Publications

Background paper for 1965 Conference on Electronics Design (pp1-3)

OCRd Copy

ELECTRONICS DESIGN

H. V. BECK, M.A., B.SC., M.I.E.E.

 

The flexibility of electronic techniques has resulted in their permeation, in a remarkably short time, of most spheres of human activity. Research in both the arts and sciences has benefited, services such as entertainment and communication have been transformed, transport and defence facilities are highly dependent on electronics, and a significant and growing role is being played in fields such as medicine, materials production and administration among many others. The design of electronic equipment thus takes place in many different types of organisation and for a great variety of purposes, and this has to be kept clearly in mind when discussing electronics design. The outlook of a designer of electronic equipment for general sale differs considerably from that of one designing for use within his own organisation. Designing for quantity production is very different from designing a single working unit. Within one organisation, the approach of one design group may be quite distinct from that of another covering a different field of application or based on another electronic technique. An engineer designing general-purpose electronic equipment needs quite different attributes and would express quite different views from one concerned with apparatus for a specific application. The profusion and divergence of opinion on the design of electronic equipment resulting from the different objectives and facilities make the study of basic design processes very difficult. At the same time the level of electronics design activity is so high and its importance in such areas as trade, public services and research is so great that an understanding of these processes and their relationship to the fields of application and types of design organisation would be very rewarding. What is design? One of the most difficult problems is to say precisely what we mean by 'design'. We can get a measure of the task by taking a few statements at random: (a) 'Our designers should be concerned with cost as well as appearance. . . .' -extract from speech by the Duke of Edinburgh, May 1964 (b)  'A multivibrator may be designed as follows . . .' (c)  'There is really no difference between design and development' (d)  'The original design came out in 1948, and since then we have developed it a good deal' (e)  'When we have developed a working model, we pass it to the drawing office who carry out the design work' (f)  'At the conclusion of the design process the final result must be carefully defined so that it can be built—this is the function of layout draughtsmen and detail draughtsmen' - extract from 'An outline of the requirements for a computer-aided design system', by S. A. Coons.

(g)  'Mechanical engineering design is the use of scientific principles, technical information and imagination in the definition of a mechanical structure, machine or system to perform prespecified functions with maximum economy and efficiency' - Feilden Report. Clearly, no single definition will cover all views on design, and the only reasonable course is to formulate and promulgate a working definition that reflects the particular approach to electronics design appropriate to the study, while recognising the possible validity of other definitions based on different approaches. The working definition adopted by the committee with respect to electronic equipment is as follows: Design is the process of establishing relationships between all relevant characteristics of an equipment. We can say, quite generally, that any item of electronic equipment has a set of characteristics such as: cost performance appearance size reliability ease of use weight adaptability maintainability novelty versatility developability The working definition of design is based on the view that the essence of designing electronic equipment is the establishment of relationships between all these characteristics, taking into account the environment in which the design is undertaken. The latter includes such factors as the time available for design, the quantity required and the availability of engineers, materials and production resources. Thus the designer integrates or synthesises all the characteristics, giving different degrees of emphasis according to the circumstances in which the equipment is being designed and the use to which it is to be put. If the equipment is being made for general sale, the appearance, cost and quantity required are usually of great importance, whereas if an item with the same performance is being designed for use within an organisation, the quantity required is usually small, and cost and appearance may be subordinate to most other characteristics. For a particular research project, technical performance and the time taken to design and make the equipment may be the overriding factors, while in another case reliability and ease of maintenance may be paramount. For a piece of equipment for use on a production line, reliability and ease of use may take precedence over cost, size and weight. These varying degrees of emphasis and the different environments in which evolution of an equipment takes place give rise to much of the confusion over the word 'design'. If, however, we remember the broader picture implicit in the working definition, we are a good way towards putting the discussion of design on a rational basis.

Electronics & Power December 1964

443

 

Basic processes We are probably at the stage in the understanding of design corresponding to that of the Greeks in the understanding of matter. Information on electronics design practice can be collected fairly readily, but what are the basic elements in the design process? This is the largest area of unknown in the committee's field of inquiry and is likely to require a considerable amount of creative thought before an adequate set of concepts can be formed. Success in this area would probably transform the whole of the design field, from the training of designers to the practice of design. The design of electronic equipment is as good a field, and perhaps the best, in which to carry out a fundamental study of this sort. The large variations in design practice resulting from the flexibility of electronic techniques suggest that the outcome of the study could be applied selectively to any field of design. Decision mechanisms in the various design and user environments will also be investigated. Some results of such studies in the mechanical engineering field have been published, and some concepts, such as tractability and critical decisions in design, have been evolved. A great deal more effort is needed, with electronics design particularly in mind. Organisational procedures Another source of confusion over the interpretation of the word 'design' arises from its identification with a particular stage or area of the organisation set up to produce the equipment. Since the organisation and its attendant procedure vary from one company or department to another, 'design' takes on a variety of meanings. It would be very illuminating to study and classify the various design procedures adopted by different organisations. A company brought up in the mechanical engineering tradition and adding electronics to its activities will probably impress mechanical design procedures on the electronics design activity. The drawing office will loom large in the course of design, and indeed control of design may be through a designer draughtsman (see e above). In other companies, the drawing office will come in at a later stage when the basic design has been completed (see f above). Organisations such as one designing a single piece of equipment for internal use will not go through successive prototype stages. In others, design is not regarded as complete until 50 models have been produced. Inquiry into practice of this sort will enable us to determine if there is any correlation between procedure and the mixture of basic elements of the design process. The number of stages involved in the evolution of even a simple piece of electronic equipment is fairly large. A procedural list drawn up many years ago is given in Table 1. It was compiled by thinking back on what had taken place, without conscious organisation, during the evolution of a particular range of electronic equipment. The list, as well as indicating a procedure, delineated the responsibilities of an electronics design engineer in a particular design and user environment; but where in the Table does design begin and end and what is the essential function of the designer? Clearly, from the Table, the designer is responsible for activities that are not by any stretch of the imagination concerned with design as such. Should he be? Table 1. Example of organisational procedure initial work review literature choose method make initial calculations investigate doubtful points make detailed calculations experimental laboratory model prepare circuit diagrams specify special components prepare rough layout diagrams supervise work in shops assemble and wire carry out experimental investigation make further calculations

experimental prototype informal discussion with drawing office revise circuit diagram revise specification of special components order components and deal with any complications arising repare detailed layout diagram supervise work in shops assemble and wire carry out experimental investigation make further calculations modify to improve performance, ease production etc. investigate likely production variations write customers' instructions and dispatch equipment discuss equipment with customer modify equipment to meet points raised by customer test, or return to customer for test, if necessary production prototype detailed discussion with drawing office revise circuit diagram if necessary compile list of components for specification order components and deal with matters arising collaborate with drawing office during drawing out of    mechanical details make wiring diagram or check diagrams drawn up in shop supervise wiring and assembly in shops test completed equipment carry out further investigation of likely production    variations supervise modifications to equipment and drawings apply special tests (e.g. shock tests) that could not be    applied before supervise modifications to equipment and drawings as a    result of above revise customer's instructions and dispatch discuss with customer supervise modifications to equipment and drawings to meet customer's points production equipments order long-delivery components for first batch and deal    with matters arising deal with difficulties arising in production of first batch draw up works test instructions supervise testing of first batch revise works test instructions if necessary draw up maintenance instructions revise customer's instructions if necessary Some of the questions the committee will seek to answer in connection with design processes and organisational procedures are: • What mental processes and organisational procedures are    involved in designing a piece of electronic equipment? • Do these processes differ according to the type of     electronic equipment, the field of application (e.g. military     versus civil) or the technique used, (e.g. analogue versus     digital)? •  Could we classify types of design by the processes and     procedures involved? • How do the design processes and procedures for     electronic equipment differ from those for mechanical     equipment or, say, electrical machines? Training of designers Detailed investigation of the way in which design proceeds is urgently needed to provide a basis for the training of designers. The ability to design in the broader sense of establishing satisfactory relationships between a large number of characteristics appears to be innate in only a small percentage of engineers and fears have been expressed that such ability is often suppressed in the course of training. The analytical

444

Electronics & Power December 1964

 

approach to simplified, artificial problems is said to alienate the student from the process of synthesising in real-life situations. If this is true, application of the knowledge of design processes should do much to prevent this loss of talent. If innate design ability is insufficient to meet needs, we must ask ourselves whether we can train those who do not possess it. There are at least three schools of thought here. One is that designers are born and cannot be made. This is exemplified by Issigonis's remark, quoted in the Feilden report, that 'the craving to design exists from a very early age if it exists at all'. Another, based on the belief that design can be carried out by a logical process and does not depend on inspiration (just as some scientific discoveries and tech­nological inventions have come from systematic investigation), assumes that design can be taught. Yet another school takes the view that design is really a process of trying out innumerable alternatives until one is found that is satisfactory and that what is really needed in training a designer is the development of an ability to persevere in the face of unusual and prolonged adversity. Some teaching difficulties Several attempts have been made to teach electronics design but none, so far as is known, has been made from the comprehensive point of view of synthesising a large number of characteristics. For example, considerations of cost, ease of use and reliability are usually left out of students' design projects. Many of the difficulties attendant on these projects arise from problems of assessing the results. Others arise from the difficulty of finding the right kind of teacher. A teacher with no design experience may not be able to teach design, but it is open to question whether designers themselves make good teachers of design. Clearly, it is important to see that students' design ability, innate or otherwise, is fostered in our educational establishments. The design link between research and production is becoming increasingly important. It is often suggested that our national character is unsuited to the design of general-purpose electronic equipment for large-scale production and that it would be better for us to look for special-purpose small quantity markets as more appropriate to our human re­sources. The truth of this is hard to discover. What can be said with certainty is that a better knowledge of design processes would enable us to determine our trading and training policies much more surely. Such knowledge would also benefit research projects requiring the design of a great deal of special equipment. With a knowledge of design processes and procedures we can ask the following questions with much more hope than hitherto of obtaining a useful answer: Who make good designers of electronic equipment? What training can and should be given to an  electronics design engineer? Who make good electronics design teachers? Should training be given at a university, at a college of advance technology or in industry? Should a national body be set up to give confidential and impartial assessments of students' design projects? Are designers' needs being met? In addition to seeking an understanding of the design process, so that improvements can be made in training techniques, it is intended to look into the possibility of helping the existing designers of electronic equipment. It is hoped that designers will tell us their needs and whether they arc being met. Are there, for example, any categories of information  that would be of assistance but are not at present forthcoming?

The information aspect of design is a very important part of the concept that design takes place by a series of decisions, each of which is considerably influenced by the information available at the time. This has particular relevance to electronics design, because of the exceptionally large number of directions from which to choose in any given design situation. The role of technical journals On the purely technical side, the Services have been responsible for much information on valves and components. Detailed information on semiconductor devices has come direct from device manufacturers. Analogous information on circuit performance, however, leaves much to be desired. Here, technical journals could do much to see that either the specific application of a circuit is described in detail, so that a designer can judge reliably whether it is relevant to his own situation, or that performance characteristics in practical conditions are given. Other categories of information useful to the designer are related to the many characteristics of an equipment. A few of the topics that might repay consideration are: studies of ergonomic principles and practice the effect of choice of system or circuit on ease of use environmental conditions likely to be met in practice and the means of catering for them measurement of reliability and its effect on design criteria for the choice of constructional form, taking into account any economic, technical and maintenance factors manufacturing techniques of particular relevance to electronic equipment techniques for controlling and predicting cost methods of reducing size and weight relationship of appearance to commercial performance. A major difficulty in solving these problems is that there is no independent body that can look into them, especially those that are not purely technical. Many organisations have issued data sheets or guides to designers as a means of providing information of this sort, but they arc usually specific to a particular design and user environment. A more comprehensive effort on a national scale would possibly be of benefit to all electronics designers. Some questions for the electronics designer In considering, therefore, what can be done to help electronics design engineers, we might ask the following: ·    What types of information needed for electronics design are not at present satisfactorily covered? ·    Are any improvements needed in existing sources of information? ·    Is a national body needed to obtain and disseminate information required by electronics design engineers? ·    Is standardisation of electronics design information desirable and can it be achieved? ·    What is the best environment for a given type of design activity? There seems little doubt that computers will ultimately be able to take over a good deal of the detailed work involved in the design of standard circuits and their layout and wiring, but will they be applicable to design in the broader sense of establishing relationships between all the relevant characteristics? We are back at our starting point, for attempts to apply computers in this way will underline the need to understand basic design processes.

Electronics & Power December 1964

445

Publications