FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1962
1962 - 2395.PDF
Texas Instruments semiconductor network, measuring ±in x tin x 0.035/n, with its equivalent in miniature components. The network is seen under test on page 632 AVIATION ELECTRONICS... who indicated what could be done and the electronics engineer who determined how the discoveries could be applied to produce the electronic effects required for a particular circuit. This is the point at which solid-state or molecular electronics (molectronics) emerged as a study in itself. It offers spectacular advances in several direc tions, but it is still in its early stages, perhaps five or even ten years from large-scale production. But a large number of companies are engaged in research and some elements are in pilot production. In Britain, the Royal Radar Establishment is conducting basic research on a scale sufficient to keep abreast with developments elsewhere. Commercial concerns active in the field include Standard Telephones and Cables, Texas Instruments, Marconi, Mullard, General Electric Co, English Electric, Ferranti, Elliott Brothers (London) and Plessey. Transistor manufacture, though closely controlled has in the past been a relatively hit-or-miss process. It is still common to see large numbers of transistors, from the same production line, being indi vidually tested and then classified as several different types accord ing to the actual performance achieved. Three or four different transistors may emerge from what is intended to be a precision production process. Some circuits call for matched pairs of transis tors but, because of the spread of characteristics, these are in many cases virtually impossible to select. New transistor types and tech niques are still being developed and present power and frequency limitations will in due course be overcome. Consistency of manu facture is just one of the benefits to be expected with the new solid- state processes. There are, in fact, many different approaches to the problem, and individual companies are trying a great variety of techniques in solid-state circuitry. Some organizations, such as the transistor and capacitor divisions of Standard Telephones and Cables, Texas Instruments Ltd and, shortly, a British subsidiary of Fairchild Semiconductors, are marketing microcircuits, but only in small quantities for more or less experimental purposes. The new tech niques are in their very early stages. It is probable that a number of the companies at present doing research will not continue to the production stage and will rely on bought-out circuit elements when they manufacture microminiature equipment. Research is necessarily concentrated at present on certain parts of the electronics field. Tiny transistors have been produced, but the related techniques required for interconnection, mounting and packaging have not progressed nearly as far. Microminiature com ponents still have to be attached to relatively standard types of wiring, so that the interconnection is in some cases larger than the component itself. Only a relatively limited number of components are at present combined in a single blob of material. When a complete radio, for example, is ultimately made in micro miniature form, the designers will have to decide whether it is to be mounted within the aerial or the pilot's control panel—and the aerial, control knobs and cables will be by far the largest portions of the installation. But that stage is still years away. Small size creates its own peculiar problems. Manufacturers are working in deposited layers and films measured in angstroms and microns, and in components a few thousandths of an inch thick. Sheets of transistors have to be cut up by ultrasonic shaking; termi nals have to be attached under microscopes by a technique which has been nicknamed "micro blacksmithing." Tiny transistors and other components are made up into circuits which are supported on 634 FLIGHT International, 18 October 1962 fine gold wires and encapsulated in standard transistor casings, with most of the internal space still left unused. Obviously, this kind of circuitry can never be inspected and main tained in the conventional sense of the word. Most of it is sealed by the very process of its formation and, if it fails, it must simply be discarded and replaced. Yet, because of its homogeneity, failure is much less likely. The components take so little space in themselves that redundant units, or integral spares, can be incorporated on a quite extensive scale without materially affecting the size of the finished product. A five-stage IF amplifier strip can be made up of six actual stages so arranged that any stage which fails will automatically be shorted out, leaving the original amplification level intact. The manufac turer and customer between them will have to decide just how much of a complete equipment will be sealed into each module and therefore ultimately discarded without possibility of repair. If the full potential of redundancy is exploited, it may become possible for the manufacturer to guarantee that a given circuit module, or even a complete equipment, will continue to function regardless of the failure of any individual components for a definite period, say five years. After that, the module would be discarded and re placed. A more advanced technique of reliability engineering and redun dancy is therefore emerging as an integral part of microminiaturi zation. Because of their very small dimensions, the number of components does not materially affect black-box size, but some quite clear-cut decisions must be taken concerning lifing and re placement. At the same time, many of the problems formerly associated with maintenance and overhaul will disappear, if only because the performance of the equipment can be much more closely controlled and guaranteed by the manufacturer and because much less will have to be done to the equipment in service. On the whole, it appears that microminiature circuits will be more tolerant to temperatures and require much lower powers than does present transistorized equipment, although power output of between five and ten Watts at very high frequencies is already being achieved experimentally. As well as "micro blacksmithing," the term "micro- power working" is coming into use and designers are talking in terms of nanoAmperes. A brief description of the manufacture of planar transistors, an interim stage between micromodules and true solid-state cir cuitry, will indicate the scale on which this type of work is done. The planar transistor is formed by treating thin slices of silicon, cut from bars, to form the active elements within the material itself, in layers of controlled depth and horizontal extent. The silicon sheet is placed in a special oven and a film of silicon dioxide is thermally grown to protect the smooth surface. Because the dimensions are so small, extreme smoothness is important. The sheet is then coated with a layer of photosensitive lacquer, covered by a photo graphic mask, exposed to ultraviolet light and photographically developed. This process forms spaces, called windows, through which silicon oxide can be etched away to reproduce the original pattern of the mask. The silicon dioxide is now itself used as a Placing a group of silicon slices in an oxidization furnace for one of the etching or smoothing stages in solid-circuit
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
