Foiling the electronic pirates with ever tinier circuits

Press, 23 August 1984, Page 21

Foiling the electronic pirates with ever tinier circuits

If New Zealand is unlikely to establish a Silicon Valley, it cannot afford to stand aloof from developments in micro-electronics. In this, and three subsequent articles, ERIC BEARDSLEY, information officer, University of Canterbury, outlines how young engineers are being trained in this new high technology field and the applications of research at Ram in both industry and agriculture.

New Zealand cannot hope to enter the high technology world of micro-electronics without a pool of highly-skilled engineers. Nor can the engineering profession provide advice on the technology without first-hand experience. That is why a special course — the most comprehensive of its kind in the country — for six engineering graduates at the University of Canterbury is crucial to success in the headlong rush to miniaturisation, which has become essential in the design of today’s high-perform-ance electronic systems. Next year, the physics and engineering laboratory of the Department of Scientific and Industrial Research at Gracefield will commission a laboratory (known as a silicon foundry) capable of producing Veiy Large Scale Integrated circuits. In the meantime, the expertise to deal with this extraordinarily tiny but complex world is being developed in the electrical and electronic engineering department of the School of Engineering at Ham. The six graduate students under the direction of Mr L. N. M. Edward, a senior lecturer in electronics, are learning to master the revolutionary changes that have taken place in integrated circuit engineering since the introduction in 1980 of design methodology and processes which make full economic custom design possible. Chips produced in this way are incredibly tiny, complex, and powerful, and they are not more expensive than those they replace. The simplest integrated circuit contains an interconnected array of active and passive elements integrated within a single chip of semiconductor and capable of performing at least one complete electronic circuit function. A personal computer uses one or more printed circuit boards carrying perhaps 100 or more integrated circuits. They are relatively expensive to make, bulky (by the latest standards), quite heavy users of power, and they may require special cooling. But their major disadvantage is that they may readily be copied and pirated. Because of this absence of pro-

duct security, general purpose micro-processor chips are far from being the final solution to all needs that they were once thought to be; and, when fast, real-time signal processing is required, general purpose chips lack the necessary performance. But with the design methodology the students are learning, it is possible to embed in a single silicon chip the equivalent of several hundred thousand distinguishable circuit elements. The design of these Very Large Scale Integrated circuits (they are designated V.L.S.I. circuits) involves many complex decisions which it is feasible to marshal into the form of computer programs for the department’s VAX computer. Mr Edward says that because of time constraints the students are actually designing L.S.I. (Large Scale Integrated) circuits rather than V.L.S.I. chips, but they are using similar design methods. Even so, each chip on which they are working has the equivalent of between 400 and 4000 transistors. V.L.S.I. chips may contain at least a million transistors. What distinguishes V.L.S.I. circuits, however, is their functional and design complexity rather than the transistor count. Compressing the circuit density in this way reduces costs. Even though the complexity and performance of the chip might have been increased a hundredfold there is unlikely to have been an increase' in price. Power usage is also reduced. But of even greater importance to the manufacturer is another virtue: the small size and the complexity protect the design from piracy. Last year, the first of the special course, four students completed designs which were fabricated by the Commonwealth Scientific and Industrial Research Organisation in Australia in its multi-project chip program. The four designs took a mere nine square millimetres of silicon on the 34.8 square millimetre chip. And, says Mr Edward, three of the four designs worked well. This year, the students are trying to complete their designs by midSeptember so they can be returned

from Australia about the end of the year. It is expected that next year the student designs will be fabricated in the Gracefield “foundry.” It was the need for New Zealand engineers to have access to local fabrication which prompted the D.S.I.R. to establish the new laboratory, an N.M.O.S. (Negative-type-cnannel metal oxide semiconductor transistor) fabrication line. It will become the first-line fabricator for all New Zealand N.M.O.S. designs. Later, a C.M.O.S. process will be commissioned which involves combining P.M.O.S. (Positive-type-chan-nel metal oxide semi-conductor) and N.M.O.S. transistors in the same integrated circuit chip to form a complementary-type-chan-nel M.O.S. system. A powerful computer with graphics capability is essential for designing V.L.S.I. chips. Their operation must be simulated in the computer because it is impossible to build a prototype to determine whther the design will work properly. With the assistance of the electronics industry, Mr Edward was responsible for completing a micro-electronics laboratory at Ham in 1979 for teaching and research in thick film technology. It uses silk screen printing to form circuits and is an advance on conventional printed-circuit boards, it complements V.L.S.I. technology well. “Thick film is an almost ideal packaging technique for integrated circuits,” Mr Edward says. Why should New Zealand bother about trying to maintain a place in the technology? Mr Edward says that reduced assembly cost, smaller size, reduced inventories, more secure semi-conductor pro•curement, lower power dissipation and cheaper power supplies, less cooling and simpler packaging, greater reliability and market acceptability make a very strong case for maximising the customdesigned content of New Zealand electronic products. “Our world-wide competitors are using these approaches and it seems clear that we must also if we are to continue to do business,” he says.