Musings on APOs
by Roland Christen
"I've been told by optical engineers whom I trust that an apochromat
brings twowavelengths to focus at the same distance rather than just one like an
achromat does (and none for a singlet)."
"After reading David Knisely's post and thinking about it I realized the
below should have said that the apochromat brings three wavelengths to
focus together and the achromat just two."
May I point out that this original definition: "bringing 3 wavelengths to a
common focus and be corrected for spherical aberration at two wavelengths" came
about historically because the very first apochromats actually did this -
thus the definition was made to fit the example. These were the Zeiss dense flint
apochromats (later re-invented by Mike Simmons and called the Good Glass
Apochromat). While these lenses did indeed meet the above criteria, they did not
have very good color correction in the violet end of the spectrum, and were
also off more in the visual than a modern ED or Fluorite apo.
Enter the Fluorite doublet (Tak, Vixen). These lenses also have 3 color
crossings, however, one crossing is deep in the infrared where it really does no
good for the visual wavelengths, especially the blue-violet end. Between C and F
(red, yellow, green and blue-green), the color correction is much better than
the traditional dense flint apo. The spherical correction is only null at one
wavelength - usually at the green visual peak.
Lastly we have the ED/fluorite triplets, of which there are now several
makers. These lenses will typically have very low color error with 3 crossings at
the ends of the visual spectrum, but again only one spherical null in the
middle. The amount of sphero-chromatism at the ends of the spectrum will depend on
aperture, focal ratio, type of ED glass used (the lower the dispersion the
better) and the length of internal airgaps (small airgaps have no effect on
sphero-chromatism, very long airgaps can reduce it to almost zero).
In summary, the definition of Apo is not as simple as 3 color crossings and 2
spherical corrections. In fact, the best possible apo would ahve no color
crossings, but would exibit a straight line. A very shallow secondary spectrum
with only 2 crossings can have better overall color correction than one that
wiggles around 3 or more times. Finally, it is not the ED or Fluorite which
determines the overall correction, rather it is the mating element. A designer will
often not choose the mate for best color correction, rather he will choose
one which balances all aberrations for best overall image, or one that makes the
objective easier to manufacture.
The choice of mate is restricted in the real world, especially for astronomical
lenses, because glass manufacturers only pour a limited number of glasses
in larger sizes. A lot of the more desireable glass types have been
discontinued, or they are made in such small strip widths that they cannot be
used for astronomical objectives. Remelting and repressing small blocks leads to
undesireable internal properties like strain, and inhomogeneity.
Return to Roland Christen Images and Essays.