Views on the photographic universe by Erwin Puts

The myth of the glass laboratory (april 7, 2009)

The design process of a lens often starts with a number of restrictions like size, diameter of front lens, length of lens, and number of glass elements. Then you try to create a configuration (surface shapes, distance between lens elements, location of stop) that brings the required correction of aberrations. This analysis is based on ray tracing, related to the power of a lens element and its diameter. Only in a later stage the required properties of the glass (index of refraction and dispersion) are added to the design. In modern times you optimise the lens to the level you want and the program will calculate the ideal glass properties. These theoretical glasses are then mapped onto the factual glass map and you have to choose the glass that is close to the ideal construct. Then you need to re-optimise the lens to stay within the required merit function. In many cases you could have to reduce or change the goals to accommodate the glasses found in the catalogue.

The Leitz glass laboratory.

In the past the same process was in place, but with much more laborious calculations and a more restricted glass catalogue. The designer was tempted to create a special glass in order to reach the goals then to settle for a glass with different properties, that required a renewal of the design process. And when finding the limits of lens design, one could experiment with exotic glass types to see how far one could go and offer suggestions to glass makers to melt this specific glass.
Here lie the origins of the famous, but short lived Leitz glass laboratory. When more glass types became available and automatic calculations were added to the design process, the role of the own glass lab was reduced and later even made obsolete. Of course it would be nice if a designer could use extremely specific glass properties to create an even better lens. But today the catalogues form Schott, Hoya and Ohara offer a wide selection of glass and the design process is highly automated.
It is now more important to study the properties of a glass and employ specific characteristics from this glass in unusual ways to get the required performance.
Leitz made the existence of the proprietary glass lab as an argument for the superior quality of their lenses and in a limited number of cases this was true. But progress of the glass companies made this argument obsolete.
It does linger on however in books and articles and the common knowledge of the Leica world.
The modern Leica M lenses are certainly world class lenses and all glasses employed are made by the well known companies. In a few cases Leica ave the required specs and the glass company melted a glass to conform to these specifications.

The modern way.

The quality of the Leica lenses is derived form the expert choice of glass types, the polishing and centring of the glass surfaces, the treatment of the glass and the careful assembly. The design process is the main issue as is the expert handling of the glass. Leica has invested millions in new equipment to create aspherical surfaces, to minimise tolerances, to get the best surface treatment and so on.

The Cosina glass laboratory.

Recently a claim was made that Cosina owned a proprietary glass melting factory where glass is being made for use in the Voigtlander line of lenses. The corollary of this claim was the inference that because of this fact (creating and melting their own glass) the Voigtlander lenses were better than others who had to use glass form the main manufacturers. The basis argument is the one I have sketched above in the paragraph about the Leitz glass lab.
It is indeed true that Cosina makes their own glass. They do this in small amounts, just for their own demand. It would be not possible to create the same range of glass as is available on the world market, so the glass that Cosina melts is of a limited range. If it would be the case that the Cosina designers are restricted to use the glass available, their design choices would be biased in this direction. Some designs of Voigtlander lenses look really daring and this solution may be needed to design around the glass limits. Better results would be possible if the glasses on the world market could be integrated in the design.

The fact that a company makes their own glass may not be an advantage, but could be a limiting factor for the design.

It could be quite cost effective as the manufacture of this glass is freed form the overhead and profit margins of the main companies.

The conclusion

Cosina does indeed own a glass manufacture plant that produces some glass types for their own consumption. There is no published catalogue of the glass types that are manufactured so no one outside the company can assess the characteristics and quality of the glass.
There is no theoretical or inherent advantage nowadays in using glass from their own glass manufacture and in fact it may be a limiting factor for getting the best possible performance.
Leica and Zeiss are good examples of companies who work with glass form the main companies in the world and can deliver outstandingly good lenses. The optical design process and the selection of glass and the handling of the glass are more important that the fact that a company makes its own glass.