‘Straight-sighted’ Binocular Prism Systems,
   with little or no offset in the optical axis.
Geradsichtige Feldstecher-Prismenumkehrsysteme
   ohne oder mit nur geringem Achversatz.
by R. Liebmann; Carl Zeiss, Oberkochen
from Optik, Zeitschrift fuer Licht und Elektronenoptik
   (Journal for Light and Electron Optics)
Special Reprint, Optik 26 (1967), pp264-272
Translated by Ilse Roberts and Peter Abrahams
Text of this translation, copyright 1996


The best optical characteristics do nothing for a field glass, if it is 
so large and heavy, that it is not even taken on a hike or to a sports 
event.  Besides faultless optics, lightness and handiness are some of the 
most important requirements of a field glass.  A very important influence on 
these last two qualities is the prism erecting system.  For a long time, the 
outside appearance of most field glasses was determined by their Porro I 
prisms.  The desire for a more compact, lighter and more elegant 
construction has, in the course of the development, elevated other prism 
reversal systems to prominence.

To begin with, here are some criteria which shall serve to evaluate 
different prism systems.
1. Number of reflections.
2. Number of Prisms
3. Number of surfaces with optical effect (surfaces with total reflection + 
silvered surfaces + glass/air passage surfaces + cemented surfaces)
4. Number of silvered surfaces (important because of lessened reflectivity 
and higher price, compared to surfaces of total reflection)
5. Number of air spaces, and angle of light path through spaces
6. Length of the path of light through glass
7. Deviation of the angle of the ray between incoming and exiting bundles, 
(capacity of varying from zero to a few millimeters is desirable)
8. Length, measured on axis of light path
9. Minimizing instrument length (determined by the overall length of the ray 
passage and the length of the path through the prisms.  A ray that is 
‘coiled up’ in the glass to shorten the instrument results in heavy prism 
systems, less ‘coiling’ of the ray in the prisms causes long field glasses.  
Desirable is a medium measure.)
10. Type of glass (determines the critical angle of total reflection, partly 
by the f ratio of the objective and partly by the geometry of the prism 
system. High index of refraction glasses such as F2 are heavier, more costly 
to work and easier to damage than lower index BK 7.)

Criteria 1 to 9 and perhaps 10 can be expressed in numeric values when 
comparing normal systems, as was done (with the exception of number 10) in 
the chart at the end of this report.  Of course, each number value would 
have to be multiplied by a factor accounting for the importance of the 
characteristic in the framework of the total evaluation.  But one shouldn't 
proceed so schematically since many imponderables which cannot be expressed 
in numbers are contained in the total evaluation. Some of the criteria that 
are extremely important, but hard to express in quantitative form, are:

11. Slimness, smallness and weight of the system (best judged by a visual 
comparison)
12. Cost of the construction (number and size of the surfaces to be worked 
on, requirements for precision of the surfaces and angles (extreme in roof 
prisms), the type of glass, the methods of fabrication, possible mounting 
devices inside the housing, etc.)

The various forms of the prism systems and reflections within them:

A side reversal system, where the chief ray stays in one plane and 
reflects an odd number of times, can be taken as an example.  If one then 
replaces one of the reflecting surfaces by a roof ridge, the number of 
reflections increases by one and the result is an erecting system.  Four or 
six reflection systems are most practical.  ‘Straight-through’ prism systems 
with less than 4 reflections are not possible and those with more than 6 
bring disadvantages with no advantages.  At 6 reflections, a larger variety 
of constructions are possible, which does not necessarily a better 
instrument.

Because of the required precision for the angle between the two planes of 
a roof ridge, and because a very sharp edge is required, roof prisms are 
more expensive to manufacture than other types.  Also, light is diffracted 
at the roof ridge, though this is only visible when observing very bright, 
point sources of light (bright stars).  When the narrow tolerance for the 
angle between the roof surfaces is exceeded, double images become visible.  
It is therefore interesting to contemplate whether straight through erecting 
systems without roof prisms are possible or efficient in use.  One solution 
is to place one ‘side reversal’ system behind another, offset by 90 degrees.  
Since a side reversing straight sighted prism has to have at least 3 
surfaces, the result would be 6 surfaces.  It would be simpler to use one 
side reversal system and to replace any of the planes by a roof ridge.  This 
system has only has 4 reflecting surfaces, and is shorter and simpler.  In 
addition, it cantilevers (part of it projects over the other), in only one 
plane (if at all), while the system consisting of two prisms cantilevers in 
two planes offset by 90 degrees.  Thus, two side reversal systems, one 
behind the other, does not lead to advantageous solutions.  Another 
‘straight-sighted’ erecting system, without roof ridge and without axis 
deviation, is shown in a patent drawing from Wild.  Without positioning two 
systems behind each other and without roof ridge, an image reversal was 
achieved.  But this system cantilevers substantially in two planes which are 
turned to each other by about 40 degrees, while the roof systems do this in 
only one plane.  The Wild prism system therefore cannot be used in slim 
housings.  The author has found neither in literature nor in his 
imagination, solutions without roof ridge, which approach these 
requirements, and therefore believes that there are none to be found.  Of 
course, the correctness of this opinion is not thereby proven.

The systems shown in the drawings are all shown with cylindrical ray 
bundles of 2 cm diameter for comparison.  This is reasonable for a first 
example; but it means a deviation from the real, often strongly constricted, 
ray passages and the simplification will have to be taken into account at 
the final selection and utilization.

The depicted prism systems are arranged as follows:
1.1 to 6.2 have 6 reflections with an overlapping or loop of the ray 
passage.
7.1 to 8.4 have 6 reflections but without a loop.
9.1.1 to 9.2.3. use 4 reflections (Abbe Prisms).

The Pechan system is by far the most advantageous of the first group (1.1 
to 6.2.)  It has, at reasonable ray passage, the most compact form, and has 
only 7 optical surfaces, compared to 8 to 11 for other designs.  These 
positive characteristics are because the maximum number of planes are used 
more than once, and from the especially favorable interplay of the two prism 
forms.  It may be mentioned that, as is well known, a Pechan system can be 
made of BK 7, if one arranges the angles a little differently and if one 
forgoes a large f ratio for the objectives.  When using Pechan systems one 
has to watch out for side pupils, which can appear because 
of steeply inclined incoming rays which are not totally reflected at the gap 
between the prisms, but go directly through both prisms.  The blind 
[aperture stop ?] in the gap has to be positioned so that such rays are 
caught in front of or behind the system by the housing wall or the stop.  
Next to the Pechan system, the best prisms in this group (4.4 and 5.3) are 
shown in the chart, for comparison.  It can be seen, that the designs of 
this group have a certain relation to the Pechan system.  They have two 
adjoining planes, in connection with specific positioning of the angles and 
the roof ridge.

Number 7.1 and 7.2 show the Uppendahl prism system.  The shortening of 
the instrument which can be achieved with it is slightly smaller than with 
the Pechan system (see chart).  A slight advantage is the fact that it has 
only 2 glass/air surfaces, compared to 4 in the Pechan system.  But it 
consists of 3 cemented prisms and therefore has 10 optical surfaces, 
compared to 7 for the Pechan system.  But 4 of these 10 planes require less 
precision in fabrication, because they are cemented.  The system cantilevers 
strongly to one side.  This cantilevering can be positioned towards the 
middle of the field glass housing, but requires a much more complicated 
housing than the Pechan system.  All in all, it doesn't seem to offer 
advantages over the Pechan system.

The Rosenhagen system is special, especially in the form of 8.1.  It does 
not cantilever over the square circumscribed by the bundle diameter.  But it 
is rather long, does not allow instrument shortening to any practical 
degree, consists of 3 prisms with two air gaps, and has two silvered 
surfaces.  Because of these disadvantages, it is generally not suitable for 
field glasses, except when a narrow bodied instrument is of great import.  
The variants 8.2 to 8.4 show shorter construction but greater width.  The 
form in 8.4 needs only one silvered surface.

Those reversal systems which have an air gap, can advantageously pass the 
light through this air gap at an incline.  Inclining angles (measured in the 
glass) of up to 25 degrees, even 30 degrees, are possible.  One gains a 
degree of freedom for more compact construction, which becomes clear when 
comparing the constructions of 8.1 and 8.3.  The resulting errors of 
correction become noticeable as a side shifting of the ray bundles of 
various colors.  Astigmatism and coma are without significance.  The color 
shifting should not be larger than 1/4 minutes.  The result is that one has 
to pay attention to small gaps at higher entrance angles (15-25 degrees), 
and to the quality of glass used.  It is often desirable to install an 
aperture stop in the gap.  If only one of the two ‘bordering planes’ is used 
twice, one can paint a stop onto the other plane.  Then a smaller gap width 
(about 0.1 mm) can be achieved.  It is less favorable if both bordering 
planes have to totally reflect , as in the Pechan system.  Then the stop has 
to stand free between both planes which results in a larger gap thickness 
(0.3-0.4mm).

The Abbe prism system (9.1.1. to 9.2.3.) is the same form as 9.1.1., 
similar to a common design, in regards to the angles (but not in regards to 
the diameters because of the ‘normalization’ of the cylindrical ray 
passage).  The five forms shown next attempted to achieve a further 
reduction in size by minor variation of the angles and by using the above 
mentioned principle of inclined passage through the air gap.  An inverse 
ratio exists between the axis deviation on the one hand and the housing 
length on the other, i.e. if one shrinks the length the axis deviation 
becomes greater and vice versa.  The optimum could perhaps be achieved, when 
the variation of the angles of the ‘real’ ray passages are the basic 
foundation.  This erecting system could achieve smaller size and lighter 
weight by using a plane mirror instead of the one prism.  ‘Side pupil’ 
[kidney bean ?] has to be carefully considered, since in this case, the 
shield for the entry surface of the Bauernfeind prism is cancelled by the 
redirecting prism in front of it.  The use of a mirror is only possible at 
perpendicular passages through the entrance plane of the Bauernfeind prism.

As the chart summary shows, the experiments confirm with a high degree of 
certainty, that for the required purpose, there are no better erecting prism 
systems than the Pechan system and the Abbe system.  The Pechan system 
allows a little shorter binocular, but near the prism is a thicker glass, 
compared to the Abbe system.  The Uppendahl system is also compact, but 
requires a somewhat higher quality optic as well as complicated housings.

Which of the two designs should be considered superior in the development 
of an actual instrument, can only be found by design work on the drawing 
table, where, considering the requirements of each scheme (method of 
focusing, adjustment etc) each prism system must be adjusted to the actual 
path of the reflected rays of light.

The author thanks Dr. Rantsch for suggesting this paper as well as Mr. 
Knutti, Mr. Leinhos and Mr. Weyrauch for useful suggestions
Received July 20, 1967


4