Re: [Rollei] focus shift

[Feli di Giorgio <feli2 >]

Thanks Richard for the incredible detailed answer.

cheers,

Feli


On Jan 1, 2004, at 9:24 PM, Richard Knoppow wrote:
>   Focus shift is due to spherical aberration. while good
> quality lenses are highly corrected for spherical some still
> have a residual of it. Spherical aberration is a fundamental
> property of spherical surfaces. The effective focal length
> of a lens with spherical surfaces varies continuously from
> the center to the edge of the lens. The focal length is
> greatest for light entering near the center of the lens,
> what is called the paraxial region, and gets shorter as one
> moves toward the edge. A hand magnifying glass will
> demonstrate the effect of spherical. Most of these
> magnifiers are bi-convex. If one focuses to get an image
> with one the image will be blurred and have no definite
> point of sharp focus. In fact, it is an overlay of many
> images focused at a coninuum of distances from the lens.
> Spherical is corrected by combining positive and negative
> spherical surfaces. A corrected lens has a plane from which
> the image appears to originate called the second principle
> plane. In fact, its not a plane but a curved surface of
> approximately paraboloidal shape. A lens can be corrected
> for spherical to any desired degree but the more highly it
> is corrected the more complex the lens must be. The use of
> even a single aspherical surface can reduce the spherical
> aberration very considerably.
>   The correction for spherical in practical lenses is done
> so that there is no spherical in the paraxial region and the
> spherical is again brought to a minimum at some point toward
> the edge, ususally at the edge. This leaves a "zone" of the
> lens, typically at 0.707 the diameter, where the residual
> spherical is at a maximum. This is called zonal spherical
> aberration and is the primary cause of focus shift. When the
> lens is used wide open a sharp image is projected by the
> center and edge of the lens, but there is also another range
> of images focused at a closer (usually) distance by the zone
> of the lens. If this residual spherical is large enough the
> image will appear to be slightly soft. When wide open the
> best sharpness, especially when judged by image contrast,
> will appear to be somewhat closer to the lens, a compromise
> distance between the center and edge distance and the zone
> distance. As the lens is stopped down the zonal spherical
> has less contribution to the image so the point of best
> focus will seem to move further away from the lens to the
> paraxial focus position.
>   Zonal spherical is minimised by, of course, improved
> spherical correction, but also by the adjustment of the
> distance the point where the aberration curve crosses the
> zero point in relation to the lens diameter. A lens
> corrected this way will ave less focus shift although it may
> not be a sharp as one corrected in the usual way. The
> compromised correction is often used for lenses which are
> used in rangefinder cameras since the range finder can not
> correct for focus shift. The slight loss of maximum
> sharpness is justified by the position of best focus always
> being near that indicated by the rangefinder. This is also a
> requirement for use on a TLR camera.
>   Occasionally, lenses are compromised in the other
> direction. This results in a smaller blur spot when the lens
> is used stopped down. It can be used for lenses which are
> usually used stopped down and where resolution is important.
>   Meniscus lenses like the Goerz Dagor have inherently have
> large zonal spherical residuals, so they exhibit focus
> shift. They must be focused when stopped down enough to
> eliminate most of the zonal aberration and resulting focus
> shift. for a Dagor this is around f/11.
>   In general, the kinds of lenses used in rangefinder and
> TLR cameras do not have lot of zonal spherical. Tessars can
> be designed to have good spherical correction and lenses
> derived from the double Gauss type, like the
> Planar-Biotar-Opic type can be very well corrected for
> spherical even without aspherical surfaces. That is one
> reason that these lenses are popular for rangefinder cameras
> and TLR's. Modern lenses wtih aspherical surfaces can have
> virtually perfect correction for spherial aberration.
>   BTW, since lens aberrations are interacing some residual
> spherical is often left because it helps with correction of
> coma.
>
> ---
> Richard Knoppow
> Los Angeles, CA, USA
> dickburk  
>
>
_______________________________________________________
They that can give up essential liberty to obtain a little temporary 
safety
deserve neither liberty nor safety. 				-Benjamin Franklin

feli2  

------------------------------