Proceedings of the Institution of Mechanical Engineers

This lecture comprises a survey and commentary on the use of welding for the construction and repair of marine machinery for main propelling and ancillary purposes. The history of welding in relation to marine engineering is traced from the hammer welding of boiler seams in 1870 up to the present day, when welds made by modern processes are accepted for highly stressed components.

Descriptions of welding processes are omitted since the lecture deals with welding applications rather than welding technique. Numerous examples are given and illustrated, and many of them relate to design details which are typical of good practice and are accepted by ship classification authorities.

The lecture is divided under two main heads, namely, “New Construction” and “Repairs”. A further subdivision is made under “New Construction”, so that boilers, pipe-work, machinery components, turbines, gearing, electrical propulsion, refrigerating plant, and dredging craft are dealt with in that order. Also included in this section are some remarks of general interest on the subject of residual stresses and their relief by thermal treatment.

Problems concerning the welding of alloy steels for gas turbines are discussed. There is need for research into the weldability of these steels, bearing in mind that not only the welds should be free from micro-cracks to start with, but they should also be equivalent to the parent metal in resistance to metallurgical and physical deterioration whilst operating at high temperatures over long periods.

A field of welding which is always of interest to marine engineers is in repair work. Unlike land plant for which complete resources for repairs are usually available, marine machinery may be involved in mishaps and breakdown on the high seas and in remote parts of the world when repair facilities might be primitive, or even non-existent. Many ships now carry electric welding equipment, and sea-going marine engineers include welding in their many manual accomplishments.

Several interesting examples of weld repairs are described, which in some cases have enabled ships to complete their voyages carrying valuable cargoes with a minimum of delay. Other examples are given which emphasize the discrimination that is necessary in deciding whether or not a welded repair is desirable. Unfortunate consequences have followed the ill-advised use of welding in some cases.

The flexural vibrations of the walls of thin cylinders are considered. In this type of vibration many forms of nodal pattern may exist owing to the combination of circumferential and axial nodes. Theoretical expressions are developed for the natural frequencies of cylinders with freely-supported and fixed ends and a comparison is made with the frequencies obtained experimentally.

In practice, the ends of cylinders are subjected to a certain degree of fixing by end-plates, flanges, etc., and the natural frequencies thus lie between the corresponding values for freely-supported and fixed ends. To make possible the estimation of such frequencies, a method is devised in which an equivalent wavelength factor is used. This factor represents the wavelength of the freely-supported cylinder that would have the same frequency as the cylinder under consideration when vibrating in the same mode. The results of experimental investigations with various end thicknesses and flange dimensions are recorded, and from these the equivalent factors are derived.

Sets of curves calculated for cylinders with freely-supported ends and covering a range of cylinder thicknesses are given. From these it is possible to obtain close approximation to the frequencies of cylinders under other end conditions by the use of an appropriate factor. An example is given of frequency calculations for a large air-receiver for which two frequencies were identified by experiment.

Two 50-megawatt turbo-alternators unitized with reheater boilers were commissioned at Dunston “B” generating station towards the end of the year 1950, and ranked as the most efficient generating plant owned by the British Electricity Authority during the year 1951.

In Part I of the paper are shown the changes in design compared with the original plant which was commissioned seventeen years previously.

In Part II plant staffing and staff training are discussed and details of operating experience are given, including methods adopted for overcoming the initial plant troubles and the procedure eventually evolved for starting-up and shutting-down, and, in particular, the method adopted to permit full load to be achieved in hours from starting-up.

Performance and “outage” figures for 1951 are discussed and analysed and results of tests are tabulated, including load throw-off tests and monthly performance tests for the two Units 5 and 6.

It is concluded that unitized generating plant can be controlled so closely as to achieve test conditions and performance in commercial operation, but that attention needs to be drawn to detailed improvements in design and workmanship. In particular, the need is stressed for more care in making joints and tightening bolts for high-pressure work.

Future economies and improvements proposed as a result of experience are mentioned.

Materials handling is one of the fundamental activities of mankind, and the common denominator of all industry, yet it has been relatively neglected by engineers and managers generally. In the United States realization is growing that materials handling accounts for the largest single activity cost in much of manufacturing and distribution. Although many individuals are fully conscious of this, companies differ widely in this recognition and in the way they are trying to improve their materials handling activities.

The aims of this lecture are to clarify the major objectives of a materials handling programme; to show how progressive companies are organizing for materials handling; to describe the scope of the good materials handling programme; the type of individuals to be placed in charge; the authority and responsibilities to be delegated to them, and the ways in which these activities co-ordinate with, and contribute to, other aspects of the plant.

Techniques of analysing materials handling problems and the characteristics of basic materials handling devices are presented briefly.

Some fifteen significant trends in American materials handling practice will be discussed and briefly illustrated by case examples.

Nantgarw colliery is the subject of a major reconstruction scheme, the details of which are of interest to mechanical engineers. Many of the major features including the shafts, winding engines, air compressors, mine-car handling equipment, and stowage material preparation plant are outlined in this paper. Because of the necessity to wind from more than one level in the pit, the shafts have equipment which is a departure from the conventional types in use in Britain.

The method of mining to be used is known as the “Horizon” system, well favoured on the Continent for steeply inclined seams. The chief feature of the system is that horizontal roadways are driven to the seams at convenient vertical distances apart.

The use of compressed-air power is mentioned, together with reasons for the choice of the air- compressors installed. The system of transporting coal and materials within the colliery is dealt with at some length, because of its originality and its improvement over the systems in being at the older collieries.

A fairly full description has been given of the shale preparation plant for stowage material for underground use. This is the first plant of such a size to be designed in Britain, and it has been developed from experience gained in smaller units built at nearby collieries.

This paper begins with a general summary of the problem of interpreting interchangeability and a limit of size; the general nature of a fit and of the limits of size which determine it are briefly examined. Some statistical problems of assembly and inspection and their effects on interchangeability and on the practical interpretation of a limit of size are investigated. The standard of measurement is considered in some detail as defining gauge tolerances and errors of measurement, and as giving guidance in relation to certain varying degrees of interchangeability that have been shown to exist. The effect of these arguments on the standard of limits and fits is discussed and certain practical pitfalls in the statistical arguments are briefly examined. Low degrees of interchangeability are discussed and also interchangeability in relation to complex forms, position, and angular dimension. It is shown how the various degrees of interchangeability can be indicated by means of suitable provisions in the standard of drawing practice. Finally, the general attitude of mind to design, which prompts solutions of the type suggested, is set out with special emphasis on the matters discussed.

Though the application of steel castings in the field of the prime mover is selected as a basis for treatment, the opportunity is first taken of mentioning developments inside the foundry that appertain to manufacture and inspection. The trend towards more stringent service conditions has augmented the responsibility of the foundry for the production of castings free from defects. There are indications that the designer and user of castings can with advantage pursue and develop the tendency to break down designs into relatively simple constructions that can be fabricated together at a later stage, and examples of this type of construction are given. Where this trend is impracticable, a strong case can be made for the concentration of allied types of work in relatively few foundries.

Mention is made of the trend of research as applied to steel castings, and the lasting importance of methods departments is stressed.

There is at present no written matter in the English language on watch production, the industry having died out in Britain some thirty years ago. The aim of this paper is to provide the engineer who is not a specialist in the subject with a background knowledge of the methods used in the production of watch components and the special difficulties encountered in watch manufacture.

In the first part of the paper the historical development of the watch and the general principles of design are briefly reviewed. A fifteen-jewelled man's wrist-watch with lever escapement is taken as a typical example, in which the features commonly encountered are examined in more detail and the small size of the pinions and pivots is emphasized.

The latter part of the paper is devoted to the techniques used in the production of the components (the plates and bridges, turned parts, wheels and pinions, escapement, and balance); their final assembly; and the oiling, springing, and timing, of the finished watch. A general note on the three main categories of watches is amplified by figures illustrating the performance actually attained by a wrist-watch of the highest grade.

This scheme, now in its third year of operation, is designed to bring to Great Britain every year thirty-eight Canadian graduates in engineering for post-graduate studies.

The Fellowships have a duration of two years and are tenable in industry, in universities, or partly in each of these.

The author has been connected with the scheme from its inception; he assists in the selection of the graduates in Canada and arranges for their location and training in the United Kingdom. The paper describes in some detail the origin, purpose, and operation of the scheme and raises many issues of interest and importance.

The author also gives summaries of the reactions of Athlone Fellows who are now receiving postgraduate training in Great Britain. The purpose of the paper is twofold: to give information concerning a new and important system of Fellowships, and to evoke constructive criticism from professional engineers of its objects and operation.

The design of a machine for solving heat-conduction problems by their electrical analogy with the flow of current in series-resistance, shunt-capacity, networks, is described in this paper. Consideration is given to the representation of cooling by radiation and convection, and the detailed design of the circuits is discussed.

The loss of power due to the deposition of fuel-oil ash on the turbine blades at present limits the use of boiler fuels in open-cycle gas turbines, and therefore prevents the more widespread application of this form of prime mover in the marine and industrial fields.

The nature and occurrence of the ash-forming constituents are discussed, followed by consideration of the possibilities of removal of these from the oil. There appears to be no solution along these lines nor by removal of the ash from the gas stream. The basic factors controlling deposition of ash are still not fully understood and therefore further experimental work is required. However, a method which gives a considerable reduction in deposition has been discovered. In this, combustion of the fuel droplets is controlled so that each droplet burns down and leaves the combustion chamber as a hard, dry, carbon particle containing an appreciable proportion of the ash. The combustion loss due to this unburnt carbon is less than 1 per cent. Long-term engine tests are now required to assess the practical use of the method.

Another means of reducing deposition which appears to offer considerable promise is the use of various additives to the fuel or gas stream. Of those tested the oxides of silicon, zinc, and magnesium were the most effective.

The paper, of which the following is a summary, is a Supply Section paper of the Institution of Electrical Engineers, and was presented for discussion at a Joint Meeting of that Institution and the Institution of Mechanical Engineers on 5th March 1953 at the Institution of Electrical Engineers, London. The paper in full, together with a full report of the discussion, will be published by the Institution of Electrical Engineers.

Requirements for stability are formulated mathematically and, through the “transformatory operations of mathematics”, yield a series of “stability equations” of ascending order which are generally applicable, for example to control mechanisms, electronics^†, nuclear physics, etc. From these stability equations, the equation of the stable characteristic curve of a governor, and the differential equations of the oscillations of a governor-engine system, are derived. It emerges that the first part of the new oscillatory equation is identical with the whole of the differential equation in the literature to date (unchanged since Maxwell 1868)^‡, while the important second part, which consists of terms of the same order of magnitude as the first part and which is the only one containing the equation of the stable characteristic curve, is lacking in literature.

The stability equations classify all possible constructions of variable-speed governor according to “order of stability”, which signifies important operating properties. This classification accounts for the known shortcomings of conventional types.

The stability equations, combined with the mathematical formulation of practical requirements (speed-adjustment with only one actuating motion, etc.), lead to new basic types of variable-speed governor, with complete systems of design equations. In addition to determining all unknown dimensions, this set of equations is important because it derives constructions of which the complexity increases with order of stability and, furthermore, a simple construction which provides any required high order of stability with the minimum number of adjustable components.

To facilitate future development of the cooled gas-turbine more test information is needed on the effectiveness of cooling in an actual turbine operating at high gas-temperatures. Part I of this paper deals with some design aspects of a single-stage experimental turbine built to enable an experimental investigation to be carried out on the air cooling of nozzles and blades.

The turbine, built for operation at high gas-temperatures, was fitted with internally air-cooled blades having a large number of small cooling passages running the whole length of the blades.

A description is given of the pressed powder technique used to introduce the small passages in blocks of heat resisting material from which the blades could be machined.

Mention is made of some of the difficulties encountered in this method of manufacture and also of the need for careful consideration of suitable methods of disposal of the cooling air when internally cooled nozzles and blades of this form are used.

A comprehensive series of tests have been made on an experimental single-stage turbine to determine the cooling characteristics and the overall stage performance of a set of air-cooled turbine blades. These blades, which are described fully in Part I of this paper had, internally, a multiplicity of passages of small diameter along which cool air was passed through the whole length of the blade. Analysis of the, test data indicated that, when a quantity of cooling air amounting to 2 per cent, by weight, of the total gas-flow through the turbine is fed to the row of rotor blades, an increase in gas temperature of about 270 deg. C. (518 deg. F.) should be permissible above the maximum allowable value for a row of uncooled blades made from the same material.

The degree of cooling achieved throughout each blade was far from uniform and large thermal stresses must result. It appears, however, that the consequences of this are not highly detrimental to the performance of the present type of blading, it being demonstrated that the main effect of the induced thermal stress is apparently to transfer the major tensile stresses to the cooler (and hence stronger) regions of the blade.

The results obtained from the present investigations do not represent a limit to the potentialities of internal air-cooling, but form merely a first exploratory step. At the same time the practical feasibility of air cooling is made apparent, and advances up to the present are undoubtedly encouraging.

In this note is suggested a modification to overcome the difficulty met in the application of Bland and Ford's simplified theory of strip rolling to passes where heavy back tensions are applied.

An equation for the roll force is evolved and compared with those of Orowan and of Bland and Ford. Experimental and calculated values of roll force given show an accuracy to within 10 per cent.

The object of the investigation was to provide easily handled data for the design of helical torsion springs from round patented plain carbon spring-steel wire. For this purpose, the validity of the curvature correction to the stress formula was investigated and the effects of variations in low-temperature heat treatment and wire size were explored.

It was found that the theoretically derived curvature correction for maximum stress (stress at intrados) was the dominating feature in fully heat-treated springs under all conditions, and for patented wire springs when the maximum stress was tensile. In patented wire springs, however, when the maximum stress was compressive it was found that, although with springs of small index the maximum stress was still the dominant feature, with those of large index yielding was due, not to the maximum compressive stress at the intrados, but to the smaller tensile stress at the extrados, and that a critical ratio existed where yielding occurred simultaneously at both extrados and intrados; a finding that has considerable practical value. It is suggested provisionally that the phenomenon may be due to the combined effect of the low-temperature heat treatment, which relieves textural stresses (present in the wire as the result of the drawing operation and the presence of cementite) and at the same time allows the strain-age-hardening effect to become evident.

Design charts are submitted incorporating the results of the above investigation.

In this paper is shown how in the case of steam transmitted entirely for use in process heaters, it is the total heat energy that finally counts; and hence a pipe-line can be designed for a high pressure drop and a high velocity of flow, since the friction work and the kinetic energy will be reconverted into heat which is also available at the outlet of the pipe-line. The size and cost of the piping and lagging can therefore be kept low, and radiation losses will also be reduced.

In a long transmission, however, the very rapid increase of velocity towards the end of the pipe-line may cause the steam velocity to reach the “acoustic” value; and when this occurs a “choking” effect results in the discharge being lower than the expected value.

The method of design proposed is to calculate the size of pipe required so that the given length of pipe-line is an assumed fraction, say, 0.80 to 0.90 of this “choking” length. This ensures that the required discharge will be obtained with the minimum size of pipe, and with no danger of “choking”.

Simple formulae are derived with the help of the approximate Callendar expression for total heat, which is applicable with reasonable accuracy at low pressures; and a case is worked out for a combined power and process steam installation, which shows that the problem may have several solutions of equal heat utilization efficiency.

In this paper are put forward the basic principles under which all grinding machines, irrespective of size or type, must be constructed. The chief problems facing the designer of precision grinding machines are discussed, together with the essential elements and physical components. The choice of solutions which seem to offer the most promising future developments is suggested. The factors affecting the design of grinding machines—cost, ease of maintenance, and the influence of the user—and the essential elements—rigidity, accuracy, and constancy, are discussed. Mechanical and hydraulic traverse mechanisms are compared and a brief description is given of a typical hydraulic system. Although mechanical methods of traverse have been giving way to hydraulic systems, the author considers that with the improved electrical control gear available there may be a reversion to mechanical traverse mechanisms on account of their simplicity and reliability.

The influences of the wheel-head and wheel-head spindles are discussed. The two lines of approach that wheel-head design and wheel-head speeds will follow during the next few years is suggested. Work-head and work-head drive arrangements are described; followed by the main features of universal machines and machines arranged for automatic operation.

Finally, methods and mechanisms are described for obtaining concave and convex camber with roll-grinding machines.

A review of the work carried out for the Steels for High Temperature committee of the British Electrical and Allied Industries Research Association between 1930 and 1952.^†

In this paper the achievements of the J/E committee between the years 1930 and 1952 are reported.

Development of steam power plant largely depends on the physical properties of the materials. Realization of the effects of creep in metals has caused creep data to supersede Hooke's law in the determination of design methods for steam power plant.

The study of the effects of creep properties in different steels led to the adoption of chromium- molybdenum, and chromium-molybdenum-silicon and, to a less extent, molybdenum-vanadium steel for the high-temperature components; superheater tubes, steam pipes, and superheater headers for plant using temperatures above 900 deg. F. From creep tests made, the stress-time relation for each component was obtained. Methods of creep testing are compared, and abnormally high creep rates are investigated.

The testing of the reliability of long-time creep tests deduced from short-time creep tests is described. Relaxation tests carried out enabled a mathematical relation between relaxation time and stress to be derived.

Investigations into the effects on the properties of the steel, of the presence of minor elements in the steel, and the method of manufacture, are described, and also the cause and effects of grain growth.

The causes of, and means of avoiding, cracking of steel are studied.

Future developments are outlined.

FOREWORD

C. H. Desch, D.Sc, Ph.D., F.R.S.

Chairman of the Steels for High Temperature Committee of B.E.A.I.R.A.

Since 1930 the Steels for High Temperature committee of the British Electrical and Allied Industries Research Association has fostered work on the properties of steels used in steam power plant. The committee was formed from the committee of the Department of Scientific and Industrial Research on the behaviour of materials at high temperatures, at a time when the importance of the phenomenon of creep in designing high-temperature steam power plant was being realized.

The committee has always included in its membership representatives of leading steel makers, tube manufacturers, turbine builders, and boiler makers. The Admiralty has also been well represented and, also since their formation, the British Electricity Authority and the British Iron and Steel Research Association. The views of metallurgists and engineers concerned with the production and use of high-temperature steels have thus been combined in furthering co-operative work which has effectively contributed to advances in the design and efficiency of British steam power plant. The work of this committee has thus formed an integral part of the development of high-temperature steels which has been made by British industry over the last twenty years.

The committee has throughout the work had the co-operation of the Engineering and Metallurgy Divisions of the National Physical Laboratory where most of the investigational work has been carried out. This work has been supported financially by the steel-making and the appropriate user industries.

Since the inception of the committee in 1930 a number of detailed technical reports have been issued on various problems which it has investigated. The following report has been prepared to provide an overall picture of the major achievements of the committee.

Although he has not been concerned with experimental work, Mr. Sage has had access to all reports and papers connected with the research. The value of the contributions of numerous investigators at the National Physical Laboratory is acknowledged, and Dr. Jenkins and Mr. Tapsell should be specially mentioned as having taken a leading part in the investigation since its inception.