Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.448377
Title: Wafer-scale integration of semiconductor memory
Author: Aubusson, Russell Croston
Awarding Body: Middlesex Polytechnic
Current Institution: Middlesex University
Date of Award: 1979
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
This work is directed towards a study of full-slice or "wafer-scale integrated" - semiconductor memory. Previous approaches to full slice technology are studied and critically compared. It is shown that a fault-tolerant, fixed-interconnection approach offers many advantages; such a technique forms the basis of the experimental work. The disadvantages of the conventional technology are reviewed to illustrate the potential improvements in cost, packing density and reliability obtainable with wafer-scale integration (W.S.l). Iterative chip arrays are modelled by a pseudorandom fault distribution; algorithms to control the linking of adjacent good - chips into linear chains are proposed and investigated by computer simulation. It is demonstrated that long chains may be produced at practicable yield levels. The on-chip control circuitry and the external control electronics required to implement one particular algorithm are described in relation to a TTL simulation of an array of 4 X 4 integrated circuit chips. A layout of the on-chip control logic is shown to require (in 40 dynamic MOS circuitry) an area equivalent to ~250 shift register stages -a reasonable overhead on large memories. Structures are proposed to extend the fixed-interconnection, fault-tolerant concept to parallel/serial organised memory - covering RAM, ROM and Associative Memory applications requiring up to~ 2M bits of storage. Potential problem areas in implementing W.S.I are discussed and it is concluded that current technology is capable of manufacturing such devices. A detailed cost comparison of the conventional and W.S.I approaches to large serial memories illustrates the potential savings available with wafer-scale integration. The problem of gaining industrial acceptance for W.S.I is discussed in relation to known and anticipated views- of new technology. The thesis concludes with suggestions for further work in the general field of wafer-scale integration.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.448377  DOI: Not available
Share: