main-
tain at least this 6-arcsecond separation or a little
bit more with the mesh mask, keeping in mind
the pixel resolution at the effective focal length
of the system. The tilt angle should be lower than
40° for fi ner meshes than the classic mask, or else
the three spots appear too far apart for accurate
judgment. For a 5- to 7-count plastic mesh (5 to
7 periods per inch), the best tilt angle is roughly
10° to 20°.For example, with a 7-count mesh with two tilted regions
diverted to ±10°, the triplet is separated by about 11 arcsec-
onds, delivering two divided separations of 5.5 arcseconds
each. For a generic 1,000-mm focal length imaging system,
the fi rst-order triplet is separated by twice 26 micrometers.
Depending on the size of the camera’s pixels, this may be
tight or loose. A camera with 5-micron pixels has about 5
samples between the dots, and so should work fi ne.Making a Mesh Mask
Assembling a mesh focus mask couldn’t be simpler. Perforate
a thin sheet of cardboard with the major holes that com-
prise the three large openings of the mask. There’s no need
to be particularly neat cutting the edges of these large holes,
because they don’t contribute to the desired linear diffrac-
tion. However, the alignment of the three mesh screens is
important. The two smaller tilted regions must be mounted
at the same opposite angles. An error in angle will result in
a focus position with a systematic offset, producing slightly
out-of-focus results.
Using a mesh is not a kludge. They even have a built-in
self-test for alignment accuracy: The fi rst-order maxima in
the spots seen horizontally from the main star will imme-ASSEMBLE YOUR OWN MASK
(A) Plastic canvas can be found at most any craft
store or online and can be cut with scissors. (B)
Draw the pattern for your mask holes on a sheet
of cardboard, being careful to place the registra-
tion lines of the two smaller holes at precisely
the same angles relative to the larger aperture.
The edges of the apertures don’t need to be
particularly neat, since they don’t contribute to
the image shift. (C) Assemble your mask with the
mesh aligned to previously marked registration
lines. (D) Remember that it’s important that all
three mesh screens are precisely aligned to your
registration lines. (E) When completed, add three
holes to install long nylon screws that will secure
the mask to the front of your telescope or lens.uAnother excellent focus mask called the Oleshko two-
frequency mask uses a bisected aperture with 5- and
7-count mesh grids.uuAssembling an Oleshko mask is even easier than craft-
ing a Bahtinov. Simply abut 5- and 7-count plastic canvas
along a single registration line.A B